So we all hate Bill Gates. Apparently for being ruthlessly successful at exploiting the (fairly) free, capitalist system we all hold dear.
Hey cool! You mean, like Robin Hood? Yeah man, I can see that. He like robbed from the rich to give to the poor!
Robin Hood was an outlaw as well, found guilty by the Evil Sherrif of committing terrible crimes and stealing for poor defenceless rich people.
Despite some evidence that suggests he had a taste for violence, and in fact helped rich people as well, of course everybody loves Robin, don't they.
Anyway, back to the point, Bill made a lot of money out of the markets precisely by so effectively removing that freedom. He should be punished like anybody else who attempts to play the markets, but he hasn't been. Giving away lots of money to charity doesn't make it OK by the way, if that were the case all drug dealers would have to do to get off the books would be to give away some of their personal fortune to good causes.
Considering he basically did steal that money and even got found guilty of it, I don't see why his charitable preferences should override mine. He gave $10 billion to India to fight aids yes? He also gave a lot more than that to fight Linux in the very same country, not a good cause I'd have chosen to donate to (and I do donate to charity by the way).
$24 Billion is more than most developed countries in the world will put into that sort of work in our lifetimes.
Since when? Governments give huge amounts of money away as part of aid initiatives and so on. Britain still pays out large sums of money to help prop up parts of Africa, as well as supporting many charities through grants. Other governments do similar things. Often it has strings attached of course - Bill can give away all his money and see it disappear down the drains through long term corruption and mismanagement but governments who represent the people need to be a bit more careful, which is why such organisations often require governmental reforms to go along with aid.
Oh and finally, don't forget that if him or his company had paid income tax, then a portion of that money would have gone towards such aid, and (at least in theory) the people would have chosen where the aid went or at least had some influence over it.
The British ones are pretty hilarious, victory after victory against the French until the loss of Calais appears in a footnote.
Completely off topic, but I really wish that was so. When I did History GCSE it was ALL about the flipping industrial revolution and the poor old farmers. International politics barely got a lookin, let alone wars with the French. It's far more politically correct to study the economics of the weaving industry, or medieval farming methods.
If something interesting happened at some point in British history, you can be guaranteed it will not appear in a reasonably advanced history course.
Edison invented the electric chair as part of his marketting campaign for DC - the chair used AC.
He went around electrocuting elephants as well I think. Well, it just goes to show, in case of FUD vs market economics, the market usually wins.
i can't help wonder if the announcement was timed to take some attention away from the extremely significant move by apple to move it's sizable number of users from a proprietary based browser to one at least based upon a free software engine
I think you're paranoid. Apple is no threat whatsoever to Microsoft, OK? They are "opening" up their code because their customers are telling them that this is an advantage of Linux, and Microsoft have woken up and are trying to compete. Their own internal memos say this.
I'm sure Microsoft knew full well that most Mac users were already on Chimera anyway, and the IE for the Mac was a failure. They don't seem too concerned. As for "sizable number of users", I'd like to see the statistics for that. The installed user base of MacOS X seems to vary wildly depending on who you ask, but the actual big statistics companies (who get paid to compile figures as accurate as they can make them, usually) say that it's either behind or roughly equal to Linux on the desktop, which seems reasonable seeing as the latter is free and works on PCs, but the first is a better desktop OS at the moment.
I can't remember Microsoft ever taking defensive moves against Apple, least of all now. And FYI using an open source rendering engine doesn't make Safari open source itself, so really nothing much has changed, I'm sure they're actually more concerned about Mozilla as that's the only browser that really competes with IE in any real terms.
p.s. i have been using the term 'free software' in the above post but am unsure of the exact license that khtml is under (i searched and couldn't find the info) - anyone who knows?
It's under the LGPL, which is why Apple had to contribute their changes back when they started redistributing them (in the form of a patch dump unfortunately) but they don't have to make Safari free software.
Well, I think the Linux kernel is now pre-emptable to some extent (not everywhere, but most places). Robert Love did the work necessary. I don't know how well it stacks up against the NT kernel, but I'd guess over the years it'll close the gap.
# Why are we using xpcom considering the huge bloat/threading issues on non-win32?
Because XPCOM allows plugins to have some semblance of binary compatability, and because it enables XPConnect which makes it trivial to create cross platform UIs. Note that the large amount of code written in XUL/JavaScript is very easy to hack, a lot of contributors to Mozilla started this way. The development time costs were probably worth it alone.
Why do the signatures on our api make almost no sense to outsiders?
Signatures? If you mean function prototypes, they are fairly self explanatory usually. Anybody with a good grasp of C++ who wants to understand them can find out what the portable typing system is.
# Why do we compare our performance almost exclusively to IE?
Because Gecko is feature-comparable to IE (Trident) and KHTML isn't? Also remember that nobody uses Konq and everybody uses IE from a statistical viewpoint.
# If Apple wont use our code because it's too big, do we have any real chance of being used on small devices?
I dunno if they really target very small devices any more. For starters, very small devices probably aren't going to need fully featured web browsers anyway.
Why are we still using xul now that we ifdef [hixie.ch] out platform-specific ui code?
