ModEnc edit count visualization

[Renegade]

Good day!

I just created these two graphs out of personal curiosity, and I figured people like D or other old-timers might find them interesting.

Essentially, what I did was create a set of all days since the first revision on ModEnc (10.08.2004) and join/group the actual revisions to that set, getting a count of revisions per day.

Since I wanted to measure outside contributions, I filtered out my own edits, D's, the bot's and Alex's and pd's, since the latter two have artificially inflated edit counts due to working on RockPatch/Ares and being able to read the disassembled code.

In other words: These counts show the number of edits done by people who truly have a choice about editing ModEnc.

The first graph shows the number of outside contributions on a timeline, i.e. the direct edits-per-day view:


For the second one, I grouped the edit counts into six ranges, and measured each range's share:


As you can see, almost two thirds of the time, no outside contributor edits ModEnc - and on 92.6% of the days of ModEnc's existence, five edits or less were made by outside contributors.

This last number as well as the long, flat lines of 0 edits for 2010 and 2012 in the first graph have confirmed my view that interest in ModEnc is low, and maintaining it myself is not a good use of my time.

This is especially considering that many of the outside-contributor-edits aren't even constructive edits: These numbers include all edits, anywhere. In other words, if a user registers and then edits his own user page after that, that's one edit for the day. If you limit the selection to just pages from the main namespace, you get 3527 total rows. This can still be spam, bullshit or otherwise unnecessary edits, but even with the pointless edits in there, that's an average of just 1.18 outside contributions a day.
Or, since I know from the previous dataset that there were 1935 days with no contribution at all, that's 3.36 outside contributions a day, on the 1048 days someone actually contributed...those 35% of days.

So yeah...when filtered down to just outside contributions, just in the main namespace, even including all bullshit edits, you still end up with less than 4 edits a day, on 35% of the days.
On 65% of the days, ModEnc is just dead¹.


If you'd like for ModEnc to stay up despite the low amount of contributions, feel free to volunteer as a maintainer. ~22 days left.

---

¹ At least when it comes to outside contributions, and those could still be spam.

[zzattack]

Do you have any statistics about views? I rarely have anything to contribute to ModEnc myself, but I use it very often as a resource instead. I think it's only natural that we see less edits as the game just ages and nearly all of the interesting knowledge is already there.

[Renegade]

Each page's view count is in the footer of the page - there's a text that says "This page has been accessed x times". The main page currently has 151770 views, that's about 50 views a day since the start. Which sounds like a lot, until you look at the bounce rate: While the main page has gathered over 150000 views over its lifetime, an integral flag like Damage has only been viewed 3174 times - that's a little more than once a day. Who_crushes_whom, a ModEnc-unique page explaining the interaction of crushing-related flags, has even less with 3016 views. And this was written exclusively for ModEnc, so the low view count cannot be explained by people reading it in The Guide or something - it's just plain a lack of interest. If you click through random pages, it mostly gets worse:
SpiralDeltaPerCoord - 381 views
ControlledAnimationType - 491 views
Using XCC Mixer to split a singleframe PCX into multiframe SHP - 2047 views

To put this into perspective:
The first ten pages, together, have accumulated 545181 views since August 2004.
In order to reach that number again, that is, accumulate another ~545000 views, you need the next 61 pages.
After that, you need the next 146 pages.
After that, you need the following 250 pages.

Or, phrased another way: The top 11.57% percent of actual content pages accumulated half of all views on content pages since 2004.
Half of all views on content pages we got went to less than 12% of ModEnc's content pages.

And several of those top pages aren't even relevant anymore: The Legacy RockPatch Wishlist‏‎ is in third place with 61591 views. (Notice the sharp drop from place one and two with ~150000 views)
Place 7 is List of mods using Rock Patch‏‎ with 19436 views.

So while, yes, a few view counts do seem impressive at first, the truth is that ModEnc mostly consists of a very long tail.
And, still, these aren't even monthly or yearly counts. These are counts for the entire 8 years ModEnc has been up.

Ultimately, however, it just plain doesn't matter. Because what's important isn't how many people view ModEnc, what's important is who's stuck with the work. As the stats above showed, little of what has happened was the work of outsiders. Most of it was done by D, me, and a handful of others. Case in point: If you select all revisions over all namespaces [i.e. all edits in total] for all users except for the group excluded above, you get 4925 revisions committed; if you do the same narrowed down to D, me, Alex and pd [the inverse minus the bot], you get 5518. The four of us together, even without the bot account, have accumulated several hundred edits more than the entire rest of the community. Include the bot account, and you get 6129.
Let me restate that: 55.45% of all edits on ModEnc were done either by D, me, Alex, pd, or the bot after my preparation.
49.92% of all edits on ModEnc were done either by D, me, Alex or pd, without the bot.
(The bot is mostly used for mass-template changes or category moves, if you were wondering.)

And these are just pure revision commits¹.
It doesn't take into account software maintenance, plugin- and template development, spam fighting, host migrations, and so on.

That is what this is about. Workload, not view counts.

It doesn't matter whether we get a hundred or a hundred thousand views, what's important is who does the work?

For the past eight years, that was a handful of people around RenProj. Now, either someone else takes the lead, or I'll take a lack of engagement as a final confirmation of a lack of interest, and draw the logical conclusions.

---

¹ In addition, not all revisions are created equal. Not all of those thousands of revisions the community committed were actual, useful page edits. There were user page edits, talk page comments, bullshit edits, spam, etc., etc.

[pd]

So, your plan is to stop hosting ModEnc?

I am not willing to maintain it, as in edit articles etc, because there is simply no time. I also have limited knowledge when it comes to defending it against spam. However, if nobody takes it over and before you are going to delete it, I am ready to host it.

[Renegade]

It's not a question of hosting. I can host it just fine. What want is someone to commit to maintaining it, and you just ruled out doing exactly that.

[Graion Dilach]

In case of editing and such, when I have decided to arise from the nothingness, ModEnc felt complete to me. Take that into account too.

I am thinking about taking over, but I am not enough sure about myself considering that I never ever worked with MediaWiki (besides some small edits in ModEnc and OpenArena's regarding bot logics)... while I'd really see ModEnc saved, I think with my current state, skills and timetable... I just simply can't be the one.

[Nighthawk]

I do have experience working with MediaWiki, and at any other time would immediately volunteer... but with two research projects on the go at the moment, my free time's currently running low (that, and the university proxy is making IRC impossible to connect to), so I'm not entirely certain whether I'd be of any use in the role right now. But I don't want to rule myself out just yet, especially since I'll hopefully have one project wrapped up in the next few weeks.

[Nighthawk]

And naturally, one day too late, IRC starts working again for me. If you're still wanting a maintainer, I'd be happy to put myself forward. One of the projects I'm currently doing ends tomorrow, so I'll once again have free time.

[Renegade]

I would appreciate a watchful eye until I have the time to properly rework the site. After that, we'll see how life looks for everyone.