But this software implementation can't be used as a guide to make an generic hardware-accelerated implementation? Better yet, can't this be used for a clear-room implementation? One that is safe to incorporate on Wine, Cedega, SDL, etc...
Apple is wrestling with the studios to get their films online, the price doesn't matter at this point. They need the content, and they need it to be first on ITMS, to keep themselves ahead of the competition...
Once the content is there, Apple may even sell it at a loss... only to drive more sales of the iPod Video and AppleTV.
It's a worth fight... because the "Linux OS" is an imaginary entity, that has all the faults and merits from every distro out there, without being none. Of course when most of the IT drones talk about "Linux", they are really talking about RedHat... so in the end this misconception might be good for RH, but it's bad for the other distros.
It's hard to argue with your boss when he doesn't realize that there are HUGE differences between SUSE, RedHat, Ubuntu, Debian, Gentoo etc... If every distro was equally acknowledged for what it is, it would be much easier to pick the best for a particular task, and deploy it without having to argue why Debian will do fine in the server... despite the fact that Mandriva sucked balls when running at your boss notebook.
Why do I have to browse the web on something that wants to be an applications platform...
Consider for a moment that currently, the web itself is an application platform. But as cool as some websites might be, HTML + Javascript wasnt designed for complex interface design... Flash and JavaWebStart are too heavy and intrusive.
We *need* a sane standart way to define rich user interfaces that can be deployed using HTTP, something designed to better support assincronous events, and that look and behave consistently. Lets face it, AJAX as nice as it might seem, its nightmare to implement and mantain... and it only gets worse as your project grows.
Firefox uses a XML format to define its interface, the widget placement and behavior, its called XUL. And actually it can interpret XUL documents loaded at runtime, even over the internet... and it uses JavaScript to define the behavior, so if you use AJAX today it will be very easy to learn and use XUL.
Take a look at this proof-of-concept application: Mozilla Amazon Browser - http://www.faser.net/mab/ - its just fantastic! It behaves and looks like an ordinary desktop application, but every bit of the processing happens at the server-side.
Im not saying that Firefox and XUL are the right choices... but theyre a step in the right direction, and a much saner solution than aberrations like ActiveX.
This only means that some internet radios will move their servers to somewhere out of the USA... somewhere with more sane copyright laws, or with no laws at all.
I guess it's good news for the internet hosting business at Russia, China, etc...
What bugs me most, is that I can't change settings and apply them without having to restart my whole session! And configuring a second monitor on my laptop involve so much tinkering that I gave up... Windows got those right a long time ago:-(
Xorg was a huge step in the right direction, but there are lots of usability issues that remain to be fixed.
Depending on the level of access they have to the systems, that may not actually mean anything. Were the interested parties able to do a source code audit? Does the system provide a means to ensure that the audited code is also the code running on the device? The device can be silently eating votes and you'll never know.
IIRC, representatives from every political party had access to the software sources... and several exploit points where detected at that time. Yet, no attempt to actually exploit the machines was actually heard of.
AFAIK, the votes are stored using cryptography AND a MD5 hash during the voting process. The internal storage is only accessible by opening the machine, something that is hard to do in front of several people and policemen. The contents are copied to an CF card, again using crypo and a MD5 hash to ensure nothing would be altered.
So, while there are some flaws in the software... the way the entire election process is conducted prevents one from actually exploiting these machines.
Take for example our last presidential election. Everything went smooth, and without any problems... sure, some voting machines had to be replaced, and at few remote locations people had to use paper ballots. But even with these minor problems, we had the elections result within HOURS.
Note that the average education among the people here is very low... lot's of people here can barely read. Yet, IIRC, most voters spent less than 1 minute to vote... thanks to educative campaigns made by the government.
And while the voting machines use a rather flawed system... no fraud attempt was detected. I guess that's because every political party can send representatives to watch the entire process... and there's LOTS of political parties around here, so there's also lots of people watching.
Another thing that might contribute for our success, is uniformity. The elections are handled by the Electoral Justice (Justiça Eleitoral), a federal department, and every state must follow it's rule.
What I find amusing, is how much success we had using electronic voting machines here at Brazil... we have been using these for almost 10 years now. The last presidential election was almost entirely conducted using these machines... and only a few on the entire country had to be replaced due failure.
Of course there are some issues to be sorted, but overall it was a huge improvement over the old paper-based system.
The funny thing is... VMWare actually adds value to Windows.
I think that this is a very bad move from Microsoft... as usual, they can't be satisfied with just a piece of the pie, they want everything. VMWare is a mature application, that adds real value to their product. If people can't use VMWare + Windows probably they'll switch to VMWare+Linux, or KVM, or Xen.
