Re:Hacking PS2 not a problem for Sony
on
PS2 As PC
·
· Score: 2
The kernel and various other components will be open source, so Sony will gain the benefits of a rock-solid, continually evolving platform. Thats the advantage of open source - take what the community does well and use it.
But why they would open their GUI, application libraries etc. to the public? Sony have more than enough development muscle to create what they need themselves.
When it comes to a desktop framework for the average computer user, its clear that they can't leave *any* of that to the open source community, since we really suck at producing anything like that. The open source community is a hindrance, rather than a help in this area.
I don't know about you, but i use a linux desktop exclusively, at home and at work, and i can't recommend it to anyone else i know, because it just sucks so badly at providing accessible desktop functionality. Of course, there are lots of benefits with a Linux/UNIX desktop (modularity, remote display, customisability etc.) but almost none of them are particularly relevant for a console.
Maybe we'll get there one day, but I'm not holding my breath, and i'll bet Sony won't be either.
Sony also want to control the apps that ship for the PS2. Quality Control is extremely important with regard to consoles, while it has been completely forgotten on the desktop PC. They will need the legal clout to enforce this quality, and they will have it by licensing their GUI and toolkit libraries accordingly.
It may be possible to install 3rd party stuff on a PS2, but you can bet your ass you won't be able to ship a program for the PS2 without first getting (paying for) Sony's approval, or requiring the user to replace the kernel, windowing environment etc. and hence sacrificing compatibility with the Sony-sanctioned apps.
Hacking PS2 not a problem for Sony
on
PS2 As PC
·
· Score: 2
The number of people who will buy a PS2 so they can run only free software on it is statistically insignificant. IT probably will be possible to run X on the PS2, if only to avoid the bad press from hordes of whining Slahdotters.
The most likely scenario for consumer use, however, is that Sony will use the linux kernel, but write a proprietary framebuffer-driver like DirectFB for the PS2 hardware.
They will also probably write their own GUI toolkit, also under proprietary license, that also runs on an X Windows or Linux framebuffer backend.
Meaning that while the core OS is, and always will be free, PS2 applications and the 'Official' GUI on the System will be proprietary, much like Apple's MacOS X GUI.
You will have to pay Sony to develop apps to run on a PS2 (since you need the toolkit API and libraries), but you will be able to develop and run those apps on a Linux desktop PC, as long as you have the (proprietary) runtime and devlopment libraries installed.
There are other markets for a Linux-running PS2 platform - the film, broadcast and sports industries for realtime graphics (realtime mocap-driven characters, virtual sets, advertising overlays etc.) These apps need a solid OS and devlopment tools, which is what Linux is perfect for.
Fisheye lens-based panoramas are no better than panoramas carefully constructed from a large number of rectilinear images.
They are, however, faster and easier to produce - 2 images, as opposed to the 12-16 you need to cover a full 360 degrees with a non-fisheye lens.
Thats really the only benefit - in fact taking lots of pictures will give you a better quality final image, assuming you can orient and align the images perfectly - i use a tripod with calibration marks on it to do mine, with excellent results.
Panorama Tools lets you use both methods to produce panoramic images, also has a free viewer, runs on several OSes as plugins for the GIMP and Photoshop, and has high-quality output.
Why not use digital I/O on soundcards?
on
DSLBlaster?
·
· Score: 1
Soundcards with SP/DIF or Optical I/O are pretty common these days.
Why would you use analog I/O at all?
16bit samples at a 48khz sampling rate= 768Kbps which is pretty respectable.
Panorama Tools is undoubtedly the best toolset for producing panoramic imagery available today.
It is a sad day when a guy as committed to providing great software to the community as Mr. Dersch gets bullied by the U.S. Patent Office and a litigious company who have obviously failed to come up with anything as functional as Mr. Dersch's Panorama Tools.
One can only hope the European Patent Office will look a little closer at the many cases of clear prior art with regard to IPIX's patent, and then throwing it in the bin where it belongs.
4-in1 drivers? Well, i was going to try that as my next step under Win2k (i used to dual boot), but..
In attempting to make my fucking VIA USB controller work, i flashed the BIOS on my mobo using the ASUS-supplied utility.
