But Microsoft is pulling the same shit Adobe is with Flash. Claiming it's "open" but gaming the spec release schedule in such a manner that it is impossible for anything but their implementation to actually work - By the time someone has implemented to the open spec, MS and Adobe have released new versions and content providers are using those new versions.
For example, there was hope that Monolight combined with Netflix transitioning to Silverlight would bring Netflix streaming to Linux. However, Netflix tracks Microsoft's Silverlight releases pretty quickly, meaning that on a consistent basis, by the time Monolight has caught up to wherever Netflix might have been, Netflix has moved on to the next Silverlight release. Same for nearly all other SL content out there.
It's the same sad situation for any of the alternative Flash players - They continue to remain novelties because Adobe never seems to update the spec until they have released the next version of Flash, meaning Gnash et al are always at least a generation behind. Let's not forget the fact that the Flash spec is missing critical stuff such as RTMPE documentation. (The only public RTMPE documentation out there was obtained via reverse engineering Adobe's implementation.)
Problem is, it is VERY difficult to implement content-provider-friendly DRM on Linux.
Content providers still insist on DRMing what they can, despite the fact that many of their delivery methods are known to be fully compromised. (They're dumbasses like that.)
I mean, who cares if Netflix streaming can be ripped when the higher-quality DVD is 100% compromised, and even Blu-Ray is compromised?
Their selection now is MUCH better than it was six months or so ago. (They have recently made quite a few deals with content providers, those deals are why the bumped up prices a bit.)
For me, the FF series has actually driven console purchases for me. I didn't buy a PS2 until FFXII came out (Worth it IMO! I did not buy many other PS2 games but still don't regret the PS2 purchase), I didn't buy a PS3 until FFXIII came out - NOT worth it! (At least not for FFXIII alone - I still have a great Blu-Ray player/UPnP frontend, and I have picked up a few other PS3 games.)
I'm going to be a lot more careful with future FF releases though.
It could be damage control - some of the characters did have potential, and Square has shown that sequel gameplay can be VASTLY different from the original game. See X-2.
(Note: X-2 was a massive step down from X, but XIII-2 could be a major step up - it would be hard to go down.)
The linear gameplay in XIII was AWFUL. I hope XIII-2 is significantly different. (Good news: the last -2 version of an FF game was vastly different from the game it was a sequel to. Bad news: X-2 was a major step down from X. On the other hand, it'll be hard to go downwards from XIII.)
As to change in team - This is definately a factor. The Final Fantasy team is still capable of putting out some great games (like XII), however the percentage of duds is increasing. It's a combination of "milking the franchise" (although mainline games should get the quality focus, it's more acceptable to have spinoffs like the various VII spinoffs be "meh") and the original Final Fantasy team members taking a less active role and more of an advisory role.
Nobuo Uematsu has been reducing his involvement in many of the FF games over the past few releases, but it is obvious from the quality of FFXIII's soundtrack that he had VERY little involvement with that game. One of the things the Final Fantasy series has always been known for is great music (much due to Uematsu being an amazing composer), however FFXIII's music was so bad that one area had elevator music as the background music!
A good point here - Google has a lot of "green" initiatives (reduced-power computing, huge solar cell farms on their roof, etc.)
This approach is NOT a "green" approach - a "green" approach is one that makes use of the large amount of hardware acceleration infrastructure now deployed for the existing standard codecs.
WebM/VP8 will force a non-accelerated CPU-only rendering path on ALL existing hardware. This eats power compared to hardware acceleration. (Look at how well most Android devices handle H.264 thanks to hardware accelerated decoding.)
Google is being hypocritical and inconsistent here. Great summary at http://daringfireball.net/2011/01/simple_questions - Key here is, HTML5 was supposed to at least partially break Adobe's stranglehold on the web by moving some content away from Flash. Google just killed any hope of that - They talk about supporting open codecs, but they still bundle Adobe Flash (which includes H.264 support) with Chrome?
As a result of this mess, content providers are starting to shy away from HTML5 and stick with what "just works" (for the most part) - SmugMug was starting to consider HTML5, but Google's latest decision has them moving back to Flash.
Ordinarily, I'd side against Microsoft in anything, but in this case, they're right.
They tried to do exactly what Google did (release a non-patent-encumbered video codec, VC-1) and as soon as it started becoming used, a couple of patent holders crawled out of the woodwork and asserted patents against VC-1.
I wouldn't be surprised if WebM gets patents asserted against it in 6 months to a year once it becomes widespread.
(A lot of shadier patent holders like to wait for infringement to become widespread - hence the term "submarine patent")
The value of a patent goes up with the difficulty of circumventing it.
