Using same data for both "ref" & "name" tag fields.

I have recently noticed quite a number of roads appearing where the road number (or ref) appears in both the name and ref tags. If the data in name is a copy of the data in the ref, I have been deleting it.
This was on the basis that I thought name was reserved for the actual name of the road, which in rural Thailand, is often impossible to verify as we all know the bigger rural Highways just don’t get names.
So, should I carry on ? Or stop and reverse the few changes so far ?

And what about this sort of hypothetical example :

ref=นค.1234
name=Rural Road Nong Khai 1234

I suppose it is the “name” of the road, but should we put it in that format ?
Of course, if the answer is affirmative, I might be asking Johnny to write another one of those scripts to rectify the entire country !!

Rgds, Russ.

I’m not sure what their motivation is for doing this but I do not think that type of naming convention is either proper or correct. There are many such named ways in northeastern Thailand. You should ask EndlessRoundabout about those as he is responsible for a large majority of them. User:dilwyn is another mapper who has done this but only in a few cases.

The example you posted is perhaps close to being right but most of the ways I’m looking at look more like this, with the “name” consisting of the exact same number as the ref:

Way 124745686
highway=tertiary
name=2281
ref=2281
source=Bing
surface=paved

My 2 baht

I’d say it depends. Many numbered highways do have no name and are actually referred to that way (Rural Road Nong Khai 1234), and I’d generally support entering the name tag as such if the fact is confirmed by surveys / local knowledge, since that would be how most people will expect the road to be called. That should be better than leaving the tag empty. But some roads do indeed have names, and it would be amiss to pre-fill their name=* tags with a formula value instead of first actively looking them up. I’d recommend against such mass edits.

Removed

I believe the tags were added by EndlessRoundabout, and though the tags are incorrect, they are somehow helping him while mapping. He assured me this was only a temporary solution.

I considered it a bit like the “name=Village” discussion we had previously and decided to leave it alone.

Removed

I think we can all agree on that, question is whether to allow it or not.

Removed

I wouldn’t go so far as reverting any changesets. If we think this is unacceptable, then a fix can be scripted rather quickly to remove the bad name tags.

But first we need to decide whether the data is completely unacceptable and needs to be removed now, or if it is a bit like the name=Village (and name=Ban ?) and can stay as long as some people find it helpful.

I haven’t been able to find my original communication with him where he explained why he did it, maybe we should start by asking why he needs it, and whether he still needs it.

EDIT: I just dropped EndlessRoundabout a message asking him to take a look at this thread, hopefully he’ll explain it so we can make the right decision.

We have to consider that while we (as the experienced mappers) tolerating this know well it’s wrong tagging it still serves as a bad example to all the newcomers.

If they see that it’s sufficient to put everything in the name tag, we’ll end up with a map full of inconsistent tagging. OSM has tags to describe the features. And the “name-tag-abuse” is a serious problem.

As we now have a quite powerful machine at hand: what is the exact reason of wanting to have the name tag duplicating the ref? Another case of a garmin unit not having a proper style? We can produce specific renderings if this is of any use to make mapping easier without manipulating the data.

I vote for removing the duplicate tagging. Rather sooner than later.

The name tag is for the name of a feature. “second ring road”, “super highway” in Chiang Mai might be examples of this. That could be name=* or loc_name=*.
But a ref is a ref. If people call a road by “this is highway 1001” than they refer to the ref. Not giving it that as a name.

OK, thanks for everyone’s input … I think Stephans last post sums it up, and I agree that its a bit heavy handed to just reverse changesets.
Johhny - A month ago, I tried the personal approach to Endless with this message~

Hi there,
Thanks for all your mapping in Thai.
One question, however…I noticed a few of your roads are being tagged like in this example : นค.5020 (2005) both as ref=
and name=* Firstly, I was under the impression that if the road reference was นค.5020 then this should only appear in the ref= tag. The name= tag would be reserved for its actual name, such as “Payolothin Road”, “High street” etc etc.
Secondly, I also wonder what the (2005) means ?
Finally, if you are unsure of a road ref, please feel free to add the reason why using the tag, note=. That way other Thai mappers will be inclined to help u out and clarify. A question mark in the ref= is not really the way it should be done, as I noticed in another nearby road you called “6020 ?” I tend to add the source=sign tagging if I have been there and physically seen a road sign.
I trust my comments are helpful, and carry on the great work. Russ.

Maybe you will get an answer, but I never had the courtesy of a reply, so on that basis, if we have no input back from him, then we must decide. So :
1/. In the short term, when I come across them, I will delete the name when it duplicates the ref, and suggest others do the same.
2/. Can someone write a script to do the above automatically ?

Rgds, Russ.

I can do that, but I’ll wait a few days to see if ER gets back to us.

I suggest a manual fix by reviewing every entry in JOSM. This reduces the risk of breaking something else.

The case of using name=Village or name=Ban? that we discussed happened a while ago and the map is now littered with these “temporary” name tags. I didn’t like the concept then and I don’t like it now. I think we should strongly discourage such practices. Once you give the go ahead for these exceptions, you end up with many more poorly tagged items. As Stephan correctly points out, and I agree, it sets a bad example for new mappers. Not only that, it sets a bad example for experienced mappers as well.

Dave

PS: I’m back in Thailand after a 3 month long stint of mapping in Alaska. Hoping to get together with some of you Chiang Mai locals before too long.