What if Microsoft's Virtual PC just don't catch on? They're risking to loose this entire virtualization market to Linux, both as host and guest OS.
Dell can just build 100% Intel boxes, CPU, GPU, NetWorking etc... And they'll just work with the majority of the distros out there, and using opensource drivers.
I'm hoping that this will put some pressure on AMD/ATI and nVidia to release opensource drivers to their products.
These laptops can connect to the net, sure... but I think the main network they'll be acessing is the mesh network formed by the other kid's laptops.
Also, they access internet trougth a gateway placed at the local school, a simple content filter for squid-proxy like Dans-Guardian or Chastity will do the trick. They can filter who can access the internet by filtering MAC or IP addresses, so the laptops owned by children under a centain age wouldn't pass.
These laptops won't be directly connected to the internet.
I have a notebook with an i915G video chipset, and making beryl work on my system was a straight forward process. Now I'm using it all the time, with very few (if any) stability issues, and even when anything goes wrong it falls back to the default window manager.
What I'd like to see is a more serious effort to show what hardware is fully supported under Linux. Of course you have some listings at the Ubuntu wiki, and other distros provide similar info... but I don't think it's enough. It would be nice to have an unified online database of products that work out-of-the box with linux, complete with user reviews and per-distro issues listed. And, of course, a hall of shame... listing all unsupported hardware, and hardware that only have proprietary drivers.
BTW, low bar to entry is one of the reasons JavaScript is such a way cooler language than anyone ever gives it credit for. Say what you like about its shortcomings; there is an immense benefit to being able to point someone at a URL and they can run your program.
Yeah,
That's why I think Java WebStart is a underused feature... As arcane as SWING might be, I find it much more sane than the equivalent complex AJAX code.
Actually I enjoy Java quite a bit. Things like SpringFramework, WebWork and Hibernate take a lot of weight from the developer's back, and are very flexible... Java has a really nice environment for Web development.
Swing on the other hand is a major PITA. But with Java now under a GPL licence, I'm expecting to see more and more native bindings to commom GUI toolkits.
Not to mention that with full access to the Sun libraries, gcj will probably do a much better job at compiling Java to native code!
Now that Java is OpenSource, and that it has bindings to both GTK (as in SWT) and QT (as in Jambi), will we see it on more desktop applications? I'm asking because I feel that Java is a better choice than C#, because of its extensive libraries and frameworks.
Also, Java is already a major player on the server side, if KDE and Gnome had a better integration with it than Windows... it would be a major push for the adoption of a FOSS Desktop...
Also one may notice that many applications will see a boost in performance, even being sequential. That's because it will suffer less from context switching, since there's more than one processing core to share the load.
I can see this trend as a big win for the OpenSource OSes out there. Remember how the Linux scheduler was slow, and how Apache2 on Windows was faster? That doesn't last long, a new scheduler was written, and the Linux performance for treads got a big boost, and just kept on getting better. Windows, on the other hand, just couldn't keep up... because Microsoft doesn't have the necessary agility, and because the Windows kernel is tied with too much stuff.
This depends on how the voice signal is processed at the stations. Remember that your conversation has at least one intermediary, in the case that the both persons are in the same cell.
So, if the station processes your voice data-stream to further increase it's compression, you won't be able to apply any crypto to it.
Write on OpenOffice, then upload it to the servers.
Hell, the minute this thing is out someone will appear with some OOo plugin to save/edit the documents directly to the servers...
One might as well write a KIOSlave for KDE, or a plugin for GnomeFS... that way both KOffice and Abiword will be able to work transparently on the documents stored on the server.
And, maybe nobody has to anything at all... a simple WebDAV interface to the server will do the trick.
I'm pretty sure that Epic is able to provide a compatibility layer between UE2 and UE3, so they can share models, maps, texture formats, etc... Also, if the API is similar enough, the same game logic should work with both.
Of course the graphics will be less impressive on the Wii, as it has less power... but does it really matters? I really don't enjoy playing FPS with joysticks... this kind of game is much better served with a keyboard/mouse combo, and the Wii controller is what most resembles this. So the Wii version might have the weaker graphics, but it might have the strongest gameplay.
Anyways, all the three consoles cost an arm and a leg around here at Brazil, and I'm talking about 5x the USA price! I guess I'll be keeping my PS2 for a looooong time, as probably most of the world outside USA/EUROPE/Japan.
I guess nowdays most of the film downloads are from series such as Lost, Heroes, etc... Who would bother to download an entire DVD when is so much easier to rent one, and them rip it?
And, yet again... why rip it when is soooo much easier to rent it again?
Slashdot Mirror
Ok,
But this software implementation can't be used as a guide to make an generic hardware-accelerated implementation? Better yet, can't this be used for a clear-room implementation? One that is safe to incorporate on Wine, Cedega, SDL, etc...
