Page MenuHomeMiraheze

Global Wikibase proposal and a few technical questions
Closed, InvalidPublic

Description

I and @Ugochimobi are discussing a possible addition to the Miraheze farm: we would like to present this idea to ask for feedback and then discuss a few technical topics.

Our idea would be to setup a wiki with Wikibase extension activated in order to act as a common data repository for any other wiki hosted at Miraheze. This would work similarly to the way Wikidata works for the sites of the Wikimedia universe (note that data hosted at Wikidata are not accessible by wikis outside Wikimedia).
The rationale behind this is that Miraheze has a Commons wiki, corresponding to Wikimedia Commons, but it currently doesn't have a Wikidata equivalent.
So our basic questions are:

  • how do you feel about this project?
  • Would this something useful and well-received?
  • Are there any user that can see a practical application of it right away?
  • Was something like this already tried and what was the outcome?

If this project is indeed something deemed useful, then there are some technical point to address before start (something similar, but not identical, was discussed in T6041).
In order to make the direct access work in the client wikis, on this proposed Wikibase Repository two things should be setup:

  • sites table should contain the link to the client wikis. In T6041 they already explained it is not an option to include external sites to the table, but I am right assuming all the Miraheze-hosted wikis are already in a shared sites table?
  • again as mentioned in T6041, there are a few variables to be customized in the LocalSettings.php file of the Repository wiki: would this be possible? It is unclear whether this is something feasible or not, as the discussion in T6041 stopped before reaching a clarification on this topic.

Thanks for any comments or contribution on this proposal.

Related Objects

Event Timeline

New wikis will appear in other wikis site tables every few weeks (5AM 5th & 20th) I've not timed how long a run takes but we could consider running it more often.

In T8630, @Lucamauri wrote:

The rationale behind this is that Miraheze has a Commons wiki, corresponding to Wikimedia Commons, but it currently doesn't have a Wikidata equivalent.
So our basic questions are:
how do you feel about this project?
Would this something useful and well-received?
Are there any user that can see a practical application of it right away?
Was something like this already tried and what was the outcome?

If you wish to establish another "official" Miraheze wiki to host this global Wikibase installation, this proposal would likely require an RfC to back it. Commons was established following this RfC in 2017 and Developers Wiki following is this Community noticeboard thread. I suggest you propose it on the Community noticeboard and request feedback from users on active forums such as Discord and IRC to see what others think before making a fully fledged Request for Comments should you wish to go forward with this.

Agent_Isai triaged this task as Normal priority.Jan 19 2022, 01:50
In T8630, @Lucamauri wrote:

The rationale behind this is that Miraheze has a Commons wiki, corresponding to Wikimedia Commons, but it currently doesn't have a Wikidata equivalent.
So our basic questions are:
how do you feel about this project?
Would this something useful and well-received?
Are there any user that can see a practical application of it right away?
Was something like this already tried and what was the outcome?

If you wish to establish another "official" Miraheze wiki to host this global Wikibase installation, this proposal would likely require an RfC to back it. Commons was established following this RfC in 2017 and Developers Wiki following is this Community noticeboard thread. I suggest you propose it on the Community noticeboard and request feedback from users on active forums such as Discord and IRC to see what others think before making a fully fledged Request for Comments should you wish to go forward with this.

You're right Agent, a formal Community noticeboard would make sense.
But what do you think personally? As a CES, would you want to ask us some necessary questions for us to enlighten you more so that you can carry it yourself to the CN or would you still prefer us to take it ourselves?

New wikis will appear in other wikis site tables every few weeks (5AM 5th & 20th) I've not timed how long a run takes but we could consider running it more often.

Thanks for your prompt answer: regardless the update interval, this will address the first topic.

Beside that, regarding the customiziation of LocalSettings.php for the Repo and Client wikis, do you think this is something feasible?

Assuming they work, it should be. I'm not much of an expert with WikiBase.

