Toolserver CleanupListing Migration
- Volunteered to migrate over at Wikipedia:Bot requests/Archive 60#Migrate CleanupListing tool from Toolserver to Wikimedia Labs --Bamyers99 (talk) 20:33, 15 May 2014 (UTC)
- Sorry for not being active and thank you very much for doing this. I'll try to be at least somewhat active in the future, so if you need some help (e.g. with maintaining the list of projects, thought that's much easier now that it's just a wiki page), I could help. User<Svick>.Talk(); 12:25, 29 June 2014 (UTC)
- Also, I have now added redirects from my old cleanup listings to your new ones, just so you know. User<Svick>.Talk(); 15:32, 29 June 2014 (UTC)
Hello, would you be able to help with creating a simple advert for the Today's featured article requests page (WP:TFAR)? I can write a pretty decent 1,200 character blurb but I tend to leave image work to those that understand them! All I'd like it to say is something along the lines of - Do you know of a featured article that ought to be on the main page? Then nominate it at the requests page. Any help gratefully received - no rush - and if you can't fit this into your busy schedule, don't worry. Yours, BencherliteTalk 14:32, 20 May 2014 (UTC)
The latest update to the database report was generated yesterday, but it contains redirects such as File:DI.JPG.jpg which were deleted back in 2012. Could you check what's wrong? --Stefan2 (talk) 13:21, 2 June 2014 (UTC)
- @Stefan2: I think that's some issue in the Toolserver database. I plan to move the report generating to Tool Labs soon (since Toolserver is shutting down tomorrow). Hopefully, that will solve this issue too. If you find another error in the next report, let me know and I'll have a more thorough look. User<Svick>.Talk(); 13:13, 29 June 2014 (UTC)
Hi. I just discovered <http://itn.svick.org>. The output looks pretty good. I'm going to start using it for an IRC bot or two. If you can share the underlying source code at some point, that'd be nifty. --MZMcBride (talk) 01:06, 10 June 2014 (UTC)
- @MZMcBride: The C# source was always available at https://github.com/svick/Wikipedia-In-the-news-Feed/. Most of the logic is in Models/ItnFeed.cs.
- I realize there was no way for you to find the repo, so I have added a link to it to the feed source; hopefully that will be enough for others to discover it.
- User<Svick>.Talk(); 13:48, 29 June 2014 (UTC)
- Ah, beautiful. Thanks! :-)
- I should have mentioned, I found the feed by some Google search (I can't remember which) leading me to Wikipedia talk:Syndication#RSS/Twitter of In The News. Of course I now notice that the feed is linked from Wikipedia:In the news (at the bottom of the page). I swear I checked the talk page... I should add a link there as well. --MZMcBride (talk) 14:09, 29 June 2014 (UTC)
digital object identifiers
I'm interested in digital object identifiers, because I think they can be used to materially improve the ease of adding references.
I see you write some of the Universally unique identifier article, so I thought I would ask you some questions, although I see you haven't worked on it recently.
My goal is to persuade our main reference sources, such as newspapers and magazines, to use a DOI for each article, which could then be used to generate a reference, without having to fill out any other field.
This requires two things:
- Mediawiki ability to process the DOI (in progress)
- Media adoption of a convention, assigning a DOI to an article, and storing the citaiton fields in an accessible database
Obviously, some of our reference sources already use a DOI, whether doi, PMID or ISBN, but that still leaves many sources without DOIs. I see five hurdles to convincing sources to implement a convention:
- No value, if no one will utilize them
- The cost of registration
- The cost (IT) of building a system to assign values to articles
- The soft cost of writers/editors having to do something to add the DOI to articles
We cannot do much about items 1,4 and 5 If Wikimedia will use the entries, we can address the second item.
My question to you relates to the third item. Am I correct (see this site) that UUID are free? This might help, if some potential users balk at the cost of registration. (Of course, there is no free lunch. With non-free options such as doi, I assume registration includes addition of the entry to a databasse. With a UUID, your own shop has to handle this, I think.)
- @Sphilbrick: Yes, UUIDs are free, everyone can generate nigh unlimited number of them from their computer.
- Theoretically, I think a better option would be to use the URL of the article as URI, though I'm not sure that's feasible, since some sites like to break URLs when redesigning.
- But if you want to promote something like this that doesn't have any form of central database yet (like DOI does), I think the question of what ID to use doesn't really matter, there are going to many technical decisions and issues that are going to be much more important.
- Also, have you considered other approaches that already exist, like RDFa or hNews?
- User<Svick>.Talk(); 14:09, 29 June 2014 (UTC)
- Thanks for responding. I'll partially respond now, then come back. I do not want to use the url as a URI. My main area of interest is women's basketball, and I have a few hundred links in to various pages on their site, because they record the results of all-star games, and many awards. A few years ago, they reorganized their site, and all my links went dead. I had 50 or so at the time, so it was a pain, but not a big pain. I emailed them last week on another topic, and they mentioned in passing that they were planning a new site reorg. I didn't know whether it would break the links, but I checked and see that ::I have 400-500 links. I just checked minutes ago, and see they have updated their site, and yes, links such as:
- are now dead.
- An anecdote to be sure, but I believe I recall other situations where a major newspaper revamped its site, and all the old links went dead. My goal is to avoid that problem.