Third round of Project X completed.

[Poetlister]

This was the third round of WikiProject X, intended to wrap up the project through the deployment and testing of our final product, CollaborationKit. In this, progress was made. Much of the development work to get CollaborationKit ready for live deployment was indeed completed. But overall, the project has failed, through this and every iteration thus far.

Let this report serve as a warning, to all who might consider requesting some funding to get your project going, knowing that what is available to Grantmaking is limited, and that coming from a volunteer background you are willing to put in some amount of time as a volunteer as well.

Do not do this.

You will burn yourselves out, as we have. For all the love that drove you to begin, there will come a time when none will remain to see it through.

The initial goal of WikiProject X was as follows:

To address the causes of WikiProject failure on the English Wikipedia. By addressing these causes, we hope to increase the activity on WikiProjects and make WikiProjects more central to the experience of editing Wikipedia. This is based on the hypothesis that WikiProjects help facilitate Wikipedia-editing in a given subject area by organizing contributors around a cause and by providing resources and social support.

Meta

[Eric Corbett]

More money wasted then.

[Osborne]

Wikipedia: the sum of all human failed projects.

[Midsize Jake]

Really, it's worth reading the whole thing - apparently the two main guys working on this "Collaboration Kit" totally underestimated the amount of work involved and the almost-total lack of volunteer interest in it, and completely burned out on it to the point where at least one of them, User:Isarra (T-C-L), came close to some sort of breakdown.

We should have had a full team from the start. We should have had a budget perhaps orders of magnitude greater, in terms of time, funding, and all other kinds of expenses that would normally go with an extended contract-based development project. We began with what we knew, but when it became clear that more would be needed, we should have brought in the full requirements. We are professionals, and should have treated ourselves as professionals.

And now it is clear that we have caused ourselves even as-yet unknown damage, not just in terms of our quality of life and own professional development, but our very mental health.

To be fair, Wikipedia and the WMF are unusual entities in terms of both what they are and how they have to operate in order to drive WP's expansion, or at least how they think they have to operate in order to expand. Nobody wants to believe or accept that Google is really driving the bus, and nobody wants to admit just how crucial the dual roles of dramamongering and general narcissism are, in terms of both recruitment and retention of participants. It thrives on what might be called "attitudinal heterogeneity" - IOW, if it became more unified and purpose-driven, it would be less "fun," and people would lose interest in it. So it's not too surprising that people who try to do things like devise a team-building "kit" would run into problems and underestimate the amount of effort required.

[Vigilant]

I love how they chased the changing technical specs around from the WMF instead of building an abstraction layer.

[Poetlister]

To be fair, Wikipedia and the WMF are unusual entities in terms of both what they are and how they have to operate in order to drive WP's expansion, or at least how they think they have to operate in order to expand. Nobody wants to believe or accept that Google is really driving the bus, and nobody wants to admit just how crucial the dual roles of dramamongering and general narcissism are, in terms of both recruitment and retention of participants. It thrives on what might be called "attitudinal heterogeneity" - IOW, if it became more unified and purpose-driven, it would be less "fun," and people would lose interest in it. So it's not too surprising that people who try to do things like devise a team-building "kit" would run into problems and underestimate the amount of effort required.

Yes, how many people would edit if they had to do so only to build an encyclopaedia?

[Midsize Jake]

Yes, how many people would edit if they had to do so only to build an encyclopaedia?

Assuming that's not a rhetorical question, I would say that as long as Google continues to give them preferential page-rankings, some form of "civility miracle" would probably only reduce head-count by 30 percent or so. They might lose about 15 percent of their current "active" user base, maybe as much as 40 percent of the "highly-active" users over time, but they might make some of that up with a small influx of people who previously were either intimidated by the existing user base in some way, or who just didn't want to waste their time arguing on the internet with complete strangers. (Depending on whether or not the WMF et al can effectively get the word out that things have changed, and people actually believe them.)

I'm not going to say this is an "absurd hypothetical," because it isn't totally impossible for them to achieve this if they really wanted to. I just think they're grossly overestimating the number of people who would come in to replace all the "bad apples," and one of the main reasons would be that every "bad apple" they get rid of would be another voice out there telling everyone who will listen that Wikipedia has "gone completely to shit." Because in many ways, that's exactly what it will have done - improvements in civility have never really been shown to make a significant positive difference in general article quality, but a serious reduction in experienced users, regardless of how horribly some of them tend to behave, could make a significant negative difference - quite easily, I would imagine. And if the overall effort is clearly failing in that (IMO crucial) aspect, fewer people will want to get involved and stay involved.

