Proposal talk:Move proposal voting from wiki to the new forum

From OpenStreetMap Wiki
Jump to navigation Jump to search

Separating discussion from voting


I agree that the barrier of entry for voting on the wiki is too high. Having a separate wiki account doesn’t really encourage people to participate. That said, the number of editors that participate in forums and on mailing lists is already a significant minority of overall contributors. This proposal wouldn’t fix that.

However, I’m concerned moving the voting into a different forum will inevitably fracture discussion. By de-emphasising the discussion proposals will be more akin to a simple voting exercise, rather than a genuine attempt to gain consensus with the community. Diacritic (talk) 08:55, 24 September 2022 (UTC)

This is indeed important to consider but I doubt it will be much different from the current situation. Current proposal discussion happen on various places like the talk page, mailing list, current forum, discord etc. If you properly link the wiki/talk page and other discussion near the poll, people can access them. I can change the template to bether emphasize to take to to read the proposal and discussion before voting but it is not very different from the current situation in my opinion--Cartographer10 (talk) 12:53, 24 September 2022 (UTC)
In an ideal world, the voting would be on the same platform as the proposal with an easy to use interface and no separate login required. Removing the separate login barrier is a priority as noted below. People have also brought up the idea of installing a different MediaWiki talk page plugin that would improve the discussion/voting interface. However, neither of these things has happened yet. What has happened is the launch of with OSM login. Despite it's problems, it has a much more usable interface for discussions and voting than this wiki currently does. That could change in the future, but as it stands I think moving voting to the platform that currently has fewer barriers to participation is a good idea. That said, I think this proposal is a bit too early. Some testing and examples are needed to show the community how voting on the forum would actually work, and how it would be linked back to the proposal page. -- Ezekielf (talk) 14:53, 24 September 2022 (UTC)

Logging into the wiki with OSM accounts

That is something that is very much planned and being worked on, see Top Ten Tasks#OAuth login to wiki.

So I don't think that the rationale of the proposal should focus so much on the account barrier since it can be addressed without migrating the process to the new forum.

--Push-f (talk) 09:29, 24 September 2022 (UTC)

@Push-f: Thanks for this addition. I added that an oAuth login is being worked on. Is there already a date of when this is implemented? If more people can participate, are more modern, less conflict sensitive system is needed though. I put more emphasis on that. --Cartographer10 (talk) 09:35, 25 September 2022 (UTC)

Requirements for a voting system

I would postulate the following requirements:

  • you need to be able to attach a comment to your vote (and for "no" votes explaining why should be mandatory)
  • the proposal author(s) should be able to respond to such comments
  • voting has to take place on the same page as the proposal (so that people actually read the proposal and it's clear what is being voted on)

All of this is possible with the current in-wiki voting system. As far as I can tell Discourse's voting system doesn't support any of this.

--Push-f (talk) 09:29, 24 September 2022 (UTC)

I agree with all of this. Additionally, it comes handy that the current system is very flexible, you can change your vote should you have changed your mind during the discussion, and all of this is visible to everyone. And it is not just the proponent, also other participants can reply to comments during the voting phase. --Dieterdreist (talk) 15:16, 24 September 2022 (UTC)
I also fully agree with Push-f. --Martianfreeloader (talk) 13:01, 26 September 2022 (UTC)
The points you address are indeed important requirements. I do think however that this can also be achieved with Discourse:
  • The current system on wiki is also not fool proof. During some of my proposals I saw plenty of oppose votes with no reactions explaining their vote. Votes on Discourse can be made public so you can see who voted for what. You can then see who commented or nor, elaborating their vote. Another feature of Discourse are thumps-up reactions. With those you can indicate you agree with an already given reason, removing the need to re-iterate an already given explanation.
  • A proposal author can reply to the comments.
  • I understand your point here but you have to do your due diligence before a vote. The current wiki system does not prevent you from just scrolling to the bottom without reading and indicate your vote. As explained in the proposal, the topic template includes a clear link to wiki so you can see what is voted for. I can update the template if you have suggestions for this.
--Cartographer10 (talk) 09:26, 25 September 2022 (UTC)
By attaching comments to votes I meant that comments are displayed directly next to the respective vote. Discourse cannot do that. Furthermore Discourse does not keep a history of vote changes, which I consider quite important. If you can just change your vote without anybody noticing that results in a very intransparent voting system. I do not consider Discourse's post reactions to be relevant at all. Also the ability to link the proposal does not satisfy the criterium that voting takes place on the same page. --Push-f (talk) 09:30, 8 October 2022 (UTC)

