Andreas Schneider wrote:
I've talked on IRC with mnencia. He is the maintainer of the debian
package. He helps fixing cmake bugs. We discussed some things which
should be done in the near future.
Restructuring of the source tree
---------------------------------
The structure of the current directory tree is a mess. We have a lot of
workarounds in the cmake files to get it building.
With a cleaner structure which mirrors the deps too, a lot of things
would be easier.
phapi project
--------------
phapi should have it's own tree with the needed headers. So that we can
build a phapi package and install it as a system library.
3rd party source
-----------------
owcurl or other similar crap shouldn't be in the source. For every used
library there should be a developer responsible to get changes upstream.
3rd party libraries
--------------------
There should be a "Wengophone developer bundle" with the precompiled
binaries you need on windows to build wengophone. It would be easier in
cmake to reference to that bundle.
-- andreas
------------------------------------------------------------------------
Very valid points,
I'd like to jump on the occasion and suggest that we think on the issue
of the overall project management.
I think we lack official 'subsystem maintainers'.
In order to introduce them we need a 'subsytem list'
Once we construct this list we'll need to adapt directory tree structure
to confrom to the list
And We need to have much more discussion on project evolution with
respect of the feature set.
Of course anyway, Wengo as main underwriter of the project get the last
word on the feauture list
but i believe that the discussions of the issues could have very
positive effect for Wengo too.
The almost complete lack of discussion IMO has very negative
consequences for spread of wengophone.
For example:
Untying wengophone from Wengo servers IMO will seriously boost the
number of people who use it.
The Wengo management does not see this a priority, but myself i
believe that early implementation of this feature would cause more
people testing the software in various setups which would accelarate the
bug fixing.
Or the decission to suppress 'Call Transfer' it was not discussed with
the community....
I understand that there are problems related to the cost of this
service, but i'm sure the discussion of the issue could raise couple of
creative ideas, like for example to keep the service only between 2
Wengo numbers....
I'm personally aware of a number of VOIP providers who ask for this
feature. VOIP providers means more funding for the project.
Thanks
Vadim
_______________________________________________
Wengophone-devel mailing list
Wengophone-devel@lists.openwengo.com
http://dev.openwengo.com/mailman/listinfo/wengophone-devel
