Do you know how HARD that is to do? Essentially Dell would have to QUALIFY every permutation of kernel (2.6.17,.18,.19), open source drivers, closed sourced drivers, etc. It's not just figuring out which "drivers work", but also verifying they work through test cycles. The last thing anyone wants is some driver that 'works', but under some stress or certain combination of I/Os causes hangs/stability problems.
Well, I'm not asking them to check *every* kernel. I'm just saying it'd be nice if they pick a fairly recent kernel and test against it. If they can tell me that some piece of hardware has an open-source driver marked "EXPERIMENTAL" in 2.6.19, that's a lot of googling I won't have to do right there.
One of the issues that I feel is holding back Linux support is the perception that vendors have to exhaustively and thoroughly support Linux for their effort to have any value. That's really not the case. Having a wireless card package state that "there is a Linux kernel module introduced in 2.6.17 which supports this device" would be HUGE... I would pick that card over another that I had to bring home and plug in and play around with before knowing if it would work or not. I don't think exhaustive corner-case-checking is necessary, as it completely fails to take into account the enormous advantages of open-source development: bugs in open-source software get fixed remarkably quickly and easily once identified, in my experience.
Seems perfectly reasonable to me! I'd be happy enough if Dell would simply support the hardware without charging me for a Windows license, which is basically the current situation if you order a Dell box and put Linux on it.
It would be *really* nice if Dell would do some basic work to document device-driver compatibility for their systems. So if I was configuring a Linux system online, I'd like to see something like:
Information in bold shows the availability of Linux device drivers for the selected components, based on Linux kernel version 2.6.19, x.org version 1.2.3.4, and CUPS version 5.6.7.8.
Video Card:
NVidia GeForce Yadda Yadda (open-source driver for 2D graphics, closed-source vendor-provided driver for 3D graphics)
Intel Extreme Graphics Foobar (open-source vendor-assisted driver for 2D and 3D graphics)
Broadcom Whatzit (no native Linux drivers, may work with ndiswrapper)
If Dell could do something like this, I'd give them *huge* props... and I imagine a lot of other Linux folks would to. I'd gladly order my next box or ten from them. It wouldn't even be that hard... I would guess that one guy working, say, 10 hours a week on this could easily document driver availability for all the hardware Dell sells with its desktop systems.
Hmm, interesting. Where did you read that? I was under the impression that defense procurement is universally slow, inflexible, rigorous, and paranoid (make a $100 radio cost $5,000 by building it to withstand a nuke).
It'd be interesting to hear about a military that has managed to reduce cost and delay without sacrificing reliability of critical systems.
I don't know about you, but I connect to a mail server using SSL, and the server is not operated by my ISP. Are they going to log some unintelligible bits? Are they going to force people to use their ISP's mail server? Who is an ISP? Anybody who resells bandwidth? How will they know you're reselling bandwidth? Etc...
Bingo. Even if the government gives you bad SSL certs and otherwise attacks and cripples every KNOWN secure protocol, it'll only get them so far.
If that happens, some company will spring up outside the USA that will charge a monthly fee to tunnel your Internet traffic through their servers via SSH. And they'll send you the server's public key fingerprint via postal mail so that you can verify that there's no man-in-the-middle attack. That will be foolproof unless the US govt decides to start opening mail and altering anything that looks like a public key fingerprint or SHA sum or whatever. And then the foreign companies will start broadcasting their public keys via short-wave radio. And then the govt could ban short-wave radios. And then... this is beginning to look like North Korea...
Note that I do not believe any of this will really happen. I do not believe we Americans will accept a totalitarian government. I don't even believe we'll accept small steps in that direction in the long run. I think the proposed policy is destined to fail and is the result of (a) a power-hungry administration (whose time is up in 2 years anyway) and (b) a desire to catch terrorists and (c) an extraordinarily bad understanding of technology.
It's amazing to me how legislators and policy-makers fail to understand crucial points about technology. They believe that DRM can be effective (or, failing that, they make it illegal to break), they blithely ignore the global reach of the Internet, and they don't know how easy it is to use strong encryption. They need to pick and choose their battles differently.
I got a suit from JC Penney a couple years ago... if you spent >$80 you got a free DVD player. It was actually pretty decent, though I gave it away. Right now I have a $20 DVD player which skips and takes forever to eject, though I have seen other $20 DVD players which seem to work great. I know it's basically just the drive, a massively integrated DVD-on-a-chip ASICs, a remote, and a power supply... but I still don't understand how they make 'em so cheap.
