Re: [reporting module]

On Thursday September 8 2005 10:58 am, you wrote:
> Adam Treat said the following, On 2005-09-08 16:38:
> > What about Kugar... hmm...
> >
> > http://www.kexi-project.org/wiki/wikiview/index.php?Reporting#4._What_abo
> >ut
> >
> > Well, as the Kugar maintainer, I have a few things to add:
> >
> > 1.  Kugar _is_ being maintained.  One simply has to look at the recent
> > svn history to see this.  Furthermore, Jaroslaw, you well know it is
> > being maintained as does Alexander.
>
> Ok, in these terms it's maintained. It's not *that* level of vitality, a
> few of developers, namely me and adymo, could dream about. We talked,
> without criticizing anybody, what to do to improve things, and whether
> adding report painting engine from scratch is a good idea (design
> componenet, as reusing Kugar's designer component is not an option). More,
> I've spent hours on trying to identify the problems by looking at the code,
> again in Malaga, before being able to back and propose the suggestions.

Well, see there you go.  You say on your bullshit page that it is not 
maintained... which is presumably what you said in Malaga.  Of course, this 
is not true as you damn well know.

What's more, if you are not happy with my 'vitality' ... well, you have even 
less.  I do not see much activity in Kugar coming from you.  There simply is 
no replacement for Kugar.  You can dream all you want, but that is just what 
it is: dreaming.

> > 2.  The fact that Kugar is database platform independent is a _good_
> > thing. If Kugar were not then it necessarily limits the apps that can use
> > it.
>
> That is why I mentioned: "new reporting module, meybe coming with Kexi,
> maybe not, could be emulating this behaviour, as a one of it's use cases".

So, you agree that it is to Kugar's benefit to be database platform 
independent?  If so, then why on earth did you list this as one of the 
Kugar's current "problems"?

Maybe you should spend a few more hours getting your story straight.

> > 3.  Please explain why iterating over dom nodes is a bad thing and what
> > iterating over 'QWidget trees' gives you.
>
> Personally I see a clear object model, as an advantage, instead of having
> to depend on single-layered XML, not supporting inheritance,
> features-introspecion and so on - anything we've learnt developing
> KFormDesinger, which is awesome in it's current state.

I want a _CONCRETE_ example.  What, precisely, would this give the end user.  
I don't want anymore hazy dreams.  That is how software becomes a permanent 
alpha.

> > 4.  Yes, Kugar has a specified set of display objects.  These consist
> > mainly of labels and data fields.  What else do you need and what do you
> > mean by 'tweaking' them.  You want oval labels or something?
>
> Factories, providing an entry point to insert a graphics, charts, plots,
> barcodes and anything that 3rd party devs can dream about.

This could be done by modifying Kugar _without_ a rewrite.

> > 5.  Any talk of removing Kugar as the reporting component of KOffice is
> > wildly off the mark.  First, you have _nothing_ to replace it with.
>
> We've got a decent designer componenet, see above. And, my research is
> mostly about looking at options for rendering engine.

So, you agree.  You have _nothing_ to replace it with.  Just dreams.  Which 
might be good dreams, yes, but hardly something with which to begin comparing 
to Kugar as it is now.

> > Third, other applications _are_ using it.
>
> No matter that Kugar is going to be available outside of KOffice or within
> it, it will be available as long as it's maintained (or better, actively
> developed), what's a problem?

Exactly.  So keeping Kugar inside of KOffice will be no problem.  Glad you see 
it that way.

> > Finally, _you_ do not get to make that decision.
>
> s/decision/collective proposal/

Well, seeing as how you won't discuss this with me on IRC, and everyone else 
is telling me that you were the only one discussing Kugar at all at 
aKademy...  then this is _your_ "collective" proposal.  One, which is wildly 
premature.

Jaroslaw, this is frustrating precisely because I have talked about this with 
you many times and have been cooperating with you.  Not once did you mention 
these problems.  Yes, you mentioned that you thought your form designer would 
make a good report designer... and yes, that is why I integrated koproperty 
into the current designer.

But you never expressed these other problems.  You shouldn't have been 
spreading falsehoods at aKademy especially since you know I've been 
maintaining Kugar and you knew I wasn't there.

Please take down the known falsehoods on your wiki.

Adam

--

-- 
I'm running a Marathon in December!
HELP ME SAVE LIVES and Donate Today!
http://helpmesavelives.blogspot.com/