[cut]
If you wish to establish another "official" Miraheze wiki to host this global Wikibase installation, this proposal would likely require an RfC to back it. Commons was established following this RfC in 2017 and Developers Wiki following is this Community noticeboard thread. I suggest you propose it on the Community noticeboard and request feedback from users on active forums such as Discord and IRC to see what others think before making a fully fledged Request for Comments should you wish to go forward with this.

You're right Agent, a formal Community noticeboard would make sense.
But what do you think personally? As a CES, would you want to ask us some necessary questions for us to enlighten you more so that you can carry it yourself to the CN or would you still prefer us to take it ourselves?

I just had a look at Commons RfC and it seems to me a very good example on how we should present this project as well. Sorry I did not know about the whole RfC mechanism in Miraheze.
@Ugochimobi @Agent_Isai if it's fine with you, I would proceed creating the RfC there in the next few days.

Assuming they work, it should be. I'm not much of an expert with WikiBase.

@RhinosF1 I believe both from the technical aspect to the simplest one, everyone would be of help. So I believe we can help each other with various capabilities.

[cut]
If you wish to establish another "official" Miraheze wiki to host this global Wikibase installation, this proposal would likely require an RfC to back it. Commons was established following this RfC in 2017 and Developers Wiki following is this Community noticeboard thread. I suggest you propose it on the Community noticeboard and request feedback from users on active forums such as Discord and IRC to see what others think before making a fully fledged Request for Comments should you wish to go forward with this.

You're right Agent, a formal Community noticeboard would make sense.
But what do you think personally? As a CES, would you want to ask us some necessary questions for us to enlighten you more so that you can carry it yourself to the CN or would you still prefer us to take it ourselves?

I just had a look at Commons RfC and it seems to me a very good example on how we should present this project as well. Sorry I did not know about the whole RfC mechanism in Miraheze.
@Ugochimobi @Agent_Isai if it's fine with you, I would proceed creating the RfC there in the next few days.

Makes sense for now.
There's something I want to say, maybe in few minutes time I would remember them.

Please, spend some serious time drafting such an RfC.

This is a massive proposal, and I would highly, highly recommend you go into this prepared. Some items I can think of off the top of my head that would need to be present for me to personally even consider anything near supporting would be a policy to vote on for selecting initial administrators, and policies to vote on for selecting future administrators. Personally—and I'm sure others feel the same—I would absolutely not feel safe with such a project if it were assigned leadership arbitrarily or assigned by nature of the identity of the proposer(s).

First of all I'd like to say that I agree with @dross and think we need to take care in drafting such a proposal and giving everyone the opportunity to comment/come up with proposals and amendments before making it an RfC.

Second, since this seems to be blocked on the RfC taking place I don't see a reason for it to continue being an open task on Phabricator, since until an RfC confirms that the community wants to have a global Wikibase there's nothing SRE can do. If the RfC is successful, a new task should be open detailing what the implementation should look like (likely by the closing Steward).

If you have any more technical questions that SRE needs to answer feel free to comment here if you wish or contact us elsewhere (email, IRC, Discord, on-wiki).

I just had a look at Commons RfC and it seems to me a very good example on how we should present this project as well. Sorry I did not know about the whole RfC mechanism in Miraheze.
@Ugochimobi @Agent_Isai if it's fine with you, I would proceed creating the RfC there in the next few days.

Done as Miraheze Data - a common structured data repository for Miraheze.

I think it should be drafted first
So I'd move it to my userspace draft
Where other proposals can be added, then discussions about it can go in the talk page of the draft.

In T8630#175037, @dross wrote:

Please, spend some serious time drafting such an RfC.

This is a massive proposal, and I would highly, highly recommend you go into this prepared.

We will surely put our best effort in the RfC, but it won't possibly be perfect and polished at first attempt.
I'm not very expert on things at Miraheze, but a Request for Comments, by its own definition, looks like a place to get feedback and discuss it.
We want to spark conversation, first and foremost, we are not trying to sell any concept here.

