[Tagging] Tag:amenity=motorcycle_taxi not approved

[Andy Townsend]

On 08/05/2020 14:04, s8evq wrote:
> And then some people in this very thread suggest to just ignore a 
> rejection and start using it anyway. What's the use of the whole 
> voting system then? 

Frankly, not much.


> Why even bother writing a proposal in the first place? I'll just do 
> whatever.

"I'll just do whatever" is why OSM succeeded and other approaches 
failed.  "I'll just do whatever" allows people to just add stuff to 
their neighbourhood _right now_, which they can't if they have to 
consult a committee beforehand.*

That said, it _does_ make sense to discuss what is the best way of 
tagging a particular real-world object - and it also helps if the people 
discussing it have actually seen one of those in the real world (as was 
previously suggested, I doubt some of the "no-voters" have).

In this case clearly not all the people in favour of adding a subtag to 
"amenity=taxi" could be persuaded that it was a bad idea, but since they 
are never likely to encounter such a feature in their everyday lives 
their data is not likely to matter.  OSM should be built be people who 
are familiar with the objects that they are mapping, not people guessing 
from afar.

Best Regards,

Andy

* OSM vs (say) wikimapia isn't the only example of this - wikipedia / 
nupedia is another more famous one.  Elsewhere way back in the 1980s and 
1990s the company I was working for was telling people that a 
statistical approach to fault diagnosis was a better approach to trying 
to diagnose and fix electronic stuff than an "Expert Systems" approach - 
essentially having someone coming in and trying to design some rules 
based on a few hours "sitting with Nellie".  The statistical approach 
won out, allowing you to read this message easily in your inbox, with 
the Bayesian spamfilter having moved all the undesirable stuff into 
"junk"**.

** but not in electronics, unfortunately, as no-one repairs that any 
more - it's (currently ) cheaper to buy more stuff from 
$low_wage_economy elsewhere.