[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-kuml-devel
Subject:    Fw: Question...
From:       "Darius Stachow" <dstachow () gmx ! net>
Date:       2001-11-22 18:57:47
[Download RAW message or body]

Hi guys,

I just would like to forward a kUML related mail so you get something to
read ;-)


----- Original Message -----
From: "Darius Stachow" <dstachow@gmx.net>
To: "Dan Marinescu" <TheVeryLittleOne@PacBell.net>
Sent: Thursday, November 22, 2001 7:55 PM
Subject: Re: Question...


> Hi Daniel,
>
> I guess you're refering to the compiler warnings on the sourcecodes for
the
> interactive graphical editor. The reason for these compiler warnings is
the
> way the sourcecode was made. I am using Together from TogetherSoft for
> sw-engineering and the sourcecode was generated out of a model. I mainly
> worked on the model and not on the sourcecodes so they aren't over-worked
by
> hand and mainly not implemented.
> There is just a rudimentary test implementation of a general editor to
test
> if the design works. Fortunately it does :-)
> The model for the editor's graph/network extension isn't tested yet fully
> for semantic correctness but I think the design should also work. The same
> But I remember there was some changes necessary I haven't done yet ...
> The same applies to the UML extension of the editor.
>
> The design is functional but the implementation has to be expanded. A
> problem is the missing documentation that would helps to do this. As the
> design isn't trivial anymore this would be a great help. Peter Rake,
another
> developer is helping on the editor and he pointed out this topic just a
> couple of days ago.
>
> I worked a lot with Windows to work on the model because Together was
> running a bit slower on a linux machine. After the model has reached a
state
> to try to get the generated sourcecode compilable I switched to linux.
> Through this switch I've gotten some dos-unix conversion related problems,
> but they should be all eliminated.
>
> I was the one who initiated this project and I still work on it in the
time
> I have left. But this time has got rather limited for some time now :-(
> It has began as student research project but my university isn't involved
in
> its development. They are just providing us some webspace. Students as
well
> as professional developer with all having special knowledge in their field
> of action are bringing their experience into the accruement of kUML.
> Beside myself there are up to 3-4 developer working on kUML. Unfortunately
> the overall progress in development is lacking as we all have limited
> available time to work on this project :-(
>
> What we urgently need is a KDE or Qt front end application that utilizes
> what we developed so far and what would provide other developer something
to
> see and to touch. That would make them potentielly interested in helping
in
> this project and this could speed up the development process.
>
> You're right. This product really could rock ... when it gets finished a
> day.
> The design and architecture of the whole application is advanced and
around
> this project has cumulated a number of pretty cool and experienced
developer
> which are interested in getting a really cool and powerful open source
cool
> CASE tool.
>
> It's really a pity that we don't make progress faster.
> More free time for the developer and a fresh and charged maintainer could
> soften this.
>
> If you're interested in this project you may subscribe to our mailing
lists
> and send your mails to this list.
> How to subscribe is described at kUML's homepage.
>
> Don't hesitate to drop me a mail if you have more questions ...
>
> Bye
> Darius
>
> ----- Original Message -----
> From: "Dan Marinescu" <TheVeryLittleOne@PacBell.net>
> To: <stacho_d@informatik.fh-hamburg.de>
> Sent: Thursday, November 22, 2001 1:16 PM
> Subject: Question...
>
>
> > Hi Darius,
> >
> > I got your source code and I am compiling it... I am just wondering,
> > what Operating System are you using in there? With all due respect, but
> > the source is "raining" with trivial warnings like "no newline at the
> > end of file" and variable which are declared and never used, some
> > operator overloads are commented out, some includes are definitely
> > deliberately omited, same thing about many casts (uncasted void* ),
> > integers receiving floats and even functions with no returns...
> >
> > So, my question is: "The general class desing and is almost perfect, but
> > the source is definitely very bad maintained, just like the initial
> > author is not with the project and some probably students are having fun
> > with it. Is that true?"
> >
> > The product could rock! However, it requires so much work to even get
> > compiled that I didn't believe my eyes... Why is that?
> >
> > PS.
> > I am using RedHat 7.2 (gcc 2.96 & QT 2.3.1) However, it seams
> > VERY CLEAR to me that a lot of problems do not have anything to do with
> > that... It's just bad programming practice overriding a brilliant
> > initial design, which is anything but amateur work! Period.  So, what
> > happened with this project?!? Who wrote it initially and who the h*ck is
> > "maintaining it now"? Don't you feel that it's pitty that such a product
> > to be so bad maintained? Gee...
> >
> > Regards,
> > Daniel
> >
> >
>
>

[prev in list] [next in list] [prev in thread] [next in thread]