User talk:Jtmorgan: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
Nanoatzin (talk | contribs)
→‎Teahouse welcome after 4 vandalism warnings: Question about how to deal with racist content including question about how to avoid having account canceled in the process.
Line 202: Line 202:


::Please ping me if you see the bot failing to recognize these warnings again, or if it does anything else weird [[User:Jmorgan (WMF)|Jmorgan (WMF)]] ([[User talk:Jmorgan (WMF)|talk]]) 01:46, 15 September 2012 (UTC)
::Please ping me if you see the bot failing to recognize these warnings again, or if it does anything else weird [[User:Jmorgan (WMF)|Jmorgan (WMF)]] ([[User talk:Jmorgan (WMF)|talk]]) 01:46, 15 September 2012 (UTC)

:::Hello. I seem to have been invited to this location after attempting to correct several pages that have offensive content. The [[Self-deporation]] article says that any immigrant that crosses a border leaving the United States has "self deported." [http://www.google.com/?q=self+deport+23.2+site:www.uscis.gov That deviates from the text of the actual law in a way that that I find to be highly offensive (click here, search, then click the top link for the law)]. Crossing the border "while Mexican" is not "self deportation". The [[Self-deporation]] article is firmly rooted in [[South park]] intellectual territory for that reason. This is the simplest and most obvious example of racist content because you can directly compare the two-word title of the article to the two-word legal definition of "self deport" and see the content is intellectually compromised because it fails to mention deportation requires a crime (i.e.: being Mexican or Canadian is not equal to being criminal). The legal difference between self deportation and forcible deportation is determined by whether you leave the country voluntarily or after you are arrested following a deportation order. All deportation involves a crime of some sort. The article deviates from that definition in an offensive way by implying that certain races or nationalities are illegal - i.e.: you are not subject to deportation from the US just because you are "being Mexican" (nor "being Nigerian", nor "being Chinese", ...). Race is not a deportable crime (self or otherwise). Referring me to this location makes it sound a little bit like this racist content might have been intentional, which seems hard to believe. This kind of thing is scattered across dozens of articles. I don't want to have my account cancelled just because I've tried to point that out or fix it. Can anyone make a recommendation about how to deal with this kind of thing in a more constructive way? Is racist content protected by Wikipedia policies? Perhaps I should elevate my concern? Suggestions? Opinions? Thank you for your time and effort. I hope this finds everyone well. Best regards.[[User:Nanoatzin|Nanoatzin]] ([[User talk:Nanoatzin|talk]]) 04:48, 15 September 2012 (UTC)


== ''The Tea Leaf'' - Issue Six ==
== ''The Tea Leaf'' - Issue Six ==

Revision as of 04:48, 15 September 2012


Welcome to my talk page! Drop me a line.

  • If you're here because you got an invitation to visit something called the Teahouse from someone called HostBot: double welcome! I run HostBot; let me know if there's anything I can do to help.
  • If you're here because you have a question about HostBot, or want to report a bug: please leave a detailed message and I'll respond as quickly as possible.

Thank you so much

IRC HostBot

Hey, J-Mo, I had an idea for HostBot, and I wanted to run it by you first. Would it be possible to have the chatbot read and interpret a leading comment in a Teahouse question and use that to ping users in chat? Like, I would put <!-- Writ_Keeper --> at the top of the question's section, and then, when the bot notifies IRC about changes to that section, it would ping Writ_Keeper. How doable is that? I know you can get the source of a section with an "action=raw" query to index.php (example), but I don't know how much hostbot knows about what's been edited. Writ Keeper 13:43, 16 August 2012 (UTC)[reply]

This idea sounds interesting, but I'm not sure I see the big picture yet. Also, I've been wanting to chat with you about TH dev stuff for a while now. Want to talk about it on IRC sometime tomorrow? I'm afk most of the day today. - J-Mo Talk to Me Email Me 16:17, 16 August 2012 (UTC)[reply]
Sure. Writ Keeper 16:26, 16 August 2012 (UTC)[reply]

Add teahouse to welcome template?

As you were involved in a previous discussion regarding this issue, I am informing you of a new discussion proposing that the Teahouse be linked from the Welcome template(s). The discussion can be found here. KillerChihuahua?!? 23:54, 16 August 2012 (UTC)[reply]

Thanks

Thanks for the invite. Val sarah25 (talk) 20:33, 19 August 2012 (UTC)[reply]

Thanks

thank you for the info. :)

Val sarah25 (talk) 15:35, 21 August 2012 (UTC)[reply]

Teahouse invite

Hi Jtmorgan,

I don't think this invite by HostBot was appropriate. I know that with bots there is always going to be the possibility of this sort of thing but I thought I would bring it to your attention. Maybe there is a way to tweak the bot's conditions for posting the message. Maybe after the new user has made a certain number of edits to article space and is a certain age, the bot could check with WP:WikiTrust before posting the invite.

