On Sat, Mar 8, 2014 at 7:45 PM, Kevin Kofler <kevin.kofler(a)chello.at> wrote:
Rahul Sundaram wrote:
> Yes and I am convinced that not enough effort was taken to address the
> concerns expressed. This is probably lost opportunity already but one
> could strive to do better on future specs.
So you think the KDE developers should have renamed those D-Bus methods and
signals, requiring all their code to change, and also making new KDE
applications incompatible with old KDE Plasma workspaces and the other way
round, just because the GNOME developers did not like the names? To me, this
sounds like attempting to rename "Referer" to the correct "Referrer"
in the
HTTP spec; there's a reason this has never happened! ("Referer" came to be
because the British inventor of HTTP thought that that was the correct
American English spelling, when in fact it is just wrong everywhere.)
What is the lesser evil:
* that a handful toolkit and shell developers have to cope with historical,
slightly suboptimal names, whose meaning is well-documented in the spec, and
that are not visible to end users nor to application developers at all, OR
* a major compatibility break in both KDE Plasma and Unity to accomodate
those cosmetic name changes?
Why are people surprised at all that the KDE developers rejected the
proposed changes???
You make it sound like Dan's review only talks about names. But that's
not the case.