Well, that link goes to a simple preprocessing tool, it doesn't make any mention of XUL I can see. And more to the point, XUL is an abstraction system so if anything removing platform-specific code would make sense. Of course Moz does use some platform specific code, like common dialog boxes.
Using XUL makes a lot of sense btw. Other than Qt which is only free software on X11 platforms, there weren't really any good C++ cross platform toolkits back then. The nearest is wxWindows which wasn't anywhere near as well developed as it is now, and still isn't really up to the quality needed of Mozilla from what I've heard (not used it myself, might be wrong).
The choice was simple - either XUL or Windows only.
Mozilla is complex at points, the use of XPCOM in all parts of the app was a mistake (which is now being rectified in de-comtamination, ho ho), but that's because the web is a complex thing. I think people malign Gecko too much really...
Er, yes? If one group of volunteers can make a well integrated and fast browser based on Gecko, what stumped Apple so bad? The Mozilla codebase is large, but that's becuase it's a hell of a lot more than a rendering engine. In fact, the rendering engine part of it isn't so bad at all, I've seen it and understood most of it and I'm hardly an ace coder.
There has been an enormous effort gone into Mozilla and it shows, but I think it still has a way to go.
Well, like where? Gecko is the most standards compliant, the most powerful and supports the most technology. It's also very fast, the FUD about it being slow here simply is not true. I've been blown away by the speed of Galeon. I think part of the problem is that the default front end "feels" slow because of the default theme to some extent. Oh, while I'm thinking about it, try adding this line:
user_pref("nglayout.initialpaint.delay", 0);
to your prefs.js file (shutdown moz first). You'll find it feels a lot snappier if you're on a decent machine.
Yes, and of course KHTML is not used in the "real" world.
Well of course KHTML is used a bit, but on my Linux specific site about 60% of the hits are from Mozilla, about 25-30% are from IE and about 5% are from Konq. So really KHTML is not used much at all in the real web, although it's used plenty inside KDE for their help system etc.
Let's call it like it is -- Gecko, while a noble effort, is really a failure. It was YEARS late, and completely missed its goal (a lightweight, fast. cross-platform rendering engine). One bit of that (cross-platform) does not a success make.
Actually its goal was to be useful and powerful. The fact that they thought they could also be fast and light is a common mistake amongst coders, the two arne't necessarily mutually exclusive but often are in real life.
Gecko is standards compliant, fast (no, really), supports many standards and is extremely powerful. So, it's larger than KHTML. Most importantly, it actually renders the vast majority of the web.
Apple have a problem - their machines are slow. I compiled GNOME2.2 with Galeon today, and the speed blew me away. I have never used such a fast browser. Tabs opened and rendered near instantly (I was using the paint-delay trick) and I never found myself waiting for the browser, it was just there. I'm sure other people who've used Galeon2 can corroberate this. This is not a particlarly fast machine, an Athlon 1400 I think, and Gecko hasn't been optimized for Linux as much as it has for Windows (on which it's also very fast), so this Gecko is slow BS seems to be more a Mac problem than anyhting else.
I mean, if the Galeon team can produce an insanely fast browser out of Gecko, what's stopping Apple?
Safari's what a browser should be -- small, lightweight, and out of my face. The interface is slim & sleek, and, like the rest of Apple's software, lets me focus on the CONTENT rather than the delivery.
Oh boy, that's funny. So that's why it has a textured window (that cannot be themed to something less distracting), along with all the rest of the usual Apple eyecandy - but no tabs?
Apple is all about presentation. See how all the talk here is of speed, not accuracy in actually rendering the contet?
I really think that's why OSX is so wonderful -- it just stays out of my way and lets me do what I gotta do. And I have to admit, running a DVD authoring program alongside several terminal windows on a Mac (!) is still impressive to me.
Safari has a ton of room to grow before it achieves Mozilla's mammoth size.
Oh please. That's such a pile of crap.
Developers always start off thinking they can do what the competition does, except faster and smaller. The Mozilla project themselves started off that way. I remember in the early days them proudly announcing their rendering engine would fit on a floppy disk.
Then they started making it actually work and be useful on the web. They added support for the latest technologies, they made it cross platform (which itself has quite a bit of overhead) and so on.
Getting to about 80% of the features of your nearest competitor while staying small and fast (relatively) isn't hard, but what you always find is that after you've done the last 20% and you have enough compatability to be useful in the real world, and your software has all the hairs necessary to make it work on grans bizarro ancient setup, and then you find you made a mistake in the design that wasn't obvious at the time so you hack around it and so on... by the time you've done all of that you're just as big and "bloated" as the competition.
The idea that somehow the KHTML have magically produced something better than Gecko is fallacy. Don't get me wrong, KHTML is a fine piece of work, but to pretend it'll remain fast and light when it has to deal with enough web pages to be useful and support all the new tech (XSLT, XForms, SVG etc, XPath, SOAP) that's beginning to filter down into the general purpose web is insane.
Joel Spolski wrote a good article on rewriting software in this way, and despite the fact that KHTML was already there, it fits into his theories quite well. Sometimes you don't have much choice, the old Netscape codebase was SO bad it could never have gone further, but it's something that's done in dire straits only.
