[Tagging] A system structured approach rather than piecemeal
61sundowner at gmail.com
Sat Jun 9 23:30:17 UTC 2018
On 09/06/18 21:53, Thilo Haug OSM wrote:
> I think what Mark wanted to express
> is more about some general guidelines which could be established.
> The tagging systems are very inconsistent, just have a look at the
> different "service" syntax (namespaces)
> of a car vs. bicycle vs. motorcycle shop (and none for others).
> There could be a system of different levels of discussions, for example.
> I'm personally not keen on reading every single mail of every topic,
> but there's no possibility to choose some topics.
> To stay with the "shop" example :
> There should be a consensus first what all shops have in common,
> THEN the single shops may be discussed, not bottom-up.
Agree whole heartedly! Add sub tags to individual shops is not the way to go, there needs to be a common system for all shop values.
> Same for the namespace syntax.
> It would be good to have some simple voting system
> without having to edit the page code.
> So when people are able to choose their topics
> and an easy way to vote, more will participate
> and the "consensus" of currently some dozens
> which are willing to withstand the current mess
> would be on a broader base (really democratic).
> So let's discuss the possible structural enhancements
> instead of presuming political opinions
> (he didn't really mention "strong leadership, did he ?)
> Ideas ?
One of the problem is that these main tags don't come through the tagging group .. they arrive through common use that sees a demand and satisfies it.
A problem with that is that the initial users see only their local issues and don't see it on a global scale ..
so it gets used in ways that were not part of the initial use/intent.
Tight definitions are hard to do when you only see what you are wanting to map, not seeing the wider picture.
Thus we have a mess to fix up.
A structured system is far easier to learn/teach and comprehend compared to a system with lots of exception.
Look at the English languages an example of a system with lots of exceptions!
If the OSM structure is made into a systematic structure then OSM will be much better off.
Of course there will be continual deviations away from such structure give the free tagging approach,
but with time these can be adapted into the structure.
> Am 09.06.2018 um 02:22 schrieb EthnicFood IsGreat:
>> Date: Fri, 8 Jun 2018 08:29:25 +0200
>>> From: Frederik Ramm <frederik at remote.org>
>>> To: "Tag discussion, strategy and related tools"
>>> <tagging at openstreetmap.org>
>>> Subject: Re: [Tagging] The endless debate about "landcover" as a
>>> top-level tag
>>> it is a gut reaction by people when forced with difficult issues to call
>>> for strong leadership to solve them once and for all. OSM is no
>>> On 08.06.2018 01:29, EthnicFood IsGreat wrote:
>>>> I wouldn't mind if all the existing tags were replaced tomorrow with a
>>>> brand new set of "intelligently-designed" keys.
>>> Designed by... a visionary leader? A board of experts? A random draw?
>>> And if something turns out to be designed wrongly, how will it be
>> Of course any system would have to have a means of making revisions,
>> as better ways of tagging things become apparent over time. There
>> could still be innovation.
>>>> And I wouldn't mind if
>>>> these keys were enforced from now on.
>>> Not having an enforced set of keys and values was definitely a big part
>>> of OSM's success (there *were* competing projects which got stuck trying
>>> to define the one true set of keys and values that would work for
>>> Some people say that while this may be true, the time has now come to
>>> get rid of the old ways that got us where we are, and change tack to
>>> something more conservative. This is a valid argument but I am not
>>> convinced; a lot of innovation is still going on with tags, and strict
>>> enforcement would run the risk of killing that.
>>>> Someone some time ago on one
>>>> of the OSM mailing lists summed up the current situation by stating,
>>>> seems OSM is incapable of moving forward."
>>> OpenStreetMap is becoming a larger group of more diverse people with
>>> more diverse interests, and since we don't - and don't aim to - have a
>>> dictator at the top, things need to be done by consensus. These people
>>> who take to the internet complaining about how OSM is incapable of
>>> moving forward usually are people who are unwilling, or unable, to
>>> convince the "great unwashed" their idea of "forward" is a good thing.
>>> So they lament the lack of leadership and complain about "gatekeeping",
>>> but it's really just them being unable to do the work required to
>>> establish consensus in a large project.
>>> Because that takes much more than a couple of blog posts (cf. the
>>> license change).
>> I have been editing in OSM for almost four years, and I've been a
>> member of this mailing list almost since then. I read every single
>> post. During that time I have never seen what I would consider a
>> consensus reached on anything. I'm not sure it's even possible.
>> Whenever someone proposes a way to tag something, you can be
>> guaranteed that people will bring up every possible angle and nuance
>> concerning the meaning of the tag, and nobody wants to compromise.
>> Consequently there is never a consensus. Eventually people get tired
>> of the debate, when they see it's a no-win situation, and the debate
>> just dies away, until somebody brings it up again next year. Case in
>> point: the current issue of landuse versus landcover. There was no
>> consensus the last time this was brought up and there is none now.
>> I've seen several tags debated more than once in four years. I can
>> only assume that each time, a different group of people get drawn in
>> to the discussion, unaware that the issue has been debated before,
>> with no resolution. This cycle is doomed to repeat itself over and
>> over, as long as OSM proceeds the way it is. A waste of time and effort!
>> I don't see how OSM can work well when mappers are free to tag however
>> they want. Different people have diametrically opposed ideas about
>> how things should be done. For example, some people think the meaning
>> of a tag in OSM should be the dictionary meaning of the word; others
>> are okay with a tag word having a "special" meaning in OSM. How is a
>> mapper to decide? There is no consensus on this issue. Although OSM
>> has a policy of "any tag you like," based on the posts I've read, it
>> seems most mappers want some guidance when it comes to tagging. I
>> deduce this from all the posts I read from contributors having to do
>> with editing and refining the wiki. However, there isn't even
>> agreement on the purpose of the wiki.
>> It seems there are basically two camps of OSM mappers---those that are
>> fine with the way OSM is currently structured, and those that want
>> more structure. If OSM doesn't change, another thing is guaranteed to
>> not change as well: endless and pointless tagging debates, and never
>> a consensus.
>> Tagging mailing list
>> Tagging at openstreetmap.org
> Tagging mailing list
> Tagging at openstreetmap.org
More information about the Tagging