If someone asserts now, it's not too late for Google to say, "Whoops, WebM is actually covered with patents, we'll switch to H.264."
If someone asserts after WebM takes off, it is a LOT harder to switch.
Hence the term "submarine patent" - It's out there, the holder knows it is applicable, but they wait for a high-value target. ("High-value" meaning people are using the patent widely.)
Not really. In the past, hardware acceleration of video was done partially, with a few semi-general-purpose blocks.
However, most modern H.264 hardware acceleration is full bitstream decoding - feed it a bitstream and it displays. It's not partitioned into easily "gluable" blocks like it used to be.
"While both have an obvious bias in this, I'd rather trust the word of Microsoft with its armada of IP lawyers than one x264 developer. Especially since Microsoft will take the possible courtroom fight."
See, I just replaced one company with another in your post. Microsoft made the SAME promises as Google did with VC-1 ("It's patent-free! Use it!")
Six months later a bunch of patent holders crawled out of the woodwork and asserted them against Microsoft and there is now a VC-1 patent pool. (Note: Microsoft is NOT the holder of any of these patents. In this case VC-1 remaining patent-free was in Microsoft's business interests.)
Yup, there is very little evidence other than Google's claims that WebM is really patent-free.
There was a VERY good analysis of WebM from one of the x264 developers (admittedly there could be bias there, but my opinion was that it was high on technical content but low on bias.) - http://x264dev.multimedia.cx/archives/377
There was at least one component of WebM that the x264 developer felt WAS patent-encumbered. However there is a potential for a prior art challenge on that one.
The WebM/VP8 spec is apparently AWFUL. Almost as bad as, if not worse than, Microsoft's OOXML spec.
Good question. I know a Sourceforge database glitch or other outage caused the disappearance of files from a (long-defunct) project I briefly contributed to.
Five years later, I was trying to find that code (mainly curiosity) and found a bunch of rants from a guy claiming GPL violations because the project no longer existed.
Note that in this case: 1) The binaries disappeared in addition to the source. This guy was complaining that not being able to find ANY component of the project was a GPL violation. 2) Very early in the project, there WAS a fork of commercial code - the commercial branch forked even before the GPL release. The guy was under the false assumption that the project had been "taken closed" when in reality, the GPL fork just got abandoned. (There was a LOT of duplication of effort on the commercial side to reimplement the few things that had been added to the GPL branch since release.)
Yeah, it IS possible to relicense GPL code, but you need to be able to prove in a court of law that you attempted to contact EVERY single contributor and give them plenty of time to reply.
(As I understand it, there were some Mesa GPL contributors that could not be contacted, but the Mesa developers went through a lot of trouble to try and contact everyone and gave them many months to object.)
In this case, it sounds like there were "no external contributions for 10 years", but even ONE contribution still present in the code from 10 years ago is enough. Based on the description (I can't access TFA from my current location), it sounds like they made no attempt to contact the original mtr authors.
There is absolutely no reason Intel can't supply an open source driver that is properly modularized so that it is as "drop-in" as the NVidia/ATI drivers.
That's pretty much why Windows/Alpha and Windows/PPC were duds.
Yes, desktop Windows (NT specifically) did have native Alpha and PPC variants. There were basically no applications so they were a novelty and Linux pretty much killed both of them.
It is obviously possible for a manufacturer to do an architecture switch or support dual architectures (see Apple), but I really don't see Microsoft pulling off what Apple did.
Actually, the way I read one of the articles - Intel has completely and totally fucked up their driver architecture.
NVidia and ATI can drop a single driver "module" into an existing system and have it work great. No new kernel (just a tiny bit of kernel glue), no new Mesa, no new X.org in nearly all cases.
Meanwhile, Intel is requiring at least FIVE different base operating system components to be changed for their drivers to be updated?
It's just another example of "Intel graphics in Linux sucks" because of all of the horrific interdependencies their driver architecture has. Look at how badly broken Intel GMA support was in Ubuntu 9.x or 10.04 (can't remember which exact Ubuntu release, but it was basically unusable for 3D - Google Earth would stutter massively when it's silky smooth in 8.x or 10.10. Oh, and this was on a GMA950 - not by any means a new chipset.)
It isn't a challenge in general to deliver "launch day" GPU support in Linux - NVidia and ATI have been pulling this off smoothly on a routine basis. It is, however, a challenge for Intel since they seem to have no clue how to make their driver a nice modular drop-in.
Unfortunately, it's pretty hard to packet-inspect 3G data usage of a closed proprietary OS without provider cooperation...
Even if you had one of those AT&T femtocells, you likely would just see traffic between the femtocell and an AT&T server that handled femtocell traffic and then routed it to the real world.