[Randy from Boise]

Although Sucks currently, ummm, really sucks, there is a pretty interesting illustrated thread there about Project X. It was started June 15 by BadMachine. Here is the link if you wanna skip the rest of the monkeyscat flinging...

https://www.wikipediasucks.co/forum/vie ... f=5&t=1261

Project manager is/was James Hare, the guy who brought you the WP essay "Maintaining a Friendly Space" (linkhttps://en.wikipedia.org/wiki/Wikipedia ... ndly_space[/link])

Top financial beneficiary of the project, to the tune of $20K, is Isarra the Pie Lady of Arbcom election fame... (linkhttps://foundation.wikimedia.org/wiki/User:Isarra[/link])


dodging the monkey shit so you don't have to,

RfB

[Midsize Jake]

Well, look. This grant came directly out of the fact that the WMF, and to some extent The Faithful themselves, have really started believing their own bullshit. They've got this idea that improving their "collaboration tools" will lead to more collaboration, and collaboration somehow naturally leads to more civility, and more civility will naturally lead to more diversity and - finally! - better media coverage. Personally I think they're wrong, but it doesn't matter - the WMF has to try everything they can short of banning 40% of the user base to get to where they want to be, and this is just one of those things they have to try.

They won't even necessarily accept that this is a "failure." Sure, it almost mentally broke one or two of their long-term users completely, but from their perspective, that doesn't mean the whole thing was a bad idea, it might only mean they didn't throw enough money at it. They might have intended to "take it in-house" all along, actually - having grantees do the initial heavy-work might have just been an internal PR move, for all we know.

Ultimately though, this has very little to do with the Women in Red, other than the fact that the WMF likes the Women in Red and actually wouldn't mind seeing other "WikiProjects" with a similar level of dedication. (WiR is one of Mr. Crowsnest's pet peeves, so from his perspective it's all about them and their systematic attack on WP's notability criteria, which means it's really all about Jess Wade, and so on. He calls WiR a "cult within a cult," which isn't completely off-base, but very few, if any, of the other WikiProjects are that single-minded and insular.)

What's really going on here is that the WMF knows that "walled gardens" are an inevitable aspect of the system they have in place, so they're taking a "can't kill it, so let's regulate it instead" approach. This "Collaboration Kit" was supposed to be a way of preventing, not encouraging, new WikiProjects from becoming too insular and cult-like while still retaining the "wikilove" they think is part-and-parcel of the whole concept. (Hint: It's not.) But it's only the new ones they can really do anything about, and they probably know that. There's nothing they can do about WiR at this point, even if they wanted to (which they don't) - that horse has left the barn, and while they really liked that horse, it's halfway to the next county by now.

[Randy from Boise]

Sure, it almost mentally broke one or two of their long-term users employees completely....

Fixed it for you.

RfB

[Poetlister]

Well, look. This grant came directly out of the fact that the WMF, and to some extent The Faithful themselves, have really started believing their own bullshit. They've got this idea that improving their "collaboration tools" will lead to more collaboration, and collaboration somehow naturally leads to more civility, and more civility will naturally lead to more diversity and - finally! - better media coverage. Personally I think they're wrong, but it doesn't matter - the WMF has to try everything they can short of banning 40% of the user base to get to where they want to be, and this is just one of those things they have to try.

Indeed, the best thing would be to reduce the amount of collaboration. The less interaction there is between editors, the less scope there is for peopleto get annoyed. Of course that would be very difficult to achieve, and it would go against the idea that crowdsourcing is the best way to get good articles. It's a pity that Google Knol never worked out.

[Osborne]

I can't tell which part is sarcastic in your comment.

[Poetlister]

I can't tell which part is sarcastic in your comment.

Wikipedia is often beyond sarcasm.

[Osborne]

I can't tell which part is sarcastic in your comment.

Wikipedia is often beyond sarcasm.

Lol, so which part?

[Poetlister]

I can't tell which part is sarcastic in your comment.

Wikipedia is often beyond sarcasm.

Lol, so which part?

Certainly "It's a pity that Google Knol never worked out" is not sarcastic.

[Osborne]

Certainly "It's a pity that Google Knol never worked out" is not sarcastic.

For a moment I thought the other parts are serious too. If sarcastic, then we agree.

"crowdsourcing is the best way to get good articles"

I wanted to discuss this topic for some time. Let's see a few openness models, that comes to mind:
* closed (eg. Encyclopaedia Britannica) - made by vetted employees (professionals)
* corporate open-source communities (eg. Mozilla) - core is professional employees + anyone can contribute by VETTED pull requests
* grassroots open-source communities - core is an amateur startup + anyone can contribute by VETTED pull requests
* crowdsourcing (Wikipedia) - core is ad-hoc/random jobless (online 7*8 hours) people + anyone who can tolerate their oligarchy (self-regulation)

The crowdsourcing end of the spectrum gets more articles, generally worse quality (with exceptions), but also enables the creation of articles by people, who wouldn't get a job at Enc.Br., and also on fringe subjects, that wouldn't fit in Enc.Br. (besides porn actors).
Imho Wikipedia went downhill around ca. 2006 by letting anybody in, without building an open-source community, that is sustainable, and has values. Building such community is damn hard work; not building it results in a mess like WP.