Oh and finally, considering Phoenix is smaller than that, but does more, I'm not particularly impressed anyway.
It's getting very, very difficult to find new technologies in OS X that are proprietary
You know, I think this is going to be like the "X is slow" argument - utterly full of it, but never goes away.
OS X is chocka with proprietary tech. Oh and no, using XML for preferences doesn't make it suddenly not proprietary, the next version of IIS uses XML config files too. Is IIS not a proprietary web server?
Anyway, if it's so hard to find proprietary stuff, where can I find the implementations of Carbon and Cocoa (the bulk of the platform). Yes, I know about GNUstep. What about the artwork?
If MacOS isn't proprietary, where is the PC port?
Why do Apple insist on STILL using Sorensen for all the videos on their website? And considering it's being given away for free anyway, what did they have to lose by making Safari free software?
The "MacOS is open" line is really getting quite old.
The reason being of course that it's generally accepted that the American people, left to their own devices, basically don't care about anything happening outside their own borders - they care only about the economy and whether guns are legal or not.
So I think it's fair enough for the government to take the initiative on this one, because from what I've seen the American people wouldn't (in fact that's true generally of western societies. grass roots campaigns tend to be reactionary and against change, not for it)
I would now recommend FreeBSD as the unix of choice for any purpose, it may not have a fancy graphical install program, but you will really appreciate this simplicity when you come to make changes/ do something a little out of the ordinary.
Well no offence but I hope you don't recommend it to newbies. I've had friends tell me Linux was still in the dark ages because it lacked a friendly install program and they couldn't figure out how to configure it. It turned out some smartass had recommended Debian because "it's so cool, everyone uses Debian, and it's free", ignoring the fact that newbies want simplicity perhaps at the expense of reliability.
When i bought my powerbook with osx it was too soon, things didnt really roll until the 10.0.4 release. I was sucked in by their excellent marketing of the powerbook g4 running a gorgeous open sourceOS.
Since when is OS X open source? I must have missed the point where they released the other 90% of the code.
No matter what I run on it or do with it it conveys an image that I am stylish, that I value quality over other considerations such as cost and speed. That I think different.
This is one of the most basic advertising strategies around. The "think different" campaign and in fact most of Apples marketing is little more advanced than draping hot girls over a car, or trying to associate a beer with a lifestyle.
Drinking fosters doesn't make you even remotely Australian except in your own mind, and it's the same with Apple. Where you look in the mirror and see a stylish hip guy with the latest accessories, who's a radical cutting edge thinker (oh please) others may look at you and think "jeez, what a shallow elitist snob".
Apple did something with its brand that very very few companies have done. They created incredible value; Apple appeals to people. You dont get that with your dell or toshiba or even an alienware rig.
Sorry, but if you really need the crutch of a marketing assocation to get people interested then you should probably work on either your skills or your style, because that's usually a lot more important than what logo you have on your laptop case. Remember the markup you're paying for that branding too.
My "terribly slow" Dual 1 Ghz Macintosh is limited by its slowest part... me.
If only that were true, computers would probably be a lot cheaper than they are today. Nonetheless, for the same reason people buy cars that have snappy response, people buy fast computers not because they need huge quantities of raw calcuating speed but because it makes their desktop feel snappier.
About a year ago (when I first started using Linux) KDE had a big issue with people saying it felt slow. Investigation revealed that actually the reason it felt slow was application startup speed, when the user hit a button to open Konq or Kate or whatever, it took a long time to respond. The issue was traced mostly to fixup speed for C++ apps inside the linker and I believe a solution was found and is now being implemented.
The actual speed of a machine does matter, because although most of the time it's idle, when you want it it needs to be there.
Apple understand this which is why they wind up the UI servers to 10,000 priority IIRC - they have put a lot of effort into making MacOS X feel fast, despite the relative slowness of the underlying hardware. So for instance it uses vsycned video, it uses hardware accelerated graphics and all the GUIs are rendered synchronously by the ui servers (so you don't get windows/x style lag as the the screen is momentarily damaged before the app repairs it). All this takes up even more CPU time though of course, leaving even less for the actual work.
So in short, although Macs might be slow, they don't feel as slow as you might expect, but nonetheless they are slow, and you notice it the moment you do anything more complex than moving some windows around.
Not really.. as great as the GIMP is, it still has a ways to go before it can pry photoshop out of the cold dead hands of the people who use photoshop what it is intended for rather than just for general cropping and resizing.
Indeed, a better comparison would be to Paint Shop Pro, which is in fact what I'd gues 90% of the Photoshop users actually should be using. I know so many people who just pirate Photoshop so they can feel "pro" and use "the best" that it's not even funny. Get over it!
Not to mention the GIMP looks horrible on every OS
Looks OK to me, running in GNOME on Linux (which is in fact its "native" OS) - note that screenshot is quite old now.
Considering that the GIMP will run on basically anything, and Photoshop runs on Windows or Mac OS (unless you count Wine), I think the:
I guess you get what you pay for though.
line is extremely old. No, hard to believe though it is, there's this thing called charity and it means sometimes you get something great for absolutely nothing.
