Wikipedia talk:Wiki Guides/Archive 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 1 Archive 2

Resources for Guides

What can we use? What could you use both to help you with new users and to make the process easier?

I've been thinking of a couple things to start with:

  • A tool to help see the contributions of the users you're following at a glance (probably on toolserver)
  • I would really like to work on the welcome emails we want to send to the new users at the start. If possible I'd like to get a bunch of drafts that get emailed to us to save but before that we should probably talk about what we want to say. Do we want it to be personalized? If all the users have done an edit or 2 at least do we want to add a bit to a template about those edits?

Jalexander--WMF 05:10, 3 February 2011 (UTC)

We send out welcoming emails on this project? I didn't know that. What I do know is that I was freaked out to get a pile of emails from other wikis when I "created" accounts on them by following the interwiki links on a specific article shortly after I created my SUL account, but I doubt there's anything to be done about that. :-) Risker (talk) 05:36, 4 February 2011 (UTC)
We don't :) The other wiki's don't either actually the email you're getting is that the default is to send you an email when they edit their talk page (which we don't have set up on EN) and so you get the email when you get a welcome message ;) Of course some of those DO have automatic welcome bots when you create an account which will trigger the email. The idea for this though would be the active outreach so that when a Guide reaches out to a new user they get a somewhat personalized email from the Guide (rather then the obviously bot/ random language email you get from those other wikis ;) ) Jalexander--WMF 07:52, 4 February 2011 (UTC)

Invitation to edit

People involved in this might like to see WP:Invitation to edit, a tiny trial on the effectiveness of an extra "you can edit this article" button. WhatamIdoing (talk) 23:23, 3 February 2011 (UTC)


Ok what else? Questions? Other discussion we should have? We can break them into more sections if needed.

What are the possible situations we could run into when we reach out and are helping them? Jalexander--WMF 05:10, 3 February 2011 (UTC)

IRC chat with Chzz

I got permission from Chzz to post this discussion I had with him about the whole project, helping users in general and ideas. I'm really interested in what people have to say! I'm putting the discussion under a hat to make it a bit easier to deal with.