Even more difficult to believe, the claim is they did it in order to spy on another organization within the DOJ.
Legal or not, any truth to this would have ignited a political shitstorm within the DOJ.
Also note: If there were anything possibly illegal about this, the fact that the alleged target organization of the backdooring (EOUSA) is FULL OF LAWYERS, you can bet someone would've been torn a new legal asshole over this.
The original email sort of gave hints to this, referencing side channel/key leaking vulnerabilities. Side channel attacks can be VERY esoteric and difficult to identify - Look at Adi Shamir's work with abusing the Pentium 4's HyperThreading implementation.
However, I believe within the first days of the audit, some of the code contributions from Netsec appeared to, if anything, be an attempt at eliminating a potential timing-based side channel attack.
Honestly, I still can't figure out why Theo even believes that this company might have been contracted. It would be basically impossible for the FBI to contract out an organization to backdoor the FBI's parent organization's communications system as claimed - you'd have to have some VERY creative accounting to be able to hide that from the parent organization, and if there were even the slightest bit of truth to Perry's claims, you can bet there's a massive political shitstorm going on within the DOJ over this.
Please show me the confirmation that there actually HAS been a hole for 10 years, vs. some guy who hasn't been involved with the project making wild and inconsistent claims to spread FUD.
Read his email in more detail - lots of inconsistencies that scream "bullshit" to me.
This whole thing screams classic FUD in the purest form to me.
Which is why (in addition to some various inconsistencies in the email that started this all) I think there is no problem, just a desperate grab for attention by someone no longer relevant (this George guy).
First he claims the EOUSA is the FBI's parent organization - it is not. Both are, to my knowledge, part of the DOJ. Second, he claims the backdoors were put in so that the FBI could monitor the EOUSA's communications - Really, the FBI illicitly monitoring the lawyers it works with?
Also, some of the audits that have kicked off have come up with exact opposite results of what this guy claims - Looking back through the conversation, it looks like at least one of the commits was a defense against side channel attacks, not adding a side channel attack vulnerability.
Slashdot Mirror
SL had the potential to be better than Flash.
But Microsoft is pulling the same shit Adobe is with Flash. Claiming it's "open" but gaming the spec release schedule in such a manner that it is impossible for anything but their implementation to actually work - By the time someone has implemented to the open spec, MS and Adobe have released new versions and content providers are using those new versions.
For example, there was hope that Monolight combined with Netflix transitioning to Silverlight would bring Netflix streaming to Linux. However, Netflix tracks Microsoft's Silverlight releases pretty quickly, meaning that on a consistent basis, by the time Monolight has caught up to wherever Netflix might have been, Netflix has moved on to the next Silverlight release. Same for nearly all other SL content out there.
It's the same sad situation for any of the alternative Flash players - They continue to remain novelties because Adobe never seems to update the spec until they have released the next version of Flash, meaning Gnash et al are always at least a generation behind. Let's not forget the fact that the Flash spec is missing critical stuff such as RTMPE documentation. (The only public RTMPE documentation out there was obtained via reverse engineering Adobe's implementation.)
Problem is, it is VERY difficult to implement content-provider-friendly DRM on Linux.
Content providers still insist on DRMing what they can, despite the fact that many of their delivery methods are known to be fully compromised. (They're dumbasses like that.)
I mean, who cares if Netflix streaming can be ripped when the higher-quality DVD is 100% compromised, and even Blu-Ray is compromised?
It's improving rapidly.
Their selection now is MUCH better than it was six months or so ago. (They have recently made quite a few deals with content providers, those deals are why the bumped up prices a bit.)
For me, the FF series has actually driven console purchases for me. I didn't buy a PS2 until FFXII came out (Worth it IMO! I did not buy many other PS2 games but still don't regret the PS2 purchase), I didn't buy a PS3 until FFXIII came out - NOT worth it! (At least not for FFXIII alone - I still have a great Blu-Ray player/UPnP frontend, and I have picked up a few other PS3 games.)
I'm going to be a lot more careful with future FF releases though.
It could be damage control - some of the characters did have potential, and Square has shown that sequel gameplay can be VASTLY different from the original game. See X-2.
(Note: X-2 was a massive step down from X, but XIII-2 could be a major step up - it would be hard to go down.)
The linear gameplay in XIII was AWFUL. I hope XIII-2 is significantly different. (Good news: the last -2 version of an FF game was vastly different from the game it was a sequel to. Bad news: X-2 was a major step down from X. On the other hand, it'll be hard to go downwards from XIII.)