Fix the root of the problem

I think:

  • Anyone who's tech-savvy enough to make a significant contribution to OSM can figure out how to create a wiki account.
  • The proposal author usually puts a lot of effort into creating the proposal, replying to comments, editing their proposal accordingly, and so on. Compared to this, it's minimal effort to copy-paste a template line for your vote. If this is too much effort for someone, then it's likely a sign they're not sufficiently interested in the proposal. I'd rather have 15 people vote on a proposal who really spent some thoughts on it than 100 people vote on it who merely read the headline, thought "don't like this", leave their one-click vote and disappear.

Should the community disagree with the above, then rather than further fragmenting the decision building process (wiki + mailing lists + forum) because of existing technical issues, I advocate fixing the issues at their root:

  • Make it even easier to create a wiki account.
  • Implement an easier in-wiki voting system (make sure people can and must comment on their no-votes).

--Martianfreeloader (talk) 13:19, 26 September 2022 (UTC)

@Martianfreeloader: thanks for the feedback. Point 1) It is not always about not being able to create but also about actually wanting to create (yet another) account.
2)It is not really about the copy paste exercise but more about a clumsy UI to vote. It prone to edit conflicts (edits at the same time), not mobile friendly. To answer point 4) True but since this is not happening, it makes sense to move to a platform that already has a voting system that solves these issues. Point 3)They are working on OSM login for wiki. It is not clear when this is happening though.--Cartographer10 (talk) 17:01, 26 September 2022 (UTC)
I think there's been minor discussions about improving the UI less clumsy, more mobile friendly, Etc. Etc. but from what I remember (it's been a minute) they haven't and probably aren't going to ever go anywhere. These aren't things that things that even Wikipedia has figured out either. Voting on proposals has the same issues there that voting for them here has despite their vastly superior resources. Really, this is kind of an archaic, outdated system in a lot of ways. One that isn't likely to improve. I don't think the answer to that is requiring people to leave in-depth vote rationales either. Since there's zero way to enforce it. Plus, that would kind of be putting a band aid on a bigger issue instead of just solving the bigger issue, that is to obtuse and inefficient of a system for most people to want to participate in it beyond making superficial comments. If they even do in the first place. So, why not modernize? --Adamant1 (talk) 12:40, 3 October 2022 (UTC)
Who says that an easier in-wiki voting system is not happening? Talk:Wiki#Introducing a more user-friendly UI for proposal voting via an extension. --Push-f (talk) 08:42, 7 October 2022 (UTC)
It's good there's movement in that direction. But there's plenty of half-baked ideas around here that never get off the ground, or if they do it takes multiple years for them to implement. Even when it comes to extremely important changes. As an example there's the discussion/PR around making a unique namespace on the Wiki for proposal pages that hasn't gone anywhere in 8 years despite being a pretty critical improvement to the Wiki's search function IMO. Given that and many other examples, I don't think the existence of something that might happen in the future (but realistically probably won't at least not anytime soon) should hold up current proposals to improve things. Especially in this case where there's other reasons that the Wiki isn't an optimal platform for voting on proposals besides problems with the actual voting. You can either endlessly patch an old tire until it's shredded into nothing or you can buy a new tire. It should be obvious here what the better one of those two options is. --Adamant1 (talk) 10:33, 7 October 2022 (UTC)
The Proposal namespace has recently been introduced. And yes I think that in-wiki voting is obviously better (even though I know that's not what you meant). --Push-f (talk) 14:34, 7 October 2022 (UTC)
The last I checked the PR for it hadn't been merged yet. Although it was a few weeks ago. Cool if it has been. That doesn't negate the fact that it still took an inordinate amount of time to be implemented. In this case, I'm sure the "easier" voting system will be implemented eventually. You can't tell me it will be implemented immediately though. IMO it's worth weighing the benefits of waiting versus implementing the other options in the meantime. It's not like it's an either or thing anyway. You can make it easier to vote on the Wiki in 15 years if that's what you prefer. I don't think it means everyone else should have to wait that long for an improved voting system when there's already another option available that wouldn't involve waiting endlessly for you to implement a band aid on what's already an extremely outdated, obtuse system to begin with. As far as if in-wiki voting is "better", I haven't seen any evidence that it is. Mostly just appeals to tradition.
In the meantime, multiple people in this discussion agree that barriers for entry for voting on the wiki are too high. I'm sure you'd at least agree that making the "easier in-wiki voting system" is only one aspect of that. So how many barriers and how much time/effort would it have to take to fix them before your willing to say the people should use the new forum? Or are you just categorically against the forums? Like say I get back to you about this in a year and it's still at exactly the same spot. What then? How about 4 years? How long should it take for the band aids and how many people should be turned off doing proposals in the meantime for you to be willing to say proposals should be moved to the forums? --Adamant1 (talk) 00:28, 8 October 2022 (UTC)
Thanks for your concern. I do agree that removing such barriers is an important matter and should be done as soon as possible. However I do not think that the noble goal of removing barriers is worth the sacrifice of fundamental features, such as being able to attach comments to votes and being able to reply to specific votes. And no I do not think that getting an extension installed takes years, e.g. the PR to add the calendar extension was merged within 51 days, and the PR to add the CodeMirror extension within 2 days. --Push-f (talk) 09:51, 8 October 2022 (UTC)
I haven't look into it that much but I assume people can attach comments to votes and are able to reply to specific votes by just writing a new message and pinging the person you want to respond or whatever since that's usually how most forums work. While I'd agree that's as optimal as what the current system allows for, 99% of no one uses features as things currently are anyway. Like the only current proposal being voted on, Proposed features/Privacy, has three votes and only 1 person commented. They could have just as easily made the comment in a separate message. To give another example, Proposed features/Mining, 1 person out of 8 disapproved it. Like two of the people who approved left comments, but it's not like their comments ultimately matter since it was approved either way.
More broadly IMO for the comments to matter, or at least be useful enough to go against voting taking place on the new forum, you'd have to prove that they are ultimately taken into account or otherwise matter specifically in relation to how the person is voting. At least more so then someone just leaving the same message in separate message on the forums. At the end of the day if I vote to approve something and it's approved then what does the message I leave with it ultimately matter? Really, my guess is that in most cases comments are either ignored outright or have very little impact on the outcome of the vote if at all. It's not like there are official people who close the proposals while weighing the strength of every argument some one makes or whatever like they (barely) do on Wikipedia. --Adamant1 (talk) 04:55, 9 October 2022 (UTC)