When the WMF was created, it should have made this movement into a corporate open-source community. That includes professional, employed editors, community managers, moderators, etc., who would have created the rules that achieve the "anyone can edit" target.

[lonza leggiera]

Certainly "It's a pity that Google Knol never worked out" is not sarcastic.

For a moment I thought the other parts are serious too. If sarcastic, then we agree.

"crowdsourcing is the best way to get good articles"

I wanted to discuss this topic for some time. Let's see a few openness models, that comes to mind:
* closed (eg. Encyclopaedia Britannica) - made by vetted employees (professionals)

…

Encyclopaedia Britannica is not quite quite completely closed. At the top of each article page there's a button labelled "FEEDBACK". To submit feedback via this button you need to register an account, and if your feedback includes suggestions for changing an article which the encyclopaedia's editors end up making use of, you will be credited as a contributor to the article in its history, and in its list of contributors. To see these you need to click on the "INFO" button at the top of the article, and then select the appropriate item from the menu. If you do this for the article Christopher Columbus, you will see my name listed as the only entry under the "OTHER CONTRIBUTORS" heading.

[Osborne]

[Poetlister]

* crowdsourcing (Wikipedia) - core is ad-hoc/random jobless (online 7*8 hours) people + anyone who can tolerate their oligarchy (self-regulation)

I think that the problem with Wikipedia is that it has got so big that monitoring it is an overwhelming task. If you have a small wiki with a limited number of contributors and some good admins, it can work quite well. Of course, such a site would attract little attention so not many people would benefit from it.

[Osborne]

I think that the problem with Wikipedia is that it has got so big that monitoring it is an overwhelming task. If you have a small wiki with a limited number of contributors and some good admins, it can work quite well. Of course, such a site would attract little attention so not many people would benefit from it.

Exactly. The model of editing, admining (and "editorializing") was developed before WP got big around 2006, and it did not scale (with IT term). Those models operate badly with a community this size, resulting in good editors banned, wp:frams, admin abuse, etc.
"Monitoring" the work of a community this large, needs structure, consciously developed with a purpose in mind. Open source communities develop these structures.
Wikipedia also has power structures, but those were built along personal interests, in the shadows, not for a common goal, a community purpose.

[BrillLyle]

Project manager is/was James Hare, the guy who brought you the WP essay "Maintaining a Friendly Space" (linkhttps://en.wikipedia.org/wiki/Wikipedia ... ndly_space[/link])

RfB

I met James at the first WikiCite in Berlin, though I had heard people talk about him in the WM NYC chapter. Hare had set up an off-Wiki interface to keep track of metrics that he let our chapter clone and use. James was semi-responsive after I put in a year's worth of event info -- which was a lot. I think that was at the peak period of activity for Wikimedia NYC. I couldn't get the reports to run and we needed the info for our reporting for our annual grant so finally at the last hour I exported everything to Excel and word processed the results. Which Winifred Oliff, our grant officer, said didn't tell enough of a "story". Hours and hours of work, tons of graphs and raw data.

Essentially, James Hare said that I broke his system there (wish I could remember the name of the online tool but am blanking now). Anyway, he said it wasn't able to handle the data and number of events.

But really why I am describing this here, is that he was not responsive to emails and requests from me. I learned about his work on Project X during the few times he did email me back because he mentioned it was taking up all of his time and attention. I looked at the implementation and had a lot of pretty basic functionality questions. I have done a lot of beta testing from the end-user / power-user perspective in real life. I was concerned about the user interface and square peg into round hole issues that I could see right off the bat in how it was in the WiR build.

I know James Hare is very young (as a person) and had a lot of enthusiasm. But I wasn't impressed with his responsiveness (and lack thereof). Or ability to translate concepts into workable outcomes. But then again, I thought Project X was "his" project. That's how it sounded like it was to me. Had no idea he was "just" (just?) a project manager / cog in the wheel on it.

I'm not surprised at the bumps on the road re: Project X. Editing Wikipedia and managing task lists for initiatives -- which I was doing using Wikidata (and which was brilliant when it was working) -- is a LOT of freaking work. The curation of that should be structurally supported (again, I think Wikidata is great for this) without taking up so much time managing by volunteers the way it exists now. Editathon in a box concepts spring to mind. Art+Feminism could've done this given the good will and so much press they had. But instead put their focus outside Wiki, stopping their use of on Wiki meetup pages, etc.