Yaris678 (talk) 09:19, 22 August 2012 (UTC)[reply]

An alternative/additional suggestion is to check the user's page for text that is used in templates that relate to vandalism, removing notices and sock puppetry.
Yaris678 (talk) 09:35, 22 August 2012 (UTC)[reply]
Hi Yaris. Thanks for the feedback. I see your point. I'm trying to tread lightly, when it comes to deciding which users to invite and which ones to exclude. Currently, hostbot invites any editor who meet certain minimal requirements, such as having joined within the past 24 hours, made at least 10 edits, and not been blocked. Vandalism is a lot more ambiguous than blocking, and unfortunately (as you know) a bot can't make the subjective judgement a human can and decide whether someone's editing is a problem or not. A lot of good faith newcomers get hit with warning templates for things that were not intentionally harmful, and a lot of folks who will go on to become productive Wikipedians make test edits in their first few sessions that look like vandalism, but are really just simple goofing around. Since TH is inclusive, I don't want to blacklist editors just because someone (or some other bot) decided their edits were vandalism.
Sockpuppetry seems qualitatively different to me, though. My impression is that someone is not accused of sockpuppetry lightly; a fair bit of research goes into it. Is that correct? So I can see a case for excluding editors from automated invites if they have, say, a link to WP:SOCK on their page.
I don't know enough about the algorithm WikiTrust uses to know whether it would be useful for assessing the quality of very new editors. If someone only has 10 edits and has been around for less than a day, I would expect the algorithm to rate them with less confidence. In fact, I wouldn't be surprised if it showed some bias against new editors, since it's meant to be a reader-facing tool, and in general recently-added content by recently-joined editors is less vetted. But like I said, I don't know the code... or how to integrate with it, for that matter :) - J-Mo Talk to Me Email Me 21:44, 22 August 2012 (UTC)[reply]
Hi Jtmorgan,
Thank you for your reply.
As far as I can tell from my use of the WikiTrust queue for WP:STiki, WikiTrust finds it difficult to distinguish between a brand new user and a user who has had most of his/her edits reverted... so maybe that was a red herring by me.
I agree with your point about sock puppetry being very different so excluding people with such accusations seems like a good first step.
For vandalism warnings, maybe the answer is to exclude such people from the bot message but keep them on a list so that human users can look at the profile and determine if they should be invited. This approach may give a bonus - someone who has been warned by automated or semi-automated messages may appreciate the human touch in an invite.
Yaris678 (talk) 07:17, 24 August 2012 (UTC)[reply]
Ooh, that's an interesting idea. I can definitely see that folks who have received lots of warnings might be good candidates for 'triage' or personalized follow-up. - J-Mo Talk to Me Email Me 07:52, 27 August 2012 (UTC)[reply]
Another example: User talk:EugeneDiamond. :)
Of course, maybe it's a good thing to invite those users in particular, to nudge them towards more constructive edits.
Amalthea 12:13, 29 August 2012 (UTC)[reply]
Amalthea, I agree. Thanks for the heads-up. And if you see any instances during the course of your work in which being invited to the Teahouse seems to have encouraged bad behavior, on the part of a new editor, definitely let me know. - J-Mo Talk to Me Email Me 18:56, 2 September 2012 (UTC)[reply]
And an example of a user blocked more than 24 hours prior, getting an invite.[1] I'm not quite sure how this one got in under the radar, but you might want to see if there was something unusual about the situation that caused an unexpected result. Thanks for looking into this — and for your work in this area. Risker (talk) 06:39, 1 September 2012 (UTC)[reply]
Yeah, this was caused by a bug in the code. Thanks for bringing it to my attention! I've now fixed that particular error, but I always appreciate notifications like this. Cheers, - J-Mo Talk to Me Email Me 18:56, 2 September 2012 (UTC)[reply]
Probably the same thing with User talk:Bertsbigbrother. Although it was just barely 24 hours, so maybe it was a simple time difference in this case. Just noting in case it's helpful :) -- Quiddity (talk) 18:58, 2 September 2012 (UTC)[reply]
Very helpful :) This does look to be the same case: my script wasn't checking far enough back in the logs, so I failed to detect when some users were blocked. Thanks & keep 'em coming! - J-Mo Talk to Me Email Me 20:25, 2 September 2012 (UTC)[reply]

I'm a real person

Yes, I'm a real person. Just really tired and need some sleep. Please email me at dnelson30@gatech.edu *sleep is good*

MrDrBiochem 05:13, 23 August 2012 (UTC) — Preceding unsigned comment added by MrDrBiochem (talkcontribs)

I've taken the liberty of moving this section to the bottom of the page and adding a header. Hope you don't mind. Graham87 06:01, 23 August 2012 (UTC)[reply]


Hostbot invitation template tweak

Hi,

Just a quick note to tell you I have made a minor tweak to the Teahouse invitation template, so just keep an eye on it, and let me know if I messed it up.