Very well said! I suppose that in my more rational moments I too just want Microsoft to compete on a level playing field. Though when they're doing stuff like strong-arming OOXML through standards bodies and adding DRM everywhere, I sometimes wish they'd just go away.:-)
Microsoft had a rethink of the strategy they were using, dropped the prices for the government, threw in some more support for servers and all sorts of other things and all of a sudden, the government has a change of mind and also givs MS some tax breaks.. go figure.
You (and others) often use this as evidence that open source initiatives don't really work. In my mind, that's not really clear. I mean, in your example, competition from open source has in fact forced Microsoft to reduce prices, improve support, and maybe provide better docs. Of course you and I would prefer Microsoft get dumped all at once, but I believe that progress has being made. The fact that government initiatives in favor of open source have forced Microsoft to negotiate better deals DEMONSTRATES that open source has influence and strength.
Prior to the last few elections, red was always used for the democrats. At the time the republicans favored freedom more and the democrats were all about big government, so the association of the democrats with the reds (communists) was fitting.
No, this is not true. The colors used to draw political divide on maps are not meant to symbolize the political leanings of the parties in any way. They were chosen arbitrarily by TV networks, and there was no agreed-on color scheme prior to the year 2000... that's why nobody talked about "red states"/"blue states" until then.
Well... if you live in DC you can see that the White House perimeter has excellent visibility and is crawling with DC metro police, Sec Serv Uniformed Division, and plenty of undercover SS too, probably. The Naval Observatory is a larger, more sprawling facility surrounded by woods.
Not that I think this is any excuse for Google to cave to censoring, mind you. You can see the Naval Observatory uncensored in Yahoo Maps anyway =)
Yeah, it's funny they are still pushing performance/per/watt. I certainly don't hope they are doing that on the desktop space. Consumers don't give a crap about power efficiency of their products. You can't even get people to replace 2, 60 watt light bulbs in their house, why would they care about performance per watt? It only matters for servers.
Well, *I* for one care about power efficiency of my desktop. I've gone over to compact fluorescent bulbs (12 W, 60 W equiv) exclusively. I imagine anyone who cares about their electric bill will sit down and realize that their always-on computers consume several hundred watts (several times that when the AC's on and must remove all that waste heat).
Besides, performance per watt is about much more than the power bill. Higher performance per watt means more performance for less HEAT. And getting rid of heat requires larger desktop cases, with more noisy fans... greater performance-per-watt will allow us to have smaller, quieter cases. On a microscopic level, heat actually LIMITS the attainable clock speeds! So if you want to see higher clock speeds, then you'd better be pushing for better performance-per-watt as well.
Also, there are several classes of applications and problems that cannot be handled well with multiple cores, no matter how much you wish it would. You could have a 3Ghz single core vs. a 2GHz 128 core, and the 3GHz machine may be faster. I think the GHz race will need to continue someday, just not with silicon. Quantum computers here we come!
Hmmm... I can't think of a single REAL WORLD task that is so massively un-parallelizable that I would prefer a single 3 GHz core over many 2 GHz cores. Of course there is software that hasn't yet been optimized for parallelism, but there are few real-world problems that resist it entirely.
I do agree that the GHz race will continue at some point in the future, with silicon even. But for now performance-per-watt is a more significant bottleneck in terms of using computers efficiently. This is trivially obvious in the case of server systems. And even on my desktop I can't think of many tasks that I do which are CPU-bound... except for some heavy matrix crunching with Octave. And that, of course, is easily parallelizable. I'd much rather have a few more cores than a few more GHz.
A production E4300 running at 1.8 GHz can be run at 3 GHz on stock air cooling.
Didn't you get the news? The clock speed race is sooooo 2005. For everything but the most CPU-bound number-crunching applications, increasing clock speed is no longer very desirable.
Today it's all about PERFORMANCE PER WATT (crucial for server farms and portables) and on-chip parallelism/SMP (useful for everything from desktop GUIs to web serving to RTOS embedded systems).
Thank you! You've clarified everything in my original post, much better than I could have.
I guess for the manufacturers of "trusted" chips, there's a tradeoff between uniqueness of keys and ease of finding them:
If the keys are hard-encoded into the mask of the chip, then they will be difficult to locate and recover. But cost considerations (e.g. the pressure to make $20 DVD players as you pointed out) will mean that the number of unique keys is quite small.
If the keys are stored in some mutable memory, then they will be easier to locate thanks to semiconductor processing considerations. The process for making flash memory is different from that for making CMOS logic, so the key will be in an isolated and distinct part of the chip, maybe a separate die. But on the other hand, there could be a unique key for each player.
