On Mylar we generally try to follow these guidelines as closely
as possible we have always strived to minimize what we start, how long we take
to start it and what we contribute to the UI. But it will be impossible to
add something like Mylar to Eclipse without violating some of these guidelines
(this is partly why it is so difficult to add something like Mylar to IDEs
other than Eclipse). For example, we consider the Open Task action to belongs
in all perspectives just as the SDK considers the Search to do so (fyi, that’s
the only one).
But this is a slippery slope and I couldn’t agree more
with the “death of a thousand cuts” argument. So I wonder if a
project should need to justify and bring up for debate any violations of
guidelines like those listed below. I don’t think that the goal of
this should be to minimize how different the Europa install ends up being from
the SDK install, but to make sure that there is a good reason and justification
for everything that’s added and started and that the end result is as cohesive
as possible. For example, regarding Mylar’s Open Task, we could
limit this to just SDK perspectives and rely on authors of other perspectives
to add it themselves. I’m not sure if that’s worse or better,
but definitely debatable. Wrt to other Mylar related concerns, I tried to
cover our decisions in my previous email and would be happy to discuss any other
suggestions and alternatives.
Mik
From:
cross-project-issues-dev-bounces@xxxxxxxxxxx
[mailto:cross-project-issues-dev-bounces@xxxxxxxxxxx] On Behalf Of Eric
Jodet
…
As a reminder,
I'll quote you:
- Your plug-in is not the most important
thing:
- do not start your plug-in on startup.
- do not gratuitously put your views in random
perspectives
- do not contribute menus / action sets to all
perspectives
- do not create projects in the workspace on startup.