User talk:Sphilbrick/Top ten editors

So I have a variety of thoughts on this; I'm going to start by sort of setting out how I understand the plan.

To help improve editor retention, the top 10 editors to each article (by editcount) will be posted on the talkpage where readers can see it - presumably the editor will also be notified that they're in the top 10 contributors to [article]. This is a great idea that touches on a couple of key parts of editor motivation; by classing these edits as an achievement, you're associating them with the editor's identity, wedding them to the site. By thanking the editor, you're building a sense of gratitude and community and recognition, which is awesome :).

I do have a couple of concerns, however. The first is that 'edits' isn't a very good metric for contributions here, particularly edits-over-all-time; with edits you're not touching on the underlying quality or size of contributions, and with edits over all time you're going to get a lot of 'buffer room' on old articles - users who made a lot of contributions some time ago and, under this calculation, effectively prevent anyone moving up the ranks. The second, and this is a minor quibble; I think the 'showing it to readers' thing is probably not helpful :). I think most readers get through their reading experience without really paying attention to talkpages or knowing they exist.

That being said, it seems like an interesting idea. I note that with "Echo" the Foundation is building automatic notifications when people hit N edits (for a given value of N). Ironholds (talk) 20:37, 2 March 2013 (UTC)

Thanks for your feedback, and my apologies for not realizing you had responded promptly. I somehow missed that the talk page had been created.

We are in complete agreement that the metric is not the best. As I note below, what I like about it is that is already exists. If it didn't, I would probably lean toward some other measure, such as size. I'll still support size if someone steps up ad says they can do it, but my fear is that we decide to do something spiffy, and then do nothing because it is too much work.

You aren't alone in observing that all-time edits has a downside. As I responded to the other editor who mentioned it, I think I can have my cake and eat it too. Why not an all-time list, and a last year list?

As for showing it to readers, I fully agree many readers have never seen a talk page. But I assume there would be extreme opposition to showing this on the article page itself, so this was the best I could come up with. Throwing out something off the top of my head, if this were implemented, I could imagine a banner "Ever wonder who writes these article? Click here." and the link would bring them to the talk page. Not permanent, but run as an occasional banner.--SPhilbrick(Talk) 18:58, 12 March 2013 (UTC)

Which metric?

I agree that edit count is suboptimal as a metric. Much better would be size in kilobytes, especially if editor contribution size can be compared to current article size. As well, vandalism involving massive addition of kilobytes should somehow be thrown out. Binksternet (talk) 18:34, 12 March 2013 (UTC)

I agree that size is better, albeit still imperfect. My rationale for count is that Contributors example exists, so it might be easy to program a bot to start from that and create a list.

If someone tells me it isn't much harder to write a bot that measures size contributions, subject to the obvious issues that need to be addressed (a revert of a blanking is a trivial edit, yet would be measured as a huge contribution), I could get onboard. However, I would be unhappy if we did nothing, because we decided that edit size was better, and it turned out no one steps up to write the bot.--SPhilbrick(Talk) 18:44, 12 March 2013 (UTC)

I have a serious objection to size. Subtractions from an article can also be a form of contribution (cutting cruft, for example), and I have no wish to see any metric that furthers this 'all editors are equal, but content contributors are more equal than others' malarkey. Ironholds (talk) 19:58, 12 March 2013 (UTC)

Good point. I see count as easier to access, and while imperfect, the challenges of designing a perfect measure are considerable.--SPhilbrick(Talk) 15:50, 14 March 2013 (UTC)

Edit count is certainly easier to automate than trying to determine who has had the most effect on an article by judging additions that are still in the article, and deletions that are still not in the article. Binksternet (talk) 05:42, 20 March 2013 (UTC)