The gui in Eclipse is not multithreaded at all. A single thread can easily lock the entire gui, preventing it from redrawing itself. Most often seen with buggy plugins which just freezes the entire gui for 10 seconds at a time.
The reason the problem is much worse in FireFox then it is in eclipse is because so many of mozillas operations are gui operations. Running javascript in it's own thread sound easy, until you realize that almost all javascript operations, modify the DOM and thus need to run as gui operations because they trigger gui callbacks.
The correct, but also very costly(In code size and performance) way to do it is that all gui operations must be put in a object/struct which are then inserted in the gui queue, where the gui thread will then perform the actuelly gui operations.
But that is much work(And quite diffucult to get right), because when a javascript changes the dom, insted of just performing the needed drawing operations by calling the drawing operations in the Gui toolkit, it need to save all needed information for the drawing operation to an object which are then put in the gui drawing queue. But now your javascript is running out of sync with the gui, because the javascript continues without knowing the status of the drawing operations it have started. (Which can be solved by the gui thread, calling back to the javascript, but then things get really complicated).
Now if only the gui toolkits were allowing multi threading, things would be much more easy, but all known toolkits require that only a single thread is the gui thread, and only that thread, can send drawing/gui operations to the toolkit, and only that single thread can recieve callbacks from the gui.
btw: Does anyone know how to scale the TEXT in firefox 3? Ctrl +/- now scale the entire page, insted of just the text(Which looks bad with many images) -(
This might also be a problem with your current session. Try to start firefox with firefox http://www.google.com/ That always worked for me, when firefox3 beta2 did its start/crash thing
]In a rendering competition between the latest geforce and the latest quadro in maya or 3ds max or something, the quadro would completely obliterate the geforce I know that's how its suposed to be, but I have newer seen a benchmark between 'workstation cards' and 'gaming cards' which included example images from the different cards, that showed the difference.
This benchmark don't even include any example images, which I don't understand because it might be the biggest difference between the cards. Having a benchmark of 'workstation cards' that are suposed to look better then the gaming cards, and then not even including anything about the image quality is wierd.
It's a shame they don't test them against 'game cards'. It would be really interesting to find out how theese cards differ from the normal gaming cards, when doing realtime 3d.
But even if ie8 starts in "super standard mode" a site is only giving problems with ie8 if it have a doctype, and don't work in standard browsers, such as firefox. (If firefox can render it, so can ie8 super standard mode).
And it is many years ago I saw such a buggy site. (In fact I can't think of any site where there would be problems right now.
They would only need to change the site, if the site currently don't work in firefox and other browsers.
If the site work in firefox now, alll microsoft have to do is change internet explorer, so it don't trigger all the hacks that ie6/7 need, and your site works just fine in ie8 "super standard mode". (If "super standard mode" is a working implementation of html4+css2.1 that is)
So all Microsoft have to do is to change the user agent string to something that will not trigger ie6/7 workaround. That way they don't need a third render mode, but will get the same code as firefox and other browsers.
That does ofcause only work if ie8 can show code as well as firefox.
Then someone are fucking with you. Here in denmark i can get international bandwidth to the price of 5GB/$ (That is from a datacenter, if I transfer more then 2TB/month, so I guess its pretty close to whole sale price)
I use 5.0, and the point is, I should not have "check the success of all queries before"
If any part of the transaction fails, there is NO WAY mysql should allow the rest of the transaction to be performed. (That is, a commit on a transaction which contains queries that failed, should be a nop)
The A in acid is ATOMIC, that is "all or nothing".
Quote: 1. Lack of flash plugin. Yes, they totally side-stepped the legal problems, but how about a script to do the job on startup??
What legal problems would that be? Adobe allows you to distribute the flash plugin with your software or service, so preinstalling flash would be legal. (See http://www.adobe.com/licensing/)
Including a link to youtube, without including flash is stupid.
That your database can't generate unique id values in that situation sounds like a problem with your database software, not with the concept of auto numeric keys.
