[mailto:e4-dev-bounces@xxxxxxxxxxx] On Behalf Of Kevin McGuire
Sent: January-27-09 6:50 PM
To: E4 Project developer mailing list
Subject: Re: [e4-dev] What *are* we doing here??? <g>
It's great that
you're thinking of this. Boris, McQ and I were discussing something
similar just the other day and to be honest we were really struggling.
With respect to
what you have below, this is the first we've used the word "Cloud".
It's pretty trendy today so maybe we should, not sure. I will note that
at first you're saying how we're extending the reach of the platform (good),
the second is a discussion about adoption of technologies. The problem
with the latter is that they are only as useful as the advantages they provide,
so saying we'll evolve the architecture and adopt technologies ... to what aim?
My belief is
that we need to speak in terms of practical advantages so that a company will
see real benefit in e4, either in terms of reduced application development
cost, integration with a new breed of applications, or ability to reach new
audiences by targetting the web without having to completely abandon their
Eclipse investement. If they see benefit, they may be will be willing to
spend some developer resources contributing. Thus that value must be
Things we said we'd do, and how I think they'll provide value:
0) Be more open
as an organization: That's important for Eclipse, and for people to know
they can get involved, but I think is only interesting with respect to
perceptions of the past which hopefully we're changing (hence #0).
1) Make it
easier to write applications, make it easier to maintain applications:
This is a big win for anybody choosing Eclipse as an application platform.
In fact, arguably that's the whole point of an application platform, that
you have to write less stuff because of all the hardened libraries at your
disposal. Modelled UI, declarative UI all come into play here.
2) Make it
easier to contribute to the platform: based on the notion that by cleaning up
the code base and picking known popular technologies like EMF, people can step
in to contribute where they could not before. Maybe more of an
organizational item, but also a benefit to consumers in the sense that they
believe that if they find a platform bug they can fix it, and if they find
missing features they can add them.
3) Enable new
kinds of UIs: This touches on both the "shape" of the
application, so RCP on steriods through modelled UI (which, by the way, I think
we should start calling "Flexible UI", because the model aspect is a
technical choice only and does not on its own ascribe goodness) [sorry Ed!
:>]. It also touches on the CSS work in making it easier to create new
looks for your applications. Presumably this has appeal to application
developers because they can modernize their apps, they are less restricted in
look, they have more opportunity for product branding, etc.
about flexible resources.... sorry haven't been keeping up here <sheepish
the reach of the platform: web to desktop, desktop to web. Reuse through
desktop all come in. The carrot here is reduced development cost through
reuse, and parity of application look and feel for applications with both a web
and desktop component (increasingly popular).
I'm sure I've
missed some stuff.
Please respond to
E4 Project developer mailing list <e4-dev@xxxxxxxxxxx>
Project developer mailing list <e4-dev@xxxxxxxxxxx>
[e4-dev] What *are* we doing here??? <g>
Yes, what I wrote was weak in that respect. Take 2. <smack/>
Beyond the Enterprise, into the Cloud
Tomorrow's applications will require integration of hand-held devices, desktops,
enterprise server applications, and applications that are hosted "in the
cloud" and are accessible from anywhere. E4 makes Eclipse the best
platform for delivering integrated applications that scale into all of these
environments. With E4, you can now target tiny devices, all the way up to
cloud services like Amazon's EC2, all from a single code base.
Take Eclipse's Architecture to the Next Level
In order to accomplish the above objective, Eclipse will fully adopt
technologies--such as Eclipse RAP--that have been in incubation for some time,
and evolve its core workbench architecture so that it can fully participate in
distributed and web 2.0-enabled applications.
2009/1/27 Boris Bokowski <Boris_Bokowski@xxxxxxxxxx>
What I meant was a position statement as defined by: http://www.ericsink.com/Positioning.html
(and probably many other marketing text books...)
