Long, prickly post ahead. On Dec 24, 2020, at 7:35 AM, Anders Torger <and...@torger.se> wrote: > By not rendering valleys and peninsula and possibly other tags often used on > "non-verifiable geometry" it's a signal to mappers that those are not > considered important or desirable. We could say that we shouldn't care about > what OSM-Carto does, and for advanced users that have their own renderers > that makes some sense, although it risks further increase fragmentation in > mapping methods.
False. This is pure speculation. OSM asks that Contributors enter on-the-ground-verifiable data into the map. Period. This is what I mean by "map well." If you get a rendering that pleases you, great! If you don't, "at least" the data (accurate, verifiable — to the extent they can be verified — I agree "verifiable" can be slippery at times) are "in OSM." I put "at least" in quotes not to diminish its importance, but rather to emphasize that THIS IS THE INTENT! There is no "signaling" by renderer authors "tag like this so you see in renderings what is considered important or desirable to map." Big "no," right there. Do renderer authors render data which ARE entered into OSM? Sure. Period. Full stop. Anders, please: DON'T TAG TO ACHIEVE PARTICULAR RENDERINGS! I applaud good dialog here about "fuzzy" and natural-naming, as those are interesting, even necessary discussions. But these value-judgements you make like above are specious and even pose danger to OSM should they gain traction. My passion against this shines brightly here, as I don't want to see that: it drives the desire for particular renderings to guide what we map. No! That's not what OSM is! Entry of good DATA is the goal, RENDERINGS are a nice-to-have, particular expression of them. OK? > But how many of us mappers have our own renderers? A few on this list have it > of course, but broadly speaking it's probably less than 1% of all mappers. > And if you don't have your own pipeline or is simply interested in the free > end products that anyone can access for free all over the world, one have to > take into account what OSM-Carto does. Continuing to complain that OSM doesn't render as you like (or that few have or can or do make renderers, or that making a renderer is challenging...) further illustrates the distance of your understanding what of OSM is. OSM absorbs good geographic contributions of volunteers (like you, me and readers of this list). This list absorbs questions about tagging and improving strategy for better tagging and possibly results in improving or making tools or better process for achieving that. It strives to offer answers, better strategies and guidance. Getting hung up on how our tagged data render seems it needs a different list like OpenStreetMap-Carto, where "renderer discussion" takes place and where people might have far more patience with you than I do (I've extended a great deal here with you). You don't seem to discuss tagging (the point of this list), you seem to repeatedly, exhaustingly, frustratingly-to-many-here complain about rendering. I ask you explicitly: please stop. > Even if the mappers community end up with a consensus to map in one way (or > at least a consensus that it is an okay alternative), and those in charge of > OSM-Carto choose not to render it, then it's really not working out... > because OSM-Carto is the only renderer that can represent "the community". You do realize that OSM is about the data that we put into a database, right? I distinctly tire of repeatedly making this point to you. Repeated complaining about particular renderings is truly unwelcome traffic here. If you wish to discuss better TAGGING strategy IN THE CONTEXT OF a renderer, one where you do not seem to act like a child who wants to be given a cookie of your particular rendering desire, that's one thing. But that's not what you continue to do here. I believe I have been kind and patient with you, but patience is finite: I repeat myself that you seem to entirely miss the point of what this list (and even OSM as a project) really is. Better the data, better the tagging, better the strategies / tools / documentation, yes. But repeatedly complain? You get me who says "stop that" and others who support me. As you explicitly state "it's really not working out" (for YOU, not OSM): OK, see you later and thanks for coming. I don't like being harsh, but we're going around and around, yet you persist. Stop. You are being more than Kevin kindly put it, "a bit confrontational," you go far beyond that. > I know many think that we should not care about rendering at all, and the way > to see our own work is to download it in JOSM and enjoy the geodata objects > we've made, as OSM is supposed to be a geodatabase, not a geoservice > provider. I don't think that is how the typical mapper see it though. OSM cares about renderings: we offer four on our map page as a start of our data's capabilities. OSM certainly improves, though pipelines / feedback loops for renderer improvement are complex and sometimes lengthy. Evidently you don't understand this. I don't fully understand it either, even over many years of email conversations with renderer authors — and I believe the vast majority of OSM Contributors don't understand it. But the point of OSM is the data, not renderings. Repeating myself, the easiest way for OSM to salve the sort of boo-boo you seem to have is to say "don't tag for the renderer" (and OSM doesn't often say why, or what complexities are behind that). But as you repeatedly pick at this scab (as a long-time Contributor, I say OUCH!) we evidently have to bleed all over — again and again — with exhortations like these between us in a public forum. I sincerely wish this weren't so ugly, Anders. What part of "please stop complaining that renderings aren't to your liking" do you not understand? OSM has a problem (I'd say moderate-sized) in that sometimes people tag to achieve particular renderings. We shouldn't, as this results in poor, usually incorrect data entering the map. To remedy this, we say "don't tag for the renderer." You say "there's a bunch of differing opinions on how we should map and no uniform decision can be made...". Anders, I assure you, despite your attempts to gaslight us into believing otherwise or redefining the project, it is an ironclad, uniform decision of OSM that we shouldn't tag for the renderer. It won't work to continue complaining. Do I make myself clear enough? I have little energy and no desire to continue to defend against this list being baited or further misunderstandings of persistent, dead-end, ineffective approaches. Anders, if you want to do something constructive on this list, I wish you would (following its established tenets, which I summarize above), but what you have been doing and how you do it (over and over again) isn't it. Your point that "typical mappers" see things a certain way contradicts that OSM's data entry continues to grow, without them complaining that renderings aren't to their liking. I'm reluctant to say it: be constructive (ask relevant, judgement-free questions, offer relevant perspective...), or be gone. I'd like to see this list extract some value from this discussion (I'm going to go take a refreshing shower). I hope we can gain some value from the topics discussed. In my opinion, we should not dwell on the mechanics of what has happened here, but rather its potential fertility, rather than its actual futility. SteveA _______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