IRC Discussion

Feb 03 13:46:05 <Jamesofur> hiya :) Im starting to poke out for my study. I thought I'd show it to you as getting ready
Feb 03 13:46:07 <Jamesofur>
Feb 03 13:46:38 <Jamesofur> (haven't advertised it many places yet but going to fix that today most likely. Pages were created late last night)
Feb 03 13:48:18 <Chzz> er, how does this synch up w/ 'ambassador' ?
Feb 03 13:49:14 <Chzz> how / who will choose them ? what happens when they fuck up?
Feb 03 13:49:54 <Jamesofur> at some level not at all. Those are very much focusing just on people from PPI specifically. This is supposed to be a much broader group (and probably overall more people assigned to each helper but with less time required for each). In general I want to allow almost anyone who wants to to help. We can't just hand choose, god knows I don't know every single person.
Feb 03 13:50:21 <Chzz> "at least 1 edit" might be a bit restrictive. Maybe 3+ edits. There's so much utter utter shite
Feb 03 13:50:24 <Jamesofur> it's quite possible some of them will fuck up at some point, it is inevitable if you open it up but I think the benefits out weight the risks
Feb 03 13:50:37 <Chzz> do we exclude (or just dump out) vandals? have a go at winning 'em over?
Feb 03 13:51:31 <Chzz> hmm - well, I'm thinking of the ambassador thing; I advised Sage a lot on that, and told him it'd cause big problems when people fucked up - and, it did; the selection process is *important* - and it's really gotta be nice, honest and open - no fucking cabal
Feb 03 13:51:42 <Chzz> if you allow anyone, you'll have a shitload of massive problems
Feb 03 13:52:07 <Chzz> you'll get loads of people with no fucking clue, who will mis-advise and drive people away; so you won't really get decent stats
Feb 03 13:52:29 <Jamesofur> hmm that's interesting. The 'not blocked' was a bit towards that (though clearly someone who's edited only once even if vandalism may not be blocked). The more of a selection process you do the more you force a cabal whether you like it or not. It breeds it
Feb 03 13:52:44 <Chzz> depends
Feb 03 13:52:55 <Chzz> you could set higher criteria - considerably higher, I'd say
Feb 03 13:53:48 <Jamesofur> if we want to do this on a large scale then we need to try and work with people and there are a lot of people who I honestly have no clue how good they are and telling them "sorry can't help" will cut out a huge amount of people and waste their skills (and waste our ability to help guide editors who do fuck up)
Feb 03 13:53:55 <Chzz> Rollbackers who've been around 6 months, or something. No significant blocks or DRAMA. Evidence of giving helpful and valid advice to new users. Have a quickie !vote on choosing 'em
Feb 03 13:54:39 <Jamesofur> well obvioulsy you can go a whole lot higher then 'nothing' which is basically what it is now :)
Feb 03 13:54:44 <Chzz> Not everyone *Can* help - it's a real skill; it needs a certain talent for teaching, plus considerable tech know-how, a lot of drive and determination - clue - to find answers, plus immense patience
Feb 03 13:55:28 <Jamesofur> Those aren't bad requirements though I really would prefer to avoid a !vote if we possibly can... I think those are a big part of the whole cabal breeding thing..
Feb 03 13:55:55 <Jamesofur> would you mind posting some of your ideas on the talk page? I'd really like to keep as much of it open as possible and try to guide it some but not control it totally
Feb 03 13:56:10 <Chzz> trouble is, it's pretty impossible to kinda list exact critera
Feb 03 13:56:16 <Jamesofur> yup
Feb 03 13:56:17 <Chzz> e.g. you can't say "never blocked"
Feb 03 13:56:21 <Jamesofur> it is
Feb 03 13:56:42 <Chzz> but, it depends *what* they were blocked for, how long ago, and a zillion other things you can't just list
Feb 03 13:56:49 <Jamesofur> and wouldn't want to. Certainly seen a lot of people who have 1. been blocked badly and 2. been blocked correctly but are still good now etc etc
Feb 03 13:56:54 <Chzz> hence, you end up either a) having a !vote, or b) a cabal to choose
Feb 03 13:57:29 <Chzz> otherwise *you* - or somebody - is making a judgement call. Which is cabal; non-wiki, and opens such a can of worms... can't do it
Feb 03 13:58:17 <Chzz> Sorry to raise all that, but I just *know* it'll cause problems, pretty well from day #1
Feb 03 13:58:40 <Jamesofur> I think they sadly end up being the same choice too frequently. I would like to lean towards less selective if necessary to avoid that if at all possible
Feb 03 13:58:51 <Chzz> it's all v nice to say "let anyone help" but... you'll get utter incompetence, arguments, and DRAMA
Feb 03 13:59:15 <Jamesofur> there will ALWAYS be problems, it's how our community works at some level. If I need to be the bad guy I'm semi ok with doing that
Feb 03 13:59:24 <Jamesofur> we get utter incompetence after giant RfA's :)
Feb 03 13:59:39 <Jamesofur> and of course I'm ok with you raising things, I want that why do you think I asked you
Feb 03 14:00:05 <Chzz> yes, - RfA - we *do* but, we actually have documented ways of dealing with it, too.
Feb 03 14:03:02 <Chzz> I can imagine from the beginning: there would be all these over-keen silly children, all posting "ooh, please pick me, let me be your guide" on over 9000 talk pages of new users - probably using some non-auth bot
Feb 03 14:03:15 <Jamesofur> this is a study
Feb 03 14:03:24 <Jamesofur> they don't get a choice in who their assigned
Feb 03 14:03:30 <Jamesofur> it's a random group
Feb 03 14:03:46 <Chzz> so you 'allocate' kinda stuff?
Feb 03 14:04:06 <Jamesofur> Do I think it may be nice to continue this type of thing if it works? Very much so but yeah for this part it will be an allocate type thing
Feb 03 14:04:15 <Jamesofur> and we will have group sizes based on how many guides we can get
Feb 03 14:04:19 <Chzz> (in some form yes)
Feb 03 14:04:32 <Jamesofur> one of the big things I want to see is what works/what doesn't
Feb 03 14:04:41 <Jamesofur> this is very much a 'get data and see what works' type thing
Feb 03 14:05:03 <Chzz> OK; communication issue. 'x edits' ? A *hell* of a lot of new users only come to Wikipedia *one time* - or, maybe, for a couple days. They're not gonna know shit; not how to use talk pages, or anything. So, think about process: what you'll do;
Feb 03 14:05:32 <Chzz> --if you post a 'do you want a guide?' on their talk, they may never answer. They may never log in again, by the time it has been arranged. This is actually extremely likely
Feb 03 14:05:37 <Jamesofur> which is why I'm going to want people to give us their welcome email drafts etc etc. Hash things out
Feb 03 14:05:42 <Jamesofur> We anticipate emailing them
Feb 03 14:05:52 <Jamesofur> which is why they are required to have an email on their account
Feb 03 14:06:01 <Jamesofur> an active specific outreach
Feb 03 14:06:01 <Chzz> what %age of new users put in an email addy?
Feb 03 14:07:00 <Jamesofur> I don't know off the top of my head actually... I know that the new users who put in an email addy are more likely to edit... We should probably look for that stat
Feb 03 14:08:22 <Chzz> hm
Feb 03 14:08:35 <Jamesofur> But the idea was to have welcome emails sent out to their users (probably along with a talk page thing but as you say just a talk page is going to be an ENORMOUS drop off)
Feb 03 14:08:45 <Jamesofur> we'll still have a large drop off obviously and we know that
Feb 03 14:08:54 <Jamesofur> but that's part of the stat
Feb 03 14:09:03 <Chzz> Lets say, I was allocated some 20 new users, who'd said they were interested. Hm. I reckon, *I* would be lucky to get anywhere (as in, any interest in ongoing editing) from more than 1 or 2
Feb 03 14:09:26 <Jamesofur> how many would we actually lose, is it worse then the random group I'm following who don't have users assigned? (but obviously if they reach out for help we still help)
Feb 03 14:09:42 <Chzz> well, you can see the kinda thing you'd get at either AFC or FEED - I've done hundreds of both of those, in the past few days
Feb 03 14:09:46 <Jamesofur> I was thinking about 50... which sounds like a lot but I think as you say we'll see a big drop
Feb 03 14:10:47 <Jamesofur> yup, and some won't response, some will respond but not be interested, some will be interested but fail, some will lose interest etc
Feb 03 14:10:58 <Jamesofur> *won't respond
Feb 03 14:11:07 <Chzz> so - apart from plain vandalism; 80%+ of new users want to write about their own company/band/book/themself, have 0 interest in anything else. Of those, BTW, less than 5% will actually pass N. Maybe 1-2% will actually end up as an article. But, they won't bother to continue anyway
Feb 03 14:12:10 <Chzz> of the other 20%, I guess that >10% of them will be incompetent in editing, and/or can't be bothered to learn
Feb 03 14:12:30 <Chzz> So, from 50, you'd be v lucky to get 5
Feb 03 14:12:38 <Chzz> more likely, 1 or 2, to be honest
Feb 03 14:12:49 <Chzz> not gonna give much useful stats
Feb 03 14:13:18 <Jamesofur> right, but if we show some of them other options can we get them doing those too? And is that totally true? That is certainly what we see the most in a lot of places, no doubt, but we have a lot of legitimate edits from new users and ip's as well. Anti-vandalism, backlogs,gnomiing etc are all places where we can try and point people
Feb 03 14:13:25 <Jamesofur> and yup. We're going to to lose a shit ton
Feb 03 14:13:36 <Chzz> If I, for example, put my 'welcome' on the page of 100 new users who've joined today, and made at least 1 constructive edit - then, I hear back from about 2 of them
Feb 03 14:13:42 <Chzz> I've done that, actually, loads of times
Feb 03 14:14:05 <Jamesofur> I think if we continue the study we may do a couple batches, depends on how things go. If you think we want to start with bigger groups (or do it in rounds of large groups) that's a distinct possibility
Feb 03 14:14:22 <Chzz> hm
Feb 03 14:14:32 <Jamesofur> aye, I think the emails will be a lot more responsive then talk page messages but that isn't saying much
Feb 03 14:14:36 <Chzz> gotta think about what the actual goal is, 'coz it seems a bit mixed;
Feb 03 14:14:44 <Jamesofur> it is mixed
Feb 03 14:15:06 <Chzz> a) to help new users develop - yep, that's a good goal, but to do that, I'd suggest a totally different approach to,
Feb 03 14:15:21 <Chzz> b) getting info to discover why people don't get involved in the project
Feb 03 14:15:39 <Chzz> I think, you have to choose. Trying to do both at same time will be v problematic
Feb 03 14:16:36 <Chzz> for b), actually a better approach might be via stat analysis of existing behaviour. Because any a-type experiment is not going to provide 'typical' stats - because, most new users do not email, do not get help, do not stick around - and a) will only be looking, really, at the tiny % that do
Feb 03 14:16:54 <Chzz> Do you see what I mean?
Feb 03 14:17:33 <Jamesofur> I disagree. I think we can work on both and at some level have too. Just throwing out survey's isn't going to help, though yes I will probably do that at the end . We're going to be doing stat analysis on both users we're not actively helping and users we are. The point of this is to see the difference though when we actively reach out and try to help them
Feb 03 14:17:41 <Chzz> *** If you wanted to find out why people didn't learn Japanese, you wouldn't start by creating new, free Japanese classes
Feb 03 14:17:51 <Jamesofur> of course some aren't going to get their email, some not get help, not stick around etc. But that is PART of the test
Feb 03 14:17:55 <Jamesofur> *study
Feb 03 14:18:11 <Chzz> 'some' being about 99% - thus, devaluing any meaning of stats
Feb 03 14:18:21 <Jamesofur> the idea is to try and guide people to actively help, some of that will fail
Feb 03 14:18:39 <Chzz> 99% will fail, honest.
Feb 03 14:18:55 <Jamesofur> Wouldn't shock me
Feb 03 14:19:01 <Jamesofur> wouldn't totally disappoint me either
Feb 03 14:19:16 <Jamesofur> but I think it will be less for the group we're trying to help then the group we're ignoring
Feb 03 14:19:22 <Jamesofur> and if it isn't? That's very improtant data
Feb 03 14:19:26 <Jamesofur> *important
Feb 03 14:19:33 <Chzz> OK - can I tell you about an idea I had, please? This will take... 15 mins.
Feb 03 14:20:17 <Chzz> are you in WMF now, BTW?
Feb 03 14:20:25 <Jamesofur> of course (I should point out that I have some ability to change this, I'm running it, but the basics are a top down "run this" which is why some of it will be harder then most. The idea of trying to do this study wasn't mine at the start though I think we could get a lot of info)
Feb 03 14:20:42 <Jamesofur> I'm not, I'm at home. I'll be back in SF end of the month and will hopefully move in the next couple but not yet
Feb 03 14:21:22 <Chzz> Rrrright. /me rolls up sleeves... look, I had this idea - and there's a few things to it; hence, I gotta just try and explain it, and then we can chat about it. Here goes;
Feb 03 14:22:44 * Jamesofur listens
Feb 03 14:22:52 <Chzz> I - Chzz - write several 'lessons' aimed at new users. And contact 10 new users. Ones that want a 'guide'. I have it all nicely organized, so that they can do the lesson whenever they want. #1 might be, to leave a message on a talk page. signing it. getting a reply. (maybe a bot will reply)
Feb 03 14:23:12 <Chzz> they get a snazzy template - like a barnstar - showing which lessons they've completed
Feb 03 14:23:45 <Chzz> lessons will progress to reasonably high levels of competence - template code. writing and adding a DYK. that sort of stuff.
Feb 03 14:24:16 <Chzz> The group of 10 will have a name. A team name, or some shit, IDK. Team Aardvark will have their team discussion board - their talk page, and they'll chat and also help each other
Feb 03 14:24:37 <Chzz> they'll actually go through things 'together', and hence they'll have an instant community of 'friends' on the wiki
Feb 03 14:25:06 <Chzz> they'll also get real live practice in on-wiki debate, civility, and so forth; I, as coordinator of the team, would ensure smooth operation
Feb 03 14:25:26 <Chzz> checking they did lessons right; helping further when needed; and all the house-keeping shit when they post to the wrong fucking place
Feb 03 14:25:37 <Chzz> BUT - this is the good bit;
Feb 03 14:25:50 <Chzz> ---when they finish, *they* are new guides.
Feb 03 14:26:01 <Chzz> And they fuck off and create their own teams of 10.
Feb 03 14:26:10 <Chzz> and take 10 new users through the lessons
Feb 03 14:26:19 <Chzz> OK - that's the basic idea.
Feb 03 14:26:48 <Chzz> one of the major aspects being, the built-in ongoing development of it
Feb 03 14:27:08 <Chzz> I teach ten; they all teach 10; *they* teach 10.
Feb 03 14:27:13 <Chzz> And then I take over the world
Feb 03 14:27:19 <Jamesofur> heh
Feb 03 14:27:24 <Jamesofur> I like it
Feb 03 14:27:36 <Chzz> it wouldn't be easy
Feb 03 14:27:49 <Chzz> it's a lot of work, to make good lessons which are 'resilliant'.
Feb 03 14:27:59 <Chzz> The first lot would be a pretty-much alpha test. But, still.
Feb 03 14:28:15 <Chzz> And the guide would need to be *very* active. To say the least.
Feb 03 14:28:35 <Chzz> ---at *least* definitely being online *every* day for a few hours.
Feb 03 14:29:03 <Jamesofur> This is one of the things I'd like to see at some level from this. People coming up with their own ideas, some of them working (and clearly being seen as working) and being copied/expanded on etc etc. It's why I'd like to do some basic expansion "Here are your random starting 50, send out an email, let us see your welcome email template etc" but give a freer hand when we can and hopefully get some discuss
Feb 03 14:29:03 <Jamesofur> ion about what is working/what isn't
Feb 03 14:29:13 <Chzz> I'm actually thinking of a small team of guides, in charge of each group - BUT a small, GOOD group. say, 3 of us. Me, you and ...IDK...someone fucking good, pref working at WMF
Feb 03 14:30:12 * Jamesofur nods
Feb 03 14:30:14 <Chzz> see - the template and maybe 'awarding stars' or whatever... is important, too. They'd see they were making progress; people like a target, people like to get pretty stars and shit. (put the 'tick' into a prot page)
Feb 03 14:30:41 <Jamesofur> they do, awards like that are underappreciated
Feb 03 14:30:42 <Chzz> "This user is in TEAM AARDVARK, and has completed these lessons: * * * * o o o o o o" type of shit
Feb 03 14:30:58 <Chzz> with nice links to their team talk page and the main coord page, and so on
Feb 03 14:31:51 <Chzz> For some lessons, I might well find a specific article - and say, "Fix the refs and formatting on this"
Feb 03 14:32:19 <Chzz> if they did all the lessons, they'd be pretty much ready to be an admin, to be honest. (possibly, needing time and edit count)
Feb 03 14:34:32 <Jamesofur> yeah but done a lot to prove themselves and show what they've done
Feb 03 14:35:26 <Chzz> yeah; and all nicely documented and ready
Feb 03 14:35:49 <Chzz> "I have CSD'd 100 articles - LINK, I've made 3 DYKs LINK, I've proof-edited 20 articles LINK"
Feb 03 14:36:23 <Chzz> participated in AfD, reported vandals, NPP'd, reported bad user-names... etc etc etc. The whole kaboodle
Feb 03 14:36:42 <Chzz> 'coz...all of these things, I've *already* taught. Most of 'em, over 9000 times
Feb 03 14:36:56 <Chzz> so, I know how to teach 'em
Feb 03 14:38:04 <Chzz> e.g. AFC -- I have shown about...IDK, 30? people...exactly how to review them. Took 'em through the whole process; made an example page; they accepted and declined it. Chose some real article; went through it with 'em. Etc.
Feb 03 14:38:34 <Chzz> Having done it 30+ times, I know how to teach it; I know how to tell 'em each step. And, what will fuck up along the way
Feb 03 14:40:19 <Jamesofur> yup
Feb 03 14:40:34 <Jamesofur> and some of that fucking up is good tbh it's just better if you have someone there when you do it
Feb 03 14:40:52 <Chzz> yeah; you can use it, to demonstrate how to fix things - absolutely; I do that all the time
Feb 03 14:41:18 <Chzz> When I do my 'standard intro to editing', usually at some point someth goes wrong, which is great, 'coz then I do all the stuff about history pages and undo
Feb 03 14:41:29 <Jamesofur> and they learn it better, when they've made the mistake rather then when you say "well do it this this and this way because I know you're goin to fuck up"
Feb 03 14:41:37 <Jamesofur> yup
Feb 03 14:41:41 <Chzz> [19:39] <Chzz> "how many new users put in an email addy"
Feb 03 14:41:42 <Chzz> [19:39] <Chzz> can that q be fairly easily answered, w/ some SQL thingy, or something
Feb 03 14:41:42 <Chzz> [19:41] <Delta> Chzz: its possible, not sure I have access to that info though
Feb 03 14:41:42 <Chzz> [19:41] * Delta looks
Feb 03 14:42:50 <Chzz> y'see... whatever process you develop, the hard part really is 'teaching the teachers'
Feb 03 14:43:02 <Chzz> which is why I've come up w/ this ongoing 'pass it on' type idea
Feb 03 14:44:00 <Jamesofur> yup, but I also think there are alot of teachers who could do it 'well' (and obviously better but I'll start with well) with very little help
Feb 03 14:44:05 <Chzz> it's something I've kinda had in the back of my mind, for a long time. I could probably just go ahead and do it. I know though, it'd be a lot of work. And being honest - I'm not sure if I can be bothered. The pay isn't great :p
Feb 03 14:44:16 <Jamesofur> which is why I don't actually want to start the new user part of this study until the end of the month at least
Feb 03 14:44:38 <Jamesofur> hey, the pay isn't great even when you're GETTING paid ;)
Feb 03 14:44:40 <Chzz> [19:43] * Delta goes to bug river for access to user_email_authenticated
Feb 03 14:44:49 <Chzz> heh; no, but it's better than 0
Feb 03 14:44:57 <Jamesofur> this is true
Feb 03 14:45:47 <Jamesofur> and it's the idea of the fellowships and a lot of community hires. Pay people to help do the things that they do as volonteers but especially for the things that they wouldn't have the time to do as volonteers if they had to do real work at the same time too etc etc etc
Feb 03 14:46:05 <Chzz> I know you won't misunderstand, if I say that... well, if the world made sense - WMF might consider paying me some nominal fee to make the thing work. Enough for beer, sausages and interweb. $2000 / month, or some shit
Feb 03 14:46:06 <Jamesofur> *would do as volonteers
Feb 03 14:46:15 <Jamesofur> heh
Feb 03 14:46:29 <Jamesofur> no I would totally understand :)
Feb 03 14:46:42 <Jamesofur> you know there are grants that are going ot be opening to everyone ;)
Feb 03 14:47:06 <Chzz> I reckon that, to move forward, WMF needs more professional 'help'. And that is gonna mean paying people - just to get the necc. level of commitment and professional ...well, 'onus'
Feb 03 14:49:23 <Jamesofur> I'm hoping if this project works even a bit (and remember works can happen in a lot of ways. Data is works) we'll be able to expand different things. I think at some point that will happen and Zack knows that
Feb 03 14:49:36 <Jamesofur> obviously we can't just 'do it' straight out but that's a very distinct possibility
Feb 03 14:49:51 <Chzz> oh, yeah; absolutely. I was just sharing ideas
Feb 03 14:50:01 <Jamesofur> totally
Feb 03 14:50:18 <Chzz> I'm not quite sure about what 'this project' is, to be honest; and I think you do need to think about that, and come up w/ specific aims
Feb 03 14:50:36 <Jamesofur> The aims are multiple. It's going to be very hard to get away from that
Feb 03 14:50:46 <Jamesofur> that is ingrained in the very idea
Feb 03 14:50:52 <Chzz> a or b though, really, to be honest
Feb 03 14:51:02 <Chzz> they're mutually wossname. not compatible
Feb 03 14:51:07 <Chzz> or, at least, they're different projects
Feb 03 14:51:37 <Jamesofur> at some level their different projects that are running concurrently
Feb 03 14:51:58 <Chzz> I think, probably, what you want to do - for now - is more b) - "Find out why we do not retain many new editors"
Feb 03 14:52:08 <Jamesofur> my 'stats' will be public obviously but the guides themselves don't really have to worry about them. It's out of their hand
Feb 03 14:52:12 <Chzz> mmm
Feb 03 14:52:19 <Jamesofur> *hands
Feb 03 14:52:41 <Chzz> but if what I just wrote is the goal, then I'd suggest very differnt things to ...well, the mentorship thing. It's just not the best way of getting that data.
Feb 03 14:52:57 <Jamesofur> b) is an enormous part of what we want, maybe the bigger one. But a lot of that is going to come from metrics and the survey etc
Feb 03 14:53:06 <Chzz> scratch the 'guide' idea, for a sec; just listen up;
Feb 03 14:53:12 <Jamesofur> I'm listening
Feb 03 14:53:40 <Chzz> -email 'em, yeah; make it clear you are doing an independent survey, to help improve Wikipedia, and ask for their participation
Feb 03 14:53:54 <Chzz> From respondents to that, find 'key members'. By which, I mean this;
Feb 03 14:55:20 <Chzz> ---90% of them, just ask 'em to fill in a few questionnaires, at about 3 - 5 stages in their Wikipedia 'career' - asking how they're getting on, what problems they faced. Lots of "on a scale of 1-10, how hard was it to add references" and stuff
Feb 03 14:55:53 <Chzz> ---for the chosen, special 10% - those who seem really keen to get into the whole 'independent survey' stuff - build a personal relationship, and *watch* their progress
Feb 03 14:56:08 <Chzz> watch, but do not directly intervene
Feb 03 14:56:14 <Chzz> don't actually help 'em
Feb 03 14:56:48 <Chzz> keep a step back from it all. Ask 'em not to mention anything about the thing.
Feb 03 14:57:01 <Chzz> look at their every edit; make extensive notes
Feb 03 14:58:12 <Chzz> then, at the end of - what? a month, or something? collate it, make a report w/ lots of summary shit and bullet-points, and a pretty graph or two showing "drop out rate" and average scores showing that "the wizard is absolute crap" and "CSD is horrible" - but, with all the details available; the diffs, of the 'special' ones
Feb 03 14:58:13 <Chzz> ???
Feb 03 14:58:15 <Chzz> profit
Feb 03 14:58:50 * Jamesofur nods
Feb 03 14:59:00 <Jamesofur> I understand what you're saying
Feb 03 14:59:14 <Jamesofur> Unfortunatly I can't really "just scrap this whole guide thing"
Feb 03 14:59:18 <Jamesofur> I can't do it
Feb 03 14:59:24 <Chzz> well, y'know, every line is prefixed w/ IMHO YMMV
Feb 03 14:59:30 <Jamesofur> of course :)
Feb 03 14:59:33 <Chzz> no; don't scrap it
Feb 03 14:59:36 <Chzz> split the idea in two
Feb 03 14:59:58 <Jamesofur> and I really really appreciate it. Well part of it is included ;) Though in a slightly diferent format at least in my head so far
Feb 03 15:00:00 <Chzz> 1. a survey, like I just said. 2. An alpha version of, pretty much, what I suggestied - about making a team, led by a guide
Feb 03 15:00:37 <Chzz> it's really 2 projects. If you try and force it into one, you're gonna have problems
Feb 03 15:01:10 <Jamesofur> aye
Feb 03 15:01:24 <Chzz> projects need goals. If goal a is to help noobs, that's great; but if goal b is to assess how hard it is for noobs, then that goal is tangential
Feb 03 15:01:44 <Chzz> related, sure - but, tangential
Feb 03 15:01:50 <Jamesofur> so here out some of what our ideas have been. Much of it is clearly flexible though not all
Feb 03 15:02:03 <Chzz> mmm, sure
Feb 03 15:02:36 <Jamesofur> (and I'm going to ask you to put as much of this as you can on the talk page. If you want I'm happy with just our discussion going on there but I'd really love to try and get it open and see if we can get others ont he discussion too especially once I get time to start reaching out.. which will haev to be soon so I don't get killed)
Feb 03 15:02:48 <Chzz> I'm v interested in all such ideas. Anything to improve the 'new user experience' and the project as a whole
Feb 03 15:03:19 <Chzz> yep, that's fine - you can copy-pasta all of this, I give full permission
Feb 03 15:03:33 <Chzz> public domain
Feb 03 15:03:53 <Jamesofur> <3 thank you :)
Feb 03 15:09:41 <Jamesofur> So in the end there is a bit of a) help new users and b) figure out how hard it is for new users though b may be the lower end. The original goal was to have 1 group who we followed from the start (the control) but didn't actively outreach too. They would end up getting a survey at the end asking how their experiences went/demographics etc (basically everything being optional). Then there would be 1 or mu
Feb 03 15:09:41 <Jamesofur> ltiple other groups with randomly assigned users to guides. The guides would send out welcome emails to their users and try to pick up on those first edits the new user made
Feb 03 15:10:28 <Jamesofur> many would get lost obviously but have tools to show them all their users contributions at a glance etc to help them follow their 'group' and reach out with resources and advice as it went on
Feb 03 15:10:57 <Jamesofur> following both groups for metrics such as edits, articles, blocks, talk page discussion, user page edits (?), emails etc etc
Feb 03 15:11:10 <Jamesofur> the same metrics we would actually follow the control group for
Feb 03 15:11:13 <Chzz> if you make 1 into "my b)" (in this discussion), then whatever happens wih the other groups can't do any harm. And yes, you can assess how well it went, sure.
Feb 03 15:12:05 <Jamesofur> in the end EVERYONE gets surveys. I'm almost as interested in what they guides said as what the new editors. One of the things I'd really like over the next month before we start as we recruit guides is them to share their own experiences
Feb 03 15:12:41 <Chzz> I *suspect* that the 'trial' of having allocated guides will be far more trouble than it is worth, and merely generate info on how NOT to do it, the need for a more professional approach, and the need to re-think how to do it. But, yeah; that's fine.
Feb 03 15:13:07 <Jamesofur> possibly. I actually suspect we will find some people do damn well and some people fail miserably
Feb 03 15:13:08 <Chzz> (basically, I think the whole 'guide' thing - if done in this way - will be more trouble than it is worth; but that's fine too; you might need to do it, to prove that)
Feb 03 15:13:16 <Jamesofur> aye
Feb 03 15:13:21 <Jamesofur> I think there is a bit of that
Feb 03 15:13:45 <Jamesofur> Sue really really likes the idea of active outreach. We need to find good ways to do that, which means at some level we need to find bad ways to do it too
Feb 03 15:13:48 <Chzz> If you allocate guides to 1000 users, then you'll be v lucky to get 5 decent editors from them. That's my prediction
Feb 03 15:14:14 <Jamesofur> My original goal was control of 1000 and then guides for 1000 but it depends ont he amount of guides we have
Feb 03 15:14:46 <Chzz> you'll have a lot of work to do in keeping check on the guides
Feb 03 15:15:09 <Chzz> 'coz that's what - maybe 50 guides?
Feb 03 15:15:14 <Jamesofur> doing a survey for some of the control from the start and watching them when they know they're being watched is interesting
Feb 03 15:15:26 <Jamesofur> oh no doubt, it's going to be a nightmare
Feb 03 15:15:39 <Jamesofur> luckily I don't have a whole lot of fundraising duties for at least a couple months
Feb 03 15:15:47 <Jamesofur> and I'm already used to 18-20 hour days
Feb 03 15:15:48 <Jamesofur> :)
Feb 03 15:16:06 <Chzz> it absolutely will drive you insane, trying to manage it. You know what its like, trying to manage Wikipedians is like trying to herd cats
Feb 03 15:16:56 <Chzz> they're BOLD, they'll fuck off and do shit. Lots of it. And they won't understand the underlying objectives you have. And then, there's the DRAMA...oh yes.
Feb 03 15:17:08 <Jamesofur> I also luckily have a couple other staff who will likely be interested in helping out. Christine especially who just loves this shit (and was purposely kept off it while she gets to look into OTRS alternatives and how to blow that up nicely... which I was purpsoely kept off of. Philippe wants us to help eachother but not have as many preconcieved notions)
Feb 03 15:17:23 <Jamesofur> herding cats with cattle prods attached to their butts
Feb 03 15:17:32 <Jamesofur> (and yes, that's going into the talk page post)
Feb 03 15:17:36 <Chzz> also, of course, lots of the 50 will unfortunately get ill / get a life / not actually turn up and so on.
Feb 03 15:17:45 <Jamesofur> yup
Feb 03 15:18:00 <Chzz> so you'll have 'guide drop-out' too. And need to plan for that
Feb 03 15:18:02 <Jamesofur> but again, that's part of doing things with volonteers. In the end if they decide they don't want to volonteer...
Feb 03 15:18:07 <Chzz> yep
Feb 03 15:18:18 <Chzz> or, they decide they don't agree w/ your remit, and do their own thing
Feb 03 15:18:54 <Chzz> simply *stopping* would be no big problem. But, they won't. They'll head off in their own (and quite possibly 'good faith') directions
Feb 03 15:18:55 <Jamesofur> I'll totally admit that there is a side effect thing of this where it would be nice for us as staff and other community members to start to figure out more people who "have a clue" and can help out when things come up
Feb 03 15:19:02 <Jamesofur> oh of course
Feb 03 15:19:12 <Jamesofur> but that's data too. Recording what they do, how it works
Feb 03 15:19:18 <Jamesofur> hey if they go on their own path and it WORKS?
Feb 03 15:19:22 <Jamesofur> fuck I want to know
Feb 03 15:50:46 <Chzz> one thing, to bear in mind;
Feb 03 15:50:46 <Chzz>
Feb 03 15:51:32 <Chzz> (worth you spending 5 mins looking over all that tl;dr shit)
Feb 03 15:51:39 <Jamesofur> oh that's nice
Feb 03 15:51:40 <Jamesofur> yeah
Feb 03 15:52:17 <Chzz> I showed the IRC discussion, 'coz I hoped to 'teach the teachers' - but, that's kinda not happened
Feb 03 15:52:21 <Chzz> need better teachers

