[Osmf-talk] February OSMF Board Meeting

Gervase Markham gerv at gerv.net
Fri Feb 26 17:46:55 UTC 2010


On 26/02/10 11:24, Frederik Ramm wrote:
> department meaning at least a 2 week delay if not more). Compare this
> with how we deal with our map: Find a problem, fix it; miss some data,
> add it - no review, no delay.

But the difference between the front page UI and the map data is like
the difference between software UI and code.

If you want code to do something additional, you can just add the
feature, and some sort of configuration option to turn it on. If you
code it competently, it doesn't affect anyone else. And if someone else
refactors it, as long as it keeps the same function, you don't mind.
Similarly, if you want to tag the map with some specialized data, you
can go ahead, because your tagging doesn't affect other people's use of
the map.

But if you are changing the UI of some software, your changes affect
everyone else, whether they care about your feature or not. And if
someone refactors the UI, no-one knows where to find a feature any more.
Changing the front page of OSM is much more like this than it is like
hacking the map.

This is why UI cruft is a much bigger problem for software projects than
code cruft. And why a single-minded UI design is much more important
than having a single overarching code architect.

Gerv




More information about the osmf-talk mailing list