WebOS Project seeking for developers

R

Rorist

SAM Wrote:
You just need to drag down, the pointer leaves the title bar.
From there you can quit the mouse button, the drag will restart coming
back over the title bar.

Ok, not observed on Debian/Firefox3 tough, my bad. It's strange
because there is an event listener that checks if the window is draged
out of 'desktop' Element, the position is modified inside the
'desktop'. I look into that. The event is on the window itself,
looking for its position and compared to the desktop position (still
mootools based anyway).
Open the word thing. Mouse down on the black header. Move mose down,
quickly. The word thing stops dragging. Release mouse button. Move mouse
around. The word thing continues to drag. It's not noticing the mouseup,
probably because the event listener may be (I didn't look) attached to the
wrong element. Hint: document or body.

Oh yes, it happens with Word because it uses an ... iFrame. I won't
say what library Word is using, no matters.
Same will happen with "Media Player" as it is flash. But the iFrame in
Word takes all the space of the Window Content, so the mouse is going
into it very quickly. This Word thing is going to disappear anyway, a
friend is going to implement his Editor called Beacon:
http://dev.gentooexperimental.org/~n9986/beacon/trunk/editor/


David said:
One entry sticks out:

" - Apparently, I was drunk when I committed this. I sincerely
apologize to everyone losing their <head> and <body> because of me. "

I suspect the original author(s) were drunk as well.

Human factor is terrible, "realy".
 
J

jhuni

In TakOS, widget depends on the core

Why? Why can't you have a seperate Widget namespace like in JSAN and
like I have in my own OS? Is there some utilities or something? Why
don't you do this: JSAN.use("utils.DOM") or some such and make that a
dependency for the widgets instead of making it require the whole
core.
Real interesting library, I'm not sure I will directly use it

Well how do you handle packages and getting JavaScript on demand/on
the fly? Another thing is without JSAN how do you handle an overload
of RAM in the DOM? You see if you load 500 different applications at
the same time your OS is screwed. Here is what I am working on:

A way to unload a JSAN package from the window namespace after having
loaded it (using delete). This way if you go in and you have the first
250 applications inactive you can load another 250 without having any
memory from the first 250 this way you will actually be able to have a
stable kernel and actually be able to call yourself a web operating
system. My kernel only deletes memory from closed applications right
now, another thing that will be like some futuristic kernel is virtual
memory ( if an app is minimized then save its HTML-DOM memory ).

Anyways another thing that really helps the kernel like a lot is
http://search.cpan.org/~pmichaux/JavaScript-Minifier-1.05/lib/JavaScript/Minifier.pm
do you have any concept of compiling the OS code before hand? The one
module that the kernel should never unload is JSAN itself which is 6.7
kb which is super light I don't see why anybody wouldn't want to use
it. Anyways compiled it is 4.4 kb. That is so small
it should not be much of a problem for anybody. Anyways you get a 60%
increase in general in speed after compilation that is very
significant just look at Pysch Desktop it takes a long time to load
http://www.psychdesktop.net/desktopdev/desktop/index.html and it is
really slow it is not something you will be running on say a mobile
phone.
 
D

David Mark

Why? Why can't you have a seperate Widget namespace like in JSAN and
like I have in my own OS? Is there some utilities or something? Why

What OS would that be?
don't you do this: JSAN.use("utils.DOM") or some such and make that a
dependency for the widgets instead of making it require the whole
core.


Well how do you handle packages and getting JavaScript on demand/on
the fly? Another thing is without JSAN how do you handle an overload
of RAM in the DOM? You see if you load 500 different applications at
the same time your OS is screwed. Here is what I am working on:

A way to unload a JSAN package from the window namespace after having

From the what?
loaded it (using delete). This way if you go in and you have the first
250 applications inactive you can load another 250 without having any
memory from the first 250 this way you will actually be able to have a
stable kernel and actually be able to call yourself a web operating
system. My kernel only deletes memory from closed applications right
now, another thing that will be like some futuristic kernel is virtual
memory ( if an app is minimized then save its HTML-DOM memory ).

That really sounds like something.
Anyways another thing that really helps the kernel like a lot ishttp://search.cpan.org/~pmichaux/JavaScript-Minifier-1.05/lib/JavaScr...

A minifier?
do you have any concept of compiling the OS code before hand? The one

Do you?
module that the kernel should never unload is JSAN itself which is 6.7
kb which is super light I don't see why anybody wouldn't want to use
it. Anyways compiled it is 4.4 kb. That is so small

Oh brother.
it should not be much of a problem for anybody. Anyways you get a 60%
increase in general in speed after compilation that is very

And just look at the difference!
significant just look at Pysch Desktop it takes a long time to loadhttp://www.psychdesktop.net/desktopdev/desktop/index.htmland it is
really slow it is not something you will be running on say a mobile
phone.

Nor FF3/Windows apparently. Dojo threw a "/" error. I'm not
kidding. Console also displayed a cryptic "js:flash loaded" message.
Page was blank.
 
J

jhuni

What OS would that be?

Its in alpha (nobody has seen it yet except yours truly)
From the what?

The window namespace contains all JavaScript data...
That's nice! Five hundred and sixty two K of javascript to produce a blank
page.

Ya the server has too much traffic I presume. If you could download it
and
run it in FF3 you could see how slow it actually is.
 
L

Laurent vilday

rf :
jhuni :


That's nice! Five hundred and sixty two K of javascript to produce a blank
page.

It just gets worse.

IE8b1 produces a lot of error dialog. That's not blank :D
 
R

rf

Laurent said:
rf :

IE8b1 produces a lot of error dialog. That's not blank :D

FWIW I was using FF which did produce a rather fatal error before giving up
on the whole lot.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
473,769
Messages
2,569,578
Members
45,052
Latest member
LucyCarper

Latest Threads

Top