My guess...
Apple is wrestling with the studios to get their films online, the price doesn't matter at this point. They need the content, and they need it to be first on ITMS, to keep themselves ahead of the competition...
Once the content is there, Apple may even sell it at a loss... only to drive more sales of the iPod Video and AppleTV.
Still,
... If every distro was equally acknowledged for what it is, it would be much easier to pick the best for a particular task, and deploy it without having to argue why Debian will do fine in the server... despite the fact that Mandriva sucked balls when running at your boss notebook.
It's a worth fight... because the "Linux OS" is an imaginary entity, that has all the faults and merits from every distro out there, without being none. Of course when most of the IT drones talk about "Linux", they are really talking about RedHat... so in the end this misconception might be good for RH, but it's bad for the other distros.
It's hard to argue with your boss when he doesn't realize that there are HUGE differences between SUSE, RedHat, Ubuntu, Debian, Gentoo etc
* Sorry my poor english, I'm no native speaker.
Sorry to disagree... but a Monopoly on what? Music distribution?
If you want to feed music into your XYZ branded DAP, do as everybody else and buy a CD!
Consider for a moment that currently, the web itself is an application platform. But as cool as some websites might be, HTML + Javascript wasnt designed for complex interface design... Flash and JavaWebStart are too heavy and intrusive.
We *need* a sane standart way to define rich user interfaces that can be deployed using HTTP, something designed to better support assincronous events, and that look and behave consistently. Lets face it, AJAX as nice as it might seem, its nightmare to implement and mantain... and it only gets worse as your project grows.
Firefox uses a XML format to define its interface, the widget placement and behavior, its called XUL. And actually it can interpret XUL documents loaded at runtime, even over the internet... and it uses JavaScript to define the behavior, so if you use AJAX today it will be very easy to learn and use XUL.
Take a look at this proof-of-concept application: Mozilla Amazon Browser - http://www.faser.net/mab/ - its just fantastic! It behaves and looks like an ordinary desktop application, but every bit of the processing happens at the server-side.
Im not saying that Firefox and XUL are the right choices... but theyre a step in the right direction, and a much saner solution than aberrations like ActiveX.
Just my $0.02
Uh...
How it is different from raising a cow to make hamburger?
This only means that some internet radios will move their servers to somewhere out of the USA... somewhere with more sane copyright laws, or with no laws at all.
I guess it's good news for the internet hosting business at Russia, China, etc...
Perhaps he is talking about the desktop?
Maybe porting DBUS and HAL to Solaris... A recent KDE and Gnome wouldn't hurt either.
Agreed,
:-(
What bugs me most, is that I can't change settings and apply them without having to restart my whole session! And configuring a second monitor on my laptop involve so much tinkering that I gave up... Windows got those right a long time ago
Xorg was a huge step in the right direction, but there are lots of usability issues that remain to be fixed.
This sounds like we have a huge general purpose CPU, while the animals have a tiny one and several special purpose DSPs...
IIRC, representatives from every political party had access to the software sources... and several exploit points where detected at that time. Yet, no attempt to actually exploit the machines was actually heard of.
AFAIK, the votes are stored using cryptography AND a MD5 hash during the voting process. The internal storage is only accessible by opening the machine, something that is hard to do in front of several people and policemen. The contents are copied to an CF card, again using crypo and a MD5 hash to ensure nothing would be altered.
So, while there are some flaws in the software... the way the entire election process is conducted prevents one from actually exploiting these machines.
Well,
Take for example our last presidential election. Everything went smooth, and without any problems... sure, some voting machines had to be replaced, and at few remote locations people had to use paper ballots. But even with these minor problems, we had the elections result within HOURS.
Note that the average education among the people here is very low... lot's of people here can barely read. Yet, IIRC, most voters spent less than 1 minute to vote... thanks to educative campaigns made by the government.
And while the voting machines use a rather flawed system... no fraud attempt was detected. I guess that's because every political party can send representatives to watch the entire process... and there's LOTS of political parties around here, so there's also lots of people watching.
Another thing that might contribute for our success, is uniformity. The elections are handled by the Electoral Justice (Justiça Eleitoral), a federal department, and every state must follow it's rule.
What I find amusing, is how much success we had using electronic voting machines here at Brazil... we have been using these for almost 10 years now. The last presidential election was almost entirely conducted using these machines... and only a few on the entire country had to be replaced due failure.
Of course there are some issues to be sorted, but overall it was a huge improvement over the old paper-based system.
So, why did Brazil succeed where the USA failed?
The funny thing is... VMWare actually adds value to Windows.
