RE: [Dynapi-Dev] Counting so far

[Pascal B. <pa...@dy...>]

Voice of reason reaching us.

thanks.

Pascal Bestebroer
pa...@dy...
http://www.dynamic-core.net

> -----Oorspronkelijk bericht-----
> Van: dyn...@li...
> [mailto:dyn...@li...]Namens Richard Bennett
> Verzonden: dinsdag 30 januari 2001 18:36
> Aan: dyn...@li...
> Onderwerp: Re: [Dynapi-Dev] Counting so far
>
>
> Is this a case of win or loose?
> Eytan against Pascal??
>
> We've had all the theoretical explanations, every one has been able to
> demonstrate their in-depth knowledge and their preference regarding object
> model, a discussion is great, but lets get back to work now! Instead of
> breaking up the group with petty rivalries.
>
> There are different things to be done, and several groups of
> people showing
> interest in them:
>
> * NS6 debugging, starting at the core of the API, later the widgets.
> several people showed interest here.
>
> * Mac debugging, now we have some Mac users, their efforts should not be
> shunned.
>
> * Memory management, coupled with the basic structure of the API, OOP etc.
> This is what Doug and Eytan showed a lot of interest in, why not set up a
> test case, of a very basic API, so stability and memory use can be tested.
> If there is an improvement, the benefits should be clear. Also this would
> give people like me a chance to see what you are actually talking about, a
> lot of technical abbreviations make the whole thing a bit confusing.
>
> * User interfaces, several projects going on here, either web
> based or not,
> Dan's Java Builder, or Henrik Våglins DynBuilder.
> These projects might or might not take of, we won't know if we don't try.
>
> * Documentation, Robert and Pascal have put in a great effort
> here, keep up
> the good work.
>
> * Designer access, here I mean access for the normal javascript coder, who
> wants to check out the DynAPI, without necessarily knowing the code inside
> out.
> That's the part I'm working on as much as I can, supplying examples, and
> answering user questions as I can. These people will become our future
> user-base, and the step to DynAPI use should be as low as possible.
> See what happened to PHP in 5 years time, after year one they
> were where we
> are now.
>
> * Widget development, if the widget model has stabilized, it
> would be great
> to see some functional, error free, and logical widgets, we need window
> management, timeline, etc + a whole bank of pathanim add-ons to
> do all kinds
> of animation. There's a lot out there already which I'm trying to group as
> it is.
>
> * Group moderation, this is what Dann Pascal and Robert do (among others)
> and is basically taking the decisions.
> Someone has to do that, otherwise everybody would be overwriting
> each others
> code in CVS 'cause they didn't like the "style".
> There are a few things that could be improved here to, as I feel a lot of
> knowledge gets lost as it's simply ignored, but it should come from two
> sides, submitters should be able to accept at face value that
> their "great"
> fix, wasn't accepted, but the moderator should at least acknowledge having
> seen the submission, if not giving a reason for every line of code they
> reject/accept.
> Also, we should try to restrict any bug reports to something which can be
> reproduced, I mean, if you say, mouse events don't work in a nested layer
> containing an image, while the parent is moving across the screen, you
> expect the moderator to spend time reproducing that specific
> example. If it
> helps I'll open a public ftp account, with the latest snapshot
> preinstalled,
> so bug examples can be uploaded, and are all tested against the latest
> snapshot.
>
> Anyway, I think we have to much work to do, to waste time counting...
>
> Cheers,
> Richard Bennett
>
> ma...@ri...
> www.richardinfo.com
> (Everything running on, and ported to the 19/12/2000 snapshot of DynAPI2)