As to change in team - This is definately a factor. The Final Fantasy team is still capable of putting out some great games (like XII), however the percentage of duds is increasing. It's a combination of "milking the franchise" (although mainline games should get the quality focus, it's more acceptable to have spinoffs like the various VII spinoffs be "meh") and the original Final Fantasy team members taking a less active role and more of an advisory role.
Nobuo Uematsu has been reducing his involvement in many of the FF games over the past few releases, but it is obvious from the quality of FFXIII's soundtrack that he had VERY little involvement with that game. One of the things the Final Fantasy series has always been known for is great music (much due to Uematsu being an amazing composer), however FFXIII's music was so bad that one area had elevator music as the background music!
A good point here - Google has a lot of "green" initiatives (reduced-power computing, huge solar cell farms on their roof, etc.)
This approach is NOT a "green" approach - a "green" approach is one that makes use of the large amount of hardware acceleration infrastructure now deployed for the existing standard codecs.
WebM/VP8 will force a non-accelerated CPU-only rendering path on ALL existing hardware. This eats power compared to hardware acceleration. (Look at how well most Android devices handle H.264 thanks to hardware accelerated decoding.)
Google is being hypocritical and inconsistent here. Great summary at http://daringfireball.net/2011/01/simple_questions - Key here is, HTML5 was supposed to at least partially break Adobe's stranglehold on the web by moving some content away from Flash. Google just killed any hope of that - They talk about supporting open codecs, but they still bundle Adobe Flash (which includes H.264 support) with Chrome?
As a result of this mess, content providers are starting to shy away from HTML5 and stick with what "just works" (for the most part) - SmugMug was starting to consider HTML5, but Google's latest decision has them moving back to Flash.
Ordinarily, I'd side against Microsoft in anything, but in this case, they're right.
They tried to do exactly what Google did (release a non-patent-encumbered video codec, VC-1) and as soon as it started becoming used, a couple of patent holders crawled out of the woodwork and asserted patents against VC-1.
I wouldn't be surprised if WebM gets patents asserted against it in 6 months to a year once it becomes widespread.
(A lot of shadier patent holders like to wait for infringement to become widespread - hence the term "submarine patent")
Ever heard of a "submarine patent"?
The value of a patent goes up with the difficulty of circumventing it.
If someone asserts now, it's not too late for Google to say, "Whoops, WebM is actually covered with patents, we'll switch to H.264."
If someone asserts after WebM takes off, it is a LOT harder to switch.
Hence the term "submarine patent" - It's out there, the holder knows it is applicable, but they wait for a high-value target. ("High-value" meaning people are using the patent widely.)
Not really. In the past, hardware acceleration of video was done partially, with a few semi-general-purpose blocks.
However, most modern H.264 hardware acceleration is full bitstream decoding - feed it a bitstream and it displays. It's not partitioned into easily "gluable" blocks like it used to be.
"While both have an obvious bias in this, I'd rather trust the word of Microsoft with its armada of IP lawyers than one x264 developer. Especially since Microsoft will take the possible courtroom fight."
See, I just replaced one company with another in your post. Microsoft made the SAME promises as Google did with VC-1 ("It's patent-free! Use it!")
Six months later a bunch of patent holders crawled out of the woodwork and asserted them against Microsoft and there is now a VC-1 patent pool. (Note: Microsoft is NOT the holder of any of these patents. In this case VC-1 remaining patent-free was in Microsoft's business interests.)
Yup, there is very little evidence other than Google's claims that WebM is really patent-free.
There was a VERY good analysis of WebM from one of the x264 developers (admittedly there could be bias there, but my opinion was that it was high on technical content but low on bias.) - http://x264dev.multimedia.cx/archives/377
There was at least one component of WebM that the x264 developer felt WAS patent-encumbered. However there is a potential for a prior art challenge on that one.
The WebM/VP8 spec is apparently AWFUL. Almost as bad as, if not worse than, Microsoft's OOXML spec.
Probably a lot of them were "customer service" - Handling offensive content complaints, etc.
Likely minimum wage sort of stuff. Potentially a lot of them were part-timers.
Ow. My stomach hurts from laughing. :(
If I recall correctly, that is part of the EDGE standard as an optional feature - few phones or networks implement it.
It is required for UMTS, so all phones/networks support it when on a UMTS network.
Good question. I know a Sourceforge database glitch or other outage caused the disappearance of files from a (long-defunct) project I briefly contributed to.
Five years later, I was trying to find that code (mainly curiosity) and found a bunch of rants from a guy claiming GPL violations because the project no longer existed.
Note that in this case:
1) The binaries disappeared in addition to the source. This guy was complaining that not being able to find ANY component of the project was a GPL violation.