Thanks,

Mdann52 (talk) 14:44, 25 August 2012 (UTC)[reply]

Mail!

Hello, Jtmorgan. Please check your email; you've got mail!
It may take a few minutes from the time the email is sent for it to show up in your inbox. You can remove this notice at any time by removing the {{You've got mail}} or {{ygm}} template.

Ocaasi t | c 14:57, 25 August 2012 (UTC)[reply]

Thank you for the Invite

Thank you for the Teahouse Invite. Im currently making all articles on the Colombo crime family, Genovese crime family, Gambino crime family, Bonanno crime family and the Lucchese crime family. It look's like a whole page full of blood, It will take some time and I will have to decline your Invite. Thank you very much though! :)

King Luciano (talk) 10:57, 26 August 2012 (UTC)[reply]

Don't forget to make your new Host profile!

Hi J-mo! Thanks for being a host at the Teahouse! We're working on the Host lounge renovations and we've created a new way for hosts to become hosts. Please take a few minutes and test it out here, by creating your new host profile. It's also a good excuse to update your image, quote, and information about yourself :) You can join in on the host discussion about the new feature here. See you at the Teahouse! SarahStierch (talk) 19:43, 27 August 2012 (UTC)[reply]

Articles you might like to edit, from SuggestBot

SuggestBot predicts that you will enjoy editing some of these articles. Have fun!

Stubs
Eodelphis
Druglikeness
Didelphodon
Femtochemistry
Phylogenetic bracketing
Plagiomene
Christian Samuel Weiss
Docodon
Lynn Steen
Journal of the Royal Society Interface
Gottlieb Conrad Christian Storr
Journal of Chemical Information and Modeling
Kogaionidae
Superferromagnetism
Alopecocyon
Priscileo
Northern treeshrew
Perutherium
Lower critical solution temperature
Cleanup
Discrimination testing
TRIUMF
Institute of Chemical Technology
Merge
Cat intelligence
Palynology
Combustion
Add Sources
Manmin Central Church
Isotopic labeling
Dasyuromorphia
Wikify
Contextual cueing
Irina Posnova
Iraqi Light Armored Vehicle
Expand
Vivipary
Carbohydrate synthesis
Index of geology articles

SuggestBot picks articles in a number of ways based on other articles you've edited, including straight text similarity, following wikilinks, and matching your editing patterns against those of other Wikipedians. It tries to recommend only articles that other Wikipedians have marked as needing work. Your contributions make Wikipedia better — thanks for helping.

If you have feedback on how to make SuggestBot better, please tell me on SuggestBot's talk page. Thanks from Nettrom (talk), SuggestBot's caretaker. -- SuggestBot (talk) 17:06, 31 August 2012 (UTC)[reply]

Teahouse welcome after 4 vandalism warnings

See User talk:Wedoeditingforfun. Any chance that Hostbot can be revised so it doesn't do this? Dougweller (talk) 08:16, 5 September 2012 (UTC)[reply]