Re:Four years and half too late.
on
Ark Linux
·
· Score: 2
Judging by the number of comments, that must have been one hell of a Qt flamewar:)
Re:counterproductive
on
Ark Linux
·
· Score: 4, Interesting
Customers have no trouble learning a new interface. They do it all the time.
Amen to that brother. That statement reminds me of a "debate" (actually more of an argument) I had with my best mate a few years ago, a die hard Mac head. We were discussing interface consitancy, with Paul attempting to show that Linux would never go anywhere because it had 2 major widget toolkits that had different themes, no universal HIG etc. He ended up trying to prove that people couldn't use other peoples mobile phones.
The hilarity that ensued kept us amused for a looong time. The sight of him walking around the 6th form centre with his phone asking people to open up the address book on it, then looking dismayed when after studying it a moment they did it as fast as he would have done, was excellent. The "what are you smoking, man" looks he got were even better.
People are smart, people are adaptable. Yes, a completely consistant UI is good for user efficiency which is why people are trying hard with unified themes and user interface guidelines etc, but at the end of the day with a few notable exceptions (software installation/resolution switching) it's been possible to figure how to work Linux for a long time now and anyway no other OS has a completely consistant GUI, least of all Windows (not even the mac).
Hmmm? If you use a decent distro this should be entirely automatic? I've used several distros and never had any issues with OpenGL on decent cards. If you're thinking of the nVidia drivers, it's normally just a case of installing the RPMs.
On the other hand. Well, I haven't heard of Grin, or it's games. Sorry. I can't recall seeing anything in a PC Gamer, or a box in a software shop.
Me neither, but I don't care, I'm an absolute sucker for fast racing games. Pansy ass cars are no good for me, I need, and I mean medically need speed. Up until now I got my fix from Supreme Snowboarding, which lets you go at a simulated 130km/h on a snowboard in places (if you're leet enough;) but Ballistics looks damn cool.
I played their port of Creatures 3 and was very impressed, it was a high quality port. I didn't buy the game, Creatures just wasn't interesting enough (i used to play v1), but I'll be looking forward to the demo of Ballistics.
I stopped reading gaming mags a long time ago, so even if it'd had loads of coverage I still wouldn't have found it. Now I've seen those screenshots I'm interested.
I think it's possible for us both to be right. You could say that Apple is selling a platform that gives a certain user experience. If Apple was selling a platform only, then it is more likely that it would have gone out of business a long time ago.
Sure, their user experience is what defines them, it's what they market themselves as.
What I was thinking about was being able to click System Preferences, click Sharing, click "Start Web Sharing" as an interface frontend for running "sudo/usr/sbin/apachectl start" in the shell.
Ah yes, that's fine, not part of the platform as such, it's just part of the OS. That sort of stuff doesn't really need to be open, it's not a platform - people don't build stuff upon gui front ends to apache:) The Mac APIs are closed though, which makes the Mac a closed platform from the perspective of a developer writing applications.
IMO, this is OK because with Aqua (the user front end for OS X) Apple is really selling the user experience, not just a powerful tool.
No... they're selling a platform. If MacOS X the platform was actually open source, where are all the API implementations we'd need to make Mac apps run on Linux?
The idea that MacOS is a closed source frontend to an open source backend is fallacy - the backend is made up of many proprietary APIs just like Windows. I mean if the Windows kernel was open sourced, woop, but you'd still need to pay Microsoft to run Windows apps.
.NET also fully supports exporting and importing these objects via SOAP.
Actually.NET uses a series of extensions to SOAP called.NET Remoting which adds object export. Otherwise you're limited to flat APIs, despite the name SOAP does not allow object transport via a network. Embrace and extend... sigh
Anyway, I'd love to take a look at any.NET app that ran slowly -- it'd be a first for me to see that...
Try implementing a 3D engine in C#. It won't have acceptable performance most likely. The speed of.NET apps is fine for desktop apps, because most desktop apps do not actually need high performance in the logic area, and because the vast majority of the code you use in such a project is not in fact.NET, it's inside the implementations of the.NET apis.
Java is also quite fast when used in this way. See Eclipse for instance. They all have their uses. Note that.NET/C# has some abilities Java doesn't, like being able to temporarily turn off the garbage collector so allowing direct memory access which can improve speed quite a lot.
Yes, we've considered those kind of solutions, but they are all enormous hacks. Considering the patch to the linker is all of 20 lines including context (something extremely small) it'd be easier to simply put an autopackage provided ld.so on the system and have the ELF binaries set their INTERP field to it.
We don't want to do that, but working around this problem is far harder than actually solving it, and I want to avoid nasty hacks wherever possible.
Quicktime is indeed open, the fact that a huge majority of quicktime encapsulated videos use proprietary codecs does not change that fact.
IMHO if the company that makes an open container format chooses to encode all the videos it produces in a closed codec, semantic distinctions about whether the format is open or closed are meaningless.
The fact is that QuickTime is basically dead outside of apple.com and the occasional interactive CD. If Apple didn't encode all the movie trailors using Sorensen, most people would never even use it.
