openoffice-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From jan i <>
Subject Re: Fwd: Time to propose a CS Capstone Project!
Date Sun, 10 Aug 2014 14:09:47 GMT
On 9 August 2014 23:23, Andrea Pescetti <> wrote:

> On 04/08/2014 jan i wrote:
>> Is this something for us ?  I had a capstone project (build system for
>> windows) last term, and had quite some fun with the students.
> Since you were the only one to participate in the program so far, you
> probably have some useful additional information.
> For example (imagining that you had time instead) would you re-propose the
> project about build system? Was it too complex, too isolated, too...? I
> remember seeing some very welcome updates and wiki pages at the beginning,
> then I lost track. Did some code get committed at the end? Can you give an
> evaluation of your group of students (an overall evaluation, without
> discussion about individuals), and also estimate the effort required from
> you? With more information, maybe someone will show up.

I would not re-propose the build system, for a couple of reasons:
a) We have proven the a visual studio build system is possible, but the
community expressed doubt that a third build system is a good idea.
b) Even though quite a lot about the build system, its hard to mentor
student, without having active cooperation from other developers
c) It was basically not too complex, but it was a very hard job to split it
into pieces understandable by the students.

I have about 15 converted modules, and a script that can generate VC++
solutions. I have opted (until now) not to commit it, because its too
scattered for somebody else to build on.

The students are in their final year, and use this as a preparation to get
out in real life. The commitment and knowledge level of the students were
very different. The distance (portland USA - spain), did not offer a big

The initial efford was very high, I was so lucky to have a lot of help from
Steve H. (even though he is not at all involved in our project but in
XALAN-C). I think the two of us together used 2 days a week the first
couple of weeks to make VMs, build plan and help the students define the
project. After the initial round,  I used a couple of hours every week
(replying to email and having a google hangout).

A new project that builds on the one completed, would require a lot less
startup efford, and if someone volunteers then I will be happy to help.
Bear in mind any volunteer for that kind of project need to know how to
change our build system, not only use it. There are however plenty of other
possible themes.

>  There should be enough projects within AOO to pick and choose from
> If you have some ideas of possible projects this could help in selecting.
> For example, a "Summer of code" project is for one student for 2-3 months.
> Here, if I understand correctly, one can tackle much bigger issues, and the
> expected workload is 2-4 students for 8 months, so this could be an
> occasion to get significant contributions.

I think the problem is not finding a project, but finding somebody willing
to be mentor. Lets not discuss project, before someone volunteers, then the
project should be made to fit the experience of that person.

jan I.

> Regards,
>   Andrea.
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message