I think that this is a very bad move from Microsoft... as usual, they can't be satisfied with just a piece of the pie, they want everything. VMWare is a mature application, that adds real value to their product. If people can't use VMWare + Windows probably they'll switch to VMWare+Linux, or KVM, or Xen.
What if Microsoft's Virtual PC just don't catch on? They're risking to loose this entire virtualization market to Linux, both as host and guest OS.
Well,
Dell can just build 100% Intel boxes, CPU, GPU, NetWorking etc... And they'll just work with the majority of the distros out there, and using opensource drivers.
I'm hoping that this will put some pressure on AMD/ATI and nVidia to release opensource drivers to their products.
Well,
These laptops can connect to the net, sure... but I think the main network they'll be acessing is the mesh network formed by the other kid's laptops.
Also, they access internet trougth a gateway placed at the local school, a simple content filter for squid-proxy like Dans-Guardian or Chastity will do the trick. They can filter who can access the internet by filtering MAC or IP addresses, so the laptops owned by children under a centain age wouldn't pass.
These laptops won't be directly connected to the internet.
I can confirm that for the Intel drivers.
I have a notebook with an i915G video chipset, and making beryl work on my system was a straight forward process. Now I'm using it all the time, with very few (if any) stability issues, and even when anything goes wrong it falls back to the default window manager.
What I'd like to see is a more serious effort to show what hardware is fully supported under Linux. Of course you have some listings at the Ubuntu wiki, and other distros provide similar info... but I don't think it's enough. It would be nice to have an unified online database of products that work out-of-the box with linux, complete with user reviews and per-distro issues listed. And, of course, a hall of shame... listing all unsupported hardware, and hardware that only have proprietary drivers.
Yeah,
That's why I think Java WebStart is a underused feature... As arcane as SWING might be, I find it much more sane than the equivalent complex AJAX code.
Well,
Actually I enjoy Java quite a bit. Things like SpringFramework, WebWork and Hibernate take a lot of weight from the developer's back, and are very flexible... Java has a really nice environment for Web development.
Swing on the other hand is a major PITA. But with Java now under a GPL licence, I'm expecting to see more and more native bindings to commom GUI toolkits.
Not to mention that with full access to the Sun libraries, gcj will probably do a much better job at compiling Java to native code!
I wonder,
Now that Java is OpenSource, and that it has bindings to both GTK (as in SWT) and QT (as in Jambi), will we see it on more desktop applications? I'm asking because I feel that Java is a better choice than C#, because of its extensive libraries and frameworks.
Also, Java is already a major player on the server side, if KDE and Gnome had a better integration with it than Windows... it would be a major push for the adoption of a FOSS Desktop...
Agreed,
Also one may notice that many applications will see a boost in performance, even being sequential. That's because it will suffer less from context switching, since there's more than one processing core to share the load.
I can see this trend as a big win for the OpenSource OSes out there. Remember how the Linux scheduler was slow, and how Apache2 on Windows was faster? That doesn't last long, a new scheduler was written, and the Linux performance for treads got a big boost, and just kept on getting better. Windows, on the other hand, just couldn't keep up... because Microsoft doesn't have the necessary agility, and because the Windows kernel is tied with too much stuff.
Just my $0.02
This depends on how the voice signal is processed at the stations. Remember that your conversation has at least one intermediary, in the case that the both persons are in the same cell.
So, if the station processes your voice data-stream to further increase it's compression, you won't be able to apply any crypto to it.
Write on OpenOffice, then upload it to the servers.
Hell, the minute this thing is out someone will appear with some OOo plugin to save/edit the documents directly to the servers...
One might as well write a KIOSlave for KDE, or a plugin for GnomeFS... that way both KOffice and Abiword will be able to work transparently on the documents stored on the server.
And, maybe nobody has to anything at all... a simple WebDAV interface to the server will do the trick.
Also,
I'm pretty sure that Epic is able to provide a compatibility layer between UE2 and UE3, so they can share models, maps, texture formats, etc... Also, if the API is similar enough, the same game logic should work with both.
Of course the graphics will be less impressive on the Wii, as it has less power... but does it really matters? I really don't enjoy playing FPS with joysticks... this kind of game is much better served with a keyboard/mouse combo, and the Wii controller is what most resembles this. So the Wii version might have the weaker graphics, but it might have the strongest gameplay.
Anyways, all the three consoles cost an arm and a leg around here at Brazil, and I'm talking about 5x the USA price! I guess I'll be keeping my PS2 for a looooong time, as probably most of the world outside USA/EUROPE/Japan.
Actually,
I guess nowdays most of the film downloads are from series such as Lost, Heroes, etc... Who would bother to download an entire DVD when is so much easier to rent one, and them rip it?
And, yet again... why rip it when is soooo much easier to rent it again?