Some items I can think of off the top of my head that would need to be present for me to personally even consider anything near supporting would be a policy to vote on for selecting initial administrators, and policies to vote on for selecting future administrators.

I wonder if this is the right moment to discuss about those topics. Isn't it premature?
Shouldn't we first assess publics interest, then discuss technical feasibility and only afterwards decide such details?

Personally—and I'm sure others feel the same—I would absolutely not feel safe with such a project if it were assigned leadership arbitrarily or assigned by nature of the identity of the proposer(s).

Would you please better specify what you mean with "leadership"? You mean organizational leadership, or technical leadership? Or anything else?
Are you worried we are going to steal data or something like this?
Please rest assured our only intention is to put expertise at the service of other users for a tool we guess might be of help.

[…] we need to take care in drafting such a proposal and giving everyone the opportunity to comment/come up with proposals and amendments before making it an RfC.

I would be glad if you can please explain how and where this process should take place.
I read a suggestion above to start a RfC for this topic, but it is my impression now that RfC should be produced at a later stage of the proposal. Previous steps are unclear to me.

If you have any more technical questions that SRE needs to answer feel free to comment here if you wish or contact us elsewhere (email, IRC, Discord, on-wiki).

There are several technical issues to clarify, but I understand this is not the right moment to do it.
Should the RfC, or any other process necessary, be successful, then we can maybe reopen this task or create a more focused one.

I want to use this medium to drop a note for the System Administrators, I mean, without y'all Miraheze, in general, wouldn't be.
I want to also let us know that, this Structured Data project is a "made up overnight" project that was brought up. I want y'all to understand the fact that it is important we have a central wikibase instance where we can structure pages and media (from miraheze wikis in different fields to media from miraheze commons). This is where the internet is driving us to, and this is what the Internet is getting used to.
This project will not be a success if system administrators do not support it (as well as the community at large). Your help is highly needed to drive this project to its purpose.

It seems like you two are seeking some advice, so I can be very straightforward about all this.

We will surely put our best effort in the RfC, but it won't possibly be perfect and polished at first attempt.
I'm not very expert on things at Miraheze, but a Request for Comments, by its own definition, looks like a place to get feedback and discuss it.
We want to spark conversation, first and foremost, we are not trying to sell any concept here.

Requests for Comment here on Miraheze is used as a means of receiving community input to establish consensus for project-wide changes, those changes which affect a mass amount of users, or those changes which change the trajectory of the Miraheze community. A successful RfC should, in fact, accomplish its goals the first time. A properly formed Request for Comment should not require several separate iterations to reach success; brute forcing a project idea is not an indicator of success or community support.

Feedback regarding the formation of a proposal within an RfC should be sought prior to presenting said RfC. The official Request for Comment is not itself for developing a proposal. Your proposal should be mostly complete, with room for changes to the details as put forth by the community. The foundational ideas for a Request for Comment should not be changing during the course of discussion.

If you're trying to get more folks involved in the initial drafting process of this proposal, it is likely that you will be required to reach out personally (try user talk pages, Discord, IRC, etc.), rather than just going for a RfC or a draft and simply seeing if it flies.

I wonder if this is the right moment to discuss about those topics. Isn't it premature?
Shouldn't we first assess publics interest, then discuss technical feasibility and only afterwards decide such details?

No. This needs to be in your proposal. If a project is to be official as you desire, the community must establish (through the process of drafting and commenting on an RfC) the process and standards to be used to select initial bureaucrats, sysops, and policies.

Failing to include a proposal regarding the selection of project leadership and functionaries is sufficient reason to oppose the proposal entirely.

Would you please better specify what you mean with "leadership"? You mean organizational leadership, or technical leadership? Or anything else?
Are you worried we are going to steal data or something like this?
Please rest assured our only intention is to put expertise at the service of other users for a tool we guess might be of help.

When I mention "leadership", it is my intention to include bureaucrats, administrators, and any other functionary type users on the project.

