Andy Robinson (blackadder) schrieb: > While we are talking about tags I want to thank all those who take the time > and effort to put up proposals and make comments on them. The process is > very useful in seeing what users need and wish to tag. Personally though I > feel the process is more complicated than it need be, so I have some ideas > to simplify it without loosing the ability to propose and comment. I also > believe that we can learn a lot form the tags that are in the database > already. Not all tags need to a proposal/comment process if they are logical > and users simply get on an use them. > Unfortunately, because the proposer thinks it's a logical thing doesn't mean it's in fact a pretty good idea - if you take a look at the proposed features page, there are lot's and lot's of proposals lying around for quite a while that seemed like a good idea first (at least the author was motivated enough to write a wiki page about it) - so I guess about 2/3 of the proposals started gets *not approved*!
The classical problem is that a mapper tags a thing that he's very interested in (e.g. http://wiki.openstreetmap.org/index.php/Proposed_features/Red_Cross), so he tags it with a specialised tag. With some more thinking a much better tag could be invented that fit's very well into a lot more cases. Otherwise we can easily get into "tag fragmentation" with a lot of very special tags being similiar but not identical to each other. About your "tags from the database". What I would really like to see is something like the tagwatch page and getting the results on a weekly basis. Or a really cool thing would be some automated way to get: "the top hundred tags not already in the map features". > As many of you know I'm not fond of the voting process. I see why some feel > it's important but I'll always maintain that 6 votes of approval for a tag > is not a quorum of 2,000+ editors Well, if 6 voters - who care a bit about the topic - says yes (and noone says no), it gives you a good idea that this porposal makes sense. Not to say that about 1900 of the 2000 editors will simply never do anything about voting (maybe they're not even registered to the mailing list!) ;-) > so we should never blindly assume that an > "approved" tag is the best way of doing things. > Yes, I'm even pretty sure about this, but what's the alternative? > I'm also not in favour of "deprecating" tags, apart from this being wholly > the wrong word in the first place (obsolete is better) because this implies > that there was something wrong with the original tagging. The obsolete tags > are still just as valid as information so let's not blindly disapprove of > them by using the word "deprecate". > This is nitpicking. Call it "obsolete" or whatever - I don't really care unless the name makes sense (and I guess most others also don't care). Let's call them obsolete from now on - please just go on and change the page name ... Important to me is that by reading the map features page you get a good feeling how to tag things. This obviously means that inconsistencies and duplicates needs to be avoided - as this makes understanding a lot more difficult than it already is ... Regards, ULFL _______________________________________________ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