Upon reboot, Win2k claims the mobo is no longer ACPI-compliant and BSODs, leaving me unable to:
a) flash the BIOS back to the old version, and
b) unable to access the files on my NTFS drive.
Linux, however, boots and runs quite happily on the same BIOS, but the onboard USB controller is still fucking horked no matter which OS i try and use it under.
A third-party USB card seems to work OK, but now all my slots are full, my machine crashes a lot (even under Linux) and i paid good money for a worthless fucking USB controller and a shoddy chipset.
If i rowed a boat out into the middle of the pacific ocean and found a small, newly-formed volcanic island, are there any international laws/statutes/prior claims over bodies of water etc. that would prevent me from declaring sovereignty, setting up a government and joining the U.N. etc?
Or is the idea of 'sovereignty' entirely dependent on your ability to defend the territory you claim by economic, diplomatic and/or military force?
Someobody came out with a chipset for Althlon/Duron CPUs that isn't from VIA.
The VIA chipset (and USB controller) on my board (Asus A7Pro) is the crappiest, most unstable piece of junk i've ever used. - and its not even overclocked.
I just hope NVidia does a better job of delivering a chipset that works properly than VIA has.
With the amount of processing power being put on phones these days (to play games, MP3s and PDA functions mostly), it won't be entirely unfeasible to implement an encrypted IP-based phone/data system tunneled inside the standard one using the OS on the phone itself to run the encryption/decryption functions.
Anyone got Linux running on a cellphone yet?
Yup, i agree with you, and i found out the hard way.
Unfortunately, after vowing to 'Never buy Intel again' and buying AMD, if i vow 'Never to buy VIA chipsets again' I might not be buying much x86 hardware for a while.
Theres nothing new about 'TruForm', the technique has been employed for years.
In fact, ATI may even run afoul of Pixar patents, but i'm sure their legal department has dealt with this possibility.
Its nice to be able to do it in hardware, but it wouldn't surprise me if all the work was done in software by the driver, rather than on the geometry engine.
If the technique is implemented fully in hardware, then i don't see how exposing it as an OpenGL extension or whatever Direct3D uses could be a bad thing, since using sub-surfs where possible will certainly improve the appearance of most 'organic' 3D models.
However, enabling this 'by default' on all geometry is almost certainly a stupid thing to do.
Has anyone managed to fix the hopeless nightmare that has plagued my hardware in every x86 PC i've owned in the last 5 years?
Why, why are we still stuck with IRQs, and what moron thought the solution to the problem was to 'share' them???
My motherboard (Asus A7Pro) has 5 PCI slots, and an AGP slot, but is only capable of using 3 IRQs, since slots 4 & 5 share an IRQ, slots 2 & 3 share an IRQ and slot 1 and the AGP slot share an IRQ.
the 2 USB controllers on board share an IRQ. They just plain don't work, probably because of the shitty VIA chipset and the fact they share an IRQ.
Its insane. It would be nice if 'IRQ sharing' actually worked, but i get all sorts of problems - USB not working, sound glitches, video glitches, DVD playback card crashing, all because the BIOS, OS and hardware are simply incapable of allocating resources correctly.
And i'm not the only one.
Why put 5 or more slots on a motherboard if its actually not possible to put a card in each one and have the computer work?
But I wonder how much performance you could really hope to gain from this approach though..
Mozilla, being a heavily graphical app, probably won't benefit much from kernel integration, since fetching the pages from the web via the network stack, storing them in memory/disk, and reading the data back out - typically kernel operations, probably take no time at all compared to the thrashing, blocking and redundant redraws that contribute to mozilla's perceived slowness.
XML support in the kernel - hmm.. i'm not sure if you'd see much performance boost here either - building node trees and traversing them might benefit from kernel integration, but if youre worried about parsing performance, then why use XML?
If youre going to put an XML parser in the kernel, then why not embed Perl in there as well? And once you have Perl in the kernel, it makes sense to add Python too. Pretty soon, the idea of having a 'kernel' disappears.
Word processing in the kernel?? Now i *know* the crack where you live is really good.