Discussion about conversation

Creating a separate section just so that you can edit here without dealing with the discussion code. If we want to separate that into another subpage we can.

Suggestion about a survey

I'm still pretty new! Here's a post I made to the talk page of WP:BITE a while ago, which seems very relevant here. Would like to hear your thoughts. --Physics is all gnomes (talk) 14:37, 4 February 2011 (UTC) (formerly known as Hermajesty21)

Hi, I'm a 3 weeks old newbie here (edited a few times before I got an account, too). I hope I'm posting this in the right place! I was wondering if there are any processes for getting feedback from new users? By definition, the people who end up discussing these ideas are those who've survived the scary first few weeks. If it doesn't exist, it would be great to give a questionnaire or survey to a random selection of new accounts, to see what new users find helpful and what they find the most confusing or off-putting.
Personally, I've had two off-putting experiences in my first three weeks, which I think are in fact minor and normal occurences, but seemed off-putting to a newbie because I didn't understand the culture.
  • A user suggested my first article be merged into another, half an hour after it was created. To me it seemed (wrongly I now realise) like a direct criticism that my article wasn't good enough to be an article. At the time it had very little content because I was still working on it. I think regular editors forget just how slow editing can be for newbies - you write a couple of sentences, spend ages working out how to do a piped link, then try something out in the sandbox, then try to add a citation, then go make a snack, then preview it and find it looks weird, so then have to add an extra line between the paragraphs..... So I'm worried that speedy tagging, and especially speedy deletion, can underestimate how long an editor might take to make their article the way they want it.
  • I tried to improve the lead of a well established article (electron), spent a few hours trying to write a better lead, and then had my edit immediately reverted. I was really annoyed and felt like all my work had been for nothing. Luckily, by this point I knew how to contact the talk page of the editor who reverted it, and ended up having a discussion. But I wonder if this sort of thing makes a lot of new editors just give up and leave?
I don't know if my experiences are typical or not. But I do think that there are a lot of ways that wikipedia can be off-putting to newbies. I think these are things that an experienced editor might never even consider as off-putting. And I think the best way to be aware of these issues, so newcomer-friendly policy can be shaped, is to ask the newbies themselves about their experiences, maybe in the form of a survey. But I don't know how to set that up yet! What do you think?--Hermajesty21 (talk) 16:04, 21 December 2010 (UTC)

It is intimidating, but I trudge on with the memory of the Wiki philosophy and raison detre in mind. The template box is really useful. I think perhaps, a short video explaining the process in simple steps would help. if it doesn't yet exist (Jamaicancontributor (talk) 13:37, 10 February 2011 (UTC))

Preliminary survey/welcome

I suggested in Village Pump a week or so back that "we should really test a welcome bot. Welcome a thousand editors with the bot, a thousand by human and a thousand not at all. See who is still here in a year." Welcome bots have been frowned on as "cold" but without any evidence. I would be willing to do a pilot for this type of study. Rich Farmbrough, 10:36, 8 February 2011 (UTC).

I would like to see that. It certainly wouldn't take much doing and if it turns out that there's no effect I can stop pasting the welcome messages! The difficulty I see is keeping the "not at all" group not welcomed. --Danger (talk) 02:16, 9 February 2011 (UTC)
I think at some level you don't need to. We don't want to protect the page or something :) But you can leave that group "not specifically outreached too" and let the normal stuff happen. It's a similar idea to what I want to do for the control in this study for example. We're not going to email them and reach out but if they come looking for help on a noticeboard, or on IRC or something like that we don't want to deny it, that would skew the data just as much. Jalexander--WMF 02:32, 9 February 2011 (UTC)
Yes please. Rd232 talk 04:06, 11 February 2011 (UTC)
While I wouldn't object to trying this, a couple of years ago I heard from one of the folks at WikiHow that although they greet every new user, even so about one third never make more than a single edit. Someone may want to contact them & see if I remember the figure correctly -- or if they improved on their greeting & got better results. (Pete Forsyth might know the lady I learned this fact from; I don't remember her name.) -- llywrch (talk) 05:44, 9 February 2011 (UTC)
It would also be interesting to know what percentage of their readership has edited. Perhaps their relatively friendly atmosphere and non-academic nature (they don't have C-class articles, they have "rising stars") would make more people likely to try editing as they pass through, decreasing their retention rate. Interesting also to compare retention of more active editors; if a person makes 20 edits (indicating some level of interest) to wikiHow are they more likely to continue editing than if they do the same at Wikipedia. -Danger (talk) 06:28, 9 February 2011 (UTC)
Yeah I think that's one of the questions we have to ask too. Some smaller wikis automatically welcome on Wikimedia projects too (and some larger ones). An enormous enormous amount of new accounts don't make a single edit. I'm looking for broader numbers... but as an example:

December 2010

New Accounts: about 157650

New Contributers (more then 10 edits): 6675

Percentage: around 4.23%

Clearly we want a bit better numbers.. I ended up counting the new December accounts by hand in the logs. Jalexander--WMF 06:37, 9 February 2011 (UTC)

Hate to say it, but one way to improve the ratio is to reduce the number of new accounts created. Making it somewhat more difficult to create an account - by requiring a reponse from an e-mail, for instance, which very many websites do -- would cut down on extraneous accounts that will never be used, or are being stored for sockpuppetry use. I know that goes against the "anyone can edit" ethos, but I'm of the opinion that it was useful in its day, but is now counter-productive. I'm swimming upstream on that, though. Beyond My Ken (talk) 01:06, 10 February 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I'm inclined to agree. As long as anonymous editing remains possible, demanding email address verification is really a very small extra hurdle for creating articles (etc). And, not insignificantly, it would make some people take editing a bit more seriously. Rd232 talk 04:06, 11 February 2011 (UTC)

Where can we find these statistics?

The scientist side of me is coming out here I'm afraid:

The graph posted on the project page shows that the number of new editors with 10+ editors is declining. But is this because people leave before their 10th edit, or because fewer people are joining in the first place? The statistics we really need, to determine the wastage of new editors and whether it is getting worse, are things like

  • Number of people joining in each month
  • Proportion who then make more than 10 edits, and is that changing with time?
  • Proportion who go onto make more than 50 edits, and is that changing with time?

Anyone know where we could find these statistics? I got quite lost in the site.--Physics is all gnomes (talk) 17:27, 4 February 2011 (UTC)

I think most of that can be found at Wikipedia Statistics - Tables - English. It has a bunch of tables of stats. - Hydroxonium (H3O+) 18:37, 4 February 2011 (UTC)
Loads of interesting stuff there, but not the specific info I'm looking for. It did tell me that 80% of wikipedians have made less than 10 edits - so the wikiguide project may have to sample rather a lot of people.--Physics is all gnomes (talk) 19:10, 4 February 2011 (UTC)
You might be able to get those stats from a database query at WT:DBR. You may have to be specific about the requirements. For example, a user registering in Jan 2005 making 10 edits between Jan 2005 and Jan 2011 versus a new user registering in Jan 2005 and making 10 edits in Jan 2005 (same month). - Hydroxonium (H3O+) 19:56, 4 February 2011 (UTC)
That looks promising! I'll go have a think about exactly what to ask for, and then come back here and discuss.--Physics is all gnomes (talk) 21:01, 4 February 2011 (UTC)
Yeah there is a lot of data on that site and it can be hard to sort through. Over this weekend I'll sorth through some of it and get some better graphs for us too. Some of that is already there (for example the amount of people joining. We had 6675 new accounts in December (34361 new and old who made over 5 edits, 3500 new and old who made over 100 edits). For some of the other data we should have it but my have to sort little bit or do our own query's. I have toolsever access but if we need to do queries on the foundations database I'm sure I can wrangle someone to help us out there :) Jalexander--WMF 22:09, 4 February 2011 (UTC)
Hi, this is the same question that I was interested in, and when I went to a meeting in Utrecht organized by the Dutch Wikimedia regarding the Wiki Loves Monuments awards ceremony last November for the Dutch Wiki Commons users, this came up during a presentation by Erik Zachte. He makes the report cards for the NL wiki. He posted this "supposed decline in new users" thing on his blog, but now I can't find it. The gist of the story, as Physics-is-all-gnomes guessed, is that the picture per wiki is not as bad as the graph seems to convey at first glance. You can see Erik's work here and I suppose you can just ask him. Jane (talk) 13:07, 10 February 2011 (UTC)
Yeah, and it's definitely a very "per wiki" discussion. The numbers I've been using are all EnWiki only which clearly has a bit of a different dynamic though I would really like to see good ways to get new users involved in the other languages and projects as well. In some ways a project like this may even be more successful (and easy to do) on one of those projects. Jalexander--WMF 04:31, 11 February 2011 (UTC)
Actually, as I recall the numbers, I think there are more similarities than descrepancies. There is just a time-lag in new language WP's, as their article counts ramp up. One of the problems with measuring edits-per-user, is that you are not taking into account the increased efficiencies realized through bots of reverting vandalism. I am all for helping new users, and I believe that the lower the threshold, the more users will come in, so to speak. I hate to see something like "only registered users can edit". That said, I also believe that the level of "RTFM" among newbies will remain high, just because that's part of human nature. What I do see in other media is a strong increase in the use of short film clips, not just for news and education, but also for art and museums. The current difficulties of uploading pictures for newbies will quadruple for film footage, while most people want some pictures along with their text these days. My gut feeling about the numbers, though I again, I can't back this up, is that what you see is *not* a drop in new users, but an increase in a) random vandalism and b) good faith edits that don't pass the increasing amount of sourcing, copyright and notability rules. Layering the results for a 5 year period over increasingly new wikis (5 years old, 4 years old, etc) should show something interesting I think. Jane (talk) 17:24, 11 February 2011 (UTC)

Starting early?

