incubator-esme-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vassil Dichev <vdic...@apache.org>
Subject Re: The current state of ESME
Date Tue, 03 Mar 2009 11:11:55 GMT
First of all, I agree that what Erik said was spot on. And I'm not
against ESME being seen as targeted for the enterprise, I'm against it
being seen as inflexible. So, if Darren sees an organization (like a
school) as an enterprise, I think we're not in disagreement.

> Are you suggesting we should rename ESME to SME?  :-)

I'm reminding that we shouldn't let ESME lose the S as in "social" and
become EME :-) This thread started in the context of growing the
community, and we can do that only if we're taking care to be
inclusive, not exclusive.

Of course ESME cannot succeed if it tries to be just another
laconi.ca. But Bertrand has a valid point in that it should be easy to
experiment with (the last letter E in ESME still stands for
"experiment"). To expand on the example of authentication, if we only
provide Kerberos/LDAP/Active Directory, few people will commit the
time to set it up just to try it.

There's another story to clarify what I'm trying to say. There was a
small company building a product, and a big company interested in the
product buys the small company. In a couple of years the product is
integrated, but becomes so entangled in the big company's software
suite that noone else can use it, so its previous customers eventualy
shift to other products. A few more years pass and the big company
decides to correct the mistake and make the product more modular and
decoupled from the big software suite. I wouldn't want ESME to repeat
the mistake of our hypothetical product.

To summarize: if ESME is seen as too dependent on particular
enterprise features, fewer people will give it a try. If ESME is seen
as having no enterprise features, people might give it a try, but will
not see a reason to keep on using it.

Being both open-source and focused on the enterprise is tricky. As
Darren mentioned, the enterprise might be willing to allocate
resources if the software is "almost there". If the software is not
ready, a community independent of a particular company is vital.

> So small and simple is key, and I'll throw in extensible and adaptable.
A big +1. I have the feeling Erik is referring to the same thing, but
seen from a different angle.

Mime
View raw message