I think fundamentally in all of these processes there's a distinct disregard and underlying contempt for the end user. Just like when James Hare was surprised that I had diligently input so much data, to the point I broke the system (he joked), and was surprised at that, I was like, well, what do you expect? I did what you said to do, there's masses of data input, and just because I'm not a programmer doesn't mean I am a lunkhead.

I have enough technical knowledge to be dangerous and ask stupid questions, yes. But also from having a master's in library science and that formalized knowledge, plus my muscular editing / word processing skills, I get the end user interface / experience needs. And Project X needed / needs a ton of work from that perspective.

- Erika
handwringing, but also fascinated reading what was / is going on here with this project. so much T&E. not surprised folx got burnt krispy

-- edit --

The interface was called Podio https://podio.com/ -- now owned by Citrix :-(

[BrillLyle]

And digging more into the final Project X report and reading Sucks and stuff, I acknowledge I'm probably the worst person to provide feedback and wishlist items to an already stressed and beleaguered and over their head developers. I feel pretty bad about it now, as I read more and look back.

Though I think there was a lot of arrogance in how this project was set up (which was what my initial response was). And the fact that it didn't integrate existing (?) systems and practicalities of outreach very well. I think they needed more input from folks doing a lot of outreach (not just WiR) and actual editing. Which would've shown the huge scope and requirements needed.

I don't know. I feel bad, but then also think WMF pretty much threw a bit of money and very few resources at something that could've been very positively impactful, and then now can sort of shrug off how it ruined people's lives where they were set up to fail. Shocking behavior (sarcasm) by WMF. It's like they don't do this all the time (they seem to, a lot).

So feel bad. But egads, the naivety is stunning! Oy

[Poetlister]

"Monitoring" the work of a community this large, needs structure, consciously developed with a purpose in mind. Open source communities develop these structures.
Wikipedia also has power structures, but those were built along personal interests, in the shadows, not for a common goal, a community purpose.

That's the situation, precisely. Wikipedia has no editorial board and almost certainly will never have one. Admins cannot usually resolve content disputes, because they are rarely experts in the disputed topic. Thus there is often no quality control of content, which is the main thing that a sensible work of reference needs.

[Osborne]

Regarding Project X:
I've just read the report, did not research. It seems to me this project tried to do too much in one step. That's so 20th century project management. Whole Wikipedia is stuck in that era.

About getting feedback from the community, and working with the wmf:
Majority of the opinion givers know best, how to complain and criticize. That's easy. Very few ppl have usable, constructive suggestions, which takes more effort.
As I hear developers occasionally leave for internal bickering, resulting in abandoned developments.
In an atmosphere like this, I'm not surprised that developers/designers burn out. This is a serious risk for big projects. Starting with smaller targets is more likely to work.

Smaller target:
Just integrating discourse (discuss-space) into the wiki would be a great boost for projects, and inter-editor communication, possibly doable within a few months with testing and deployment, prototype in just 1 week.

[BrillLyle]

That's the situation, precisely. Wikipedia has no editorial board and almost certainly will never have one. Admins cannot usually resolve content disputes, because they are rarely experts in the disputed topic. Thus there is often no quality control of content, which is the main thing that a sensible work of reference needs.

That's what I really like about what Sanger said in this recent interview he did. How he envisioned a role for experts and academics in the community, in a low key way. I have a friend from college who is a physicist and he said their community stopped contributing because people kept challenging the experts in his community on really basic stuff. To the point that they won't participate. It's a real loss, as I think they would've been pretty active.

I was supporting outreach by the SCAR group (The Scientific Committee on Antarctic Research). They did a ton of pages and uploaded images, etc. It was very inspirational. Of course it was all pages about women because it was centered around a conference. And then male scientists complained there as an imbalance. I've probably told this story before. But it always makes me laugh. Welcome to our world, dude! :-)

[Poetlister]

That's the situation, precisely. Wikipedia has no editorial board and almost certainly will never have one. Admins cannot usually resolve content disputes, because they are rarely experts in the disputed topic. Thus there is often no quality control of content, which is the main thing that a sensible work of reference needs.

That's what I really like about what Sanger said in this recent interview he did. How he envisioned a role for experts and academics in the community, in a low key way. I have a friend from college who is a physicist and he said their community stopped contributing because people kept challenging the experts in his community on really basic stuff. To the point that they won't participate. It's a real loss, as I think they would've been pretty active.

Sanger never wanted what Wikipedia has become. His original intention was to use it as a source of draft articles that would be polished up and put on Nupedia. That would still be a good idea if it could be done, though it would of course require a vast amount of effort. Yes, many experts in just about any topic you can think of have tried in good faith to contribute and been forced out.