So after talking with Zack last night we're starting to think about the idea of starting the project early with a smaller group (basically whomever is signed up :) ) and working from there to start quickly and smaller figuring out the quirks and what we need and we can continue to expand and grow from there. What do you guys think? I'm thinking if we wanted too we could probably start by the end of the week (Friday? Or do we want to wait until the Monday to send out emails?) but certainly some things to do before that. I'm starting a list below but feel free to edit/extend/discuss. Anything else that you think we need in place before we start a trial? Anything you think we can actually put off if it isn't ready? Jalexander--WMF 02:07, 7 February 2011 (UTC)

From the front page, I have a question about: "1 group of users is just going to be followed while the other groups of users will be randomly assigned to helpers". I presume both the control group (the ones to be followed) and the users assigned a guide, will all have opted-in to participating, both to get feedback afterwards and to satisfy experimental ethics (of some sort)? Have you considered that some people offered a guide may say they don't want a guide? Also, some new users may be relatively inactive (I know I was as a new user). Carcharoth (talk) 03:37, 7 February 2011 (UTC)
Regarding your point about experimental ethics, I think because the data this study would use, a contributor's edit history, is automatically collected that it is not necessary to have the control group opt-in. Really the only part of the data that I can see as protected beyond the usual Wikipedia disclaimer that everything you do here is recorded is responses to email messages. It would be very reasonable to allow editors to opt out and to have the data not linked to accounts. I also think that editors who don't want a guide should be included as part of the experimental data; if, for example, only 5% of new users want a guide and 10% are insulted by the prospect and leave because of it, that would be important to know. (And the numbers are unreasonable, but the phenomena is not; I've certainly had some angry responses to welcome messages.) And certainly post-study feedback is helpful, but it's not entirely necessary for getting meaningful data. IANASocial Scientist: I'm unfamiliar with subjects that haven't been soaked in formalin, so this is not an expert's take. --Danger (talk) 05:31, 7 February 2011 (UTC)
Sorry for the delay in responding, I've had some computer issues ( and wanted to make sure that I checked my own opinions with my personal social scientist ). Danger has most of it. Some of the data is quite clearly open (contributions etc) but the other side is that in general none of the data is going to be personally linked. Rather it will be grouped together metrics wise (i.e of the random 200 person original control group x (y%) were blocked with an average of x edits per person [but only z people edited beyond 1, if you cut out the rest there was an average of x2 edits/person] etc. ) so that tracking the data to 1 person would be very tough even if you had the whole list of users.
Regarding the question about whether they want a guide or not you're very right, some will likely not want them (and some won't respond) and that can be incredibly important information. One of the big questions we have is if this "active outreach" actively reaching out to new users and offering help in fact is worth the effort and time. Everyone is pretty sure that if you start off with 50 you would probably only end with a couple active users at most but is that better then the group that didn't have anyone reach out? If we send out welcome emails (each user will have email activated) and they want help we can continue. If they end up not responding or asking us not to help then we'll be able to focus on those that want it.
We do intend to send out surveys at the end for both the guides and the new users (maybe earlier? We can discuss it). Clearly only a certain portion of those will get returned but that's ok. Almost any data can be helpful :) Jalexander--WMF 23:11, 7 February 2011 (UTC)
(edit conflict)As the somewhat resident social scientist I can chime in a bit here. This project is similar to the one I'm running for my Master's thesis. For that project, any info that's already publicly accessible I'm not gathering informed consent (ie an opt-in) for. In my case, it's posts and comments on another site, anything that doesn't need a special permission to view. For this project, I see edit histories, user contributions, etc. in the same category as, again, anyone can view them. And as to those who are offered a guide and don't want one, that's very useful knowledge that I know will be factored into the final analysis. Hope this helps :) Christine, Wikimedia Foundation (talk) 23:29, 7 February 2011 (UTC)

What do we need?

  1. Email templates for welcome emails
    Peter has a great example above, what other situations are we going to run into?Jalexander--WMF 02:07, 7 February 2011 (UTC)
  2. script to pick out the random groups
    We can either have the script do the randomization or maybe just grab a batch for the past couple days or week and randomize it manually which won't be that bad. Jalexander--WMF 02:07, 7 February 2011 (UTC)
  3. Tool to make it easier to see th edits of everyone you're reached out for?
    I'm thinking of a quick and dirty tool, give it the list of users and it spits out recent contributions from all of them. Zack or I can probably set that up on the toolserver but anyone else interested? Jalexander--WMF 02:07, 7 February 2011 (UTC)
  4. Tool/script to follow metrics (Edits/blocks/talk page edits etc etc)
    I can write up a script pretty easily and post results here but probably best if we can get it up with a bit of a front end so that the results are automatically public. Jalexander--WMF 02:07, 7 February 2011 (UTC)

the hammer

The main thing I still feel here is isolation. Nobody pays much attention to me and I am pretty much on my own. Possibly because the main site I work on nobody else seems to be working on!!! The other thing I receive when I finally do get attention is lots of admins trolling around wanting to delete everything!! I've never had anything deleted yet but it's trying at times having to fight them all off. What annoyed me the very most was a page I had not written but was a "neighborhood" page for the city I work on, one of 28 established neighborhoods in the city, and some admin decided he needed to delete it based on total hits, last time updated and the fact it was pretty stubby. The only problem was, the entire section of the site revolved around a map and navbar designed for the 28 established neighborhoods YET other admins were suggesting combining the page with another 3 neighborhoods (based on some political map they saw on Google)!!!! Yikes, that disturbed me. And to top off, the happy deleter posted the discussion about the proposed deletion on the page he was deleting (instead of the project page). I would have NEVER have known except by a chance search that revealed the travesty to me and was able to thankfully rescue the page before it was actually deleted! Nconwaymicelli (talk) 19:30, 10 February 2011 (UTC)

  • I often feel isolated while I'm working on content too. When you're developing new content you often want recognition and feedback to ensure that what you're doing is helpful. Personally I balance the lonely content work against the aggressively social work at AfD, GAN, DYK and other community noticeboards, but the community areas I think are precisely the hardest for new users to engage with constructively without getting into a potentially offputting argument. (Me, I love the arguments.) - DustFormsWords (talk) 02:50, 11 February 2011 (UTC)
Its in most new editor's best interest to stay way from AN, ANI etc. --Guerillero | My Talk 04:32, 11 February 2011 (UTC)
It's in most users best interest to stay away from AN/ANI :). I think recognition and feedback is something we could get a whole lot better at. I know there is a bit of a "not Myspace/Don't need barnstars etc" idea for some people but I disagree to a relatively big extent. Obviously we're not Facebook but that doesn't meant we're not a community and recognition and feedback within that community is huge and incredibly important for growth. Jalexander--WMF 04:40, 11 February 2011 (UTC)
We do need to be more welcoming. For a lot of new editors their first contact with another wikipedian is a telling-off by a more established editor - or even worse, a bot. It's better to encourage and suggest, and accept that newbies will make mistakes. we should be welcoming first, helpful second and critical later. Totnesmartin (talk) 15:19, 11 February 2011 (UTC)
Agree with all those points. On the isolation point: there seems to be a quite fierce view that "wikipedia is not a social network" to the point where off-topic discussions are frowned upon. People argue that it would distract from the business of creating an encyclopedia. I agree that a debate about politics on the Obama talk page wouldn't be helpful.
But better ways of keeping in touch with other editors, working together and forming friendships (even with the off-topic chat that this involves) would be a good thing. In real life teams, organisations, and clubs, it's partly the social aspect, the beers in the pub or whatever, that keeps people coming back and wanting to be involved in the work. --Physics is all gnomes (talk) 15:23, 11 February 2011 (UTC)
We need to something about the trigger happy RC patrollers. For example, I just started a discussion with a user who considered [this obvious good faith edit] worth an immediate lvl 4 warning. Actions like that can drive off more potential editors than all of us together can guide. Yoenit (talk) 15:34, 11 February 2011 (UTC)
re the social network point - I'm also on a much smaller wiki than WP (yeah, I know, they're all much smaller) which has an open-ended discussion page with no rules. This page has often been the strting point for new articles and ideas. true there's a lot of blather, but it helps keep people coming to the site and feeling part of what's happening. Totnesmartin (talk) 15:50, 11 February 2011 (UTC)
Wow Yoenit. *shakes head* That is a blatant AGF edit. Sometimes I think that twinkle should be a restricted tool like AWB. Its another example of people who don't get the idea that IP edits and brand new accounts are our lifeblood. Without them the project dies. Most of us started out as a SPA or making edits that are nothing like the ones we make today.
As for social networking, I think that more social networking and horse trading happens on all of the "drama boards" then by ignorant new editors. --Guerillero | My Talk 17:42, 11 February 2011 (UTC)

Advertising this study

Hi, I saw this at WP:RFF's talk page. I suggest you also advertise at Wikipedia talk:Welcoming committee and Wikipedia_talk:Please_do_not_bite_the_newcomers. --Physics is all gnomes (talk) 14:24, 4 February 2011 (UTC)

Those are both great idea's thank you, will do now! :) Jalexander--WMF 22:10, 4 February 2011 (UTC)
Don't forget the Community bulletin board! I just pulled an expired notice so there's a free spot. I was going to add the link & notice myself but I figured you might want to word it yourself. -- œ 16:29, 12 February 2011 (UTC)

My input

Got here from the meta banner on my watchlist. My two cents:

  • I am all for welcoming, and have done quite a bit of it manually. People like it. Sometimes it leads to individualized need for help which aids in creating more active editors. I think a team of welcomers is a better start than bots for this reason.
  • We have a kick ass WP:Article wizard.
  • We have a wonderful IRC channel that does nothing but help new contributors here. The more pimping of this the better in my opinion, and there's a nice little community behind that.
  • Help:Introduction to policies and guidelines / Help:Introduction to talk pages are both designed to be the quickest, gentlest and least "too long didn't read" for people just starting out with the bare basics. A soft introduction to wikipedia is important as many don't want to slog through detailed reading of the nitty gritty, nor do they know where to find it. JoeSmack Talk 18:12, 12 February 2011 (UTC)

An alternative explanation for newbie fall-off

This is a theory about this fall-off I've expressed before, but because it's been a few years since I stated it I'll express it again. Possibly this fall-off is due to the fact Wikipedia has reached all of the people who might be interested in writing an encyclopedia, & has saturated its potential market of volunteers. If so, then the only new blood we will be getting from this point forward will come from young people as they become old enough to be useful contributors.

This is not an implausible hypothesis. There are only so many people in the world who find any avocation interesting enough to pursue -- programming, writing poetry, or collecting stamps. Admittedly writing encyclopedia articles is not as popular an activity as any of those, so the relatively few number of contributors should not be a surprise. And if this test fails to show the expected result, then my theory is likely the explanation for the fall-off of new blood. -- llywrch (talk) 18:28, 7 February 2011 (UTC)

I think this hypothesis is rather implausible in light of survey data (PDF) indicating that many people who want to contribute are put off by the steep learning curve. (They also may not understand that they're allowed to. Anecdotally, the most common question I get when I say I edit Wikipedia is "How did you get that job?" and disbelief when I respond that anyone can do it.) It could be that people are just saying that and won't really edit if it's made easier to do; that's the value of trials and studies like this. --Danger (talk) 18:47, 7 February 2011 (UTC)
My theory may not be the actual case, & I can live with that. However, if this experiment -- & I expect this is a real experiment, not another farce as recently happened with "Pending changes" -- doesn't bring in more volunteers who stay longer, then I would hope everyone involved consider whether my theory best describes what has happened, & its implications. (Which is the point I ended my comment above on; I considered my comment little more than a "FYI" to file away until after this experiment has run its course.)

But remember that many people say they want to write a novel, but few write one, & even fewer write one worth reading. And writing an encyclopedia article -- assuming we could magically do away with any & all of the difficulties of Wikipedia policies, the bureaucracy, & the Wiki interface -- is a lot harder than it appears: I'd compare writing one good article to writing a term paper for an undergraduate college class. Not many people would consider that a fun way to spend a free evening. -- llywrch (talk) 01:05, 8 February 2011 (UTC)

Your theory would seem to be part of the reason. All exponential seeming curves in the social sciences seem tot end up as ogives. However we are actually seeing a decrease in active editors, this implies that we are loosing editors faster than we are gaining them, all else being equal we would expect an asymptotic curve to some percentage of the English speaking population. Therefore we must posit that editing Wikipedia is less a attractive proposition than in 2006, leading wither to less new editors, a decrease in active editor life or a combination of the two. Reasons for this might include:
  1. No longer the rebel, WP is now an institution
  2. Instruction creep, and calcification
  3. Complexity
  4. "Not invented here" syndrome
  5. Impolite or insensitive editors
  6. Inward looking, wall of text, conflict filled community pages.
  7. Ownership - this is evident everywhere to a greater or lesser degree
  8. The "easy" stuff is done: - Anton Checkov is no longer a stub, the articles I had planned on the history of numerals are all written, and so forth

I have seen a number of user talk pages - not vandals' pages - that consist solely or mainly of templated reasons why their content is going to be deleted. And in many cases the people who are templating are wrong - but even if the content is saved, the editor may well have wandered off to play Farmville, considering that, all in all, it will be more productive. Rich Farmbrough, 10:54, 8 February 2011 (UTC).

