|
Re: EMF Marketing [message #765034 is a reply to message #765003] |
Tue, 13 December 2011 09:52 |
Ed Merks Messages: 33227 Registered: July 2009 |
Senior Member |
|
|
Hauke,
Comments below.
On 13/12/2011 9:46 AM, Hauke Fuhrmann wrote:
> Hi there,
>
> are there objective arguments I could give to business decision makers
> about the use of EMF?
They could spend lots of money on commercial software or they could use
something for free. Yes of course folks will argue that nothing is free
and it takes effort to learn how to use software (tools and frameworks)
effectively, but that's true even of commercial software and commercial
software is var more likely to be changed arbitrarily in the future,
i.e., you'll need to buy a new version and need to relearn your skills.
Just look at all the versions of Windows and the versions of Word,
PowerPoint and so on to see how true that is. Commercial software is
often as much about fashionable change for its own sake as it is about
real substantial improvements.
Another great example for me was a large European airplane
manufacturer. I asked them why they use EMF and they explained they had
to build software to last as long as the airplanes supported by the
software and they could not rely on any commercial vendor to support
their software for 50+ years. With open source, they will always be
able to self serve, and if they choose a broadly adopted technology,
such as Eclipse and Eclipse Modeling, they can better rely on long term
skills to support it. (Look at the level commercial support available
today from organizations like itemis and Obeo). Then they went on to
tell a horror story about how they developed their own tools years ago.
Management decided they weren't a software company so spun off a company
for that purpose. But that company was bought by Rational, which was
bought by IBM, which decided to throw the software away and suggested
something new and improved be used instead. Needless to say, once
bitten, twice shy...
> I'm long convinced and I have technical/functional arguments but
> little non-functional ones.
In some ways the non-functional ones are more compelling. Why the big
drive toward open source in Europe? Being at the whims of a commercial
software vendor's weather patterns is a huge risk, as I explained above.
Also, good luck getting your bugs fixed. You've already spent your
money, why should the commercial vendor bother investing in fixing your
problems. Better to invest in a whole new version so you the valued
customer can spend your valuable money all over again. So good luck
with the next fashion trend when it comes out.
>
> Maybe something like: How old and mature is the EMF core?
Going on 10 years. The core of 2.8 is for all intents and purposed
binary compatible with 1.0. That's the kind of stability that should
give warm fuzzy feelings. And of course it's not as if nothing has
changed or improved in 10 years. The evolution continues with vigor,
much like Java itself. Support for generics is a great example of that
in both Ecore and Java.
> How many companies use it in commercial applications?
That's hard to count. When I was IBM I know they had well over two
hundred commercial applications. In many cases, companies use it
indirectly because it's used to build the infrastructure underlying the
software they use directly. After all, e4 uses EMF, so with Eclipse
4.x, all Eclipse applications will be using EMF whether they realize it
or not. How any of those are there? I don't know...
> How active is the EMF community?
The core is small, but the layers around it are deep. You might look at
newsgroup postings as an indicator. Do questions ever go unanswered?
In any case, modeling projects dominate the traffic, e.g., Xtext is by
far the most active newsgroup...
> Are there big examples of such companies or available case studies?
I can point at examples like Nasa and the Canadian Space Agency has
users, and of course IBM and SAP are obvious examples. There are a few
big names in
http://www.eclipse.org/org/press-release/20111102_polarsys.php. If I
had to estimate, I'd say it's well into the thousands.
>
>
> Are there such informations available?
It's not readily available information.
>
> Cheers,
> Hauke
Ed Merks
Professional Support: https://www.macromodeling.com/
|
|
|
Powered by
FUDForum. Page generated in 0.03510 seconds