Remember there are good reasons for separating kernel and user-space activities. This stuff just plain doesn't belong in the kernel at all.
Keep the core kernel as lean as possible, and focus on doing the few things you need to do extremely well i.e. hardware interfaces, memory management and synchronisation functions.
Its probably more to do with the problems finding people to actually do the port and then support it.
OpenGL on MacOS previous to X was limited to third-party support, i think (correct me if i'm wrong), and MacOS X is pretty new, meaning nobody is really well established in this currently small market.
Linux is much more attractive to most programmers than the MacOS is - since it runs on their formerly Windows machines.
While Macs are on more desktops, you have much more developer mindshare in the gaming industry with Linux than with MacOS.
MacOS X has much more potential than MacOS =9, since porting games from an existing Linux version to a PPC MacOS X version should be a lot less work than going direct from Windows to MacOS X.
As PPC-based Linux gets more support (e.g. TiVO), this porting process will be even easier.
However it's pretty clear that the games industry is looking much more seriously at Linux than it is at OS X.
Apple can choose to take advantage of this fact, and look at how they can ease the transition between Linux and their BSD flavour, or they can do what they usually do, and ignore the games market entirely, leaving customers like you wondering 'Where are the games for my Mac?'
For 3D work, Matrox cards are just rubbish, plain and simple.
They might display your desktop across multiple monitors without a problem, but thats all they are good for.
Matrox 3D drivers are buggy, released excruciatingly late and perform badly.
After owning a 2MB Matrox Millenium (great little card, still going strong today),I was burnt with a G200, and will never buy another Matrox card again.
OpenGL would be one possible base on which to build a resolution independent GUI, with support for hardware acceleration of many operations, including evaluator functions - for curves etc., and vertex level operations using NVidia and others' geometry processors.
This is happening slowly on a number of fronts, the most widely known being E's EVAS.
Other options might include HTML/CSS, SVG or even Flash/Shockwave formats.
Postscript/PDF is pretty unwieldy, from what i have seen, but is obviously quite flexible with good typography support, which is the biggest missing element in all the others.
A lot of vertical-market applications require a handheld computer that doesn't cost the earth with good connectivity.
These applications require none of the traditional PDA functionality. For example, we currently produce an electricity meter-reading application on Casio PDAs. None of the built-in apps are useful for this, and the CE GUI architecture just gets in the way.
We don't sell many, because the product is just grossly overengineered and overpriced.
The software is due for a rewrite anyway, so now its decision time - WinCE, PalmOS or Linux.
The agenda is going to halve the cost of delivering our app to our customers compared to a Casseiopia, and will come out slightly more expensive than a Palm 3-level platform.
Programming the Palm is straightforward enough, but the environment sucks compared to my familiar gcc. Being able to deliver the same app on an X desktop as well as a PDA is awesome, and i can't wait to get my hands on one.
Disclaimer: All following comments represent my personal opionion. I do not mean to slag off anyone elses preferred desktop configuration.
And here are the things that bother me the most:
The clipboard in Linux sucks incredibly. It is nice to be able to select and middle-click to paste when you're in a console, but doing something as simple as copying and pasting a URL into the browser is difficult.
You actually have to delete the existing contents of the location bar, before you can paste the new one in. You can't select the contents of the bar and delete it after you've copied the URL you want, since the clipboard gets overwritten with the new selection. This goes for any copy/paste work, and is plain stupid.
Cut, Copy and Paste should be explicit operations, not things that happen 'behind the scenes'. Why not make ctrl-middle button copy, and middle button paste?
How would one go about changing this behaviour on a typical Linux distro? What pieces of software need to be modified to make this happen?
Fonts X font handling (rendering is reasonably good) is f*cking useless. I spent about 2 hours last night going through docs, searching through my folder structure and editing fonts.dir files manually, just to get truetype fonts to display correctly (i.e. not overlapping) in the latest builds of Mozilla.
You could call the overlapping fonts problem a bug in mozilla, but it highlighted for me the ridiculous complexity of managing fonts on Linux.
