[Tagging] Feature Proposal - Voting - ISCED 2011 Education Programme

[kaartjesman at mykolab.com]

On maandag 18 april 2022 02:53:48 CEST Minh Nguyen wrote:
> But the existing isced:level=* and the proposed
> isced_2011_programme=* both have a different purpose: to
> accurately reflect an externally standardized classification
> system.

This actually brings up another issue that I have with this 
concept in general.
OSM is open source mapping, with the basic concept that we map 
what can be verified on the ground as truth.

But here you are basically copying data that is made up by a 3rd 
party. It is about a classification that is impossible for a 
mapper to know or verify as correct, other than to just assume 
that the 3rd party did things correctly.

Unlike a traffic sign or a road being paved or not, there simply is 
no possible way to make this about truth. Frankly, the 
"standardized classification" is largely made up as well as they 
can't have a person in each school checking the levels.

The proposal is trying to improve something, without actually 
acknowledging that the raw data is flawed.

Having a perfect display system for data that is never going to 
be high-quality, that is going to be political, that is going to 
be based on easy metrics instead of on-the-ground research, that 
doesn't feel very OSM to me.

:shrug:


>  It turns out that rationalizing all
> the world's educational systems is much more complicated than
> any one dialect of English can reasonably accommodate.

Maybe the effort to somehow compare the method of education in 
different cultures all over the world is the reason for the 
failure.


> If I'm not mistaken, this proposal is only trying to fix a
> technical problem with an existing approved isced:level=* key,
> that it's based on an outdated version of the standard. Either
> we accept the new ISCED standard as is, or we decide that this
> scheme is too fussy or unverifiable to tag at all and thus
> de-approve ISCED tagging. If the latter, then expect both
> school=* and grades=* to grow in popularity, meeting a
> cartographic need that ISCED doesn't completely satisfy
> anyways.

I think this makes more sense.

Those tags are verifiable but generic enough to be used locally.