Fwd: [PHP-GTK] what is going on with gtk.php.net?
| From: | Michael Vergoz | Date: | Fri, 16 Jul 2004 15:47:47 +0000 |
| Subject: | Fwd: [PHP-GTK] what is going on with gtk.php.net? | ||
| Groups: | php.gtk.dev | ||
| Request: | Send a blank email to php-gtk-dev+get-632@lists.php.net to get a copy of this message | ||
Begin forwarded message:
From: Michael Vergoz <descript@php.net> Date: 16 juillet 2004 19:32:29 GMT+04:00 To: <steph@zend.com> Cc: "Rhys Campbell" <rhys.campbell@da-systems.co.uk>, "PHP-GTK Mailing List \(E-mail\)" <php-gtk-general@lists.php.net> Subject: Re: [PHP-GTK] what is going on with gtk.php.net? Hello I would like to make a few comments and raise a few questions PHP is a software enabling access to a function reference base base organised by groups. Ex. Database : MySQL, PgSQL, Oracle As far as I know there is no stand alone php mysql project Why PHP GTK would be one ? Why shouldn’t we think of a wrapper system that would run with GTK1, 2, Cocoa, Carbon and maybe Windows UI ? Why shouldn’t we call this project PHP-UI ? Indeed I have been working for six months on a wrapper system, up to now, I have drawn what would be necessary to make it My point is not to discuss what has been done up to now, I just think that PHP could have a bigger impact on final developpers. In addition, it would make doc creation easier I may be wrong but I am open to comments If you find it fine, I will dedicate some time to work on this subject. Regards -mike On 16 juil. 04, at 16:26, Steph wrote:Rhys: we already _have_ a centrally managed effort..-----Original Message----- From: Rhys Campbell [mailto:rhys.campbell@da-systems.co.uk] Sent: 16 July 2004 13:07 To: PHP-GTK Mailing List (E-mail) Subject: RE: [PHP-GTK] what is going on with gtk.php.net? OK, seems like not many people are with me on this. The ants anology is good, but like ants I think people will tend to leave the not so sweet parts. Anyway this post has generated a lot of traffic so there's obviously a lot of interest in the issue. I think the idea of a cenrally managed effort is a good one, but you may be right about it not getting used. Bit sad really as you can see the good Sun have done on the documentation for Java (a rival of PHP-GTK????). I'll start thinking about what I can do this weekend and stop pestering this list with my rantings. Regards, Rhys -----Original Message----- From: Steph [mailto:steph@zend.com] Sent: 16 July 2004 12:23 To: Rhys Campbell; PHP-GTK Mailing List (E-mail) Subject: RE: [PHP-GTK] what is going on with gtk.php.net?-- PHP-GTK General Mailing List (http://gtk.php.net/) To unsubscribe, visit: http://www.php.net/unsub.phpObviously I'm not saying throw away the present manual there is a lot of useful information there. But the old saying goes "Too many cooks spoil the broth.". What I'm saying is that central management of the documentation is needed to be effective, i.e. you look at this widget and document all \ or many of its properties, methods ect according to these guidlines \ standards, formal methods, or whatever. This work could then be submitted to a project leader(s) for review which would then be passed back for corrections or included into the manual. I'm sure, like myself, many have busy lives but if we work with common aims on this I think it can be done. Having people randomly work on sections will not produce a coherant manual - central management and review is the key.Getting any material at all would be pretty key from where I'm standing :) The more complex the process, the less likely you are to see anyone going through with it. 'Having people randomly work on sections' is _exactly_ what the php.net manual guys do; OSS documentation, like OSS anything else, is better demonstrated by watching a nest-ful of ants retrieve a teaspoonful of sugar than by anything else I can think of. There's already control there, in the form of other users, in the form of user notes, and in the form of peer review (quite a few people have CVS karma for the manual, but most of them are translators rather than initiators). It's much easier to get information corrected than it is to get information there in the first place; witness this list, and the amount of flak the PHP-GTK manual's taken over the years for incompleteness, versus the amount of people who have contributed original material to it. It's always easier to criticise existing work than it is to produce any. At php.net nowadays, the docs team request that new functionality (and new extensions) are initially documented by the person that wrote that functionality or extension. This isn't an option we have here; Andrei's contributed a couple of classes to the manual (and it took him a lot less time than it would've taken me), but we can hardly ask him to document 100+ classes in his spare time. Even if he had any. He automated what he could and left the rest for people to fill in. In other words, at present - still - if there are no signals and no properties outlined in a PHP-GTK class, it's safe to say that that class has never been touched by anyone writing documentation. I should maybe make the secondary point that the GTK+ 1.2.* project itself had the same documentation problems we face here. That manual was never completed either. The documentation for GTK 2 is similarly generated but was better thought through, with the developers getting involved to introduce new concepts. I'm blathering. Sorry :) but all this stuff does need thinking about before anyone goes to the trouble of setting up an environment that won't get used, or a second manual, or anything along similar lines. - Steph -- PHP-GTK General Mailing List (http://gtk.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php