In hindsight, its not so difficult.. you modify the fonts.scale/fonts.dir files in your fonts folders, and you can find out where fonts are being stored by looking in:/usr/X11R6/lib/X11/fs/config
However, it took a long time to actually figure this out, and WHY do you need to tell X the number of fonts in the damn file??? Can it not COUNT the LINES? Rendering a single glyph is probably more intensive than counting the number of lines in a file, which it only needs to do when it is restarted. Could they possibly have picked a more unintuitive location for font files? IF theres one thing that the LSB project needs to address, its X configuration.
Filepickers You would think that one of the most useful features of the modern UNIX shells - tab completion, would make an apperance in the GUI file-pickers. Why is this not in there? if bash can do it, why can't the GTK+/GNOME filepicker? KDE may do this, i don't use it enough to know for sure.
When i think about it, there is actually not that much that bothers me about the X desktop, just mainly the points mentioned above.
I'm pretty successful at getting my work done - mostly programming and web development using Linux, but most of the other people in the office just freak out when they see that you cant copy and paste a url into the location bar of the browser.
Thank Christ Eazel have gone out of business. Nautilus was/is a pointlessly slow and bloated application, that didn't actually do anything. I actually hated this app. I only hope the GNOME foundation will be good enough to include Nautilus as an *option*, not a required component of the GNOME desktop.
As an example, Mozilla's filepicker is the worst, slowest, crappest filepicker i've even used.
Why these people thought 'hyuck hyuck, lets rewrite the widget set' is beyond me. This is, of course, purely a Mozilla issue, and not a 'Linux' one, but the problem with Linux on the desktop is that nothing related to the Linux desktop is a 'Linux problem'.. 'Thats an issue with app Z, or toolkit Y, but never with Linux, or X itself. No desktop functionality seems to actually belong anywhere central, and is instead handled by a loose collection of sloppily interconnected components, few of which know anything about each other.
Now i have that off my chest, its also necessary to point out that despite this incredibly poor design, Linux on the desktop is actually usable, and improving all the time.
XFree86 is now pretty damn fast, and i'm sure there is much potential to make it a lot faster in future.
Mozilla is pretty much complete, and even though its still dog-slow compared to browsers on an identically specced Windows machine, it will make the grade, and hopefully keep on truckin to become better, faster and more flexible than anything M$ can squeeze out of their mighty corporate anus
Sawfish and GNOME make a pretty good GUI. I much prefer multiple desktops/panels/window options to MS's super-fast but unconfigurable GUI. Can't comment on other GUIs, since i don't run them.
Overall, i just can't agree with much of what the article says. I don't rely on MS Office to get my job done though (Actually, i haven't even used MS Office in months).
Linux is alive and kicking on the desktop, and will continue to win converts all over the world.
So while this guy proclaims Linux dead and buried on the desktop, i will continue to make my living, do all my work, increase my knowledge of computing, save $$$ on software, code some cool apps, have fun and fight the power on my Linux desktop machines.
Re:WxWindows is the de facto cross platform Standa
on
Qt for Mac
·
· Score: 2
wxWindows does kick ass, and is by far the best cross-platform API available, unless you count SDL, but wxWindows is much broader in scope than SDL.
The only drawback is being forced to use C++, which is, to me, a disgusting abortion of a programming language.
However, using a minGW32 cross-compiler and wxWindows, i can build native GUI windows apps without having to leave my comfy Linux environment, nor pay Microsoft a dollar for their dev tools. This is A Very Good Thing
Haven't needed to build a mac app yet, but since OS X has yet to really win over the mac userbase, i don't see an OS 9 compatible application as much of a problem, not to mention making your app usable on the vast installed base of macs that won't ever run OS X since theyre not fast enough.
Have you just chosen to forget that America was populated thousands of years before settlers from european countries showed up and exterminated them?
Why don't you look at the (extremely) long history of your own indigenous cultures before complaining that 'Those darn red indians are just too boring, how come we don't have cool chinese mummies?' Theyre so cool they have TV shows.
What the hell is wrong with you people.. if there isn't a TV show about it, it doesnt exist?
Slashdot Mirror
The kernel and various other components will be open source, so Sony will gain the benefits of a rock-solid, continually evolving platform. Thats the advantage of open source - take what the community does well and use it.