???? Maybe there's some intermediate case I haven't though of ????
I wonder what they're going to say when it's brutally apparent that ALL software players can be compromised.
In my mind, we're already there:-) The logical next step is to allow only hardware and partial-hardware players. For a PC, this would mean having some kind of "trusted" chip on your motherboard which can encrypt and decrypt data using keys that are hard-wired in.
Of course, hardware solutions can be broken too. I can envision a couple of ways this will happen:
If the keys are truly embedded in the "trusted" ASIC: Making custom chips is expensive. There are substantial setup costs for each new mask, so there will be enormous economic pressure to only have one or a few versions of the chip. This means once one version gets cracked, millions of computers will be freed. What will it take to read the keys off an ASIC? A scanning electron microscope, that's what. As a bored physics grad student currently sitting 10 feet away from an SEM, I can tell you it'll happen:-)
If the keys are somehow individualized to each computer, they'll be stored on a flash-based FPGA, or in some kind of microcontroller's flash memory. Manufacturers of such flash-based devices go to great lengths to make it so that the code stored in flash can't be read off of the device, but this is nothing more than the same ol, same ol security through obscurity... figure out the magic voltage that you need to apply to pin 12, and oops there goes the security. Smart card hackers have already figured out ways around the protection in the common PIC16C84 microcontroller.
Bottom line: DRM is futile because it requires the distribution of a SECRET PIECE OF DATA (the decryption keys) in UNENCRYPTED form (the keys themselves must of necessity be unencrypted). All the crap interposed between the user and the keys is merely security through obscurity. QED.
Currently I'm taking a brand new course in the U of Maryland physics department, on the topic of building digital instruments with FPGA (reconfigurable integrated circuits). On the syllabus handed out at the first lecture, the professor listed "Wikipedia" as the textbook for the class, along with random things he'd hand out.
Basically, it's such a brand spanking new topic that there's no textbook material on FPGAs. And the professor is a big fan of Wikipedia. I'm thinking it will be a pretty cool class:-)
Yep, I've recently discovered numpy and matplotlib. Much more flexible and powerful than MATLAB/Octave. Though I still prefer Octave for quick-and-dirty stuff, personally.
There's an excellent open-source MATLAB clone called Octave. I've used it for a lot of real-world physics work in my lab. Worth checking out before you shell out for MATLAB.
I figure I could get say, $100,000 consulting fee out of it, right? And it'd be well worth it to any company planning to introduce a really retarded DRM scheme.
Remember SunComm, which saw its stock price fall by $10million when someone figured out they could bypass its DRM by holding down the shift key? http://news.com.com/2100-1025-5089168.html If only they'd hired a geek to give it the idiot test...
Who are the industry people who BELIEVE the crap spouted by Sony/Macrovision/whoever has created the latest DRM scheme??? Considering the terrible track record of DRM of all kinds (basically every scheme ever introduced has been broken), it's amazing that anyone makes business decisions based on it.
Can't they just hire a geek or two to give them the honest odds on how long a DRM scheme will last before being cracked? I could use the job.
... or considering whether I'd rather have Howard Stern or Oprah, because there is no practical way to get both. Frankly, it's probably all this exclusivity that has caused me not to purchase either system."
Frankly, it's the idea of giving any of my money to either Howard Stern or Oprah that has held me back from getting satellite radio service.
I agree completely. Ultimately, SEO spam on wikipedia is the fault of the search engines, not Wikipedia. Adding "nofollow" to all the external links on Wikipedia is fixing the wrong end of the problem.
The reality is that SEO spam on wikipedia won't go away until *the search engines* figure out a way to stop rewarding it. This "nofollow" hack is just a stopgap measure...
Does anyone have any intelligent theory about HOW the pump-and-dump spammers pick specific companies to target? I mean, there are probably a lot of low-value low-volume penny stocks out there... why do the spammers seem to mostly pump the same 50 companies over and over again (the author of that site only saw 50 or so companies pumped in 4300 emails).
I have a gut feeling that there's some collusion with insiders in the pumped companies in many of these schemes. I have no evidence for it though. Any thoughts?
Yeah, it's not clear to me that they discontinue using a lot of hardware.
Just about 1 year ago, I heard from someone in the know that they were still running tons of Pentium Pro servers, simply because they delivered a slight edge in terms of performance-per-watt over servers with more modern processors. I imagine that surprising anachronism is largely due to the fact that a lot of what they do is I/O-bound rather than CPU-bound. Don't know if they're still using those...
