With this version, it becomes apparent that the end of Firefox is coming.
It has a very strong bias towards Microsoft Windows, which they justify by saying that's where most of the market is. Virtually all recent features are only available on Microsoft Windows, some of which are even only available on Windows Vista or Seven. As far as I'm concerned, Firefox has only seen regressions on Linux and Mac OS X since version 3.5. It's clearly become slower with each release, despite the announced performance improvements for Windows.
That bias is hurting them much more than they realize : they're alienating the open source community on Linux, which is where they originally came from, and all the cool and techies, the influential people, are either using Mac OS X or Linux. So if they stop using Firefox, how long before they relatives see that and try to copy them?
Also, the fact they only work on Windows is a demonstrator of their technical weakness. It seems Mozilla has now become an aggregation of marketing people and inane UI designers, with Q&A testers on top, instead of the coding gurus it used to be. Good code is portable code. Good programming practices are to organize your code in such a way that you never have to write platform-specific code more than you have to, by isolating the strictly platform-specific parts. The fact that Firefox on Windows and on other platforms is so different is a testament that a significant design mistake was made somewhere. Other platforms are not even prioritized for testing. That's also against good practices. The best way to find bugs is to make code run in separate environments, as the differences in those environments are what might trigger some otherwise rare code in the "main" environment.
What's that with using Direct2D for accelerated 2D rendering? All rendering is done by Cairo, which should be exploiting hardware acceleration. If Cairo cannot use Direct2D, it should be extended, rather than modifying Firefox itself to support Direct2D. What happened to the projects of modernizing the Mozilla codebase, which is written in C with classes and macros crap? Brendan Eich, Mozilla's CTO, had ambitious projects to do that. He has had them for what, almost ten years now? It was due for Firefox 2. Why was that delayed? So that we could have personas and tabs on top instead? Adobe, one of the world's greatest software companies, GAVE them a JIT virtual machine designed for Javascript (i.e. the hard part in getting a Javascript JIT compiler) FIVE years ago. They're still not using it in Firefox, which makes it the ONLY modern web browser lacking a Javascript JIT engine. With all the money they've got, they could have hired a couple of compiler engineers and created a new Javascript compiler from scratch that targeted that virtual machine in a couple of years at most. Google certainly didn't have any problem doing that plus the virtual machine in less than a year. What's up with HTML5? They're clearly lagging behind Webkit and Opera, even though they were clearly at the front a couple of years ago.
Firefox has lost its appeal, is lagging behind, and is now desperately trying to copy Chrome so that it has new things to put in new releases. The highlight of releases are now useless gimmicks that are really regressions, instead of the major innovations they used to be in the past. It is clear to me there is a real problem in leadership, with too much "rethinking the web experience" buzzing and not enough technology. Coders have deserted Gecko in favour of Webkit, since the codebase of the latter is much more modern, streamlined and nice to work with. They're losing all that made them what they were bit by bit. The only thing holding Firefox together is that before it became popular and amassed a ton of money, which unfortunately killed it, it used to be high-quality open-source software, the state of the art of web technology. They lost that, and with that, their end is ineluctably coming.
And replacing practical, explicit information, displayed in the idiomatic way users are used to, by some unintuitive color codes placed up rather than down, breaking typical workflow is hardly a good idea.
With the status bar, you know whether the browser is trying to resolve the host, establish a connection, or is downloading data. With some animated weird progression thingy, you don't really know anything. You barely know that it is loading, and that's it.
The attacker they really mean here, is the user who purchased the Internet-connected HDTV. Indeed, it is possible for him to trick the TV that is connected to his network infrastructure into doing things the manufacturer had been trying to prevent the user from doing.
This is not very different from jailbreaking your own phone or video game console, except it's much more trivial.
Once upon a time more companies were trying to release more reliable machines, but the costs were high - Enter Dell, eMachines, Acer and Gateway (the latter three now one and the same), and their business models of inexpensive PC's that aren't necessarily solid broke the market entirely.
And anyone with half some experience purchasing laptops knows better than to get one from Acer.
Computers are becoming disposable, much in the same way mobile phones are.
Or rather, that's what the vendors would like you to think. Smart customers realize it's better to purchase quality. But quality laptops are getting quite high to come by these days.
Although I'll readily admit I often did not appreciate it as a student. It has been after a decade of working that I now am grateful to have been compelled to learn software and hardware theory, not merely programming.
Looks like you still have a few misconceptions. Computer science has little to do with software or hardware. To learn those, you should take a degree in software engineering or hardware engineering, respectively.
Computer science is about algorithms, complexity and computability.
But the reality is (often literally, during coursework), that the platform, hardware or software, is largely irrelevant to the mathematical development of algorithms.
At bachelor's level, maybe. But at master's level, you have to design algorithms that can benefit from parallelism in the underlying hardware architecture.
But no, people protest in large groups in streets all the time, force closing of universities, prevent people from going to work, without any permit whatsoever. And they don't get fines nor get thrown to jail.
Sorry, but while functional programming style is indeed the future of HPC (with C++), functional languages themselves aren't. Read the research papers of the field and see for yourself.
Slashdot Mirror
Sorry I meant vertical space, of course.
Gaining five pixels of horizontal space (0.5% of your typical screen) does not justify that major usability deterioration.
Content is not always more important than the application to manage it.
With this version, it becomes apparent that the end of Firefox is coming.