Could you not just give each db a sequence from which to give out id numbers. So DB1 got from 0 to x and db2 got from x to y, and so on.
(I do asume that any software do not require the numeric keys to be increasing, just unique).
I know that you do layout in CSS. The problem is CSS is an inadequate way to express layout. Where is my "make a three column grid that extends the height of the page" in CSS? I don't know where yours are, but mine is in grid.css or gridhack.css (Asuming that 'extends to the height of the page" mean that you always want the grid to have height atleast as high as the browser window. (There is a reason, they made the min-height tag)
But I think the real problem, is that we are still writing html/css by hand. Html is really the only document format written by hand anymore. I mean, you might hate html/css, but just try to write some postscript or pdf, and you will love going back to html.
What we really need is a GOOD html editor, so we can describe the page as : I want 2 colums this size, and the rest of the space given to the third column. Html was NEVER intended to be a format that was at large written in hand.
Quote: I want to install VMWare on a Linux distro without having to need a compiler installed.
Then run a linux distribution that is supported by vmware. You can't expect to run vmware on some random linux distro, no more then I can expect to run my Windows version of vmware on Windows mobile.
(And vmware 5.5, don't have any problems with the newer linux kernels. I am runnig it on 2.6.22 right now), so how old exactly is your wmvare?
In other toolkits a lot of basic functions like redrawing are handled by the OS or by a separate thread that manages low-level gui 'controls' What other toolkits would that be? Both Win32 and QT have the same basic problem, that you MUST recieve events in the gui thread, and that the gui thread is the only one that draw. So if you recieve an event end do a while(true) { } neither win32 or qt will redraw your window.
The basic problem is that neither windows gui(win32) nor X11 are thread safe, and that they use the same thread to draw and handle events.
Slashdot Mirror
If you think of powray as an animation player, and have a REALLY fast computer, then the animation is only 512 bytes :}
The gui in Eclipse is not multithreaded at all. A single thread can easily lock the entire gui, preventing it from redrawing itself. Most often seen with buggy plugins which just freezes the entire gui for 10 seconds at a time.
The reason the problem is much worse in FireFox then it is in eclipse is because so many of mozillas operations are gui operations. Running javascript in it's own thread sound easy, until you realize that almost all javascript operations, modify the DOM and thus need to run as gui operations because they trigger gui callbacks.
The correct, but also very costly(In code size and performance) way to do it is that all gui operations must
be put in a object/struct which are then inserted in the gui queue, where the gui thread will then perform the actuelly gui operations.
But that is much work(And quite diffucult to get right), because when a javascript changes the dom, insted of just performing the needed drawing operations by calling the drawing operations in the Gui toolkit, it need to save all needed information for the drawing operation to an object which are then put in the gui drawing queue. But now your javascript is running out of sync with the gui, because the javascript continues without knowing the status of the drawing operations it have started. (Which can be solved by the gui thread, calling back to the javascript, but then things get really complicated).
Now if only the gui toolkits were allowing multi threading, things would be much more easy, but all known toolkits require that only a single thread is the gui thread, and only that thread, can send drawing/gui operations to the toolkit, and only that single thread can recieve callbacks from the gui.
btw: Does anyone know how to scale the TEXT in firefox 3?
Ctrl +/- now scale the entire page, insted of just the text(Which looks bad with many images) -(
This might also be a problem with your current session. Try to start firefox with
firefox http://www.google.com/
That always worked for me, when firefox3 beta2 did its start/crash thing
This benchmark don't even include any example images, which I don't understand because it might be the biggest difference between the cards. Having a benchmark of 'workstation cards' that are suposed to look better then the gaming cards, and then not even including anything about the image quality is wierd.
It's a shame they don't test them against 'game cards'. It would be really interesting to find out how theese cards differ from the normal gaming cards, when doing realtime 3d.
But even if ie8 starts in "super standard mode" a site is only giving problems with ie8 if it have a doctype, and don't work in standard browsers, such as firefox. (If firefox can render it, so can ie8 super standard mode).