When the only company that uses it in any meaningful way chooses a proprietary codec and labels them as "QuickTime files", as far as I'm concerned, QuickTime is closed. I'm perfectly aware that technically that statement is inaccurate, but in spirit it is true.
Xiph didn't create an open container format and a closed audio codec when they made ogg vorbis. They went all the way. Apple should do the same.
I'd say MPEG-4 is Open, but not Free, in FSF terminology. Different strokes for different folks.
Indeed... in fact Alan Cox himself has said that the licensing of the code doesn't matter so much as open interfaces, so if people want to charge for implementations that's fine by me as long as free implementations are allowed as well...
And I expect a lot of "black sheep" apps ala MP3, to exist for Linux. Check out MPEG4IP for a LAME-equivalent.
Yes, well that's the worry isn't it - it's open now, and hopefully it'll stay open, but can the licensing be changed in future? Everybody thought you didn't need a license to decode MP3s until recently, and now people aren't so sure. That kind of legal vagueness is something to be warey of.
As for Theora, who knows? It isn't even in beta yet. It's VP3 based, and unless they enhance that code a LOT
According to the FAQ they have replaced the fixed lookup tables with dynamic ones that they can vary and tweak after Theora is actually released, and can possibly be altered on the fly. I don't know enough about codecs to say, but this approach seems to have worked well for Vorbis with the codec approaching and then surpassing MP3 for compression quality (though not by a huge amount).
Slashdot Mirror
Hey cool! You mean, like Robin Hood? Yeah man, I can see that. He like robbed from the rich to give to the poor!
Robin Hood was an outlaw as well, found guilty by the Evil Sherrif of committing terrible crimes and stealing for poor defenceless rich people.
Despite some evidence that suggests he had a taste for violence, and in fact helped rich people as well, of course everybody loves Robin, don't they.
Anyway, back to the point, Bill made a lot of money out of the markets precisely by so effectively removing that freedom. He should be punished like anybody else who attempts to play the markets, but he hasn't been. Giving away lots of money to charity doesn't make it OK by the way, if that were the case all drug dealers would have to do to get off the books would be to give away some of their personal fortune to good causes.
Considering he basically did steal that money and even got found guilty of it, I don't see why his charitable preferences should override mine. He gave $10 billion to India to fight aids yes? He also gave a lot more than that to fight Linux in the very same country, not a good cause I'd have chosen to donate to (and I do donate to charity by the way).
$24 Billion is more than most developed countries in the world will put into that sort of work in our lifetimes.
Since when? Governments give huge amounts of money away as part of aid initiatives and so on. Britain still pays out large sums of money to help prop up parts of Africa, as well as supporting many charities through grants. Other governments do similar things. Often it has strings attached of course - Bill can give away all his money and see it disappear down the drains through long term corruption and mismanagement but governments who represent the people need to be a bit more careful, which is why such organisations often require governmental reforms to go along with aid.
Oh and finally, don't forget that if him or his company had paid income tax, then a portion of that money would have gone towards such aid, and (at least in theory) the people would have chosen where the aid went or at least had some influence over it.
Completely off topic, but I really wish that was so. When I did History GCSE it was ALL about the flipping industrial revolution and the poor old farmers. International politics barely got a lookin, let alone wars with the French. It's far more politically correct to study the economics of the weaving industry, or medieval farming methods.
If something interesting happened at some point in British history, you can be guaranteed it will not appear in a reasonably advanced history course.
Edison invented the electric chair as part of his marketting campaign for DC - the chair used AC.
He went around electrocuting elephants as well I think. Well, it just goes to show, in case of FUD vs market economics, the market usually wins.
I think you're paranoid. Apple is no threat whatsoever to Microsoft, OK? They are "opening" up their code because their customers are telling them that this is an advantage of Linux, and Microsoft have woken up and are trying to compete. Their own internal memos say this.
I'm sure Microsoft knew full well that most Mac users were already on Chimera anyway, and the IE for the Mac was a failure. They don't seem too concerned. As for "sizable number of users", I'd like to see the statistics for that. The installed user base of MacOS X seems to vary wildly depending on who you ask, but the actual big statistics companies (who get paid to compile figures as accurate as they can make them, usually) say that it's either behind or roughly equal to Linux on the desktop, which seems reasonable seeing as the latter is free and works on PCs, but the first is a better desktop OS at the moment.
I can't remember Microsoft ever taking defensive moves against Apple, least of all now. And FYI using an open source rendering engine doesn't make Safari open source itself, so really nothing much has changed, I'm sure they're actually more concerned about Mozilla as that's the only browser that really competes with IE in any real terms.
p.s. i have been using the term 'free software' in the above post but am unsure of the exact license that khtml is under (i searched and couldn't find the info) - anyone who knows?
It's under the LGPL, which is why Apple had to contribute their changes back when they started redistributing them (in the form of a patch dump unfortunately) but they don't have to make Safari free software.
Well, I think the Linux kernel is now pre-emptable to some extent (not everywhere, but most places). Robert Love did the work necessary. I don't know how well it stacks up against the NT kernel, but I'd guess over the years it'll close the gap.