Indeed, if an editor decides to contribute images and doesn't understand the intricacies of the templating and copyright on the first go around, they can have a 40kb page of nothing but "We're about to delete what you contributed!" --Danger (talk) 11:35, 8 February 2011 (UTC)
Aye, I've always had concerns about the templates too (especially the bot templates. There is nothing worse then seeing a talk page filled by bot's saying "we're going to delete your image because you didn't fill out the form correctly" ) and I saw that too frequently even when I was going to look at images because we had received permission for them. I know there are people working on trying to make the upload process easier (the beta upload wizard at commons for example) but when we have a confusing system and all you get back is being yelled at (which happens a lot sadly) it is bound to scare people off. I think personal welcomes and notes could be one of the biggest things. There are obviously a lot of different issues (for both older and newer users) that can be improved in a lot of different ways. Jalexander--WMF 02:24, 9 February 2011 (UTC)
Seeing if the Twinkle folk know of a way to deal with this. If they come up with something, presumably that can be ported to the bots.--Danger (talk) 02:59, 9 February 2011 (UTC)

In my opinion, the editor drop off is related to a shift that Wikipedia is currently undergoing, the shifting being from having more poeple write more stuff to have experienced people who can cite and clean up what we have. This, by its nature, leaves those who were not part of the community during phase 1 out in the rain. That's my theory anyway. TomStar81 (Talk) 13:47, 10 February 2011 (UTC)

(At the risk of being {{tldr}}'ed:) I can only speak to my own experience, but I think that llywrch is closer to being right (though not quite there) than Danger and I think that's borne out by the survey data cited by Danger. In each of the categories cited in the study (which are, BTW, insufficient to give an overall picture), i.e. over 60 years of age, female, and highly educated, the two highest reasons for each of the three groups for not editing are "I don't think I have enough information to contribute" and "I am happy just to read it; I don't need to write it". The latter can be written off as "I just don't want to bother," but the former one - which gets answered consistently two or three times more often than the ones saying that it's too hard or too intimidating - is the key: in terms of attracting new editors, WP is a victim of its own model and its own success.
The model contributes to the new editor problem by largely attracting volunteers, most of whom are dilettantes only interested in writing about subjects in which they have a preexisting interest, not interested in becoming long-term Wikipedians, and most of whom have only an experiential or surface knowledge about the subjects in which they do have a preexisting interest. The success of Wikipedia contributes to that problem by the fact that we now have 3.55 million articles: Articles already exist about most subjects in which a new editor may have a preexisting interest and, due to most new editor's limited or superficial knowledge about their area of interest, they're not interested in documenting or improving the content of existing articles except perhaps to correct glaring errors or omissions (or, heaven help us, what they perceive or POV-contend to be such). What, in my opinion, the average potential editor is not interested in doing is researching/learning more about their subject of interest just to contribute to Wikipedia, much less to contribute to some subject about which they have no preexisting interest.
Thus, as the number of articles grows, the pool of potential editors grows smaller and smaller and becomes more and more limited to:
1. editors only interested in writing about increasingly-narrow niche subjects (the pool of which that have a reasonable chance of being verifiable and notable growing smaller every day), the vast majority of whom will, once having writ, either be driven away by having their work deleted or rewritten (I'm not implying that such deletion or rewriting is a bad thing) or by losing interest in WP once they have written about their subject of interest,
2. newcomers only interested in inserting personal comments and opinions, original research, and other unencyclopedic material,
3. POV-pushers and other ideologues,
4. true experts, often limited because true experts - especially professionals - frequently have demands on their time which are, or which they choose or perceive to be, more important than volunteer work such as editing Wikipedia,
5. individuals more interested in bureaucracy, process, or governance than content,
6. individuals who are attracted to Wikipedia as a goal in itself and who are interested in working on articles about which they have no personal interest, and
7. vandals.
The potential number of editors falling into classes 1, 4, and 6 are, I would contend, very small. The potential editors who formed the first big boom, are simply not there any more, in my opinion, because most of the "easy work" has been done about most subjects.
Let me hasten to add that I do not intend to be critical of Wikipedia by any of the foregoing statements: I see this as the normal life cycle of any project which is entirely volunteer-supported and in which there is no easy way for individual editors to benefit by working here other than through self-satisfaction. I'm also not saying that the oft-claimed technological, bureaucratic/policy, and trollishness barriers do not play a part in the downturn in new editors, but I do strongly feel that they are not the primary cause of the downturn. IMHO, therefore, even if those barriers were magically eliminated overnight (and in a manner which would not allow the inmates - classes 2, 3, and 7, above - to take charge of the asylum), I do not think that the number of new editors would in the long run substantially increase but would instead continue to decrease.
If I am correct, then we can only hope that the number of new editors in classes 4 and 6 is sufficient to make up for the number of experienced authors in those classes who are lost through age, disability, burnout, or other attrition, otherwise Wikipedia will devolve into either a museum with a few caretakers absently maintaining an exhibit here and there while the rest of the exhibits slowly gain a coat of dust or into a battleground where 2, 3, and 7 reign or, perhaps, both. Can we encourage, lead, or train new users into classes 4 and 6? Your guess is as good as mine, but I would suggest that the claimed technological, bureaucratic/policy, and trollishness barriers are not nearly as much a disincentive to potential editors in those categories, especially category 6, as they may be for the dwindling number of folks who just want to write an article about their interest du jour.
Best regards, TRANSPORTERMAN (TALK) 19:37, 10 February 2011 (UTC)
Very interesting post! A lot of wisdom and insight there. I myself would definitely fit in category #6. -- œ 19:16, 13 February 2011 (UTC)
We are trying to attract more people in group 4 via the ambassador program. I seem to be part of group 1 on some days and group 6 on others.--Guerillero | My Talk 21:09, 13 February 2011 (UTC)

How's my driving / user quick dashboard tool

If we are going to drop some advice on a newer user, it would be smart to ensure that we have checked the basics about their contributions and are not offering advice too early or to someone who is immediately making unusual edits and does not fit the profile of a new contributor. I was wondering if someone would care to knock up a script (or identify one existing) that can generate a small wiki-code edit review dashboard showing things like:

  • Date account created and any other key user log events
  • The range of contributions (such as a pareto analysis) to check if they are a single-purpose account or if there are one or two common categories for 80% of the articles they edit (and so might be interested in a particular WikiProject)
  • Velocity (a user with 25 successful edits in a day might be approached differently to someone making 25 edits in a month)
  • Acceleration (showing the difference between a user who started with a rush of contributions but has flagged or stopped since and a user who's contributions are slowly growing as they are becoming more confident and interested)
  • Progress on a check-list of top-ten desirable set of new-user life-cycle events, for example:
    • Made article contributions (that have stuck for 24hrs+)
    • Been welcomed with a guide of where to find help
    • Cited a new reference in an article
    • Collaborated on article talk pages
    • Significantly grown an article (5x expansion) or created a new one successfully
    • Written in a WikiProject (if not 'joined')
    • Used templates
    • Used their userspace to draft articles
    • Asked for help on a noticeboard
    • Used or uploaded an image
    • Summary of improvement areas due to warnings, blocks, new page deletions etc.
    • SUL check to see if they are contributing in sister wikis

With such an edit review dashboard it would be easy to graphically check if a user has tried the collaborative features of Wikipedia and such a tool might recommend a tailored form of welcome (potentially including a snapshot of the graphical dashboard on their page). -- (talk) 05:43, 11 February 2011 (UTC)

A lot of users make one or three edits and then forget their password. I don't think you should try to do statistical analysis, instead I would try to identify common editing profiles and then describe and identify the characteristics of those profiles. You have already identified one, "the experienced editor with a new alt". Gigs (talk) 03:16, 12 February 2011 (UTC)
Sure, though I think a tool that picked up on these numbers and suggested which common profile is likely to apply would make life easy. (talk) 07:15, 12 February 2011 (UTC)
A similar, interesting idea would be to have a bot post these things as small awards on a user's talk page. Slashdot has a similar system. On the other hand, that would take a lot of bot edits, and might make it hard to maintain (the appearance of) WP:TINC. -- Ken_g6 (factors | composites) 21:03, 13 February 2011 (UTC)
I'm a fan of little awards and encouragement. I think they are under appreciated within the community sometimes despite the fact that they frequently enjoy them whent hey get it :). I know there was some extensive discussion during the strategy project about things like this (You can see a some of it strategy:Editor_awards_and_rewards here and I think we need to think locally about some of that. Jalexander--WMF 23:33, 13 February 2011 (UTC)

Discourge marketing/advertising people

Lots of new editors that the community and Admins have to deal with is People involved in marketing, advertising and public relations. See my essay WP:PEW regarding most of the problems associated with these editors. We need to present new individual/corporate users with information immediately to inform them that there rules about editing their own company/personal/client articles. Phearson (talk) 20:17, 11 February 2011 (UTC)

I usually leave them a {{welcome-COI}}. Maybe welcome-coi could be improved a little. Gigs (talk) 03:10, 12 February 2011 (UTC)
I use a nice general welcome and follow-up with User:Fæ/help/whyn which gives some friendly pointers as to what encyclopaedic notability means and why it might not be the same as their intuitive understanding rather than making it seem like an accusation of fault on their part. (talk) 13:32, 12 February 2011 (UTC)
I think it's important to remember that they can sometimes be helpful too. COI is important to remember but it's really only a problem when they start trying to use that to hide legitimate things etc. They can also be immense help correctly BLP issues, getting better photos etc and in the end they don't want an article deleted etc (and I've found that's sometimes a better tact to take, They can work with us to make the article(s) better but the LAST thing they want is to have the fact that they were deleted for notability or something like that spread around). On the WikiEN mailing list a couple days ago someone pointed out a more recent article (for Marketing/PR people) about this here. There are a bunch like that too where people are starting to realize that they are better off working with us and within the rules, most of them want to you just need to help them understand the rules. Jalexander--WMF 23:37, 13 February 2011 (UTC)


As a thought experiment, pretend you don't know anything about OTRS and try to figure out how to submit permission to use copyrighted images. Assume that someone posted a note on your talk page: "Hi I noticed that you have submitted an image that appears to be copyrighted, please contact otrs to submit proof that you own these images." Have fun. Gigs (talk) 03:05, 12 February 2011 (UTC)

Ugh, I've learned to explain these things to complete noobs in as few words as possible and still don't understand why our templates are so confusing. Forget linking to OTRS and explaining the copyright thing, just give them the email and tell the photographer/author to email that address saying whatever WP:CONSENT says. /ƒETCHCOMMS/ 03:24, 12 February 2011 (UTC)
Wikipedia:Contact_OTRS might be helpful. Gigs (talk) 03:41, 12 February 2011 (UTC)
I a created a help template to substitute that tries to be friendly - User:Fæ/help/photo - my best effort but I think it's still confusing. (talk) 07:13, 12 February 2011 (UTC)
  • If the templates are confusing, you could simplify them. Locke'sGhost 07:26, 12 February 2011 (UTC)
How about Wikipedia:Donating copyrighted material. Yoenit (talk) 17:02, 12 February 2011 (UTC)
I think there is a lot of work we could do improving templates to be honest. Perhaps outside of this project but not totally since trying to improve the process in general is the best thing. A lot of work has been put into them obviously but there is very little if any usefulness in a giant talk page of notices and warnings especially for things like image deletions. Jalexander--WMF 00:02, 14 February 2011 (UTC)
They should be linked to Wikipedia:Contact us or Wikipedia:CONSENT, not Wikipedia:OTRS. Stifle (talk) 09:52, 17 February 2011 (UTC)

Rename "Talk" to "Discussion"...

...then tell people to go there and read it.

  1. to somebody who doesn't know it (and possibly learned English in school), what's "talk" supposed to mean? Telephone? Do I need some special software? And talk to whom? The administration of this place? (and who the hell reads the rollover-note?)
  2. by reading through the discussion on a topic you're interested in, you learn a lot more about how this whole deal works than by digging yourself through all the rules and regulations.

Choyoołʼįįhí:Seb az86556 > haneʼ 23:53, 13 February 2011 (UTC)

I totally agree, I love talk pages (they were how I first got interested in the community on wiki) you can sometimes learn just as much from those discussion as the article itself. I'm not actually sure when it changed but I see "Discussion" instead of talk right now... Is it a skin thing? I'm using Vector but tried with monobook and am seeing Discussion there as well. Jalexander--WMF 03:37, 14 February 2011 (UTC)
Ohhhh You mean the name space.. That's true hadn't even thought about that. Jalexander--WMF 03:41, 14 February 2011 (UTC)
Where did you check? The tab/button says "Discussion" on meta, on it says "Talk" Choyoołʼįįhí:Seb az86556 > haneʼ 11:39, 14 February 2011 (UTC)
For me it says Discussion on enWiki as well (I checked with 2 others as well.. not sure what the difference is hmm). Jalexander--WMF 20:31, 14 February 2011 (UTC)
I think Twinkle might change it to Talk. It's talk for me, but on IE and when I'm not logged in I believe it's Discussion. demize (t · c) 20:38, 14 February 2011 (UTC)
Yeah I just checked, the mediawiki interface is definitely set for Discussion here. Which is actually the mediawiki default now. I'm not sure why Twinkle would change that.... weird.... Jalexander--WMF 21:22, 14 February 2011 (UTC)
Oh, that's weird. Well, then forget about the first point above. :) Choyoołʼįįhí:Seb az86556 > haneʼ 22:33, 14 February 2011 (UTC)
It is a browser thing. The Internet Explorer tab says "Discussion", while the Google Chrome tab says "Talk" Sumsum2010·T·C·Review me! 01:24, 18 February 2011 (UTC)

Attack of the fanboys/fangirls (and the AfDs that inevitably follow)

I think that part of Wikipedia's nature is that it will attract a large group of people who are particularly passionate about their particular area of "expertise" (codified here as WikiProjects, generally). However, an unfortunate side effect of this is that a lot of articles comprising primarily fancruft (especially those dealing with fictional subjects) are generated. I think that there is often a disproportionate reaction AfD-wise, and for some reason WP:BITE goes right out the window. I realize it can be frustrating dealing with new users (especially fans), but users who could be mentored in what Wikipedia is not and subsequently improve userified versions of those articles to the point of acceptability would be a far greater asset then those who give up and go away because they didn't understand which arguments to avoid in deletion discussion, had a bunch of things starting in "WP:" thrown at them, and then had the articles they'd been working on deleted or merged to the point of near-inconsequentiality. A lot of new users don't know about or understand the process of userification or incubation (in fact, I didn't know about incubation until I just looked up the userification link, so there ya go) and as such simply give up when their articles are deleted.

I myself initially fell prey to "what about x?" and arbitrary quantity arguments while researching deleted articles recently as a member of the Warcraft task force (which is honestly all but dead due in large part to what I'm talking about here), especially as a majority of the deletion arguments ran along the lines of "Delete Nothing but cruft." and "Delete This is not notable outside the gaming community." The sheer quantity of these arguments rather easily overpowered the one or two reasonable "Delete Article is poorly written and requires verifiable sources." The general failure to assume good faith overshadowed the true issue, and while I'm certain there were likely multiple reverts and edit warring, the fact of the matter is that the topic wound up suffering for it, rather than any kind of improvement taking place. Addition by subtraction is not necessarily the best policy.

Anyhow, the result is that virtually all interest in editing/developing a potentially notable topic (and I'm talking about one with existing references that may not have been fully researched (or researched at all), not crystal ball-type stuff) is destroyed. The default response seems to be no references equals hasta la vista, with little to no middle ground where editors are guided to potential reference resources. Another side effect of this is that if an editor decides to attempt to revitalize the topic, they now have to make sure they know about, understand, and follow the undeletion process, or risk having their work speedily tossed out again.

