At 11:40 19/07/2006, Daniel-Constantin Mierla wrote:
I will add few remarks related to project's policy
and evolution.
OpenSER is driven by a board with members from different companies which
will ensure project's independence and survival when one company changes
its interest in the public project.
As this is going to the ser mailing list, I would like to take the freedom
to present a quite different approach which has shown very viable in SER
history. That is, SER has been developed by consistent team of long-term key
developers spending their full-time job on SER's evolution. They have been able
to steer SER's direction since its inception till today in moderately
growing capacity. Despite the logo of the employer has changed over the time,
all key members remained on board, provided continuity and took care to
integrate third-party contributions in the software suite in a consistent manner.
Employer's interest has been playing role in that it has been providing strong
field feedback from customer but that's, to be clear, a feature.
Even if "openser" claims to be different, solely by the fact of taking
vast amounts of SER code, it basically acknowledges the SER results which
have been generated the way they have been generated.
Also, I'm glad the point of 'incremental easy-to-execute updates' was raised.
Clearly, I consider a revised data model in SER a superior change which was
overdue and which has greatly improved applicability of the software. The price for
it is devising good migration procedures but that appears to me a kind of one-time
pain, as opposed to the pain-level associated with living with an outdated data
model. The other downside of such fundmanetal changes is longer release cycle
but that still appears very acceptable trade-off to me.
-jiri
--
Jiri Kuthan
http://iptel.org/~jiri/