I hope the talented developers will favor [profitable] platforms over [unprofitable] ones, help create and improve [profitable] platforms, and help making the world more [profitable].
There, fixed that for ya. Really, when push comes to shove, developers want their proverbial bread on the table as much as anyone else. If openness coincides sufficiently well with developer self-interest, then openness may win out as well. If it doesn't, then there's not much hope for it; ignoring economic incentives (or disincentives) doesn't make them go away.
The IPod may have made Apple plenty of money, but the concept isn't revolutionary- its evolutionary.
The patentability of any particular innovation is a nuanced matter, but a blanket assessment that any product is "not patent-worthy" because it "isn't revolutionary- [it's] evolutionary" is utterly inane.
Here's a perspective: The iPod's design was the first digital music player that allowed quick and easy navigation of a large library. A collection of well-thought out design innovations made the iPod and its successors the smash hits they've been. Sure, Apple's had its marketing machine at work. But as Apple's varied market failures have well proven, even they can't sell a lemon.
By comparison, the contemporary players at the launch of the first iPod largely sucked. Many had UI so bad that you'd have had a hard time finding any of the music whether a few meg of flash or 20GB of music on a lurching laptop-sized drive. Others, the relatively successful ones, simply paled in comparison to the iPods relative simplicity and ease of use. This is the revolution that the iPod has ridden: that the user experience should kick ass, not just be a bunch of marketing bullet-points.
Anyone who thinks that stealing such things once laid off is a bright idea just does not have a criminal mind.
Actually, such a person DOES have a criminal mind: criminals are that because they make poor risk/reward assessments. This is part of what makes a lot of criminals seem fairly dumb, because poor judgement of risk is what led them to their crime in the first place. The myth/legend of the mastermind criminal falls down here (i.e. becomes unlikely) as brighter individuals tend to find lower risk ways to succeed in life.
Actually, the decoders don't often admit much quality optimization. Modern lossy codec formats for both audio and video typically allow for considerable implementation leeway (and computational expense) in the encoder, while the decoder's work is fairly cut-and-dried (and designed to be efficient). Consider that the encoder's job is to pick the reduced set of bits that best represents the original signal (within the format spec), but the decoder just has to handle reconstituting exactly one narrowly defined format.
This front-loading of the work has two benefits: one as seen here, where better encoders can come along and provide benefit to all decoders. The other is an efficiency concern: the media will be encoded just once, but decoded many times.
For those that are new here(tm), it wasn't that long ago that just decoding audio on a desktop computer or workstation was a fairly taxing operation. This set off this deliberate encoder/decoder work imbalance, but we continue to benefit wherever power draw is a concern.
I am under the belief that somewhere out there, black-hat organizations have some really scary databases of exploits that have never been reported to vendors.
No need for abstract belief; this is near certainty. Even "better", I've seen stuff that would curl your teeth that the vendor apparently knew about but remained quietly unpatched. That was in the toolset of a professional IT security testing company. Their stuff made Metasploit look like a Lego model of a battleship vs. the real thing. It's sobering knowing that tools exist that are the direct realization of the weakest link principle. With really well-thought out and easy to use UI, and backend code just as nice. Click, ownage, click, ownage.../shudder
Sounds like your program might not have gotten the curriculum sorted out yet. Brown and other programs had really clearly thought out the order of introduction for basic OO concepts (which came absolutely first) with control flow and basic algorithms, including making all of these things concrete through hands-on work. These schools were having immense success with their students using this approach.
You're right in that just mashing OO into the curriculum without a real rethink of the whole first year program is problematic at best. This is not what the top CS-edu schools were doing. Sadly, much of the 90's in many CS first-year programs was a nightmare of pseudo-OO concepts without solid pedagogy taught in C++. This, when in years prior CS departments shied away from C in the first year because its warts detracted from the essentials of algorithms and data structures. I spent a lot of time trying to un-break students back then...:-(
In fact, most girls that are into nerdstuff are quite resistant to all those jokes.
You are Legend..-arily clueless. Are you so blind that you don't realize this belief is self-fulfilling? I've known quite a number of women that really just don't want to put up with this disrespectful boys' locker-room crap... and they find other things to do. Spelling it out: insensitive bozos like you keep repelling bright creative minds from all manner of disciplines. STOP IT!
Speaking only to the Ivy schools listed, Brown is known for having a consistently outstanding CS program. Their undergraduate CS education in particular is reknowned. They were one of the pioneers in creating a first year program that taught using OO design from the ground up, and were great at engaging the students with interesting problems. I haven't tracked the evolution of their undergrad CS program for years, but I gather via the grapevine that it's still quite strong.
[Ob. Disclaimer: I have no personal association with Brown. I studied their CS education model back in the day, when I was in academia.]
Non free software does not get the love and attention it deserves, so it's almost always buggier an and more bloated than free alternatives.
This has got to be the most ignorant generalization I've read all week. Not to mention the "backdoors and spyware and TIGERS, OH MY!" FUD routine. Really, I'm all for successful free software... but spouting off completely unrealistic nonsense about the state of the software world helps no one.
Sure, there's a nice roster of free software that's wonderful; where non-free competitors are either non-existent, laughable, or (delightfully) moot. On the other hand, there's a LOT of non-free software that has no equivalent in terms of functionality, quality, or both in the free software world. Really, just look at free vs. non-free games, pro media authoring (audio, video, or still imaging), or scientific software markets for a plethora of examples. Yes, free software makes forays into all of these arenas, but major success stories are very thin on the ground. Some software simply requires massive domain expertise, software expertise, not to mention capital, to produce. In these areas, free software efforts to date have failed to muster the required resources to meet these needs. The open question remains: can and will this change?