Proposals are about wiki documentation, therefore should be on the wiki

The whole point of a proposal is to find out if there is a community consensus that a new tag or key is a good idea and if the wiki documentation is well developed. The only effect is that the authori of the proposal then can mark a tag or key as "approved" on this wiki, and in rare cases to mark a currently used tag as "deprecated" if there is community consensus. Anyone can document the use of a tag which is being used, and so any proposal author can start to use a new tag without going through this step. In fact, the vast majority of new tags are documented as "in use" and never go through the proposal process. Since the entire proposal process is about approving wiki documentation, it should happen on the wiki. --Jeisenbe (talk) 20:09, 9 October 2022 (UTC)

While I agree that the "process" in general is about approving wiki documentation and therefore should happen on the wiki, most of that is process is done in the RfC phase. IMO the actual "voting" is ancillary to that and therefore it doesn't really matter where it takes place. Not all proposals are created, fleshed out, or "voted" on through the Wiki/mailing list anyway. For instance there was recently a "proposal" to re-tag highways in California that was mainly fleshed out and agreed on by people in Discord. When I found out about it I asked a member of the DWG to revert the edit since it wasn't approved or discussed through "official" channels. I was subsequently accused of bullying and told to stop harassing users because proposals can happen other places then the Wiki/mailing list. So there is no hard and fast rule about this to begin with. It's mainly just a matter of using whatever platform will reduce the barriers to participation that have already been talked about. It's not like someone can't just copy and paste whatever changes are suggested on the forums into the page of the proposal either. They already do that when people on the mailing list suggest changes to documentation. --Adamant1 (talk) 06:14, 10 October 2022 (UTC)

Decisions on Openstreetmap are made by consensus, not by voting

While we do vote on proposals, this is simply a way of finding if there is a rough consensus in favor of a new tag or change to the wiki. That's why there is a 75% supermajority recommended now for marking a proposal as approved: if only 60% of commenters are in favor of a proposal, there are probably reasonable concerns with the proposal. The process right now is focused on discussion and improving the proposal and addressing any problems or concerns that come up. By moving voting to a different forum, it would make it less clear that the process is supposed to achieve agreement and improvements rather than leading to an adversarial, voting-based popularity contest. --Jeisenbe (talk) 20:09, 9 October 2022 (UTC)