It has a very strong bias towards Microsoft Windows, which they justify by saying that's where most of the market is. Virtually all recent features are only available on Microsoft Windows, some of which are even only available on Windows Vista or Seven.
As far as I'm concerned, Firefox has only seen regressions on Linux and Mac OS X since version 3.5. It's clearly become slower with each release, despite the announced performance improvements for Windows.
That bias is hurting them much more than they realize : they're alienating the open source community on Linux, which is where they originally came from, and all the cool and techies, the influential people, are either using Mac OS X or Linux. So if they stop using Firefox, how long before they relatives see that and try to copy them?
Also, the fact they only work on Windows is a demonstrator of their technical weakness. It seems Mozilla has now become an aggregation of marketing people and inane UI designers, with Q&A testers on top, instead of the coding gurus it used to be.
Good code is portable code. Good programming practices are to organize your code in such a way that you never have to write platform-specific code more than you have to, by isolating the strictly platform-specific parts. The fact that Firefox on Windows and on other platforms is so different is a testament that a significant design mistake was made somewhere.
Other platforms are not even prioritized for testing. That's also against good practices. The best way to find bugs is to make code run in separate environments, as the differences in those environments are what might trigger some otherwise rare code in the "main" environment.
What's that with using Direct2D for accelerated 2D rendering? All rendering is done by Cairo, which should be exploiting hardware acceleration. If Cairo cannot use Direct2D, it should be extended, rather than modifying Firefox itself to support Direct2D.
What happened to the projects of modernizing the Mozilla codebase, which is written in C with classes and macros crap? Brendan Eich, Mozilla's CTO, had ambitious projects to do that. He has had them for what, almost ten years now? It was due for Firefox 2. Why was that delayed? So that we could have personas and tabs on top instead?
Adobe, one of the world's greatest software companies, GAVE them a JIT virtual machine designed for Javascript (i.e. the hard part in getting a Javascript JIT compiler) FIVE years ago. They're still not using it in Firefox, which makes it the ONLY modern web browser lacking a Javascript JIT engine. With all the money they've got, they could have hired a couple of compiler engineers and created a new Javascript compiler from scratch that targeted that virtual machine in a couple of years at most. Google certainly didn't have any problem doing that plus the virtual machine in less than a year.
What's up with HTML5? They're clearly lagging behind Webkit and Opera, even though they were clearly at the front a couple of years ago.
Firefox has lost its appeal, is lagging behind, and is now desperately trying to copy Chrome so that it has new things to put in new releases. The highlight of releases are now useless gimmicks that are really regressions, instead of the major innovations they used to be in the past.
It is clear to me there is a real problem in leadership, with too much "rethinking the web experience" buzzing and not enough technology. Coders have deserted Gecko in favour of Webkit, since the codebase of the latter is much more modern, streamlined and nice to work with. They're losing all that made them what they were bit by bit.
The only thing holding Firefox together is that before it became popular and amassed a ton of money, which unfortunately killed it, it used to be high-quality open-source software, the state of the art of web technology. They lost that, and with that, their end is ineluctably coming.
That's still less information.
And replacing practical, explicit information, displayed in the idiomatic way users are used to, by some unintuitive color codes placed up rather than down, breaking typical workflow is hardly a good idea.
That's weird, since it works fine in the linux version of chrome, which is GTK+-based.
Information of what actually is happening.
With the status bar, you know whether the browser is trying to resolve the host, establish a connection, or is downloading data.
With some animated weird progression thingy, you don't really know anything. You barely know that it is loading, and that's it.
And that's something entirely different from this product.
If you want something like this, you could simply use a remote server on your home network to do the graphics rendition.
And just where would you put the graphics card?
It needs to be put in a PC.
It's just that's it's got wireless output instead of wired output.
Yes, people call this "a home network".
But not as good.
The attacker they really mean here, is the user who purchased the Internet-connected HDTV.
Indeed, it is possible for him to trick the TV that is connected to his network infrastructure into doing things the manufacturer had been trying to prevent the user from doing.
This is not very different from jailbreaking your own phone or video game console, except it's much more trivial.
HP is crap as well, yes.
The decent laptop makers are Lenovo, Asus, Sony and Apple. Dell also does some decent ones but not for end-users.
And anyone with half some experience purchasing laptops knows better than to get one from Acer.
Or rather, that's what the vendors would like you to think. Smart customers realize it's better to purchase quality. But quality laptops are getting quite high to come by these days.
Thankfully the US is only 3% of the world.
Looks like you still have a few misconceptions.
Computer science has little to do with software or hardware. To learn those, you should take a degree in software engineering or hardware engineering, respectively.
Computer science is about algorithms, complexity and computability.
At bachelor's level, maybe.
But at master's level, you have to design algorithms that can benefit from parallelism in the underlying hardware architecture.
There is no reason why the US should be in control of a global entity.
Nothing new here, let's move along people.
Thank you, captain obvious, for this incredible insight.
Do TVs still have SCART inputs?
I wish that's how it worked in France.
But no, people protest in large groups in streets all the time, force closing of universities, prevent people from going to work, without any permit whatsoever. And they don't get fines nor get thrown to jail.
Only in the US.
Thankfully there is more to the world than the DMCA.
Learn to read stats.
This does not mean at all that most linux installations are servers.
He'd still have to buy Windows.
Just take a look at tilera. It's not open though.
Sorry, but while functional programming style is indeed the future of HPC (with C++), functional languages themselves aren't. Read the research papers of the field and see for yourself.