But why they would open their GUI, application libraries etc. to the public? Sony have more than enough development muscle to create what they need themselves.
When it comes to a desktop framework for the average computer user, its clear that they can't leave *any* of that to the open source community, since we really suck at producing anything like that. The open source community is a hindrance, rather than a help in this area.
I don't know about you, but i use a linux desktop exclusively, at home and at work, and i can't recommend it to anyone else i know, because it just sucks so badly at providing accessible desktop functionality. Of course, there are lots of benefits with a Linux/UNIX desktop (modularity, remote display, customisability etc.) but almost none of them are particularly relevant for a console.
Maybe we'll get there one day, but I'm not holding my breath, and i'll bet Sony won't be either.
Sony also want to control the apps that ship for the PS2. Quality Control is extremely important with regard to consoles, while it has been completely forgotten on the desktop PC. They will need the legal clout to enforce this quality, and they will have it by licensing their GUI and toolkit libraries accordingly.
It may be possible to install 3rd party stuff on a PS2, but you can bet your ass you won't be able to ship a program for the PS2 without first getting (paying for) Sony's approval, or requiring the user to replace the kernel, windowing environment etc. and hence sacrificing compatibility with the Sony-sanctioned apps.
The number of people who will buy a PS2 so they can run only free software on it is statistically insignificant. IT probably will be possible to run X on the PS2, if only to avoid the bad press from hordes of whining Slahdotters.
The most likely scenario for consumer use, however, is that Sony will use the linux kernel, but write a proprietary framebuffer-driver like DirectFB for the PS2 hardware.
They will also probably write their own GUI toolkit, also under proprietary license, that also runs on an X Windows or Linux framebuffer backend.
Meaning that while the core OS is, and always will be free, PS2 applications and the 'Official' GUI on the System will be proprietary, much like Apple's MacOS X GUI.
You will have to pay Sony to develop apps to run on a PS2 (since you need the toolkit API and libraries), but you will be able to develop and run those apps on a Linux desktop PC, as long as you have the (proprietary) runtime and devlopment libraries installed.
There are other markets for a Linux-running PS2 platform - the film, broadcast and sports industries for realtime graphics (realtime mocap-driven characters, virtual sets, advertising overlays etc.) These apps need a solid OS and devlopment tools, which is what Linux is perfect for.
Fisheye lens-based panoramas are no better than panoramas carefully constructed from a large number of rectilinear images.
They are, however, faster and easier to produce - 2 images, as opposed to the 12-16 you need to cover a full 360 degrees with a non-fisheye lens.
Thats really the only benefit - in fact taking lots of pictures will give you a better quality final image, assuming you can orient and align the images perfectly - i use a tripod with calibration marks on it to do mine, with excellent results.
Panorama Tools lets you use both methods to produce panoramic images, also has a free viewer, runs on several OSes as plugins for the GIMP and Photoshop, and has high-quality output.
Soundcards with SP/DIF or Optical I/O are pretty common these days.
Why would you use analog I/O at all?
16bit samples at a 48khz sampling rate= 768Kbps which is pretty respectable.
Panorama Tools is undoubtedly the best toolset for producing panoramic imagery available today.
It is a sad day when a guy as committed to providing great software to the community as Mr. Dersch gets bullied by the U.S. Patent Office and a litigious company who have obviously failed to come up with anything as functional as Mr. Dersch's Panorama Tools.
One can only hope the European Patent Office will look a little closer at the many cases of clear prior art with regard to IPIX's patent, and then throwing it in the bin where it belongs.
Down with IPIX and incompetent patent offices.
4-in1 drivers? Well, i was going to try that as my next step under Win2k (i used to dual boot), but..
In attempting to make my fucking VIA USB controller work, i flashed the BIOS on my mobo using the ASUS-supplied utility.
Upon reboot, Win2k claims the mobo is no longer ACPI-compliant and BSODs, leaving me unable to:
a) flash the BIOS back to the old version, and
b) unable to access the files on my NTFS drive.
Linux, however, boots and runs quite happily on the same BIOS, but the onboard USB controller is still fucking horked no matter which OS i try and use it under.