I hold no personal concerns regarding your two's roles as proposers or participants on the project. However, neither of you are entitled to submit the wiki creation request nor receive any permissions or special status on the project if it is successful.

The summary of the point I make here is do not expect to be a sysop on Miraheze Data if the proposal is successful. The potential and process for either of you two to gain permissions on the project should be identical to that of literally any other Miraheze user. On top of this, the process(es) for selecting the first sysops and bureaucrats on the project should be included in your proposal RfC.

In T8630#175588, @dross wrote:

It seems like you two are seeking some advice, so I can be very straightforward about all this.

Surely we need advice. If this whole process is explained on some documentation, I would be glad to consult it.

Requests for Comment here on Miraheze is used as a means of receiving community input to establish consensus for project-wide changes, those changes which affect a mass amount of users, or those changes which change the trajectory of the Miraheze community. A successful RfC should, in fact, accomplish its goals the first time. A properly formed Request for Comment should not require several separate iterations to reach success; brute forcing a project idea is not an indicator of success or community support.

Perhaps this should be better explained in the RfC page: the introduction there is:

Requests for Comments are an easy way to gather community feedback and to form community consensus on certain proposals, ideas, and issues. Requests for Comments can be used for a variety of purposes where the community should be consulted.

Frankly I can't see all the details you wrote above.
Thanks for explaining.

Feedback regarding the formation of a proposal within an RfC should be sought prior to presenting said RfC. The official Request for Comment is not itself for developing a proposal. Your proposal should be mostly complete, with room for changes to the details as put forth by the community. The foundational ideas for a Request for Comment should not be changing during the course of discussion.
If you're trying to get more folks involved in the initial drafting process of this proposal, it is likely that you will be required to reach out personally (try user talk pages, Discord, IRC, etc.), rather than just going for a RfC or a draft and simply seeing if it flies.

I thought Miraheze had a place to publish ideas and receive feedback and, from the comments above, I assumed this was RfC, but I see now I assumed too much.
So we'll different routes to get feedback.

No. This needs to be in your proposal. If a project is to be official as you desire

What do you exactly means with "official"?

Failing to include a proposal regarding the selection of project leadership and functionaries is sufficient reason to oppose the proposal entirely.

This sounds a bit extreme, but we'll take note of this.

The summary of the point I make here is do not expect to be a sysop on Miraheze Data if the proposal is successful. The potential and process for either of you two to gain permissions on the project should be identical to that of literally any other Miraheze user. On top of this, the process(es) for selecting the first sysops and bureaucrats on the project should be included in your proposal RfC.

I see you have great concern about this, and I still fail to see why. Maybe I gave you some wrong impression above, so let me be perfectly clear.
Working on a project of mine in the past couple of years, I gathered some experience about Wikibase and I tried to disseminate the knowledge I got in every way possible. I am also committed to help other users where possible, so when @Ugochimobi approached me with the idea I found it both interesting per se and also an area where I might help.
So if this proposal is successful and if I can help in setting it up, then I would gladly help and I think this would be a useful addition to users.
If either of the two things isn't going to happen, then so be it, nothing will change in my life.

Please understand this idea of ours is proposed in the best interest of the users and in their interest alone: my impression from all the conversation around it is that we are being discouraged to even start talking about this.

No. This needs to be in your proposal. If a project is to be official as you desire, the community must establish (through the process of drafting and commenting on an RfC) the process and standards to be used to select initial bureaucrats, sysops, and policies.

I believe so much in DUE processes, and so I don't still understand how we should be discussing how bureaucrats and sysop and policies would come when the Initial proposal is not even approved yet.

Failing to include a proposal regarding the selection of project leadership and functionaries is sufficient reason to oppose the proposal entirely.

This particularly gives me proof that you and some other person have a wrong impression about this whole proposal and the proposers, you, opposing the proposal because how crats and sysop would be elected isn't included tells me that you have a personal issue with the proposal. If you ask me, I'd say "Why don't you ask or discuss this simply on the RfC page or even suggest how you think it should go".