"The basic idea of
positioning is that your product occupies a place in the mind of the people in
your target market. You are defined by their perceptions of you. ... ask
in which market segment you want to be known as number one. You want to
be known as the best of your breed, even if you need several qualifiers to
constrain the scope of your claim. Don't think about being fifth place in
a large market. Instead, be number one in a smaller market. ... Identify
the three parts of a position: superlative (why choose this product),
label (what is this product), and qualifiers (who should choose this product)."
I.e. something like: "Equinox is the number one componentization solution
for Java applications in the embedded, desktop, and server context." or
"RCP is the best platform for rich client applications that need native
L&F across all major desktop OSs" or "The Eclipse IDE is the most
popular IDE for Java programmers".
Except we need something about e4... ;-)
Dave Orme wrote on 01/27/2009 04:36:12 PM:
> OK, cool. How about:
> E4 will emphasize two primary themes:
> 1) Beyond the Enterprise, into the Cloud
> Eclipse has its roots as an enterprise software framework, capably
> delivering software from embedded devices through the server.
> E4 will extend Eclipse with the capabilities needed to deliver
> applications that live in the network cloud, whether on services
> like EC2 or on your own cluster.
> 2) Take Eclipse's Architecture to the Next Level
> In order to accomplish the above objective, Eclipse will fully adopt
> technologies--such as Eclipse RAP--that have been in incubation for
> some time, and evolve its core workbench architecture so that it can
> fully participate in distributed and web 2.0-enabled applications.
> 2009/1/27 Boris Bokowski <Boris_Bokowski@xxxxxxxxxx>
> Hi Dave,
> I agree with you that it is very important to think about e4 from a
> marketing point of view. Especially since we have explored a good
> number of areas and should be thinking about what we want to deliver
> (both short term and long term).
> It's kind of funny that you are mentioning marketing now. Just
> yesterday, I came across a blog posting about "marketing for
> which I found interesting as someone who had a small software
> company in the past. After enjoying all the parallels with what I
> experienced a couple of years ago, it occurred to me that many of
> the issues apply to the e4 project as well, in particular this one:
> Ideally, we'd come up with a position statement about e4, maybe
> something like your (1) below but shorter. Any suggestions?
> About (2), I don't think the phrase "architecture clean-up"
> something you can use for marketing purposes. :-P It's not about the
> intrinsic properties, it's about what you can do with it.
> Dave Orme wrote on 01/27/2009 03:22:56 PM:
> > Awhile back we put together a few paragraphs describing what E4
> > about from a (dirty) marketing point of view. Beware: if
> > onward, you might need to take a bath. ;-)
> > Seriously though, what occurred to me last night is that E4 is
> > really about two themes:
> > 1) Eclipse has always been about providing great infrastructure.
> > SWT gives us great infrastructure horizontally across operating
> > system platforms. eSWT, eRCP, however, broaden Eclipse
> > down into the embedded space. E4 is about moving Eclipse up
> > vertical space so that it can also be a platform for cloud-based
> > After E4, we will cover all major desktop and server operating
> > systems horizontally and the embedded through cloud space
> > vertically. The enabling technologies here are Equinox,
> > [[the second E4 theme]] which is:
> > 2) Code and architecture clean-up. Singletons are (nearly)
> > evil, but especially so in a multi-user environment like RAP.
> > Resources can be anywhere. Declarative UIs are nice.
> > won't re-hash any more of this here as we're all well-versed in
it by now.
> > My Question:
> > Does this sound like a good way to describe and position E4?
> > OK, maybe that's a silly question to ask a bunch of engineers.
> > But does anyone think I'm missing anything important or glossing
> > over something that I shouldn't be.
> > Regards,
> > Dave Orme
> > _______________________________________________
> > e4-dev mailing list
> > e4-dev@xxxxxxxxxxx
> > https://dev.eclipse.org/mailman/listinfo/e4-dev
> e4-dev mailing list
> e4-dev mailing list
e4-dev mailing list
e4-dev mailing list