[Osmf-talk] Tagging standards

[Martin Koppenhoefer]

sent from a phone

> On 20 Oct 2022, at 01:57, Craig Allan <allan at iafrica.com> wrote:
> 
> To avoid the same tag identifying different features we need to collectively set just one rule. "No two different mapped features may be identified by the same tag"



this is assuming we all know what a feature is and how it is defined/which are its core properties that must not miss. This is not so often the case on a global level.

For example we have such an issue with bar/cafe in Italy. „bar“s (local name) are very important here, many people go there for breakfast, you typically do not sit down (many don’t have tables at all) but just get an espresso o cappuccino and a croissant and walk out again.
There are many variations of bars (mixed with tobacco shops, where people also pay utility bills, speeding tickets, etc.) scratchcards/lottery tickets etc., those combined with a pastry shop, with a restaurant, with a cafe, with an icecream shop, etc. Nearly all of them sell croissants (of vastly varying quality, baked on the premises vs. deep frozen industrial, i.e. suitable for breakfast or only for tourists/in an “emergency”), and also some sandwiches, but some also cook for lunch (typically small selection and cheaper than a restaurant).
Some are open in the evening or even all night, mostly they open early in the morning and close at 18/19h in the evening, but not so rarely they are also offering an “aperitif” pre-dinner (typically snacks and an alcoholic mix drink).
Why I am telling you all this? Because we still don’t have a good solution to the problem yet (we are making progress, by specifying additional tags). Many mappers use amenity=cafe indiscriminately for all Italian style bars, which is ok for a part of them (when there are tables and seats, possibly table service) others use amenity=bar (which could suit if the determining property for a bar is having a bar (table with a barkeeper), and is generally what joe mapper adds as it seems to fit perfectly, a bar is a bar). Many sell alcohol (spirits, wine and beer, some also cocktails and long drinks), some don’t.

Maybe this sounds familiar to you, maybe not. Did you expect a tobacco shop to offer payment services for paying the public administration, or your water bill? Could you imagine a cafe having no seats and would you still call it a cafe? Could you imagine a bar offering (fast) breakfast, many types of coffee and homemade ice cream but few if any alcohol? As there is this vast variety of “bars”, and as the values in OpenStreetMap aren’t well defined (not clear what is required/typical/excludes the tag), it may be ok to have different tags for what is all called a “bar” locally, but as long as we don’t define well what’s the meaning of the tags, we won’t know. On top of this, there are also “cafes” and “bars” which correspond more closely to such features in other countries (or “pubs” which focus on beer, you’ll hardly find tap beer in a bar, and only sometimes in a cafe).

The question in this kind of discussion is usually: what constitutes a feature, what are the defining properties and what is optional or indicating a different kind of feature. And what should be implied by the “main class”, what is an interesting additional property to record. It mostly “works” because of similar assumptions regionally, but still you should not assume a German or Austrian cafe to be the same as an Italian cafe, or American cafe. (btw., Americans also seem to have issues mapping Diners, or the French bistrot - AFAIK the tag “cuisine” is used but the difference is not just the cuisine), and if you are looking for an Italian bar, you’ll have to check amenity=bar and amenity=cafe and you still will get partly inconsistent tagging and won’t be able to distinguish the results from other kind of cafes (that don’t have a bar) or bars (that only open at night and won’t have a coffee machine).

A similar question: can you assume from a shop=butcher that you’ll get a warm meal at lunchtime?

There are cultural differences, and our struggle in tagging is defining the basic requirements, in a generic way which allows to check for local features if they apply or another tag must be used/created. And decide on additional tags that are useful to make the distinctions people are interested in.

Generally, and because we all know that multiple tags for apparently the same or very similar features aren’t a big deal, I believe we should discourage people from “gardening” by ploughing through the tag seed in June rather than letting the germination happen. Fixing typos like missing letters or additional spaces is naturally ok and helpful, but “unification” and overstandardisation can be toxic for tag development. (and retagging happens continuously, every single day).

Ah, and presets of course. The preset concept as a whole leads to many square pegs pushed in round holes, as in “it is clearly the most pertinent feature class of all available”. I don’t have a good answer for this, we can’t realistically require from everyone intending to contribute, to acquire a Phd in tagging before they begin. Maybe a guided process (similar to how it is done to determine a species in biology) would be an improvement, having the mapper answer several questions rather than being presented with a single term. Requires a lot of development work (maybe an ai could help?) for creating taxonomies and formal definitions.

Cheers Martin 

TL;DR:
careful with standardization