I think that with some initial welcoming/mentoring and a slightly "more friendly" process than no-holds-barred AfD discussions, the passion that a lot of new editors feel for their subject matter topics can be positively channeled into creating good, well-referenced articles. I realize that there will always be fanatics, there will always be deletionists, and there will always be people who bite, but there should be some way to make the overall process a little more user-friendly.  Cjmclark (Contact) 19:34, 14 February 2011 (UTC)

I personally think that deletionists get a bit overzealous in their field; I've been thinking for some time about writing an essay on that topic, in fact. Here, I'll stick to a particular angle of the issue, and simply say that for a good deal of these articles (e.g. episode articles for TV series, especially series for which a lot of the episodes are in fact notable enough for separate pages), instead of nominating a crufty page for deletion, it should be merged in with the article covering the larger part of the topic (episode list, for example). I say this because readers trying to find the episode might type in its title in the search bar, only to find nothing on it, when they might have been directed to the list part of the list which discusses it, instead. Plus, if at any time the episode does become notable enough for its own page, the article's content can be restored and expanded on, and the original creator can still receive credit for it.
If these users try to add in crufty pages/info to Wikipedia, I think one good method of dealing with it, if they are persistently passionate about adding in this sort of content, is to direct them over to Wikia, so they can contribute the content to a wiki on the topic/found a wiki there on the topic, and then start adding in the content. Although I'm very scorned by how blog/interaction-site-ish Wikia has become in recent times, it still remains one of the best options these "experts" on the topics have. Wilhelmina Will (talk) 00:24, 15 February 2011 (UTC)
Typical response from wikipedia: round up and ship off the so called "cruft" pages to afd camps for orderly deletion. I take the exact opposite approach: we should be welcoming this kind of content here instead of trying to remove it. Who are we to decide that fiction and those who dabble in it are not welcome here as well? Is this not a site that attempts to gather the sum of all human knowledge? Jim Crowe policies and guidelines that openly discriminate against fiction are a part of the factor prohibiting us from expanding our horizons. Rather than cart this material off to other sites we should give it a chance to grow here. Wikia - regardless of how well it works - is not wikipedia, and if we send people off to another site were slapping them in the face on two fronts: first, we are telling them they have nothing of worth to contribute here, and that drives these users and thier friends away form wikipedia. Second, Wikia results are not as highly praised as Wikipedia results. Would you want to go to another wiki to write about world history, politics, art, religion, feminism, etc simply because your topic was judged to be second rate when compared to other matters on Wikipedia? If you could not contribute here then would go to another wiki knowing that the material you add would never rate higher than 2nd place on a google search? I wouldn't, and I am sure that if I wouldn't then there are others who would agree that if you could not add the material to the gold standard Wikipedia why the hell would you settle for a silver medal elsewhere knowing what you wrote would always come in second place at best? It'd be enough for me to give up and walk away from this site permently. TomStar81 (Talk) 02:00, 15 February 2011 (UTC)
Yeah, I'm hesitant on the wikia front as well (see my "early" editing experience). I made some edits that weren't in line with the "rules" of a WikiProject I had joined but didn't yet fully understand (in fact, I didn't even know WikiProjects existed before I joined it), and the editor who stopped by to correct me gave me the old "why don't you check out a wikia instead, since that's where fans belong." I was, to put it mildly, pissed. Here was something I had been putting a lot of work into (and reinventing the wheel a lot, since I never had any kind of welcome or guidance), and some editor with 35,000+ edits rolls by to basically tell me I'm wasting my time. And that was non-fiction.
I do agree that the focus of fiction articles should be on their "real-world relevance," so to other words, more of the artists/authors discussing the creative process, analysis of the character, maybe some awards the subject received or how much money a movie based on it made, rather than heavy plot summaries. This is not to say that there isn't a place for plot summary...after all, it's often difficult to create a context for a fictional character without having at least a basic understanding of its backstory. I just don't think it needs to be the bulk of the article. Arthas Menethil (under reconstruction) on Wikipedia shouldn't be as plot/lore-based as Arthas Menethil on WoWWiki/Wowpedia, for example. It should look more like Cloud Strife.
And I think we can get there for a lot of fictional (let's face it, primarily video-game and comic book related) subjects. Their editors just need to be able to get past the initial AfD onslaught that editors seem to delight in whenever they see something that they don't think is notable or hasn't been properly referenced. And like I said before, deletion shouldn't be the first answer. Userify or incubate an article if it needs work that badly. It bothers me that entire topic areas get annihilated based on the opinions of ten to twelve editors.  Cjmclark (Contact) 03:47, 15 February 2011 (UTC)
    • Pop culture stuff can be done reasonably well. There are FAs, such as Tokyo Mew Mew.Locke'sGhost 02:35, 15 February 2011 (UTC)
Very true. They just have to be run the AfD notability gauntlet to make it there. I know this is a WP:WAX thing, but my original articles were BLPs for minor-league hockey players. These are guys who don't ever necessarily make it to the big leagues, but because they're real, and I can dig up something off of to show that they've played enough games to satisfy WikiProject Ice Hockey's minimum notability requirement, they make it through without a minimum of fuss (my "why don't you go to a wikia" experience notwithstanding).
Go one state away from mine and ask them who Nate Kiser is, and chances are you'll get a blank stare. Ask around about Thrall the Orc Warcheif (I cringe not only at the misspelling but at the fact that it redirects to an article that basically just mentions him in one paragraph of plot summary), and chances are you'll get a few more hits (from an astonishing spectrum of people). And the bitch of it all is that there are reliable, third-party references out there to support generating an article for him (Thrall), but the editors who are most likely to create/edit World of Warcraft articles aren't necessarily as experienced at this as those who edit any number of non-fiction topics.
And so, their work meets its doom at the hands of AfD. We know these subjects have potential, so what can we do to let that potential be explored without the articles (and their editors) getting rapidly crushed?  Cjmclark (Contact) 03:47, 15 February 2011 (UTC)
Thrall is not cruft and I am sure a GA can be written on him easily. Cruft is things like Tourneys in A Song of Ice and Fire (I have no clue why Jclemens decided to overturn the AFD and redirect this, but it sure comes in handy now). Or how about the complete family tree of a fictional royal house from that same series. If we include this type of material, then why not history and culture of elementals in WoW, or Language of fire elementals in WoW. Where do we stop? Yoenit (talk) 08:34, 15 February 2011 (UTC)
I agree with you, but unfortunately cruft seems to be in the eye of the beholder. And if you made that argument in an AfD (even for a topic that you firmly believe to be notable), you'd get hit with the "Delete WP:WAX, therefore your argument (no matter how compelling and relevant your reasons) is irrelevant." For a new editor, that leaves him/her with one of two choices: either give up, or dig your way through Wikipedia legalese in a desperate effort to obtain a stay of execution long enough to edit the article to the point of acceptability. Sadly, most took the far easier path. All but the bare bones of the Warcraft articles were dismissed as cruft back in 2007 (and many of them were), but the actually notable ones (and there were fewer then than now) could have been saved with better writing and sourcing. Unfortunately, the baby was thrown out with the bathwater, so to speak.
I think this is where the respective WikiProjects really need to step up. Without any kind of backup from subject area (and Wikipedia) "experts," new editors are going to find themselves at the mercy of trigger-happy deletionists time and time again. I'm still relatively junior here, so I don't know all the ins and outs. Is there any recourse for a new editor who finds himself over his head in an AfD short of running around to WikiProject talk pages asking for help (and then being accused of canvassing/campaigning? I think most of them don't know the system well enough and are dismissed on technicalities in many cases.  Cjmclark (Contact) 14:22, 15 February 2011 (UTC)

AfD is hostile

I have to agree with Cjmclark. Although the WP:AfD process is necessary, from a newbie's perspective, AfD is one of the most hostile places on Wikipeda. This is especially true because they don't understand Wikpedia's processes and rules. Wikipedia does need an AfD process, but the current AfD process suffers from many problems.

  • Myth — Users have 7 days to fix an article so it won't be deleted.
Reality — Filing an AfD is the first !vote, so the process begins instantly. The majoirty of users !vote and move on. They don't come back after 7 days to re-review the article and adjust their !votes according to how much the article has been improved.
  • Myth — The AfD process is neutral.
Reality — Many AfD nominators suffer from WP:EGO and and feel defeated when an article is kept. They will actively oppose improvements to the article and argue with people about their Keep !votes. Some have even been known to revert improvements to an article in an effort to make certain an article is deleted.
  • Myth — Only articles that truly should be deleted, are nominated.
Reality — Many nominators use AfD to force an article merge or some other process.
  • Myth — Only unsalvagable articles are nominated.
Reality — Many nominators will point out both the problems an article has and the solutions, but they find it easier to nominate than to follow WP:SOFIXIT.
  • Myth — Every AfD gets a fair shake from the community AND from the closing Admin.
Reality — The AfD process produces a lot of work for closings Admins and some don't spend the extra time to review the article to make certain the !votes accurately reflect the current state of the artilce at closing.

There are many more problems with the AfD process, but there is no need to for me to swamp this page with them. We should try to improve this situation. - Hydroxonium (H3O+) 17:54, 15 February 2011 (UTC)

The process is fine, there are just some problems with rabid deletionists and inclusionists which crop up all over wiki. With regards to your myths:
  1. Truth, If substantial changes are made to an article later in the week (or sources found) this should be noted in the AFD and the closing admin will take it into account and relist if necessary. (see point 5 if it does not happen)
  2. Truth, Process is neutral, but editor behavior not always. You give some excellent examples of this.
  3. Indeed a myth, but why is this a problem of AFD? If you try to force a redirect or Merge through AFD you are doing it wrong. AFDs where the nominator does not advocate deletion get speedy kept and actual merge/redirect outcomes are rare. 9 out 10 times the result is "no consensus/keep - merge discussion can be held at the article talkpage", having achieved exactly nothing.
  4. Huh, If all articles nominated for AFD were unsalvageable then why on earth would we need the process? We could just delete them all directly! The whole point of AFD is to determine whether the article is unsalvageable.
  5. Not a myth, but a delusion. Name me a human single process which works perfectly every time? wp:DRV exists for a reason.
You did not actually address the real problem with AFD and new editors: They are not familiar with one or more of the policies/guidelines underlying the argument for deltion (usually some combination of wp:N, wp:RS, wp:NOT, wp:OR and/or wp:NPOV) and are thus unable to counter the deletion arguments, resorting to bad arguments and canvassing (more stuff they are not familiar with). Also, in most cases the article in question should indeed be deleted. My advice for a new editor confronted with AFD: State your case once, ask for userfication so you can continue working on the article and go do something else for 7 days. What will happen will happen and any actions to prevent it will likely backfire. Yoenit (talk) 21:36, 15 February 2011 (UTC)
And again, I think this is a case where more in-depth education is required. The current method of slapping a cookie-cutter welcome template on their page with links to wp:N, wp:RS, wp:NOT, wp:OR and wp:NPOV seems to have been fairly ineffective (and, as we've seen, is applied unevenly anyways). These and (WP:AADD) are all vital things that new editors need to know before they ever click a Save page button.
Now, mind you, this is aimed at those editors who have the potential to be decent contributors. There will always be vandals, but as Daniel Case points out below with WP:IAV, those folks are going to be dealt with systematically anyway. So how can we teach the horse to drink, so to speak?  Cjmclark (Contact) 14:20, 17 February 2011 (UTC)

The unintended consequences of our own successes ...

I think a lot of the new-editor hurdles are, in many ways, inevitable results of our own efforts to improve the encyclopedia.

Having, as I just noted above, been on Wikipedia for a little over six years now, I remember a different place when I started. Back near what I will, in the hope of initiating a useful metaphorica use of a physics term, call the later days of the Planck time[note 1] of Wikipedia.

A place where a lot of editors were casual, making a few edits here and there to a favorite subject, usually something related to pop culture. Where many of them were not full-time members of the community, and didn't feel that they had to be. Wikipedia was just one of many places they were active online. As much as we disliked it at AfD, the differences between Wikipedia and MySpace (God had not yet created Facebook) were not as clear as they are now.

And there were giants in the earth in those days ... Well, not so much. In retrospect it feels like there should have been ...

It was, indeed, a generally more freewheeling place. The ArbCom and many of our current policies existed then. But not all. The rules we had were enforced, but you could and sometimes did smuggle something cute through (For example, note the "Early Borg Cube" joke from when I created Orange County Government Center. That's my joke, and it never caught on. It lasted about a year.)

At that time it was just a suggestion that you list your sources in a section at the end of the article, like the Britannica does. It was a requirement only if you wanted to get the article featured.

However, at the same time media coverage of Wikipedia focused on this creepy "anyone can edit" thing, and all the vandalism. I was not always forthcoming to people about being involved so much. Yet at the same time so many people seemed to trust it and go to it that you felt proud to be a Wikipedian.[note 2]

That was before Seigenthaler. Before the Dark Times. Before the Empire ...

At the time we certainly took it as a fall from grace. In the years since we have come to see it as a major turning point in the project's history; I can't see how it won't loom larger as time goes on.[note 3]

There was much gnashing of teeth. Much throwing about of brains. One of the most voted-in AfDs I remember. We all said a lot of things. But we were sure of one thing:

We never, ever wanted to read a story like that again.

Immediately, unregistered IPs were disallowed from creating new articles. I don't quite know how that tied into the problem. But BLP was created and given teeth with Jimbo's blessing and encouragement.

Slowly, other changes took hold. Changes that were, in and of themselves, good and necessary, but had the cumulative effect of essentially seceding from the Internet culture that had nurtured Wikipedia and so many of its editors from that first "Hello world" in 2001.

  • One had started beforehand. In late 2004, as I had been working up to creating my account, the removal of all fair-use images from non-article namespace was underway[note 4]
    • Then, in May 2005 the scope of what constituted a free image was severely restricted. Wikipedia-only? Gone. CC? Only BY and SA, please. With-permission? Forget about it.
    • This set the stage for a decision I was among many opponents of at the time: the narrowing of WP:FUC to include the stricture that fair-use images were permitted only where they could be elaborately justified using a complicated checklist, and where they could not be reasonably replaced with a free image ... regardless of whether one actually existed (My attempt to seek some sanity for this latter provision failed). In late 2006 the bots and the users who like nothing so much as a newly discovered forest of exposed nails were set loose with full boxes of swinging hammers, leaving a lot of embarassed redlinks and empty space.

      The purges continued with the Main Page becoming a free-image–only zone, and all fair-use FPs losing that status as well. We also started referring to fair use images as "non-free" (Which means, I suppose, that free images were now "Media from friendly socialist countries").

      All of this because one of the five pillars is "promote free content".