I would guess that Google *does* sometimes throw out hardware en masse, simply because it's too much of a pain to maintain extremely heterogeneous hardware configurations.
Slashdot Mirror
Well, I'm not asking them to check *every* kernel. I'm just saying it'd be nice if they pick a fairly recent kernel and test against it. If they can tell me that some piece of hardware has an open-source driver marked "EXPERIMENTAL" in 2.6.19, that's a lot of googling I won't have to do right there.
One of the issues that I feel is holding back Linux support is the perception that vendors have to exhaustively and thoroughly support Linux for their effort to have any value. That's really not the case. Having a wireless card package state that "there is a Linux kernel module introduced in 2.6.17 which supports this device" would be HUGE... I would pick that card over another that I had to bring home and plug in and play around with before knowing if it would work or not. I don't think exhaustive corner-case-checking is necessary, as it completely fails to take into account the enormous advantages of open-source development: bugs in open-source software get fixed remarkably quickly and easily once identified, in my experience.
It would be *really* nice if Dell would do some basic work to document device-driver compatibility for their systems. So if I was configuring a Linux system online, I'd like to see something like:
If Dell could do something like this, I'd give them *huge* props... and I imagine a lot of other Linux folks would to. I'd gladly order my next box or ten from them. It wouldn't even be that hard... I would guess that one guy working, say, 10 hours a week on this could easily document driver availability for all the hardware Dell sells with its desktop systems.
How about a levy on wood? You can build a closet out of wood, and you can store a whole lot of pirated music in a closet!
Hmm, interesting. Where did you read that? I was under the impression that defense procurement is universally slow, inflexible, rigorous, and paranoid (make a $100 radio cost $5,000 by building it to withstand a nuke).
It'd be interesting to hear about a military that has managed to reduce cost and delay without sacrificing reliability of critical systems.
I loathe Microsoft and dislike Apple. I guess I should consider this a feature, not a bug!
If that happens, some company will spring up outside the USA that will charge a monthly fee to tunnel your Internet traffic through their servers via SSH. And they'll send you the server's public key fingerprint via postal mail so that you can verify that there's no man-in-the-middle attack. That will be foolproof unless the US govt decides to start opening mail and altering anything that looks like a public key fingerprint or SHA sum or whatever. And then the foreign companies will start broadcasting their public keys via short-wave radio. And then the govt could ban short-wave radios. And then... this is beginning to look like North Korea...
Note that I do not believe any of this will really happen. I do not believe we Americans will accept a totalitarian government. I don't even believe we'll accept small steps in that direction in the long run. I think the proposed policy is destined to fail and is the result of (a) a power-hungry administration (whose time is up in 2 years anyway) and (b) a desire to catch terrorists and (c) an extraordinarily bad understanding of technology.
It's amazing to me how legislators and policy-makers fail to understand crucial points about technology. They believe that DRM can be effective (or, failing that, they make it illegal to break), they blithely ignore the global reach of the Internet, and they don't know how easy it is to use strong encryption. They need to pick and choose their battles differently.
I got a suit from JC Penney a couple years ago... if you spent >$80 you got a free DVD player. It was actually pretty decent, though I gave it away. Right now I have a $20 DVD player which skips and takes forever to eject, though I have seen other $20 DVD players which seem to work great. I know it's basically just the drive, a massively integrated DVD-on-a-chip ASICs, a remote, and a power supply... but I still don't understand how they make 'em so cheap.
Very well said! I suppose that in my more rational moments I too just want Microsoft to compete on a level playing field. Though when they're doing stuff like strong-arming OOXML through standards bodies and adding DRM everywhere, I sometimes wish they'd just go away. :-)
For the convoluted history of the red/blue color scheme, read this: http://en.wikipedia.org/wiki/Red_state_vs._blue_s
Well... if you live in DC you can see that the White House perimeter has excellent visibility and is crawling with DC metro police, Sec Serv Uniformed Division, and plenty of undercover SS too, probably. The Naval Observatory is a larger, more sprawling facility surrounded by woods.
