I'm not a years developer (23 yo) but I never saw that, interesting,
but imao less human readable than NaturalDocs. Thanks anyway.
Okay I meant to say that is what I a lot of developers see all the
time. Anyways you clearly have never used Perl because it is used in
every single Perl module. Anyways I have never heard of this natural
docs thing so clearly we are both uh confused. Anyways I use JSAN
which uses POD ( the server for uploading to JSAN is dead BTW, but it
is still a good library just a bad repository )
http://planet.openjsan.org/doc/c/cw/cwest/JSAN/0.10/lib/JSAN.html
Anyways I just use JSAN.use("Widget.Dialog") for modules. If say you
decided to move to a JSAN package management system then we could
start to work on integrating our seperate modules. The real advantage
to JSAN is that you are only loading all your dependencies for an app
on the fly and before I added it my OS was slow.
Everything is kernel/lib is part of the core yes. For example, the
menu uses os.menu, wich depends on os.widget. The taskbar will soon be
a widget too, so we can add more taskbar, move them, etc.
What do you mean part of the core? What is the distinction between a
widget and "part of the core"? A core dependency or dependent on the
core? Take for example:
http://www.openjsan.org/doc/e/ed/edoardo/Widget/Tooltip/0.02/lib/Widget/Tooltip.html
If you look at the link you notice that the documentation has the
common POD format, so right away I understand what is going on. It
describes itself, where it has been tested, and all the important
public methods. It tells me that "Edoardo Sabadelli" wrote it and that
in this case if you put it in your widget/ directory that you did not
write it yourself. Then it also describes its license terms. Another
thing is it has private methods, you seem to not use them ever?
The most important thing I think is the synopsis you tell the user how
to handle the dependencies <script src="some dependency"> if you have
dependencies and then you can give an example of the code in action.
Clearly from that you realize if I am making my own operating system I
can use Widget.Tooltip and if you are making your operating system you
can also use it because it doesn't have a dependency on any platform,
it is modular. In my operating system you can take Widget.Dialog and
use it in any everyday application that has nothing to do with the OS.
Can you take your
widgets an use them standalone?
Another fix to come are windows, actually there is object model for
windows, there are only methods in the core (os) to create window
(createWindow, removeWindow), that have to be changed too.
Well I think you should work on seperating Windows into a
Widget.Takos.Dialog module or similar and this way you make it so that
you just can use somebody elses window system. And you can make it so
that if I do:
<script type="text/javascript" src="./Widget/Takos/Dialog.js"></
script>
<script type="text/javascript">
<!--
(function(){
var Widg = Widget.Takos;
new Widg.Dialog({
pid: 12374, // (process id)
x: 100,
y: 100,
width: 250,
height: 250,
innerHTML: function(container) {
container.innerHTML = "Alert! This is my first Dialog.";
}
});
});
//-->
</script>
That should run standalone. And it should say so in =head1 SYNOPSIS in
your docs. You should make a folder like ./Widget/ and then your own
Widgets are ./Widget/Takos and you can go and download other widgets
like Widget.Tooltip. (It is a good module). Why reinvent the wheel?
Anyways do you have a distinction between applications and widgets
yet?