  • In the wake of Seigenthaler, not only sources but inline footnotes (for a nostalgia check, go see {{ref}} and {{note}}, which apparently have some uses still and are not considered deprecated yet) were required. FA status became impossible without them, GAs eventually needed them too, and even DYK hooks were supposed to be cited. Unsourced info could be, and was, not only challenged but aggressively removed by the swinging-hammer brigades, many of whom rarely if ever did any real content editing beyond that and slapping articles with {{unreferenced}}.
  • Then we started doing the same to "trivia" (or as it was later euphemized, "miscellanea") and "in popular culture" sections. In my first summer as an admin (2007), I closed a bunch of these AfDs. I admit that things like "Terminator 2 in popular culture" were never likely to be kept. But even after it turned out some months later that some of the accounts that had agressively pushed these AfDs were, in fact, socks of a couple of other users, many were nonetheless upheld at DRV. A lot of our quirky yet admittedly crufty lists ("List of songs whose title constitutes the entire lyrics", "List of songs that begin with non-music") met the same fate.
  • The Badlydrawnjeff ArbCom case sort of cemented it that transient Internet memes were no longer worthy subjects for articles, for both notability and BLP reasons.
  • Then we really got serious and turned Bad Jokes and Other Deleted Nonsense into just ... Deleted. Yes, we still have humor pages aplenty. And there were legitimate licensing-related reasons to get it off Wikipedia and move it to Uncyclopedia. But it was the day we started taking ourselves very seriously.

And during this whole time, we were encouraged to protect Wikipedia from desecration zealously. Watch articles. Revert revert revert. When you get admin tools, "'Cause Jimbo don't like 'em ... Block the vandals, block the vandals". We decoupled rollback from the admin package and gave it to deserving young users eager to level up to admin status. We added flagged protection on some articles, to general media applause. We improved Recentchanges to give us better clues as to what was vandalism, and that wasn't enough. We set bots finally worthy of the name loose on the 'pedia to find what we could no longer ourselves and bring it to administrative attention. We gave our users badass tools with deceptively, almost-purposely-chosen-for-the-irony names like Twinkle and Huggle, augmenting our obedient and vigilant bots with an army of driven, dedicated cyborgs heading for the vandals and spammers on the horizon in full-throated cry of "Exterminate! Exterminate!"

And we got respectability. Don't get me wrong. I think that without this effort Wikipedia would have become as much of a joke nostalgia piece as MySpace is now, one of those "little souvenirs/ Of a horrible year/ That makes me smile inside".

But it did not come without a price. Daniel Case (talk) 03:05, 15 February 2011 (UTC)

This is an excellent summary of Wikipedia's life. I can't tell you how many times I've seen similar essays on retired user's userpages. This really explains the graphs of the rise and fall in users. Success, but at a price. - Hydroxonium (H3O+) 15:10, 15 February 2011 (UTC)
First off, nice links. It's nice to get a good laugh in once in a while.
More to the point, I understand and agree with what you're saying. Let me provide a hypothetical "real-world" parallel. Let's say I used to work in a highly technical and potentially dangerous industry. When it started out, there was comparatively little (by today's standards) oversight and the tech manuals fit in a 2-inch three ring binder. Technicians and operators were entrusted with using their personal experience and knowledge (which was considerable, due to their extensive training) to accomplish jobs, and life was good.
Then came our version of Seigenthaler, and the subsequent gnashing of teeth. Fast-forward 30 years, and now the same job requires the permission of five people, the direct oversight of three people, at least two workers to accomplish the job, six hours of paperwork, three days of training, a day or so of lead time, a pre-job brief, a pre-brief brief, and training to ensure the the pre-brief brief went well. Oh, and those tech manuals? Now they live in forty or so volumes worth of 4-inch three-ring binders. Did the job change? Not at all. Were fewer mistakes made?
Ah, now, there's the question. Thankfully, in our little corner of this world, we didn't have any big mistakes. You might say that's our "respectability," and it is very, very important. But curiously enough, the small mistakes (the ones you don't hear about because well, frankly, they really aren't shocking) increased in number, and the punishment for them was severe. As a result, personnel attrition skyrocketed, and we found it difficult to get people interested in employment, much less sticking around. Sound familiar?
So where did we go wrong? Well, you can all draw your own conclusions. In my hypothetical opinion, we created a process that removed accountability and responsibility from individuals and gave it over to a system. A system where it's easier to send someone packing for violating paragraph y of rule x, instead of sitting them down and helping them get better. But we had our respectability, by God.  Cjmclark (Contact) 15:17, 15 February 2011 (UTC)
Good followup. But, of course, even in them daze there was still not much of a system for helping new editors (in fact, I'd honestly say we have more now in that department), and newbie-biting was considered by some to be as much of a problem (See WP:Adminitis, and note when it was created).

The key difference then was that there was more room for new users, who weren't outright vandals, to make the mistakes from which they must surely be learning to become good and productive editors without getting forearmed across the face. If you look back at the earliest volumes of my talk page, I didn't get one templated warning back then. (I don't think). Daniel Case (talk) 17:01, 15 February 2011 (UTC)

I completely agree. It was the same for us. Back in the day, you were at least given the opportunity to learn from your mistakes, rather than simply receiving a swift and immediate boot to the head.  Cjmclark (Contact) 17:32, 15 February 2011 (UTC)
The Clash, Daleks and The Frantics all in one thread. I love it. Keep 'em coming. - Hydroxonium (H3O+) 17:44, 15 February 2011 (UTC)
I would also commend to your attention WP:IAV. An interesting read ... Daniel Case (talk) 01:34, 16 February 2011 (UTC)


  1. ^ In the sense, not explored in that article, that the laws of physics did not come into full force until one Planck time had elapsed since the Big Bang.
  2. ^ I wonder if all the "... in popular culture" and other fanboy/girl cruft actually helped build that trust. It made Wikipedia not only the first open-content encyclopedia, but also the first encyclopedia to reflect the Genx/Millenial experience of late 20th century popular culture. Readers saw not only their world but their worldview in our pages.
  3. ^ Ironically, given the present discussion, it helped us, with one of our largest influxes of new editors yet.
  4. ^ Yes, at that time you could put any images you liked on your userpage, images from the TV shows you were a fan of. WikiProjects like Star Wars and Star Trek had actual images from those franchises in their project banners and userboxen.

Resources for New Users

What do you think are some good resources for new users? What do we need to get ready for them?

I know we have some resources at the the Bookshelf project My favorite is the Wiki Cheat sheet

Wiki cheat sheet

and the book it comes from

Welcome to Wikipedia book

among a couple videos and pieces (both completed and in development). We also have a pile of pages here that can get a bit confusing for a new user.

What do you think? Jalexander--WMF 05:10, 3 February 2011 (UTC)

Anything concerning AFD, CSD, WP:DR, Arguments to avoid in deletion, and so on in that manner since most of the new people here are welcomed to wikipedia with the message "Hi, we are deleting your article." TomStar81 (Talk) 13:44, 10 February 2011 (UTC)
Very true. Sage pointed out some great resources that the Wikipedia Ambassadors put together here but we probably want to try and get some simple explanations together ready.
I think there needs to be some guidance in Wikimarkup - like headers and links - for articles and templates. I've seen some new users struggle with formatting and wikifying articles correctly. Maybe a simple, straight-forward guidebook? ► Wireless Keyboard ◄ 03:15, 19 February 2011 (UTC)

Reference Farming Apprenticeship

I'm thinking about an optional program I'd call a Reference Farming Apprenticeship. (Inappropriate acronym notwithstanding.) As I see it, there are two keys to writing a good first stub: writing brilliant unplagiarized prose and adding reliable references. If a user hasn't learned the former in school, there's not much we can do for them; but we can help and instruct them in citing reliable references. I would propose an enhanced version of refToolbar that, given a URL belonging to certain common reference sources like The New York Times, can fill in all the other relevant fields. Due to technical limitations, this would require some work on the server side, but I think it's possible. But even the existing refToolbar, or Magnus' reference generator would be acceptable. Then point 'em to Google News (and/or some other good reference search engine), and send 'em into Category:Articles_lacking_sources to be "reference farmers". This gives new users a chance to (1) contribute positively with relative ease, and (2) see a variety of Wikipedia articles and their markup, so they can learn the only-slightly-less-essential art of Wikification.

I would even suggest a modified version of this for new users who have been blocked for a short time, say a day or a week: If they could only select the location where a reference goes and the URL it comes from - the rest being automatically filled in - they probably couldn't do too much damage and might even contribute positively. If they continued to vandalize even in this limited mode, it would be pretty obvious (as the article titles would have nothing to do with what they referenced) and their blocks could be extended.

What do you think? -- Ken_g6 (factors | composites) 18:05, 10 February 2011 (UTC)

Very interesting idea, it's definitely a big thing to help with and it's also something that new users can do pretty easily. Even if you aren't so sure about the whole "writing a whole new article thing" working on references for existing articles or working on them to expand an existing stub (can you say DYK?) can be really nice ways to 'break in' and start to understand the system. Jalexander--WMF 04:28, 11 February 2011 (UTC)
Ken_g6, one of the many projects floating around on my hard drive unfinished is a piece of software to do exactly this: to basically take URLs from widely-used reliable sources and turn them into WP citations. One of these days, I'll get around to finishing it. —Tom Morris (talk) 04:53, 11 February 2011 (UTC)

The help pages need improvement

Back in the August 2010 NYC Wiki-Conference we went over some of the FAQ and Help pages and concluded that they were in need of some serious reorganization. We started with a revised draft of Help:Contents (see Help:Contents/draft), but not much progress was able to be made.

At least, from what I got out of the NYC conference was that the Help pages suit three types of people: casual readers who do not edit, brand new editors, and editors with some experience but need specialized help with something. We came up with simple language in the draft such as "How do I use Wikipedia to find information?", "I want to change, add or remove information", etc.

As far as the other Help pages are concerned, we definitely need to "simplify" them and make the wording and going through them more digestible, especially for newcomers. I think what would go a long way toward pointing newcomers in the right direction, especially when it comes to simple stuff like simply getting around. –MuZemike 18:15, 18 February 2011 (UTC)

Aye, much of the help pages actually suck pretty badly (and are really out of date) We could get a lot payback by working on some of those. Jalexander--WMF 21:20, 18 February 2011 (UTC)

Statistics (responding from AN)

Trying to move this discussion from AN where it's getting long and a bit off topic and because it fits with our discussions some. There were some questions about whether the lack of new users actually correlates to a drop in active editors and at least for English (compared to other project and all projects combined) it appears that's the case though much slower which is interesting in itself. There are obviously legitimate questions about what the important stats are (and how to improve those). Jalexander--WMF 23:45, 6 February 2011 (UTC)
ENactiveEditors5.jpg ENactiveEditors100.jpg

Any chance you can run some charts for editors with a significantly larger number of edits? I wouldn't classify "greatter than 100 edits" as being very active, considering that one can easily make that number of edits in a day or two. Perhaps 1000, 5000, 10000 and 20000? I think that might give a better insight to whether we're keeping editors who are actually content creators and revisers. Beyond My Ken (talk) 15:39, 8 February 2011 (UTC)
Yeah I'll try and throw some of those together tonight or early tomorrow morning for you. I can give you most of the 10k ones off the top of my head. There have been VERY few who have more then 10k total edits a month (which is what these numbers all are, monthly edits) there has never been more then 10 and only like 3 months where it was 8 or 9 usually it's 1-3 or none. There were 13 months with 1 user over 25k edits but makes you wonder if that was an unmarked bot (the numbers I'm using take out the bots, though we have separate numbers for those). Mmore then 1k edits has a bit more info. Jalexander--WMF 01:54, 9 February 2011 (UTC)
Also I should note that all the numbers I'm using are article edits. Jalexander--WMF 01:57, 9 February 2011 (UTC)
Ah! I didn't realize the figures were per month, I thought they were cumulative totals. So maybe you could do 250, 500 and 1000? Beyond My Ken (talk) 12:12, 9 February 2011 (UTC)
Put some below. The 2500 a month is kind of interesting. Very very small numbers but consistently going up. I wonder if that has to do with the increased use of automated tools like huggle? That would certainly make getting that many much easier. 1000 is 'relatively' consistent but I wonder if it has similar things. The 250 has the downward trend again a bit flatter. Jalexander--WMF 00:22, 10 February 2011 (UTC)

ENactiveEditors250.jpg ENactiveEditors1000.jpg ENactiveEditors2500.jpg

As you say, quite interesting. Automated tools could have something to do with it, I suppose. At first approximation, though, I think one could say that while the number of editors overall is falling somewhat, the number of high-volume editors is steady or rising. Now if we only had a way to measure the quality of editors' productivity. Beyond My Ken (talk) 01:02, 10 February 2011 (UTC)
Perhaps that should be the rate of change of the size of the database divided by the number of active users? Presumably, bad edits get deleted, so this should quantify overall good edits per user. And by the way, are you including other relevant projects in this, such as Commons? After all, a picture is worth a thousand words. -- Ken_g6 (factors | composites) 05:32, 10 February 2011 (UTC)

Could we get a graph of the number of new accounts created each month, please. This might show how many new editors are thinking about contributing to Wikipedia. Thanks. - Hydroxonium (H3O+) 13:33, 10 February 2011 (UTC)

I should be able to get one for the morning. I was having some trouble finding the 'full' count for the month but I know it's out there (I ended up counting the December ones for below by hand). I'll find it and get it up for us to look at. Jalexander--WMF 04:37, 11 February 2011 (UTC)
By hand? Ouch! I retract the request. - Hydroxonium (H3O+) 07:07, 12 February 2011 (UTC)
Honestly the number of new accounts is a very important statistic. We're talking about new users falling off. But if we find out there are fewer new users signing up, a lot of the discussion is moot, and needs to turn to recruiting. If there's a way to get even a quick estimate it would be crucial. Shooterwalker (talk) 16:16, 20 February 2011 (UTC)

Wikipedia has pointy teeth

I'm not certain why all the previous studies have not been brought up yet. Maybe all of the stuff below is common knowledge. Then again, maybe not. There are a ton of these things all over Wikipedia and in the media and they raise the same issues over and over again. Here is a sample.

Reports and essays regarding newbies

Essays and comments from experienced Wikipedians

Articles in the media

I would encourage everybody to read through these as there is a lot of good information in them. Many also have useful suggestions. Here is a very brief overview for all those that WP:TLDR applies to.

  1. Wikipedia has too many rules
  2. People use these rules as weapons to attack other people
  3. WP:BITE is violated all the time
  4. People take the easy way out and use automated tools to interact with others, rather than spending several hours teaching somebody how to do things properly. With a simple click of a mouse button, a person can template a talk page to tell somebody they broke a rule or are bad. A small amount of effort from a hostile person can destroy hours or days worth of work from another person.
  5. This creates an environment that nurtures hostile Wikipedians and drives away nice Wikipedians.

