[OSM-talk] Metrics

[Justin Tracey]

On 2021-10-16 8:06 p.m., Mike Thompson wrote:
> 
> 
> On Sat, Oct 16, 2021 at 4:47 PM Justin Tracey <j3tracey at gmail.com 
> <mailto:j3tracey at gmail.com>> wrote:
> 
> 
>     It would be nice if the recommended solutions given by these validators
>     were similar to noexit=yes, where they could be readily found after the
>     fact and distinguished from more invested mapping.
> 
> Why not just leave the "error" unless one has actual evidence as to what 
> the situation is on the ground?  That way someone else might see it and 
> be motivated to go out and survey the situation first hand.
> 
> Mike

That is the correct thing to do, of course. The point being made ITT is 
that lots of people in practice don't defer that judgement. My argument 
is suggesting tags like noexit=yes, that are clearly just catering to 
validators, makes it easier to see when something was mapped without 
that deference (i.e., suggest things that fix validator issues in such a 
way that the data is technically valid, but would be unambiguously and 
algorithmically detected as something that can be improved by someone 
with more information). It's sort of like asking "why have validators 
detect issues when we could just tell editors not to make those 
mistakes?" Sure, keeping people informed of best practices is good, but 
tooling should also just be constructed in a way that acknowledges 
people are fallible, and makes fixing things after the fact simpler.

  - Justin