It's not a matter of whether it can be revised, it's an issue of why, and how much. Should we not invite someone who has three vandalism warnings? two? one? The Teahouse is meant to be inclusive by default, so having committed some act of vandalism should not necessarily be grounds for exclusion unless we're reasonably sure that the act of inviting this person will provoke more vandalism. Lots of editors start out by experimenting, goofing around, getting into shouting matches and even making intentionally bad edits. Some of these people still grow up to become productive Wikipedians.
I'm not saying we can't draw a line for exclusion, I just don't think that line should be must not have received any warnings, and I don't yet know of any principled way to justify a particular numeric cutoff. Do you have any suggestions? I already have in my queue to revise the code to exclude accused sockpuppets (see above). That exclusion makes sense to me. But I want to have a conversation about additional exclusion criteria, one which weighs the foreseeable negative impacts of inclusivity/exclusivity, before I shut the gates. Jmorgan (WMF) (talk) 00:17, 6 September 2012 (UTC)[reply]
Any criteria will be subjective. But a final warning, followed by a Teahouse invitation, seems a bit bizarre. What message does that send? Isn't there any community input into this? Dougweller (talk) 12:58, 6 September 2012 (UTC)[reply]
There was a little bit of discussion (I guess you could call it "internal" discussion) in the "Hostbot invites" section on the Teahouse talk page a while ago. While I agree with you that it's pretty weird, and wouldn't be my choice, I don't (and didn't at the time of the discussion) think it was a big deal. I don't think it's really useful to invite them, but I don't think it really does any harm, either. Purely as information: User talk:Pinkydsos is another weird candidate for an invite. Writ Keeper 13:21, 6 September 2012 (UTC)[reply]
What do you think about 'final warning' as an exclusion criterion? Are final warnings always delivered by people, rather than bots? If this seems like a good place to draw the line, the easiest way to implement it on the technical side, I believe, would be to look for a link to Image:Stop hand nuvola.svg on the userpage before inviting. Jmorgan (WMF) (talk) 18:35, 6 September 2012 (UTC)[reply]
There are some single issue warnings like {{Uw-copyright}} that have that image so I don't think that would be the best. It might be necessary to go to Wikipedia:Template messages/User talk namespace#Multi-level templates and have it search for the text of the Level 4 and Level 4-im warnings listed there. Ryan Vesey 18:41, 6 September 2012 (UTC)[reply]
Cluebot goes from 1 to AIV. Writ Keeper 18:45, 6 September 2012 (UTC)[reply]
The reliance on links is strictly a practical consideration, btw. It's quicker to query the links table on the database than to parse the text of the page, looking for strings associated with particular warning templates. It's also much easier to code :) But I'll try to find a workable solution for getting final-warning exclusions implemented within the next week or so, and reply here. Jmorgan (WMF) (talk) 00:20, 7 September 2012 (UTC)[reply]
Bots and tools typically look into the page source and scan for the comment left by the substituted warning template, e.g. /<!--\s*Template:uw-\w*4(?:im)?\s*-->/ or some such. Amalthea 11:26, 7 September 2012 (UTC)[reply]

well, if all the other bots are doing it... :) Okay, I'll update HostBot to check for {{subst:uw-vandalism4}} in the page source. Jmorgan (WMF) (talk) 21:09, 7 September 2012 (UTC)[reply]

Thanks everyone. Dougweller (talk) 05:07, 10 September 2012 (UTC)[reply]
I've updated the HostBot invites script so that it will skip users if the following strings of characters (related to sockpuppetry or serious, sustained vandalism) appear on their talk pages at the time of invite. Let me know if there are any other strings I need to add (I'm not sure if I've accounted for all the L4 warning templates, for instance):
  1. 'uw-vandalism4'
  2. 'uw-socksuspect'
  3. 'Socksuspectnotice'
  4. 'Uw-socksuspect'
  5. 'sockpuppetry'
Please ping me if you see the bot failing to recognize these warnings again, or if it does anything else weird Jmorgan (WMF) (talk) 01:46, 15 September 2012 (UTC)[reply]
Hello. I seem to have been invited to this location after attempting to correct several pages that have offensive content. The Self-deporation article says that any immigrant that crosses a border leaving the United States has "self deported." That deviates from the text of the actual law in a way that that I find to be highly offensive (click here, search, then click the top link for the law). Crossing the border "while Mexican" is not "self deportation". The Self-deporation article is firmly rooted in South park intellectual territory for that reason. This is the simplest and most obvious example of racist content because you can directly compare the two-word title of the article to the two-word legal definition of "self deport" and see the content is intellectually compromised because it fails to mention deportation requires a crime (i.e.: being Mexican or Canadian is not equal to being criminal). The legal difference between self deportation and forcible deportation is determined by whether you leave the country voluntarily or after you are arrested following a deportation order. All deportation involves a crime of some sort. The article deviates from that definition in an offensive way by implying that certain races or nationalities are illegal - i.e.: you are not subject to deportation from the US just because you are "being Mexican" (nor "being Nigerian", nor "being Chinese", ...). Race is not a deportable crime (self or otherwise). Referring me to this location makes it sound a little bit like this racist content might have been intentional, which seems hard to believe. This kind of thing is scattered across dozens of articles. I don't want to have my account cancelled just because I've tried to point that out or fix it. Can anyone make a recommendation about how to deal with this kind of thing in a more constructive way? Is racist content protected by Wikipedia policies? Perhaps I should elevate my concern? Suggestions? Opinions? Thank you for your time and effort. I hope this finds everyone well. Best regards.Nanoatzin (talk) 04:48, 15 September 2012 (UTC)[reply]

The Tea Leaf - Issue Six

Hi! Welcome to the sixth edition of The Tea Leaf, the official newsletter of the Teahouse!

