[OSM-dev] Implied Tags

[Igor Brejc]

Hi,

>From my experience with OSM tagging, you will have to implement some kind of
rules-based system in your app which would have to be easy to amend as you
discover more and more peculiarities of free-style OSM tagging.

Like Frederik says, don't rely too much on OSM Wiki, there's just too much
data there for all users to follow and a lot of people tag things not just
based on what wiki says, but from other sources too (TagWatch, OSM help,
JOSM presets etc.). These sources are not necessarily synchronized, so
expect a lot of inconsistency, regional preferences etc. Wiki is a good
guide, but it's not a rulebook.

Igor

On Tue, Apr 12, 2011 at 9:14 PM, Frederik Ramm <frederik at remote.org> wrote:

> Hi,
>
>
> Christian Vetter wrote:
>
>> And the implied tags noted in the wiki pages?
>>
>
> Like anything else in the Wiki, these are just put there by some people who
> thought it might make sense; there is no guarantee whatsover that most
> people even know, much less make use of them.
>
> To the best of my knowledge there's no editor support for implied tags,
> i.e. when editing you will not see what you have just "implied" - for
> example, JOSM will not automatically draw one-way arrows onto something you
> have tagged highway=motorway.
>
>
>  So what would be your advice? Just ignore implied tags? They seem to
>> be used in many cases ( e.g. barrier vs barrier=bollard ) and I want
>> to avoid "fuzzy" parsing.
>>
>
> I don't think there will ever be a true, fixed set of implications, but
> there may be a growing community consensus of implications for one specific
> purpose. (I could imagine that implications for rendering would be different
> from implications for routing, say.)
>
> My advice would be to make up a set of implications that makes sense for
> your purpose and make sure that every user of your software knows about it.
> Perhaps even put it in a config file or a wiki page; it might even happen
> that others start using it. But I would always start from the bottom up
> ("this is what *I* am doing right now, if anything is wrong about that,
> let's talk") rather than try to do something from the top down ("let's all
> find a consensus about implied tags now so that I can then build the one
> true rule into my application").
>
>
> Bye
> Frederik
>
> --
> Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"
>
> _______________________________________________
> dev mailing list
> dev at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/dev/attachments/20110412/b50e278b/attachment.html>