This is just a very brief summary and leaves out a lot of important details. It really is much better if people read through all these reports. - Hydroxonium (H3O+) 19:05, 13 February 2011 (UTC)

So what should we do? Ban all those who bite the newbies? We'd virtually have no editors left. –MuZemike 21:51, 16 February 2011 (UTC)
Not at all. That's obviously untenable. I think the answer is to try to educate new editors as much as possible such that they are less likely to be bitten and better able to stand up for themselves when/if they are. I think the sort of folk who have the potential to be good editors would be amenable to that.  Cjmclark (Contact) 22:36, 16 February 2011 (UTC)
There were several things we learned from WP:NEWT, including that any attempt to measure Newby mistreatment needs to include looking at those newbies whose only edits have been deleted - by contrast I believe this study is only looking at editors who have one or more undeleted edits. Changing the warning messages that we use to tell newbies that their articles are being deleted would also help. We also need some way of identifying those speedy deletion taggers who are making an unacceptable level of mistakes, and the easiest way to do that is to get more eyes on cat:speedy and the queue of BLPprodded articles. ϢereSpielChequers 01:15, 23 February 2011 (UTC)

Moving forward

First off I want to thank each and everyone of you for all the comments on this page. They are broad reaching and incredibly interesting and show the huge amount of things we can improve on. I think we want to try and work on some of these as we move forward either as a group or individuals, a great benefit of this seems to be that we've become a good spot to discuss things like this with people who share the desire to help expand the participation on the project. Of course in the end we want to try and discuss it with those who aren't so sure about the methods either or weren't interested enough to keep following this page (I'm always a fan of a good hefty debate).

I'm probably going to send some of this via email as well later tonight to ping those who signed up but aren't coming back as often but I wanted to get this started asap so that we could get the discussion going (and maybe not even need the email/talk page notices). What do we still need to move forward?

  • I listed some ideas a bit earlier above but I'm really interested in more. I have the script basically ready to grab random samples when we're ready to go.
  • I'm probably writing up a report this weekend and Monday for my boss (User:Philippe (WMF) ) since he's coming back from vacation. I'll obviously post it for everyone on a sub page, much more interesting to see what people say and even help tweak it if you think I missed things. I'll probably be doing that about once a month though what I would love to do is post the draft so that people can comment on it and make their own thoughts known (I've always thought that's better then just posting a final thing).
  • How many users do guides want to start with? I was originally going to start off large (like 50) because I'm fairly certain a huge amount of them won't respond or won't be interested and so by the end ot the week you would end up with 5-10 or less. Of course the problem here is that sending 50 people individual emails is going to be an enormous task for 1 person and as Sage said there is also a good chance that some of them will still reply a couple times even when they aren't that interested in continuing. My thought at the moment is a bit of an adjusted version. I'm thinking giving out a batch of 10-15 users to each guide and then weekly, if the guide is ready, we can offer another group of users. Leaving the total they have up to each guide and allowing for that drop off (they can just get another collection). What do people think?
  • More letters We have some great welcome email examples above What other situations can we come up with that we need emails for?
  • Meeting?: Do we want to have a meeting on IRC or something to chat live? Either this weekend or sometime early next week? We can set up a webchat link for those who haven't used it before as well as help you set it up. Another option is chatting on something like Mumble_(software) We have a server that should be available at the foundation but I'm not sure if we would have a critical mass interested in talking live. If we do want to have a chat:
    • What times are good? - I can find a way to be here basically whenever if I know it's better for people. Though generally you can assume I'm on The US Pacific time zone (UTC -8) I may in fact be either on the east or west coast but even when I'm in Boston I essentially work on the PST time :).
    • How big do we want to make it? Do we want to publicize it on the mailing lists and elsewhere on wiki to get more people to come and talk or do we want to keep this mostly within us for now? We can always do a bigger one a bit later after we've started (trying to get more people to join perhaps).
  • Shameless plug Another great project you may be interested in looking at it the WP:CONTRIB campaign or the "Wikipedia Contribution Team". It was a project that was started up with the help of the foundation over the fundraiser but is basically totally community run now and I think could be a nice corollary with us.

Anything else!!? :)

PS: I'm listening to Mumford & Sons and especially their song The Cave while I write this. Does that say anythign? ;) Jalexander--WMF 22:30, 18 February 2011 (UTC)

10-15 people at a time sounds good. Is there going to be a datatbase pae of new users to pick from?(In case we finish our 15 and want to have even more). If we do have an IRC meeting the UTC -8 is the same for me. Sumsum2010·T·C·Review me! 05:34, 19 February 2011 (UTC)
Thoughts on the "how many users" issue -- I think a small number each day, rather than a big weekly batch, would be more manageable. 5 a day makes 35 a week, but is still a manageable task from the initial-contact perspective. 35 a week will be too many for a lot of people, depending on the attrition rate, so make it super easy for Guides to temporarily (and reversibly) withdraw from accepting new users (say, by removing their user page from a "Give me more puppies to walk" category.
On the live meeting -- great idea, smaller is probably more manageable. Being in the GMT+11 minority, I won't even attempt to suggest a convenient time, but will stick to hoping adequate notice is given so I can rearrange my nap schedule =) Katherine (talk) 08:28, 19 February 2011 (UTC)
Another thought -- do we know where new users are located? It might be nice to match up new editors and their guides geographically, or at least by similar time zone, or at least give guides the option of being matched up in that way. Katherine (talk) 08:31, 19 February 2011 (UTC)
Would Gmail (Chat) be a possible communication tool, I notice that quite a bit of people have them. ► Wireless Keyboard ◄ 15:23, 21 February 2011 (UTC)
Works for me.Sumsum2010·T·C·Review me! 04:29, 22 February 2011 (UTC)
Anyone on the east coast? ► Wireless Keyboard ◄ 21:37, 23 February 2011 (UTC)
I am most of the time ;) Though I'll be on the West coast from Wednesday next week and through the following week. I think I may just make an evening time frame Friday or something to chat especially since I think we may be able to start then. Thinking of a good time. Jalexander--WMF 02:18, 24 February 2011 (UTC)

Abhishek Singh

Dear Jalexander,

  1. I think sending even 50 welcome emails in a week is not a problem, provided we get a template. But like you suggested we should aim for moderate numbers upfront and then build upwards.
  2. Secondly can we seriously look into revamping the "Declined" template. It should use more user-friendly terms like "Feedback Provided".
  3. Thirdly, can we also collect statistics of average number of "Declines" that a new user goes through? That way we can publish the statistics to encourage re-writes from new users. This is something that Harvard Business Review tells its contributors upfront. HBR informs its contributors up fronts that each article may undergo several revisions with the editorial board after being shortlisted but before being published. So please be resilient. This gives a sense of mental-preparedness to the contributor who doesn't mind the drudgery thereafter.

For your consideration, please.
abhishek singh 23:23, 18 February 2011 (UTC)

Are fewer new editors signing up?

Is the problem that new editors are having a harder time fitting in? Or is it possible that new editors are fitting in (or not fitting in) as much as they always have been, and that the real problem is that fewer readers are becoming new editors?

Don't get me wrong. Even if we've never gotten worse at helping new editors... it couldn't hurt to get better. But if we've never gotten worse, it might indicate that we need to work harder on converting readers to editors, and that nothing we've done is particularly bad to new editors.

There's probably a mix of factors at play. It would be good to know how much the problem is lack of sign ups, and not just lack of retention. Shooterwalker (talk) 13:53, 23 February 2011 (UTC)


I joined WIKIPEDIA a couple of days ago and would cherish a mentor! Plainsman89 (talk) 13:04, 15 February 2011 (UTC)

Hi Plainsman, we have not started yet with this project, but in the mean time I am more than willing to act as your mentor. Yoenit (talk) 13:11, 15 February 2011 (UTC)
Thanks Yoenit! I saw you left a message on Plainsman's talk page as well. I know all of us will be willing to answer questions and help ig you need it Plainsman :) Jalexander--WMF 21:01, 15 February 2011 (UTC)
Yoenit's greeeting at P89's talk is great. Simple, friendly and informative. Worthy of template consideration. Buster Seven Talk 13:48, 25 February 2011 (UTC)
It is a template already: {{w-link}}, together with {{w-short}} definitely the best ones out there. Yoenit (talk) 14:01, 25 February 2011 (UTC)

Discussion hours logs

So for our first discussion hours we ended up with only a couple of us but had some interesting discussions none the less. Hopefully we'll have a couple more this evening! Jalexander--WMF 19:50, 25 February 2011 (UTC)

We had another discussion hour later in the evening which was small again but active and got some good questions answered. I'll come back and put some notes (or anyone else can!) but for now the log! Jalexander--WMF 16:39, 26 February 2011 (UTC)

Probably be a good idea to have another chat now that we've started sending out our emails. I missed the last one, but I'm reasonably free for the next 36 hours, if one happens in the near future. Katherine (talk) 01:34, 27 February 2011 (UTC)

My initial experience / How it relates today

Recalling my initial experience compared against today:

1) Treatment of new arrivals: Like thousands of others, I received the {{welcomeg}} template meant as a greeting and a "how to." If that had been my only message, I would likely not still be here. The welcome templates are a far-too-easy, far-too-impersonal solution. I feel a greater appreciation from a WalMart greeter than those templates provide. That's not saying the community isn't friendly or civil once you look more closely, but the example given doesn't say much for a first impression for the community. For reasons I don't recall, I decided to jump into a large c/e early on, and from that I received a personal thank you from one of the regular editors of that article as well as my username noted in another article. It is that kindness which resulted in my wish to continue. It's not just about getting people here; it's about keeping them here.

2) Newness is wearing off: Like all new things, you always have more interest initially than later on, particularly when it's free. Scenario: it's 2007, I have not yet joined, and I come upon an up-and-coming article on a popular subject. I join because I want to expand on it feeling like I am doing some good. Same scenario: it is instead 2011, I have not yet joined, but I go to the same article. I have no reason to join now because everything that I would have felt I could contribute has already been added. I might as well edit anon on the other minor things that I see from time to time. Although I have never been a large editor, I too feel less reason to edit today than I did my first year here. In other words, one of WP's greatest benefits—an ever-improving collection of data—can also be seen as one of its greatest weaknesses.

3) With growth comes intimidation: I would gather that many newer members are no longer feeling like equals as much as they did back when the site was fresh. "Be Bold" only goes so far when you see a full-fledged article and think you will screw things up by making more than insignificant changes to it. Or worse, taking the time to make more than insignificant changes only to have it reverted, discouraging future edits. Going back to the {{welcomeg}} template, it looks like the table of contents for a large instruction manual. Who wants to read that? In a similar respect, when you join a new forum, you are starting at 0 posts seeing all of these other members with 5,000 posts and high reputation numbers which can make you feel very small. If you do feel that way, then it takes time to build up from that, but by then you are no longer a new user and are just someone that doesn't participate much which removes you from the "new editor" numbers in the statics.

I'm not sure if any of this can help spark other's thoughts, but I hope that no one starved to death while reading this. — CobraWiki ( jabber | stuff ) 21:23, 11 February 2011 (UTC)

It's ok, the cat was on my lap while I was reading so I didn't starve. Totnesmartin (talk) 22:21, 11 February 2011 (UTC)
I absolutely hated getting a welcome template on my talk page. I had already followed the Wikipedia project for years, and had been doing a handful of edits each month as an IP. When I got welcomed the first thing I did was blank most of it. It was like a badge of newbieness that I absolutely did not want. Gigs (talk) 03:08, 12 February 2011 (UTC)
While I, in contrast, was innocent enough to be gratified by a greeting. I realise now, of course, that it was pretty pro forma, and that in some cases welcome templates might even be being used as a passive-aggressive "smile, you're on camera"-style warning, but at the time it was quite motivating. If anybody is serious about improving reception/retention of new editors, it's probably worth finding out empirically which reaction is more typical. --Andreas Philopater (talk) 13:49, 12 February 2011 (UTC)

Cobra, have you looked over the other templates, to see if some others feel less impersonal? I've recently been using {{Welcomeshort}}, which has five sentences, a pointer to WP:Introduction, and no graphics. I also use {{MedWelcome}}, but it's only useful when its specialized information is directly relevant. WhatamIdoing (talk) 21:12, 14 February 2011 (UTC)

I haven't looked them over lately, but I am unfortunately viewing this, now, from the point of view of a no-longer new or novice user. My thoughts could be a bit more clouded by "seeing them for what they are" so-to-speak. Getting money from my ATM is far less personal than going into the bank and getting the money even though the ATM will display "Thank you. Have a nice day." Walking to a chatroom and have a bot quickly display everything about the room and giving you a welcome isn't at all the same as a member saying hello to you. I guess if you never used a computer before and the ATM says "thank you" then you may be more prone to accept it as a heart-felt response to your actions, but I would suspect, generally speaking that people who receive a template welcome message see it for exactly what the name describes: a template, rubber-stamp message that everyone gets. It's helpful, but not personal. Just my view, though. — CobraWiki ( jabber | stuff ) 02:49, 23 February 2011 (UTC)
I've been having a look over the welcome templates this afternoon, thinking about what's been said here. So, my impressions of the existing widely-used templates: Massive walls of links to stuff that is incomprehensible to a brand-new user. Bossy notes warning them to sign on talk pages before they've even had a chance not to. Links to the gigantic slab of bedtime reading that is WP:MOS. That godawful phrase "Thank you for your contributions" -- bland, generic, meaningless, and often blatantly untrue (we give the same acknowledgement to a nice bit of copy-editing or reference-adding as we do to assorted bits of blathering nonsense) -- and it sounds like the Wikipedia equivalent of "No offence, but <insert offensive comment here>". Use of "we" (as in, "We hope you decide to stay")... who is this we? The welcomer isn't an appointed spokesperson for all of Wikipedia, nor are they in some separate group of Others. Bits is bits, and editors is editors.
I've been working on some messages of my own to copy-paste-tweak in certain situations, but otherwise I am hereby swearing off talk page templates. For at least a week. I'll stick to short notes from an actual human, and some good old-fashioned engagement. Who knows, it might be fun! Katherine (talk) 04:03, 23 February 2011 (UTC)
A template is obviously just that - a template. But the fact that a human user can be bothered to post one is the key thing (or was for me). A nicer template posted by a bot, which some seem to be suggesting, would, I think, be entirely counter-productive.--Andreas Philopater (talk) 10:57, 23 February 2011 (UTC)
I don't think I agree with that. The first time I saw {{sofixit}}, I thought, "Wow, she went to a lot of trouble to write a really detailed message to encourage that new editor to be bold!" Now, of course, I don't even properly read them, just glance long enough to identify which template was chosen. WhatamIdoing (talk) 16:57, 23 February 2011 (UTC)
I felt pretty much the same in regard to the welcomes - I found a lovely one on the Spanish Wikipedia - I put in English versions of the text and now I use this as my welcoming message.
User:Chaosdruid/welcome Chaosdruid (talk) 02:01, 28 February 2011 (UTC)