A third-party USB card seems to work OK, but now all my slots are full, my machine crashes a lot (even under Linux) and i paid good money for a worthless fucking USB controller and a shoddy chipset.
I am not a happy camper.
If i rowed a boat out into the middle of the pacific ocean and found a small, newly-formed volcanic island, are there any international laws/statutes/prior claims over bodies of water etc. that would prevent me from declaring sovereignty, setting up a government and joining the U.N. etc?
Or is the idea of 'sovereignty' entirely dependent on your ability to defend the territory you claim by economic, diplomatic and/or military force?
Someobody came out with a chipset for Althlon/Duron CPUs that isn't from VIA.
The VIA chipset (and USB controller) on my board (Asus A7Pro) is the crappiest, most unstable piece of junk i've ever used. - and its not even overclocked.
I just hope NVidia does a better job of delivering a chipset that works properly than VIA has.
With the amount of processing power being put on phones these days (to play games, MP3s and PDA functions mostly), it won't be entirely unfeasible to implement an encrypted IP-based phone/data system tunneled inside the standard one using the OS on the phone itself to run the encryption/decryption functions. Anyone got Linux running on a cellphone yet?
Why is it that there is no decent browser for a proprietary, expensive OS which has an immeasurably small portion of the desktop market?
Why aren't you complaining to Microsoft or Netscape that they haven't ported their browser to NeXT?
Why aren't you complaining to Apple about the fact their MacOS X software isn't backwards-compatible with your version of NeXTSTEP?
Most people on Slashdot use Windows anyway, so you'll get no sympathy here.
Maybe you should look to the GNUStep project for help.
Yup, i agree with you, and i found out the hard way.
Unfortunately, after vowing to 'Never buy Intel again' and buying AMD, if i vow 'Never to buy VIA chipsets again' I might not be buying much x86 hardware for a while.
Theres nothing new about 'TruForm', the technique has been employed for years.
In fact, ATI may even run afoul of Pixar patents, but i'm sure their legal department has dealt with this possibility.
Its nice to be able to do it in hardware, but it wouldn't surprise me if all the work was done in software by the driver, rather than on the geometry engine.
If the technique is implemented fully in hardware, then i don't see how exposing it as an OpenGL extension or whatever Direct3D uses could be a bad thing, since using sub-surfs where possible will certainly improve the appearance of most 'organic' 3D models.
However, enabling this 'by default' on all geometry is almost certainly a stupid thing to do.
Yeah, this is great when it actually works.
If, for whatever reason, it doesn't, what are you supposed to do?
Has anyone managed to fix the hopeless nightmare that has plagued my hardware in every x86 PC i've owned in the last 5 years?
Why, why are we still stuck with IRQs, and what moron thought the solution to the problem was to 'share' them???
My motherboard (Asus A7Pro) has 5 PCI slots, and an AGP slot, but is only capable of using 3 IRQs, since slots 4 & 5 share an IRQ, slots 2 & 3 share an IRQ and slot 1 and the AGP slot share an IRQ.
the 2 USB controllers on board share an IRQ. They just plain don't work, probably because of the shitty VIA chipset and the fact they share an IRQ.
Its insane. It would be nice if 'IRQ sharing' actually worked, but i get all sorts of problems - USB not working, sound glitches, video glitches, DVD playback card crashing, all because the BIOS, OS and hardware are simply incapable of allocating resources correctly.
And i'm not the only one.
Why put 5 or more slots on a motherboard if its actually not possible to put a card in each one and have the computer work?
But I wonder how much performance you could really hope to gain from this approach though..
Mozilla, being a heavily graphical app, probably won't benefit much from kernel integration, since fetching the pages from the web via the network stack, storing them in memory/disk, and reading the data back out - typically kernel operations, probably take no time at all compared to the thrashing, blocking and redundant redraws that contribute to mozilla's perceived slowness.
XML support in the kernel - hmm.. i'm not sure if you'd see much performance boost here either - building node trees and traversing them might benefit from kernel integration, but if youre worried about parsing performance, then why use XML?
