[Talk-ca] What do I poutine the name tag of a road with a suffix?

[Minh Nguyen]

Vào lúc 19:10 2022-12-15, Jarek Piórkowski đã viết:
> Is the suggestion seriously to use a dataset put together by the federal 
> government to override provincial and local norms, because some 
> non-local mappers 3000 km away think that's more correct?
> 
> Can anyone think of community issues, political issues, or for that 
> matter even practical issues with this suggestion?

Setting aside the issue of local autonomy, none the sources that have 
been cited so far in this discussion seems like a smoking gun to me, 
whether the street signs or the NRN dataset. Looking for patterns on 
signs or in a dataset, we can identify a prevailing written *style*. 
However, the global abbreviation guideline was always about ignoring a 
particular style in favor of disambiguation while retaining correct 
spelling.

If someone is able to dig up a document or style guide explicitly 
stating a policy on which is the correct *spelling* and which is not, 
that would be more convincing. Until then, we're debating a matter so 
minor to laypeople that quite possibly none of the relevant authorities 
cares as much as some of us do. That's why I tried to ground this 
discussion with information on the practical impact of either choice 
when OSM data finds its way into data consumers. The OSM ecosystem is 
not the same as the environment for which the signs or the NRN were 
designed; we have different constraints.

> The hang-up seems to be about abbreviations and OSM's policy of avoiding 
> them. No policy can be applied blindly. Every policy has areas where 
> locals believe it doesn't apply.

The global guideline on abbreviations has been remarkably stable since 
2005. Given its age, we can be pretty sure that North American interests 
were not very well represented when the policy was decided upon. But the 
rest of the world has been unaware, gullible, ambivalent, or happy 
enough to mostly leave it alone all these years.

In service of this discussion, I recently moved the guideline from a 
section of the "Names" article to its own article. [1] Splitting it out 
creates a natural space for local communities to draw attention to the 
idiosyncrasies in their neck of the wood, since the regional pages are 
too easy to overlook for a matter this important. I see something has 
subsequently been written up about Canada on this page.

> I will again bring up the example of the UK town St Ives. It's pronounced "saint" and it is accepted that the town is in fact named after a saint. But the name in OSM is St Ives. Because the name of the town is St Ives. Because that's what the people there think the name is.

"St" is a particularly interesting case, touching on the fundamental 
question of what makes an abbreviation an abbreviation. British English 
speakers have made the case that native speakers in the real world no 
longer consider "St" to be an abbreviation but rather regard it as the 
canonical spelling of the word in the context of a name.

I only learned of this phenomenon after becoming a mapper. For whatever 
reason, American English developed differently. Here, "St." is 
universally regarded as an abbreviation of "Saint", no different than 
"Rev." as an abbreviation of "Reverend", except when it forms part of a 
surname. You'll hardly find a sign in the Missouri city that spells out 
"Saint Louis". Laypeople from near and far would consider it to be 
needlessly pedantic and contrarian, but not incorrect.

If only Alberta's quadrant system were so pragmatic! Whatever the 
outcome of this discussion, I hope folks recognize the distinction 
between raw database contents and the desired presentational output.

[1] https://wiki.openstreetmap.org/wiki/Abbreviations

-- 
minh at nguyen.cincinnati.oh.us