User:Waldir

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search

[edit]

I'm User:Waldir on several Wikimedia projects and I started following explainxkcd.com back when it was a blog, where I commented as "Waldir". Due to my Wikimedia background, I was one of several commenters who insisted in turning the blog into a wiki. The move was eventually made, and was announced on the blog on 31 July 2012. Two days later, I registered my wiki account, and since then I've made 5,883 edits (and counting).

My favorite comics:

explain xkcd stats:

Fun fact: Appropriately (considering the tagline), romance, math and language are the three most populated categories within the Comics by topic‏‎ category; in contrast, sarcasm is way below! That makes me happy. :)

To do[edit]

sysop one-time tasks
non-sysop one-time tasks
meta content
  • Topics to cover/mention: our twitter account the xkcd irc channel (and its wiki), the blag, the forum, other sites explaining xkcd ([1], [2], [3], [4], maybe invite members+content of the other wikis in once we're established?), etc.
  • make sure the contents of the blog posts in the wikipedia category have all been included in this wiki's pages. Also make sure all such entries are properly categorized under Category:Wikipedia.
bot edits
  • rename comic pages to "number: title" (on hold until consensus in page title format is reached)
  • rename section "Description" to "Explanation"
  • add {{Comic discussion}} to all comic pages
  • remove unused imagesize= parameter from {{comic}} (it's on the documentation in case anyone needs it)
  • add redirects from 1.png, etc, to the actual filenames
  • Add links to original comics from xkcd (and hotlink to images). Also, format image pages in standard fashion (comic images, characters...)
server-side
  • update mediawiki. The latest update was to 1.19.x (LTS), but it is only supported for a few more months.
    • after version 1.20 the PAGESINCAT magic word accepts parameters
  • Extension:Comments -- Davidy22 says it introduces quite a bit of meta-pages, being generally messy. Maybe a different solution would be a better choice.
  • Extension:Configure -- need to test it locally first, clean up any rough edges and send them upstream.
  • Try to get a rendered feed of Special:NewPages, instead of [5], filter it to show only comics, and add RSS link on the sidebar.

Proposed changes to page naming scheme[edit]

  • Set up namespaces, e.g. Comic:, Topic:, xkcd:, Meta: (replacing "explain xkcd")
    Discussion at explain xkcd:Community portal/Coordination#Page names.
    Alternative: only namespacify non-comic pages, that way the main namespace is guaranteed to be only comics (and pages in main are already content pages by default -- not that it would be hard to set others up as content namespaces too, since we'd need to edit the server config file to add new namespaces anyway)
    Pros
    • Special:NewPages gets filterable by namespace, so we could get only new comics (not sure this is really that useful, though: they have all been created and new ones are now put in place as soon as Randall publishes xkcd, so there's not really a need for a feed / notification system)
    • Special:Random/Comic becomes possible (currently all sorts of pages can be returned. True, most pages are in the comic namespace, but occasionally people will come across a meta page or something.) In fact, we'd have other specialized "Random X" too (random character, random topic...), as well as the all-inclusive Special:Random.
    • We could embed Special:Allpages/Meta, etc., to get the list of all items in a given namespace, even if they haven't been categorized yet
    Cons
    •  ?
  • Rename comic pages to use numbers in parenthesis after the name --> "Comic title (number)", e.g. Michael Phelps (1092)
    Having the number at the start of the page name was meant to automatically sort the page in categories, but doesn't really work since the numbers aren't zero-padded. That was kind of a hack anyway. MediaWiki supports category sort keys natively, so we should be taking advantage of them (e.g. by adding SORTKEY to the {{comic}} template) rather than relying on a specific page title format to achieve the same effect.
    Pros
    1. Using colons in page names is kind of ambiguous, since that's how namespaces are identified.
    2. This would allow the use of the pipe trick when linking to a comic, since content in parenthesis is automatically stripped out: [[Michael Phelps (1092)|]] results in Michael Phelps
    Cons
    1. We can't have auto-complete in the search box by typing just the comic number (and people coming from xkcd will likely use the number for search, not the comic title).
      • Alternatively, we could use the opposite format: 1092 (Michael Phelps), which would still preserve the pros, arguably even better matching the label people tend to choose when aliasing a comic link (i.e., the comic number, rather than the comic title).

Image pages as the comic pages[edit]

Even nicer would be to merge all comic pages with the image pages, preferably renaming the "File:" namespace to "Comic:". This can probably be done with Extension:SpecialNamespaces, but I need to test to make sure. Not only this would eliminate duplication of pages on the wiki, bit it would also provide extra functionality, e.g. embedding a comic on another page, and allowing category pages to display thumbnails rather than names only.

recurring maintenance tasks[edit]