And it is many years ago I saw such a buggy site. (In fact I can't think of any site where there would be problems right now.
They would only need to change the site, if the site currently don't work in firefox and other browsers.
If the site work in firefox now, alll microsoft have to do is change internet explorer, so it don't trigger all the hacks that ie6/7 need, and your site works just fine in ie8 "super standard mode". (If "super standard mode" is a working implementation of html4+css2.1 that is)
I think that Firefox just mimicked IE when the site don't contain a doctype.
I don't think that there are any case where firefox follow ie6/7 insted of w3c when showing pages with a valid doc type.
So all Microsoft have to do is to change the user agent string to something that will not trigger ie6/7 workaround. That way they don't need a third render mode, but will get the same code as firefox and other browsers.
That does ofcause only work if ie8 can show code as well as firefox.
but all theese millions of pages pages render fine in firefox, using standard mode so why can't they render fine in ie8?
If it's not fixed width, it will be as wide as it's parent and thus centering it would not make sense. What is it you are trying to do but can't do?
Then someone are fucking with you. Here in denmark i can get international bandwidth to the price of 5GB/$ (That is from a datacenter, if I transfer more then 2TB/month, so I guess its pretty close to whole sale price)
(And before someone correct me, the atomic is acid, is not what is violated by mysql in this example).
I use 5.0, and the point is, I should not have "check the success of all queries before"
If any part of the transaction fails, there is NO WAY mysql should allow the rest of the transaction to be performed. (That is, a commit on a transaction which contains queries that failed, should be a nop)
The A in acid is ATOMIC, that is "all or nothing".
How do you make mysql reject a transaction with invalid operations?
Right now, the mysql I use(with innobase), will commit a transaction even if some of the operations(Queries) in it failed. That's not acid behaveour.
Quote:
1. Lack of flash plugin. Yes, they totally side-stepped the legal problems, but how about a script to do the job on startup??
What legal problems would that be? Adobe allows you to distribute the flash plugin with your software or service, so preinstalling flash would be legal. (See http://www.adobe.com/licensing/)
Including a link to youtube, without including flash is stupid.
That your database can't generate unique id values in that situation sounds like a problem with your database software, not with the concept of auto numeric keys.
Could you not just give each db a sequence from which to give out id numbers. So DB1 got from 0 to x and db2 got from x to y, and so on.
(I do asume that any software do not require the numeric keys to be increasing, just unique).
the grid to have height atleast as high as the browser window. (There is a reason, they made the min-height tag)
But I think the real problem, is that we are still writing html/css by hand. Html is really the only document format written by hand anymore. I mean, you might hate html/css, but just try to write some postscript or pdf, and you will love going back to html.
What we really need is a GOOD html editor, so we can describe the page as : I want 2 colums this size, and the rest of the space given to the third column. Html was NEVER intended to be a format that was at large written in hand.
Quote: I want to install VMWare on a Linux distro without having to need a compiler installed.
Then run a linux distribution that is supported by vmware. You can't expect to run vmware on some random linux distro, no more then I can expect to run my Windows version of vmware on Windows mobile.
(And vmware 5.5, don't have any problems with the newer linux kernels. I am runnig it on 2.6.22 right now), so how old exactly is your wmvare?
while(true) { } neither win32 or qt will redraw your window.
The basic problem is that neither windows gui(win32) nor X11 are thread safe, and that they use the same thread to draw and handle events.
Which just make it so much worse, that Microsoft did not implement the entire ECMAScript standard.
Read the quote again:
"division also recorded its second quarterly profit in a row on July 18 on revenue of $90 million."
The $90 million is revenue, not profit. There is no indication of what the profis is.
This would actuelly make a really cool Slashdot vote.
How old were you, the first time you had sex:
Below 15
15
16
17
18
19+
What is sex??
Are you by any change running on an motherboard with an nvidia chipset, using active armor(Or whatever they call the included firewall?)
Because that's the only thing I have ever seen, that could corrupt a tcp/ip download.