Because XPCOM allows plugins to have some semblance of binary compatability, and because it enables XPConnect which makes it trivial to create cross platform UIs. Note that the large amount of code written in XUL/JavaScript is very easy to hack, a lot of contributors to Mozilla started this way. The development time costs were probably worth it alone.
Why do the signatures on our api make almost no sense to outsiders?
Signatures? If you mean function prototypes, they are fairly self explanatory usually. Anybody with a good grasp of C++ who wants to understand them can find out what the portable typing system is.
# Why do we compare our performance almost exclusively to IE?
Because Gecko is feature-comparable to IE (Trident) and KHTML isn't? Also remember that nobody uses Konq and everybody uses IE from a statistical viewpoint.
# If Apple wont use our code because it's too big, do we have any real chance of being used on small devices?
I dunno if they really target very small devices any more. For starters, very small devices probably aren't going to need fully featured web browsers anyway.
Why are we still using xul now that we ifdef [hixie.ch] out platform-specific ui code?
Well, that link goes to a simple preprocessing tool, it doesn't make any mention of XUL I can see. And more to the point, XUL is an abstraction system so if anything removing platform-specific code would make sense. Of course Moz does use some platform specific code, like common dialog boxes.
Using XUL makes a lot of sense btw. Other than Qt which is only free software on X11 platforms, there weren't really any good C++ cross platform toolkits back then. The nearest is wxWindows which wasn't anywhere near as well developed as it is now, and still isn't really up to the quality needed of Mozilla from what I've heard (not used it myself, might be wrong).
The choice was simple - either XUL or Windows only.
Mozilla is complex at points, the use of XPCOM in all parts of the app was a mistake (which is now being rectified in de-comtamination, ho ho), but that's because the web is a complex thing. I think people malign Gecko too much really...
Er, yes? If one group of volunteers can make a well integrated and fast browser based on Gecko, what stumped Apple so bad? The Mozilla codebase is large, but that's becuase it's a hell of a lot more than a rendering engine. In fact, the rendering engine part of it isn't so bad at all, I've seen it and understood most of it and I'm hardly an ace coder.
There has been an enormous effort gone into Mozilla and it shows, but I think it still has a way to go.
Well, like where? Gecko is the most standards compliant, the most powerful and supports the most technology. It's also very fast, the FUD about it being slow here simply is not true. I've been blown away by the speed of Galeon. I think part of the problem is that the default front end "feels" slow because of the default theme to some extent. Oh, while I'm thinking about it, try adding this line:
user_pref("nglayout.initialpaint.delay", 0);
to your prefs.js file (shutdown moz first). You'll find it feels a lot snappier if you're on a decent machine.
Yes, and of course KHTML is not used in the "real" world.
Well of course KHTML is used a bit, but on my Linux specific site about 60% of the hits are from Mozilla, about 25-30% are from IE and about 5% are from Konq. So really KHTML is not used much at all in the real web, although it's used plenty inside KDE for their help system etc.
Actually its goal was to be useful and powerful. The fact that they thought they could also be fast and light is a common mistake amongst coders, the two arne't necessarily mutually exclusive but often are in real life.
Gecko is standards compliant, fast (no, really), supports many standards and is extremely powerful. So, it's larger than KHTML. Most importantly, it actually renders the vast majority of the web.
Apple have a problem - their machines are slow. I compiled GNOME2.2 with Galeon today, and the speed blew me away. I have never used such a fast browser. Tabs opened and rendered near instantly (I was using the paint-delay trick) and I never found myself waiting for the browser, it was just there. I'm sure other people who've used Galeon2 can corroberate this. This is not a particlarly fast machine, an Athlon 1400 I think, and Gecko hasn't been optimized for Linux as much as it has for Windows (on which it's also very fast), so this Gecko is slow BS seems to be more a Mac problem than anyhting else.
I mean, if the Galeon team can produce an insanely fast browser out of Gecko, what's stopping Apple?
Safari's what a browser should be -- small, lightweight, and out of my face. The interface is slim & sleek, and, like the rest of Apple's software, lets me focus on the CONTENT rather than the delivery.
Oh boy, that's funny. So that's why it has a textured window (that cannot be themed to something less distracting), along with all the rest of the usual Apple eyecandy - but no tabs?
Apple is all about presentation. See how all the talk here is of speed, not accuracy in actually rendering the contet? I really think that's why OSX is so wonderful -- it just stays out of my way and lets me do what I gotta do. And I have to admit, running a DVD authoring program alongside several terminal windows on a Mac (!) is still impressive to me.
Wake up mods, that's a -1 Offtopic comment.
Oh please. That's such a pile of crap.
Developers always start off thinking they can do what the competition does, except faster and smaller. The Mozilla project themselves started off that way. I remember in the early days them proudly announcing their rendering engine would fit on a floppy disk.
Then they started making it actually work and be useful on the web. They added support for the latest technologies, they made it cross platform (which itself has quite a bit of overhead) and so on.