I hold no personal concerns regarding your two's roles as proposers or participants on the project. However, neither of you are entitled to submit the wiki creation request nor receive any permissions or special status on the project if it is successful.

And here, you kept hammering on the fact that we (@Ugochimobi, @Lucamauri) are the proposers and that we aren't entitled to submit just a wiki creation request regarding the proposal. Which of course I feel are all opposite.

The summary of the point I make here is do not expect to be a sysop on Miraheze Data if the proposal is successful. The potential and process for either of you two to gain permissions on the project should be identical to that of literally any other Miraheze user. On top of this, the process(es) for selecting the first sysops and bureaucrats on the project should be included in your proposal RfC.

In a nutshell, Since Miraheze existence, and AFAIK, no one has brought up a proposal regarding Wikibase Installation on Miraheze wikis in general. This very proposal should give chances to those really interested in supporting via wikibase to help miraheze out. Neither myself nor Luca is standing on the ground of being a crat or sysop on the wiki if it goes well. In addition, Although, aside wikibase administrations and the rest, there are also some administrative works on a wiki, like maintenance and the rest, but What do you think about having Sysops and Crats who little or no knowledge about wikibase and its maintenance?

I don't want to start saying much because it now seem like we're trying to defend a project that would stand for our own personal good only whereas it is not.

I am totally discouraged by the way things are being handled especially the way this particular project is being handled and it's making me very much uninterested in it again by the day.

No. This needs to be in your proposal. If a project is to be official as you desire, the community must establish (through the process of drafting and commenting on an RfC) the process and standards to be used to select initial bureaucrats, sysops, and policies.

I believe so much in DUE processes, and so I don't still understand how we should be discussing how bureaucrats and sysop and policies would come when the Initial proposal is not even approved yet.

You're not being tried of a crime so how does 'due process' factor into this?

I hold no personal concerns regarding your two's roles as proposers or participants on the project. However, neither of you are entitled to submit the wiki creation request nor receive any permissions or special status on the project if it is successful.

And here, you kept hammering on the fact that we (@Ugochimobi, @Lucamauri) are the proposers and that we aren't entitled to submit just a wiki creation request regarding the proposal. Which of course I feel are all opposite.

To be quite honest, I share dross' view in that neither of you should be submitting the wiki creation request. If anything, the closing Steward should in order to effectuate the outcome of the Request for Comments. By one of you submitting the wiki request, you undermine the community by bypassing people we have elected to enforce consensus.

The summary of the point I make here is do not expect to be a sysop on Miraheze Data if the proposal is successful. The potential and process for either of you two to gain permissions on the project should be identical to that of literally any other Miraheze user. On top of this, the process(es) for selecting the first sysops and bureaucrats on the project should be included in your proposal RfC.

In a nutshell, Since Miraheze existence, and AFAIK, no one has brought up a proposal regarding Wikibase Installation on Miraheze wikis in general. This very proposal should give chances to those really interested in supporting via wikibase to help miraheze out. Neither myself nor Luca is standing on the ground of being a crat or sysop on the wiki if it goes well. In addition, Although, aside wikibase administrations and the rest, there are also some administrative works on a wiki, like maintenance and the rest, but What do you think about having Sysops and Crats who little or no knowledge about wikibase and its maintenance?

I'm sure the community will elect people who demonstrate competence in managing Wikibase. From this comment however, I do sense you distrust the community to do so.

I am totally discouraged by the way things are being handled especially the way this particular project is being handled and it's making me very much uninterested in it again by the day.

Lucamauri requested feedback and dross delivered it. I'm sure his intention was not to discourage you but even if it did, I would recommend you take this constructively and work to amend the RfC to address these concerns.

Now, this is Phabricator and an inappropriate venue to request feedback on things that affect the community. I ask that Ugochimobi, Lucamauri, and dross all move to the RfC or another venue to further discuss this instead of debating it on a Phabricator task that was closed over a week ago.