Re: annotations

[Michal H. <ms...@gm...>]

On Tue, Mar 04, 2008 at 01:06:39PM +0000, Jozef Misutka wrote:
> 
> hi,
> i will try to summarize this discussion into several points:
> 
> 0) pdfedit is a really nice project thanks to all 3(4) :) of us!!! it is used above my expectations and i am happy to be a part of this team!! making such a project in 3(4) ppl cool guys!
> 
> however, we are too slow because of many things life/work/school/... so when i devote my time to this project i do not want to ""fight"" against you (mainly michal :)) when the only problem is that we do not understand each other. for me it is really a tough nut to explain my design views to you without letting you see the actual implementation so lately i do not even bother. 
> i rather implement it all by myself (i agree this is a wrong approach indeed but on the other hand i think it is very good for the project itself because it moves forward much more quickly!!!!) 
> 
> 1) you are of course right with the cvs and branches!!! i will make a branch as we agreed. 

Thx.

> i am really sorry when i have discouraged you! but on the other hand i am discouraged by many things too and wanted to avoid them, that is why i had taken this approach.

Could you be more concrete? By what are you discouraged? What did you
want to avoid? 
It is hard to prevent something if you don't know what it actually is.

> 
> 2) i will try to put it simple. i am open gto your ideas (perhaps it does not seem so many times but i really am!), but as cpage is my responsibility (authoritative approach i agree) i will submit the code and then we can talk about it. 

How can we talk about something that I don't have big overview about? It is
very easy to overlook some details, if you can't follow commit logs and
comments. It is something different if you make small incremental
changes, but this is not happening AFAICS.

> why? simply said, cause otherways it would take ages (<<-- that is how i feel and you do not have to agree). 

Again, could you be more concrete? How many times did you need to revert
my changes? How many times did your work stucked only because pointless
discussion (for how long)?

> 
> 2.5) so basically yes, you are ""forced to follow"" (as i am when it concerns your code!). as you pointed out very nicely both are working just i find my more e.g. modular..

What would happen if all of us would be so restrictive? I don't want to
make this discussion more hot but you are using kind-of-strong language quite 
often and I don't see very good reason for that. If you don't like that
I am commenting your changes and tell you if I don't like something then
I don't know what to do about it. We have common target, so we have to
discuss...

> 
> 2.55) i do not like the idea that anybody else handles cpage than the owner, in this case - me. why? it was this way and it did no good. 

This has one basic problem, I have mentioned in previous email - we have
only one person who is able to handle important part of our code base
(gui is almost under control of Martin a quite honestly we are lost
there and also kernel has changed without documentatio to be updated so
it is hard to keep track).
If you are unreachable or you loose (I hope that won't happen) interest,
then the rest of the team have serious problem. 
We have agreed on maintains file which should give a clue who should
do critical decitions about file. There is nothing like ownership of the
file.

> 
> 
> 3) pdfedit is a nice project, BUT as we are not paid for it i do not have time to discuss every design issue consisting 90% of my code. the problem is that you do not see the whole picture, and i really do not have time to describe it here, when it is implemented i will gladly discuss any questions. 

I think that we are not in havy design changing times anymore. We should
focus on bug fixing and slowly adding some new features. So we should
give good discussion before any big change (just look at our changelog
and how many new features were added since we are on sourceforge). 

And once again, note that there is some code outside which depends on 
our changes and if you start shuffling with code then you make a lot 
of work there too. We are also developing in separate branches and 
merging comes very complicated if you shuffle with code (also minor
change can produce big problems - e.g. changing namespace name like you
suggested some time ago).

> as we agreed before, cvs branches were the best place to publish these changes but see 1) why i had not used them.
> 
> 4) after i create the branch we (michal and me) can together decided which code you want to move away from cpage

ok.

> 
> 
> hopefully i have not missed anything. 
> and guys, keep up our good work!!!
> 
> jozo
> 
> 
> PS: i have offered pdfedit as a project at charles university :) (he/she would be responsible for an independent part or feature set which we can integrate afterwards to our cvs)

Nice.
If it goes on somehow, I have same ideas what can be done, so let me
know and I will try to put them on the paper.

-- 
Michal Hocko