[OSM-talk] Validator tags

[Patrick Kilian]

Hi,

>> noname and validate are not really the same thing. "noname" say that
>> something does not have a name. The validate namespace is a lot more
>> universal in its goal. 
> 
> Exact, but I was making a comparisson between 
> * higwhay=residential
> * noname=yes
> and
> * higwhay=residential
> * validate:residential-without-name
> and
> * higwhay=residential
> * internal=noname
> 
> what are the difference between those three ?
>From my point of view internal=noname and noname=yes both tell me that
the street has no name. validate:residential-without-name=ignore tell's
the validator not to highlight the fact that there is no name.


> That's what I don't like in the validate: namespace
> An object having no name is a bit more than just "disabling all validation 
> tests" it's also an information in it's own, and I prefere proposition key 
> that carry this information and validation tools taking action based on this 
> information.
Where it is possible I prefer keys that carry information too. So in
this case I'd tag stuff with noname=yes (or possibly internal=noname).
But I want a _consistent_ way to do that for _all_ of the different
checks. Therefore there is a validate:residential-without-name=ignore
too, analogous to validate:bridge-or-tunnel-without-layer=ignore.


> ( Maybe those discussions might better fit on the wiki's validate proposal, if 
> you don't mind using the wiki for that )
Actually I'd prefer the mailing list.


>> At least for the noname case the discussion has been going on for quite
>> some time and now consensus was reached. 
> did you said "now" ?? Well, the "w" might be wrong there, no, I don't see a 
> consensus here, neither a complete proposal, neither a vote on it, so maybe 
> it's time to say "ok we now need to agree", but the "noname" page is a global 
> page with many schema of wich I don't know wich to use
Sorry. That "w" was a very bad typo.
I'm not believing that there is ever going to be something like
consensus, so I just went and implemented one sensible solution. If
another tag emerges as consensus it's easy to replace noname=yes with
that. (Way easier that to find out about maplint by yourself.)


>> And so I went with the age old 
>> pardigm of crowd sourcing "Just do it".
> So, let's break with this paradigm and make it :
> propose it -> do it -> re-talk about it -> change it -> put it on the map 
> feature -> re-do it
With that attidute we'd still have blank map and would be discussing if
it should be "highway", "way" or "street".


> because the first "do it" has at least been done by you and me, but not based 
> on the same keys, and that's bad
Yeah. It didn't work out this time.


>> It's documented on in the commit message and the announcement on this
>> list.
> Could you please point me to the commit message ? are there anyway I can raise 
> comment against it ? will that be deleted by "commit gardeners" ?
> On that list I just saw your message saying "It is working like this" but 
> not "why is it working like this"
The commit message is in svn or at
http://trac.openstreetmap.org/changeset/12060


>> because I expect that it would get deleted
>> immediately by some wiki gardener anyway because I didn't wait for votes
>> first.
> 
> Then I would immediatly undo any deletation of that kind, it has to be 
> explicit, it has to be thought of, I would raise comment on it, we would 
> probably agree on some things, and then we would both implement the same keys 
> and save anyone pain.
That would end in an edit war, I neither want nor have time for.


> Looks you are an "against wiki" guy, while I'am a "for discussion" guy, and 
> because the wiki is the only thing I know of public enough, I'm using it. 
> This talk list is too hard to be usable by newcommers.
The problem is that the wiki is not dokumenting reality but what some
people would want to be reality. As a dinosaur I find mailing list much
more comfortable than wikis. Wikis I have to poll an search through.
Mailing list tell me when something new happens. And the tools for
handling many changes (say 100-200 mails per day) are so much better
than the stuff a wiki has to offer.


> But If you'r still against the wiki, I'have no problem continuing this 
> discussion just right here, but fear we are bothering others.
We could id either discuss offlist, or hope that grown up people use
grown up software and can tell their mail client to silently ignore this
thread.


> Let's start talking about :
> validate:residential-without-name
> 
> why restrict that to residential ?, what about :
> validate:noname ?
Because it is a single test in maplint. For the case of unnamed POIs
there is validate:poi-without-name=ignore. (Which is also shut up by
noname=yes)



Patrick "Petschge" Kilian