Hi Everybody, I have tried to be open to the benefits and merits of the openser initiative, and allthough I welcome the ideas on an added "fast-track release", as well as web-based and editable documentation, I agree completely with Peter and everything he mentions in his open letter. Thank you, Peter! (I can also add to his list that the people behind openser have not contributed to the ONsip.org Getting Started documentation effort and this effort is not even mentioned on the openser website. However, Jan and Jiri have been very positive and helped us out.)
So, even though some of the things in the openser initiative are good, the chosen approach is ALL wrong.
For what it's worth, I have thus decided that I will continue to contribute on the serusers and serdev lists and here only. This includes anything related to contributions to the experimental tree.
I hereby encourage the iptelorg.com guys to take the input received from all parties and independently of the openser initiative come up with a plan for improving the aspects of the project that have been pointed out. I hope the people behind the openser initiative will bring their excellent documentation skills and drive for new features to a revitalised SER project. If they don't, let them carry on.
Best regards, Greger g-)
PS! I will do my best to ensure that ONsip.org continues to be a place for Getting Started information, links, downloads, FAQs etc without any politics (i.e. anyone registered can submit links and downloads. In addition, we will try to add any resource that can be of value). Hopefully this will contribute to reducing the confusion.
Peter Griffiths wrote:
folks --
i usually do not participate in mailing list discussions but it seems that my tls contribution became hot topic in recent ser vs. openser discussions so i think i will make an exception here.
i think that openser is a bad idea. please stop using the tls code as a hostage in this dispute, i did not write it to fork the project.
iptel guys made no attempt to hide the code when i approached them although they probably have their own commercial implementation. for me ser is an open and free project because:
- i could download ser freely and use it
- i needed tls, i was not able to buy it from iptel
so i tried to implement it 3. i sent it to the mailing lists, anyone can find it there and use it if they want 4. someone else picked up the code, improved it and committed to cvs (thank you for this, by the way)
what can be more open than this ? maybe sources stored in a wiki where anyone can change anything.
openser claims that it will be more open, but:
- over 20 people have access to ser cvs on berlios,
but only 4 have access to openser cvs on sourceforge, not even other developers.
- you took existing code from ser cvs, added your own
improvements, but you did not give other developers who work on ser any chance to say whether such changes should be included or not, although they are still listed in AUTHORS file, in module documentation and elsewhere. from the user point of view they will be responsible for bugs introduced by you and have no chance to influence it.
- i did not see any discussion about this move, it is
a bit strange that the people on the lists were not allowed to participate in the decision, especially if you claim that it is for the interest of the community. how do you know people want the project to be split when you did not ask ?
- the complete cvs history is missing in openser
repository, effectively hiding who contributed what.
I also noticed that user miconda created project named ser on sourceforge, is this a preparation for the next step - full control ? did other ser developers know about this ? i think having two projects with the same name on two sites is confusing. from my perspective the new fork is only an attempt to get more power and control over the sources. it is about who will control what, not about freedom. try to resolve it without forking the project.
if ser is bad then openser is not any better:
- iptel.org advertises iptelorg.com on its webpage -
there is an ad of voice-system on the main page of openser.org, there are voice-system copyright statements and readme states that the project is maintained by voice-systems, proving beyond reasonable doubts that openser promotes voice-systems just like ser promotes iptel.org
- there were complaints about maintainers deciding
what comes into modules - so in openser only selected people have access to cvs and they will probably form a committee and decide.
- there were calls for more discussions, but the fork
was not discussed at all
there probably is a good reason why tls was put in experimental directory first and is not in the main tree yet, i do not see this as a problem. in fact i would appreciate if someone who knows in detail how tcp code in ser works could review the code before it is committed in the main tree. there are many places in the code i don't understand. i did only limited testing because tls support in end devices is virtually non-existent. use it at your own risk.
i don't know about others, but i am on alert when people talk about "the needs of the community" and "take over when someone is very busy or unwilling to cooperate", i don't remember seeing users asking for anything like this.
-- peter
Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Serdev mailing list serdev@lists.iptel.org http://lists.iptel.org/mailman/listinfo/serdev