Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [ide-dev] what about doing less?


When I looked at this issue a few years ago, I was able to determine that the vast majority of start up time is spent in the class loader. Too many plug-ins are getting activated at start up time. Removing functionality may help that, but paying attention to what you do at startup time will pay off without hasty cutting.

From: Mike Milinkovich <mike.milinkovich@xxxxxxxxxxx>
Reply-To: Discussions about the IDE <ide-dev@xxxxxxxxxxx>
Date: Tuesday, 22 October, 2013 7:47 AM
To: "ide-dev@xxxxxxxxxxx" <ide-dev@xxxxxxxxxxx>, Discussions about the IDE <ide-dev@xxxxxxxxxxx>
Subject: Re: [ide-dev] what about doing less?

If you want to see an example of a fast starting, bare-bones development tool that's based on Eclipse, check out the Dart editor. 

Mike Milinkovich
From: Martin Lippert
Sent: Tuesday, October 22, 2013 4:24 AM
To: Discussions about the IDE
Reply To: Discussions about the IDE
Subject: Re: [ide-dev] what about doing less?

Hey Kaloyan,

> This is an interesting direction that worth brainstorming.
> What would you take out of the Java EPP to make it again the killer Java
> IDE?

I think it is not as easy as "remove bundle X or do not install feature
Y"... unfortunately...
The questions that are coming up for me are (for example): what would be
necessary to get the Eclipse IDE starting up and be ready to use in
2sec, for example? I know, it sounds a bit crazy and maybe seemingly
impossible, but those are the questions that help me to think about this.

What do you think?


> Greetings,
> Kaloyan
> On Tue, Oct 22, 2013 at 12:15 AM, Martin Lippert <lippert@xxxxxxx
> <mailto:lippert@xxxxxxx>> wrote:
> Hey!
> I would like to propose a work item for the ide-dev working group
> that focuses on doing less. By that I mean instead of adding
> features to the existing platform and/or IDE I would love to reduce
> the scope, reduce the number of features, and reduce the amount of
> code - and focus on improving the remaining parts.
> The question that drives me is not what is missing from todays code
> base or EPPs, but what is too much, and even if we strip things
> down, what is still not lightweight enough, not
> fast/slick/performant enough, not high-quality enough to make the
> Eclipse IDE a killer app again.
> I (obviously) don't have ready-to-use solutions/answers for this,
> but I think following this direction could be interesting (make
> people enthusiastic about the Eclipse IDE again).
> Just my few thoughts at the moment... ;-)
> Cheers,
> -Martin
> _________________________________________________
> ide-dev mailing list
> ide-dev@xxxxxxxxxxx <mailto:ide-dev@xxxxxxxxxxx>
> <>
> _______________________________________________
> ide-dev mailing list
> ide-dev@xxxxxxxxxxx
ide-dev mailing list

Back to the top