  • Teahouse serves over 700 new editors in six months on Wikipedia! Since February 27, 741 new editors have participated at the Teahouse. The Q&A board and the guest intro pages are more active than ever.
A lovely little teahouse nestled in Germany from Wiki Loves Monuments
  • Automatic invites are doing the trick: 50% more new editors visiting each week. Ever since HostBot's automated invite trial phase began we've seen a boost in new editor participation. Automating a baseline set of invitations also allows Teahouse hosts to focus on serving hot cups of help to guests, instead of spending countless hours inviting.
  • Guests to the Teahouse continue to edit more & interact more with other community members than non-Teahouse guests according to six month metrics. Teahouse guests make more than twice the article edits and edit more talk pages than other new editors.
  • New host process implemented which encourages anyone to get started as a Teahouse host in a few easy steps. Stop by the hosts page and become a Teahouse host today!
  • Host lounge renovations nearing completion. Working closely with Teahouse hosts, we've made some major renovations to the Teahouse Host Lounge - the main hangout and resource space for hosts. Learn more about the improvements here.

As always, thanks for supporting the Teahouse project! Stop by and visit us today!

You are receiving The Tea Leaf after expressing interest or participating in the Teahouse! To remove yourself from receiving future newsletters, please remove your username here. EdwardsBot (talk) 00:08, 6 September 2012 (UTC)[reply]

So today I decided to try and clean out some of the non-free content backlog, and stumbled across this template.

This series of changes you made back in February made it so that clicking on the image takes you to the editor's userpage, but it also removed any required attribution for using the images (such as that required by File:Telefon BW 2012-02-18 13-44-32.JPG and others which are common occurrences). VernoWhitney (talk) 23:27, 10 September 2012 (UTC)[reply]

Ah yes, I've got a fix for this. We can implement the attribution as a tooltip, like we did with the Teahouse logo. Gimme a few days to put that into place. Cheers, Jmorgan (WMF) (talk) 17:23, 11 September 2012 (UTC)[reply]

User:HostBot and substitution

While I do like the idea of it substituting the template, there is the problem that if a protection template on the invite template was not noincluded, it will include the protection template. Around 1,200 pages were removed Category:Wikipedia pages with incorrect protection templates when I noincluded the tag(s). Another 190 to 200-ish are still there, (at the time of posting this) due to substituting. Perhaps we could either not use protection templates on that page at all, or just try using my sync-pp template on it, to prevent accidental inclusion if it ever does happen to not be in noinclude tags. (the sync-pp template of mine is optionally substitutable. usually substituted on article space, it tends to not be when I use it in my userpages, so that it will automatically change the template used.) Just wanted to put that out there.

The substitution of the protection templates is no fault of your own, or your bots, mind you. It was just because there weren't noinclude tags. So if it seems I am a bit mad at you here, I'm really not. I just wanted to see what you would think about it. :) LikeLakers2 (talk | Sign my guestbook!) 19:16, 11 September 2012 (UTC)[reply]

Sorry for making a mess, LikeLakers2, and no offense taken at your tone :) I see you added the necessary noinclude tags around pp-move-indef, and added a parameter. Is the edit you made a good long-term fix, in your view? The page is unlikely to be changed by others, so I think the noincludes are stable where they are for now. However, I'm open to using your custom protection template, if you think that's a better long-term solution. Thanks for letting me know, Jmorgan (WMF) (talk) 01:41, 15 September 2012 (UTC)[reply]

HostBot detecting users already know about the Teahouse

Hi. I just spotted this invite. It's obvious from the previous posts (to a human) that the user has already been to the Teahouse and therefore the invite is redundant. However, I wonder if there's some way the bot could check a few things before posting, e.g.

What do you think? (I'm not at all conversant with bots, so haven't even attempted to investigate how HostBot works.) Thanks. -- Trevj (talk) 12:33, 14 September 2012 (UTC)[reply]

hur. I've still got a lot to learn about bots, too! I already had things set up so that HostBot checked for links to Teahouse, and skipped invites to those who had those links on their page. However, it was only looking for links to the Teahouse main page, and the talkback links point to a subpage. So I made the bot a little smarter. Now it just looks for the word 'Teahouse' on the talk page before deciding to post. And I do mean I just fixed it... like, in the last 15 minutes. Looking at this evening's invites (which I ran early), one of those sneaky talkback links slipped by there, too. Thanks for the heads-up! Jmorgan (WMF) (talk) 01:35, 15 September 2012 (UTC)[reply]