Getting to about 80% of the features of your nearest competitor while staying small and fast (relatively) isn't hard, but what you always find is that after you've done the last 20% and you have enough compatability to be useful in the real world, and your software has all the hairs necessary to make it work on grans bizarro ancient setup, and then you find you made a mistake in the design that wasn't obvious at the time so you hack around it and so on ... by the time you've done all of that you're just as big and "bloated" as the competition.
The idea that somehow the KHTML have magically produced something better than Gecko is fallacy. Don't get me wrong, KHTML is a fine piece of work, but to pretend it'll remain fast and light when it has to deal with enough web pages to be useful and support all the new tech (XSLT, XForms, SVG etc, XPath, SOAP) that's beginning to filter down into the general purpose web is insane.
Joel Spolski wrote a good article on rewriting software in this way, and despite the fact that KHTML was already there, it fits into his theories quite well. Sometimes you don't have much choice, the old Netscape codebase was SO bad it could never have gone further, but it's something that's done in dire straits only.
Oh and finally, considering Phoenix is smaller than that, but does more, I'm not particularly impressed anyway.
You know, I think this is going to be like the "X is slow" argument - utterly full of it, but never goes away.
OS X is chocka with proprietary tech. Oh and no, using XML for preferences doesn't make it suddenly not proprietary, the next version of IIS uses XML config files too. Is IIS not a proprietary web server?
Anyway, if it's so hard to find proprietary stuff, where can I find the implementations of Carbon and Cocoa (the bulk of the platform). Yes, I know about GNUstep. What about the artwork?
If MacOS isn't proprietary, where is the PC port?
Why do Apple insist on STILL using Sorensen for all the videos on their website? And considering it's being given away for free anyway, what did they have to lose by making Safari free software?
The "MacOS is open" line is really getting quite old.
So I think it's fair enough for the government to take the initiative on this one, because from what I've seen the American people wouldn't (in fact that's true generally of western societies. grass roots campaigns tend to be reactionary and against change, not for it)
Well no offence but I hope you don't recommend it to newbies. I've had friends tell me Linux was still in the dark ages because it lacked a friendly install program and they couldn't figure out how to configure it. It turned out some smartass had recommended Debian because "it's so cool, everyone uses Debian, and it's free", ignoring the fact that newbies want simplicity perhaps at the expense of reliability.
Since when is OS X open source? I must have missed the point where they released the other 90% of the code.
No matter what I run on it or do with it it conveys an image that I am stylish, that I value quality over other considerations such as cost and speed. That I think different.
This is one of the most basic advertising strategies around. The "think different" campaign and in fact most of Apples marketing is little more advanced than draping hot girls over a car, or trying to associate a beer with a lifestyle.
Drinking fosters doesn't make you even remotely Australian except in your own mind, and it's the same with Apple. Where you look in the mirror and see a stylish hip guy with the latest accessories, who's a radical cutting edge thinker (oh please) others may look at you and think "jeez, what a shallow elitist snob".
Apple did something with its brand that very very few companies have done. They created incredible value; Apple appeals to people. You dont get that with your dell or toshiba or even an alienware rig.
Sorry, but if you really need the crutch of a marketing assocation to get people interested then you should probably work on either your skills or your style, because that's usually a lot more important than what logo you have on your laptop case. Remember the markup you're paying for that branding too.
If only that were true, computers would probably be a lot cheaper than they are today. Nonetheless, for the same reason people buy cars that have snappy response, people buy fast computers not because they need huge quantities of raw calcuating speed but because it makes their desktop feel snappier.
About a year ago (when I first started using Linux) KDE had a big issue with people saying it felt slow. Investigation revealed that actually the reason it felt slow was application startup speed, when the user hit a button to open Konq or Kate or whatever, it took a long time to respond. The issue was traced mostly to fixup speed for C++ apps inside the linker and I believe a solution was found and is now being implemented.
The actual speed of a machine does matter, because although most of the time it's idle, when you want it it needs to be there.
Apple understand this which is why they wind up the UI servers to 10,000 priority IIRC - they have put a lot of effort into making MacOS X feel fast, despite the relative slowness of the underlying hardware. So for instance it uses vsycned video, it uses hardware accelerated graphics and all the GUIs are rendered synchronously by the ui servers (so you don't get windows/x style lag as the the screen is momentarily damaged before the app repairs it). All this takes up even more CPU time though of course, leaving even less for the actual work.
So in short, although Macs might be slow, they don't feel as slow as you might expect, but nonetheless they are slow, and you notice it the moment you do anything more complex than moving some windows around.
Indeed, a better comparison would be to Paint Shop Pro, which is in fact what I'd gues 90% of the Photoshop users actually should be using. I know so many people who just pirate Photoshop so they can feel "pro" and use "the best" that it's not even funny. Get over it!
Not to mention the GIMP looks horrible on every OS
Looks OK to me, running in GNOME on Linux (which is in fact its "native" OS) - note that screenshot is quite old now.
Considering that the GIMP will run on basically anything, and Photoshop runs on Windows or Mac OS (unless you count Wine), I think the:
I guess you get what you pay for though.
line is extremely old. No, hard to believe though it is, there's this thing called charity and it means sometimes you get something great for absolutely nothing.