Not that I think this is any excuse for Google to cave to censoring, mind you. You can see the Naval Observatory uncensored in Yahoo Maps anyway =)
Besides, performance per watt is about much more than the power bill. Higher performance per watt means more performance for less HEAT. And getting rid of heat requires larger desktop cases, with more noisy fans... greater performance-per-watt will allow us to have smaller, quieter cases. On a microscopic level, heat actually LIMITS the attainable clock speeds! So if you want to see higher clock speeds, then you'd better be pushing for better performance-per-watt as well. Hmmm... I can't think of a single REAL WORLD task that is so massively un-parallelizable that I would prefer a single 3 GHz core over many 2 GHz cores. Of course there is software that hasn't yet been optimized for parallelism, but there are few real-world problems that resist it entirely.
I do agree that the GHz race will continue at some point in the future, with silicon even. But for now performance-per-watt is a more significant bottleneck in terms of using computers efficiently. This is trivially obvious in the case of server systems. And even on my desktop I can't think of many tasks that I do which are CPU-bound... except for some heavy matrix crunching with Octave. And that, of course, is easily parallelizable. I'd much rather have a few more cores than a few more GHz.
Today it's all about PERFORMANCE PER WATT (crucial for server farms and portables) and on-chip parallelism/SMP (useful for everything from desktop GUIs to web serving to RTOS embedded systems).
I guess for the manufacturers of "trusted" chips, there's a tradeoff between uniqueness of keys and ease of finding them:
Of course, hardware solutions can be broken too. I can envision a couple of ways this will happen:
Bottom line: DRM is futile because it requires the distribution of a SECRET PIECE OF DATA (the decryption keys) in UNENCRYPTED form (the keys themselves must of necessity be unencrypted). All the crap interposed between the user and the keys is merely security through obscurity. QED.
Currently I'm taking a brand new course in the U of Maryland physics department, on the topic of building digital instruments with FPGA (reconfigurable integrated circuits). On the syllabus handed out at the first lecture, the professor listed "Wikipedia" as the textbook for the class, along with random things he'd hand out.
:-)
Basically, it's such a brand spanking new topic that there's no textbook material on FPGAs. And the professor is a big fan of Wikipedia. I'm thinking it will be a pretty cool class
Yep, I've recently discovered numpy and matplotlib. Much more flexible and powerful than MATLAB/Octave. Though I still prefer Octave for quick-and-dirty stuff, personally.
There's an excellent open-source MATLAB clone called Octave. I've used it for a lot of real-world physics work in my lab. Worth checking out before you shell out for MATLAB.
I figure I could get say, $100,000 consulting fee out of it, right? And it'd be well worth it to any company planning to introduce a really retarded DRM scheme.
Remember SunComm, which saw its stock price fall by $10million when someone figured out they could bypass its DRM by holding down the shift key? http://news.com.com/2100-1025-5089168.html If only they'd hired a geek to give it the idiot test...
Who are the industry people who BELIEVE the crap spouted by Sony/Macrovision/whoever has created the latest DRM scheme??? Considering the terrible track record of DRM of all kinds (basically every scheme ever introduced has been broken), it's amazing that anyone makes business decisions based on it.
Can't they just hire a geek or two to give them the honest odds on how long a DRM scheme will last before being cracked? I could use the job.
Frankly, it's the idea of giving any of my money to either Howard Stern or Oprah that has held me back from getting satellite radio service.
I agree completely. Ultimately, SEO spam on wikipedia is the fault of the search engines, not Wikipedia. Adding "nofollow" to all the external links on Wikipedia is fixing the wrong end of the problem.
The reality is that SEO spam on wikipedia won't go away until *the search engines* figure out a way to stop rewarding it. This "nofollow" hack is just a stopgap measure...
This is the first coherent explanation I've ever read of how the pump-and-dump scammers actually do it and get the money out.
If this is really the case, seems like the way to fight it is essentially by fighting phishing and identity theft.
Does anyone have any intelligent theory about HOW the pump-and-dump spammers pick specific companies to target? I mean, there are probably a lot of low-value low-volume penny stocks out there... why do the spammers seem to mostly pump the same 50 companies over and over again (the author of that site only saw 50 or so companies pumped in 4300 emails).
I have a gut feeling that there's some collusion with insiders in the pumped companies in many of these schemes. I have no evidence for it though. Any thoughts?
Yeah, it's not clear to me that they discontinue using a lot of hardware.
Just about 1 year ago, I heard from someone in the know that they were still running tons of Pentium Pro servers, simply because they delivered a slight edge in terms of performance-per-watt over servers with more modern processors. I imagine that surprising anachronism is largely due to the fact that a lot of what they do is I/O-bound rather than CPU-bound. Don't know if they're still using those...
I would guess that Google *does* sometimes throw out hardware en masse, simply because it's too much of a pain to maintain extremely heterogeneous hardware configurations.