[josm-dev] JOSM-Tested

[Dirk Stöcker]

On Mon, 14 Jun 2010, Ulf Lamping wrote:

> I'm not argueing that the JOSM releases could possibly be improved, but
> your "flame style" argumentation is not going to achieve a lot here ...

Sure, everything can be improved, but what needs improvement and how?

Maybe I should add my point of view as admin:
- Current release is down to about 130 bugs, most of them special cases,
   debatable things or very hard to fix or unreproducable issues. Most
   other reports are suggestions (before release we had about 200 bugs).
- About 70% of JOSM users have a 100% localized interface (which means
   de+en). About 20% have a 95%+ localized interface (ru,fr,it).
- We get nearly no serious bug reports for tested versions.
- We VERY seldom get exception reports and these mostly for latest after
   new features have been added.
- We have a working nightly build, which is so stable, that a lot of
   people use it instead of the recommended tested.
- Most reports we get are minor issues (thought not easy to fix). For me
   this is a good sign, as for faulty software nobody cares for minor
   issues.
- We have approx. 50% of the OSM editor "market". I can create different
   stats which show that JOSM is better, others that Potlatch is better,
   but altogether it is 50/50.

Actually I don't see a reason to change the current development model.

I agree, that communication for translations could be improved, but as 
already said I don't see how this can be done currently, as I don't know 
which people are our translators and it seems they change from time to 
time. This sounds a bit strange, but it works nevertheless.

Ciao
-- 
http://www.dstoecker.eu/ (PGP key available)