Judging by the number of comments, that must have been one hell of a Qt flamewar :)
Amen to that brother. That statement reminds me of a "debate" (actually more of an argument) I had with my best mate a few years ago, a die hard Mac head. We were discussing interface consitancy, with Paul attempting to show that Linux would never go anywhere because it had 2 major widget toolkits that had different themes, no universal HIG etc. He ended up trying to prove that people couldn't use other peoples mobile phones.
The hilarity that ensued kept us amused for a looong time. The sight of him walking around the 6th form centre with his phone asking people to open up the address book on it, then looking dismayed when after studying it a moment they did it as fast as he would have done, was excellent. The "what are you smoking, man" looks he got were even better.
People are smart, people are adaptable. Yes, a completely consistant UI is good for user efficiency which is why people are trying hard with unified themes and user interface guidelines etc, but at the end of the day with a few notable exceptions (software installation/resolution switching) it's been possible to figure how to work Linux for a long time now and anyway no other OS has a completely consistant GUI, least of all Windows (not even the mac).
Hmmm? If you use a decent distro this should be entirely automatic? I've used several distros and never had any issues with OpenGL on decent cards. If you're thinking of the nVidia drivers, it's normally just a case of installing the RPMs.
Me neither, but I don't care, I'm an absolute sucker for fast racing games. Pansy ass cars are no good for me, I need, and I mean medically need speed. Up until now I got my fix from Supreme Snowboarding, which lets you go at a simulated 130km/h on a snowboard in places (if you're leet enough ;) but Ballistics looks damn cool.
I played their port of Creatures 3 and was very impressed, it was a high quality port. I didn't buy the game, Creatures just wasn't interesting enough (i used to play v1), but I'll be looking forward to the demo of Ballistics.
I stopped reading gaming mags a long time ago, so even if it'd had loads of coverage I still wouldn't have found it. Now I've seen those screenshots I'm interested.
Sure, their user experience is what defines them, it's what they market themselves as. What I was thinking about was being able to click System Preferences, click Sharing, click "Start Web Sharing" as an interface frontend for running "sudo /usr/sbin/apachectl start" in the shell.
Ah yes, that's fine, not part of the platform as such, it's just part of the OS. That sort of stuff doesn't really need to be open, it's not a platform - people don't build stuff upon gui front ends to apache :) The Mac APIs are closed though, which makes the Mac a closed platform from the perspective of a developer writing applications.
No... they're selling a platform. If MacOS X the platform was actually open source, where are all the API implementations we'd need to make Mac apps run on Linux?
The idea that MacOS is a closed source frontend to an open source backend is fallacy - the backend is made up of many proprietary APIs just like Windows. I mean if the Windows kernel was open sourced, woop, but you'd still need to pay Microsoft to run Windows apps.
Actually .NET uses a series of extensions to SOAP called .NET Remoting which adds object export. Otherwise you're limited to flat APIs, despite the name SOAP does not allow object transport via a network. Embrace and extend... sigh
Try implementing a 3D engine in C#. It won't have acceptable performance most likely. The speed of .NET apps is fine for desktop apps, because most desktop apps do not actually need high performance in the logic area, and because the vast majority of the code you use in such a project is not in fact .NET, it's inside the implementations of the .NET apis.
Java is also quite fast when used in this way. See Eclipse for instance. They all have their uses. Note that .NET/C# has some abilities Java doesn't, like being able to temporarily turn off the garbage collector so allowing direct memory access which can improve speed quite a lot.
We don't want to do that, but working around this problem is far harder than actually solving it, and I want to avoid nasty hacks wherever possible.
IMHO if the company that makes an open container format chooses to encode all the videos it produces in a closed codec, semantic distinctions about whether the format is open or closed are meaningless.
The fact is that QuickTime is basically dead outside of apple.com and the occasional interactive CD. If Apple didn't encode all the movie trailors using Sorensen, most people would never even use it.
When the only company that uses it in any meaningful way chooses a proprietary codec and labels them as "QuickTime files", as far as I'm concerned, QuickTime is closed. I'm perfectly aware that technically that statement is inaccurate, but in spirit it is true.
Xiph didn't create an open container format and a closed audio codec when they made ogg vorbis. They went all the way. Apple should do the same.
Indeed... in fact Alan Cox himself has said that the licensing of the code doesn't matter so much as open interfaces, so if people want to charge for implementations that's fine by me as long as free implementations are allowed as well...
And I expect a lot of "black sheep" apps ala MP3, to exist for Linux. Check out MPEG4IP for a LAME-equivalent.
Yes, well that's the worry isn't it - it's open now, and hopefully it'll stay open, but can the licensing be changed in future? Everybody thought you didn't need a license to decode MP3s until recently, and now people aren't so sure. That kind of legal vagueness is something to be warey of.
As for Theora, who knows? It isn't even in beta yet. It's VP3 based, and unless they enhance that code a LOT
According to the FAQ they have replaced the fixed lookup tables with dynamic ones that they can vary and tweak after Theora is actually released, and can possibly be altered on the fly. I don't know enough about codecs to say, but this approach seems to have worked well for Vorbis with the codec approaching and then surpassing MP3 for compression quality (though not by a huge amount).