If youre going to put an XML parser in the kernel, then why not embed Perl in there as well? And once you have Perl in the kernel, it makes sense to add Python too. Pretty soon, the idea of having a 'kernel' disappears.
Word processing in the kernel?? Now i *know* the crack where you live is really good.
Remember there are good reasons for separating kernel and user-space activities. This stuff just plain doesn't belong in the kernel at all.
Keep the core kernel as lean as possible, and focus on doing the few things you need to do extremely well i.e. hardware interfaces, memory management and synchronisation functions.
Its probably more to do with the problems finding people to actually do the port and then support it.
OpenGL on MacOS previous to X was limited to third-party support, i think (correct me if i'm wrong), and MacOS X is pretty new, meaning nobody is really well established in this currently small market.
Linux is much more attractive to most programmers than the MacOS is - since it runs on their formerly Windows machines.
While Macs are on more desktops, you have much more developer mindshare in the gaming industry with Linux than with MacOS.
MacOS X has much more potential than MacOS =9, since porting games from an existing Linux version to a PPC MacOS X version should be a lot less work than going direct from Windows to MacOS X.
As PPC-based Linux gets more support (e.g. TiVO), this porting process will be even easier.
However it's pretty clear that the games industry is looking much more seriously at Linux than it is at OS X.
Apple can choose to take advantage of this fact, and look at how they can ease the transition between Linux and their BSD flavour, or they can do what they usually do, and ignore the games market entirely, leaving customers like you wondering 'Where are the games for my Mac?'
For 3D work, Matrox cards are just rubbish, plain and simple.
They might display your desktop across multiple monitors without a problem, but thats all they are good for.
Matrox 3D drivers are buggy, released excruciatingly late and perform badly.
After owning a 2MB Matrox Millenium (great little card, still going strong today),I was burnt with a G200, and will never buy another Matrox card again.
OpenGL would be one possible base on which to build a resolution independent GUI, with support for hardware acceleration of many operations, including evaluator functions - for curves etc., and vertex level operations using NVidia and others' geometry processors.
This is happening slowly on a number of fronts, the most widely known being E's EVAS.
Other options might include HTML/CSS, SVG or even Flash/Shockwave formats.
Postscript/PDF is pretty unwieldy, from what i have seen, but is obviously quite flexible with good typography support, which is the biggest missing element in all the others.
'You may only use this software on an OS that conforms to license X'
is the same as:
'You may only use this disc on a player that conforms to license Y'
And we all know how offensive such a clause is to the slashdot community, don't we?
It will immediately improve the look n feel of Sun's desktop. Those pink windows and oversized menu-bar make me squeamish.
However if Sun really wants to make a contribution to the UNIX desktop, they should release a Java UI toolkit that uses native GTK+ widgets.
Hell, Sun might even devote some development resource to Nautilus, which is (for me) unusable due to its sluggishness.
You got a million dollars of funding by a games publisher to make a game about flowers and fluffy kittens?
These applications require none of the traditional PDA functionality. For example, we currently produce an electricity meter-reading application on Casio PDAs. None of the built-in apps are useful for this, and the CE GUI architecture just gets in the way.
We don't sell many, because the product is just grossly overengineered and overpriced.
The software is due for a rewrite anyway, so now its decision time - WinCE, PalmOS or Linux.
The agenda is going to halve the cost of delivering our app to our customers compared to a Casseiopia, and will come out slightly more expensive than a Palm 3-level platform.
Programming the Palm is straightforward enough, but the environment sucks compared to my familiar gcc. Being able to deliver the same app on an X desktop as well as a PDA is awesome, and i can't wait to get my hands on one.
And here are the things that bother me the most:
You actually have to delete the existing contents of the location bar, before you can paste the new one in. You can't select the contents of the bar and delete it after you've copied the URL you want, since the clipboard gets overwritten with the new selection. This goes for any copy/paste work, and is plain stupid.
Cut, Copy and Paste should be explicit operations, not things that happen 'behind the scenes'. Why not make ctrl-middle button copy, and middle button paste?
How would one go about changing this behaviour on a typical Linux distro? What pieces of software need to be modified to make this happen?
