On Sat, Nov 5, 2022 at 6:45 PM Robin Burek <robin.bu...@gmx.de> wrote:
> And if we now get to the point of just "throwing away" the consensus of 12 > years ago. > > do we still need the proposal process at all? Because the result from 12 > years ago is also completely ignored by you. > it was already decided to deprecate in 2010, but no one has finally > implemented it. > What you are discovering firsthand, are the limits of the proposal system. Indeed I encountered the same challenges. A proposal compels nobody to do anything. A proposal is nothing more than a communication tool to demonstrate support for an idea to other players in the community. It's a signal to other parts of the ecosystem. In other words, if the participants thought something was a good idea a decade ago, but the rest of the community (mappers, renderers, editors, data consumers) didn't adopt the change, in reality, then the persuasive value of that approved proposal from 12 years ago has faded significantly. In this community, we seem to be moving further and further into a system > where improvements to the system are massively prevented and established > double tagging is simply left in place instead of finally being cleaned up. > I don't agree this has "moved" at all! The project has always operated this way. Eliminating duplicate tagging that has been supplanted by a newer approved tag is obscenely difficult. I led an effort to resolve duplicate river area tagging to 99% completion[1]. which was also a change that was the subject of an approved 2010 proposal. Despite the approved proposal, it was still controversial, with some community members disagreeing about exactly what was approved due to how the proposal was worded and debating whether the old approval was still valid or even a good idea. The river project accomplished its goal because enough mappers cared about it to have community discussions about river tagging in countries worldwide. They reached out across many channels to discuss the proposed changes and the value it brought. And even with that, some people disagreed, and we had some rough spots early on in the process. The retagging was followed up with proposals and PRs to change software support across the project's tooling to drop the old tag. I was the biggest advocate for the change, but it only happened because I was met with strong agreement. Building consensus is hard. I had to write really clear and persuasive documentation. I had to discuss the specifics of river tagging with many many people. Those people talked to other people. Other mappers generated statistics, procedures, and visualizations of our progress. At one point, I even gave a "Mappy Hour" talk[2] on the project. If that sounds like a lot of work -- it was! And just for a single tag! But THAT is the scope and scale of effort that it's going to take to change tagging that has tens to hundreds of thousands of objects tagged. You need overwhelming agreement AND enough support from motivated community members to make all the pieces of the ground game come together. Is this an indication that our community is dysfunctional? Maybe. But it's 100% the reality that we live in if you want to accomplish wide-ranging change. This, by the way, is one reason why certain companies refuse to use OSM > data. The data structure is unnecessarily inflated and complicated by such > duplications. If we don't stick to our own conventions and enforce > consensus, perhaps the consensus process should be abolished altogether? > This point would be much stronger if you could point to a specific company that refused to use OSM data. I've asked for concrete examples about why our free-for-all is a problem, but I always seem to get hand-waving instead about the general benefits of standardization[3], which is the reason I've submitted a question to the candidates in the OSMF election to see where they stand on it[4]. While I don't like duplicate tagging, I have so far not seen a convincing argument that it's particularly troublesome, and this is speaking as someone that's built and operates a service that uses OSM data. If you want to propose tagging for something that's never been mapped before, a proposal is a great way to ensure that the tag you're making up is reasonable. If you want to make a change of significant scope and scale to tagging on the project, you must understand that a proposal is only a single tool to generate support for your idea, which must be part of a broader effort towards consensus-building and community action. [1] https://wiki.openstreetmap.org/wiki/WikiProject_Waterways/River_modernization [2] https://www.youtube.com/watch?v=c5YwXDKGr2Y [3] https://lists.openstreetmap.org/pipermail/tagging/2022-October/066238.html [4] https://wiki.openstreetmap.org/wiki/Talk:Foundation/AGM2022/Election_to_Board#Question:_Tagging_Standards
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
