Re: [bme-develop] Re: Possible new project member

[Jixt <jan...@pa...>]

Ok thanx! I only have to say one thing before I start coding ;).

I dont want to say how you all should do it, you all have a vision about =
BeMSN and I respect that and I will follow that. I will tell how I think =
about some things but that does not mean that they have to be done like t=
he way I think about it. We all program here for fun ;) and lets keep it =
like that ;)

greets

   Jixt

P.S: I agree with the Be way of programming and that is good! We do not n=
eed the Windoze and Linux crap ;). I disagree in the way that we have to =
mimic it a littlebit first. With that I mean that we have to make sure th=
at we have all the features that the windoze version has ;). After that w=
e can build extra nice features for BeOs users only :p

>Hi Jixt,
>
>Very busy with exams atm, so I'll make this quick... more will
>come later. 
>
>[snip]
>
>> >I agree with you here...double work is something the beOS community i=
s good 
>> >in ;) So you already did some work on an MSN client....is there some =
code we 
>> >would be able to re-use in this project? Btw you must know we want to=
 make 
>> >this messenger the Be way, not mimic the windoze MSN, but trying to s=
urpass 
>> >it...especially in user friendliness....
>> 
>> I agree and disagree a bit with that. Comments will come later ;)
>
>Sorry, you can't really disagree with that. It's fairly
>fundamental to the whole way we have planned the project. Think
>about what needs doing, think and discuss the absolute best way
>of realising it, then implement it. I'm never going to clone any
>software in my life.
> 
>> >>I can say that I am good in c++( I love it!!) and I like solving dif=
ficult 
>> >>problems ;). Maybe some of >you know I am the creator of the BeDiGiC=
am, so 
>> >>that is the experience I have in developing in ?>BeOs. I can say tha=
t I'm 
>> >>not new in developing applications under BeOs.
>> >I know the app, doesn't work with my Fuji Finipix sad enough...good i=
n c++ 
>> >is nice...I am only starting to get the hang of it...
>
>My c++ isn't great, but just about gets the job done.
>
>> I've seen some things that could change(login procedure...) in the GUI=
. Chat screen need some bug-fixes....
>
>I personally vote to have logging in on the main window - like
>with Jabber for BeOS.
> 
>> 1) When a Windoze MSN'er sends a file the BeMSN craches.
>> 2) It is also possible to send empty messages.
>> 3) When a chat screen opens, it is opened in 'small' in the left corne=
r. You need to resize it and drag it to the middle of the screen.
>> 4) ...
>
>The new GUI will pretty much rewrite the whole chat window code;
>so don't worry too much about little bug fixes to the old GUI.
> 
>> I think a test-procedure should be created(readable for beta-testers) =
and there should be a
>> good developing-plan. Milestones are needed. This accelerates developm=
ent. A
>> bug-reporting-system is also needed to have control over the bugs. And=
 it should be very handy
>> that every developper has his own part of code to develop.
>
>For now we have sf's bug tracker - although a lot of the code
>will be reworked before bme version 1. We have milestones - it's
>called version 1 :)
>
>At the moment Daniel is working on the network side of things,
>and me and Tim are working on the GUI.
>
>> I mean that every developer does development of some component (ex: ch=
at screen) and testd the
>> component of another developer during alpha testing. In the alfa testi=
ng every feature that
>> should be in a certain release should be implemented. After alfa testi=
ng, a Beta should be
>> released to the beta testers. This is a structured way of working and =
will bring much progress
>> to this project ;)
>
>I'm sure you don't mean to sound critical of the way we are
>working, but that's kind of how I read it. What will bring
>progress to the project is us having some free time to develop
>(should come soon) - it's not a lack of structure that is
>holding development back IMHO.
>
>Testing strategies are always good though - but we have to be
>careful we don't test too soon - as I say above a lot of the
>code in CVS will probably not be in bme version 1.
>
>> It is just my way of thinking, I don't now what you all think about it=
 ;)
>
>Generally I think it fits well with our attitude about creating
>a really high quality, professional program.
>
>Welcome to the project!
>
>
>Simon