Fonts X font handling (rendering is reasonably good) is f*cking useless. I spent about 2 hours last night going through docs, searching through my folder structure and editing fonts.dir files manually, just to get truetype fonts to display correctly (i.e. not overlapping) in the latest builds of Mozilla.
You could call the overlapping fonts problem a bug in mozilla, but it highlighted for me the ridiculous complexity of managing fonts on Linux.
In hindsight, its not so difficult.. you modify the fonts.scale/fonts.dir files in your fonts folders, and you can find out where fonts are being stored by looking in: /usr/X11R6/lib/X11/fs/config
However, it took a long time to actually figure this out, and WHY do you need to tell X the number of fonts in the damn file??? Can it not COUNT the LINES? Rendering a single glyph is probably more intensive than counting the number of lines in a file, which it only needs to do when it is restarted. Could they possibly have picked a more unintuitive location for font files? IF theres one thing that the LSB project needs to address, its X configuration.
Filepickers You would think that one of the most useful features of the modern UNIX shells - tab completion, would make an apperance in the GUI file-pickers. Why is this not in there? if bash can do it, why can't the GTK+/GNOME filepicker? KDE may do this, i don't use it enough to know for sure.
When i think about it, there is actually not that much that bothers me about the X desktop, just mainly the points mentioned above.
I'm pretty successful at getting my work done - mostly programming and web development using Linux, but most of the other people in the office just freak out when they see that you cant copy and paste a url into the location bar of the browser.
Thank Christ Eazel have gone out of business. Nautilus was/is a pointlessly slow and bloated application, that didn't actually do anything. I actually hated this app. I only hope the GNOME foundation will be good enough to include Nautilus as an *option*, not a required component of the GNOME desktop.
As an example, Mozilla's filepicker is the worst, slowest, crappest filepicker i've even used.
Why these people thought 'hyuck hyuck, lets rewrite the widget set' is beyond me. This is, of course, purely a Mozilla issue, and not a 'Linux' one, but the problem with Linux on the desktop is that nothing related to the Linux desktop is a 'Linux problem'.. 'Thats an issue with app Z, or toolkit Y, but never with Linux, or X itself. No desktop functionality seems to actually belong anywhere central, and is instead handled by a loose collection of sloppily interconnected components, few of which know anything about each other.
Now i have that off my chest, its also necessary to point out that despite this incredibly poor design, Linux on the desktop is actually usable, and improving all the time.
XFree86 is now pretty damn fast, and i'm sure there is much potential to make it a lot faster in future.
Mozilla is pretty much complete, and even though its still dog-slow compared to browsers on an identically specced Windows machine, it will make the grade, and hopefully keep on truckin to become better, faster and more flexible than anything M$ can squeeze out of their mighty corporate anus
Sawfish and GNOME make a pretty good GUI. I much prefer multiple desktops/panels/window options to MS's super-fast but unconfigurable GUI. Can't comment on other GUIs, since i don't run them.
Overall, i just can't agree with much of what the article says. I don't rely on MS Office to get my job done though (Actually, i haven't even used MS Office in months).
Linux is alive and kicking on the desktop, and will continue to win converts all over the world.
So while this guy proclaims Linux dead and buried on the desktop, i will continue to make my living, do all my work, increase my knowledge of computing, save $$$ on software, code some cool apps, have fun and fight the power on my Linux desktop machines.
The only drawback is being forced to use C++, which is, to me, a disgusting abortion of a programming language.
However, using a minGW32 cross-compiler and wxWindows, i can build native GUI windows apps without having to leave my comfy Linux environment, nor pay Microsoft a dollar for their dev tools. This is A Very Good Thing
Haven't needed to build a mac app yet, but since OS X has yet to really win over the mac userbase, i don't see an OS 9 compatible application as much of a problem, not to mention making your app usable on the vast installed base of macs that won't ever run OS X since theyre not fast enough.
Why don't you look at the (extremely) long history of your own indigenous cultures before complaining that 'Those darn red indians are just too boring, how come we don't have cool chinese mummies?' Theyre so cool they have TV shows.
What the hell is wrong with you people.. if there isn't a TV show about it, it doesnt exist?