2) Very early in the project, there WAS a fork of commercial code - the commercial branch forked even before the GPL release. The guy was under the false assumption that the project had been "taken closed" when in reality, the GPL fork just got abandoned. (There was a LOT of duplication of effort on the commercial side to reimplement the few things that had been added to the GPL branch since release.)
Yeah, it IS possible to relicense GPL code, but you need to be able to prove in a court of law that you attempted to contact EVERY single contributor and give them plenty of time to reply.
(As I understand it, there were some Mesa GPL contributors that could not be contacted, but the Mesa developers went through a lot of trouble to try and contact everyone and gave them many months to object.)
In this case, it sounds like there were "no external contributions for 10 years", but even ONE contribution still present in the code from 10 years ago is enough. Based on the description (I can't access TFA from my current location), it sounds like they made no attempt to contact the original mtr authors.
There is absolutely no reason Intel can't supply an open source driver that is properly modularized so that it is as "drop-in" as the NVidia/ATI drivers.
That's pretty much why Windows/Alpha and Windows/PPC were duds.
Yes, desktop Windows (NT specifically) did have native Alpha and PPC variants. There were basically no applications so they were a novelty and Linux pretty much killed both of them.
It is obviously possible for a manufacturer to do an architecture switch or support dual architectures (see Apple), but I really don't see Microsoft pulling off what Apple did.
Actually, the way I read one of the articles - Intel has completely and totally fucked up their driver architecture.
NVidia and ATI can drop a single driver "module" into an existing system and have it work great. No new kernel (just a tiny bit of kernel glue), no new Mesa, no new X.org in nearly all cases.
Meanwhile, Intel is requiring at least FIVE different base operating system components to be changed for their drivers to be updated?
It's just another example of "Intel graphics in Linux sucks" because of all of the horrific interdependencies their driver architecture has. Look at how badly broken Intel GMA support was in Ubuntu 9.x or 10.04 (can't remember which exact Ubuntu release, but it was basically unusable for 3D - Google Earth would stutter massively when it's silky smooth in 8.x or 10.10. Oh, and this was on a GMA950 - not by any means a new chipset.)
It isn't a challenge in general to deliver "launch day" GPU support in Linux - NVidia and ATI have been pulling this off smoothly on a routine basis. It is, however, a challenge for Intel since they seem to have no clue how to make their driver a nice modular drop-in.
Unfortunately, it's pretty hard to packet-inspect 3G data usage of a closed proprietary OS without provider cooperation...
Even if you had one of those AT&T femtocells, you likely would just see traffic between the femtocell and an AT&T server that handled femtocell traffic and then routed it to the real world.
Even more difficult to believe, the claim is they did it in order to spy on another organization within the DOJ.
Legal or not, any truth to this would have ignited a political shitstorm within the DOJ.
Also note: If there were anything possibly illegal about this, the fact that the alleged target organization of the backdooring (EOUSA) is FULL OF LAWYERS, you can bet someone would've been torn a new legal asshole over this.
The original email sort of gave hints to this, referencing side channel/key leaking vulnerabilities. Side channel attacks can be VERY esoteric and difficult to identify - Look at Adi Shamir's work with abusing the Pentium 4's HyperThreading implementation.
However, I believe within the first days of the audit, some of the code contributions from Netsec appeared to, if anything, be an attempt at eliminating a potential timing-based side channel attack.
Honestly, I still can't figure out why Theo even believes that this company might have been contracted. It would be basically impossible for the FBI to contract out an organization to backdoor the FBI's parent organization's communications system as claimed - you'd have to have some VERY creative accounting to be able to hide that from the parent organization, and if there were even the slightest bit of truth to Perry's claims, you can bet there's a massive political shitstorm going on within the DOJ over this.
Please show me the confirmation that there actually HAS been a hole for 10 years, vs. some guy who hasn't been involved with the project making wild and inconsistent claims to spread FUD.
Read his email in more detail - lots of inconsistencies that scream "bullshit" to me.
This whole thing screams classic FUD in the purest form to me.
Which is why (in addition to some various inconsistencies in the email that started this all) I think there is no problem, just a desperate grab for attention by someone no longer relevant (this George guy).
First he claims the EOUSA is the FBI's parent organization - it is not. Both are, to my knowledge, part of the DOJ.
Second, he claims the backdoors were put in so that the FBI could monitor the EOUSA's communications - Really, the FBI illicitly monitoring the lawyers it works with?
Also, some of the audits that have kicked off have come up with exact opposite results of what this guy claims - Looking back through the conversation, it looks like at least one of the commits was a defense against side channel attacks, not adding a side channel attack vulnerability.