Here's a link to the Hasselblad press release... which strangely doesn't actually appear on their own site (H's press releases page doesn't appear to have been updated since 2006). Third paragraph is the most relevant to your question.
How about a drive that advertises longevity instead of storage density. Seriously, I'd take half that storage if there was more assurance of my data integrity.
Nothing any storage manufacturer will EVER do will beat buying three drives and running regular backups of the primary drive to the spares. Remember to always keep one of the spares off-site. There's nothing like the feeling of non-panic when a drive buys the farm and you've got a recent (bootable!) backup. With cheap drives and external enclosures, keeping robust and up-to-date backups has never been easier.
How on earth do zero seek time and no moving parts not qualify under "significant and intrinsic advantage"? Zero seek time alone represents a sea change for mass storage -- access many orders of magnitude slower than the rest of the system has been a major assumption encoded in much software (and hardware) architecture for decades. We'll be feeling the repercussions of the end of rotating media for decades more. Yes, the price needs to come down for SSD's to annihilate traditional hard drives... but SSD's will steadily eat up HDD territory in the mean time.
Bing! Right on the heels of Hasselblad announcing their new H3DII-50 camera (to be released in October) which presumably uses this sensor. Hasselblad has also announced a future 645 format sensor (roughly 56mm x 45mm), more details to be revealed at Photokina 2008 (major bi-annual worldwide photography trade show) later this year.
By the very principals governing the organization of open source projects [...] So if I understand you correctly, you assert that freedom of speech is not a principle of open source projects. How very ironic.
Apologies for the reply-to-self, but I forgot to mention another important special-case use of SetDllDirectory(): if passed the empty string, it removes the current directory from the DLL search path. So even if your app doesn't require some particular directory to be 'blessed', this call can still mitigate a variety of DLL-related risks.
Microsoft's library path ALWAYS goes through the current directory. This isn't true. If an application calls SetDllDirectory(), it overrides the search path. The order becomes:
The directory the application loaded from.
The directory specified as a parameter to SetDllDirectory()
system directories (order elided)
PATH directories (Worthy of an if-I-had-a-time-machine-shooting.)
The above function is highly useful to force a single specified path (it takes ONE path, not a list) early into the DLL load search order. It's pretty much necessary to use this to avoid getting screwed by some fs!%!ing asshat installing a common third-party library into WINDOWS\System32.
If you're stuck wrestling the Windows alligator, this should pretty much always be used instead of trying to play stupid PATH games. But from other posters reports, it seems that MS was too stupid to use one of the very functions they created to tame various DLL hell problems.
In every organization I've worked in, developers have to manage source. Their own source, third-party source (whether OSS, vendor code, drops from another team, etc.) I.e. their work involves more than just blindly pounding out code. The new tools really do reduce the amount of time spent screwing around with the SCM system instead of doing other things. Even really basic stuff like just having more intelligent history-based merging algorithms can save staggering amounts of time.
You first need to decide whether you want a centralized model, like SVN, or a distributed model (git / mercurial / bzr?). Only in that the new tools will support BOTH decentralized and centralized usage, while the old tools only support centralized workflows. Individual SCMs have varying support for some specialized kinds of centralized use cases (e.g. you may need to deploy an extra tool to guard commits to the central repository through integration build/test machines, etc.), but generally work fine and with the added advantages the new tools provide.
You have to vote both these parties out if you want to get rid of this stuff. THIS CAN'T WORK. The way the political systems in this country run right now, this is tantamount to asking all the air molecules to "just move to the left a bit" to give you a nice breeze.
Third parties face huge barriers to entry due to a collection of factors. The first cut: a lot of intelligent folks simply stay away from politics in the first place due to these and other issues. The second cut: those interested in pursuing all but the most minor political positions really must be career-minded the way things work today. To have a viable career, this pretty much means running with one of the two major parties. (Head of snake, meet tail.) The third cut: unfortunately, many of the folks left after the first two cuts seem to be wholly unelectable. As in, if you read their platforms in detail, you realize that they're fscking nuts. At best, they are well-meaning but lack requisite insight into human nature and/or the real-world ramifications of their lofty ideas. At worst, they're really nuts: spouting off about eliminating UFO influence on our toaster ovens and the like. (My state's voter pre-election voter pamphlets occasionally offer excellent comic-relief while researching candidate backgrounds.)
The more interesting question is why candidates in the "second cut" above don't rally around a third party. Part of the reason is simple: virtually none of the USA voting districts use a ranking-based system of election, such as IRV or Condorcet. This provides a barrier to entry most notably seen in the 2004 US Presidential election as the "Nader effect". Nader was never a viable candidate to win, so many voters felt they were forced to choose between "voting their conscience" and "voting for the lesser evil." The real effects are much deeper, however. An excellent third-party or independent candidate might win an election in such a system by garnering a lot of first and second place votes from voters across the spectrum. The effect could be rather de-polarizing, and would allow a foot in the door for new parties.
Now all of this neglects other serious issues, such as campaign funding and media influence and coverage. Third-parties have an additional barrier in the form of achieving sufficient fund-raising to win a campaign, and achieving media backing. As sad as it sounds, it is absolutely necessary today to get the word out and successfully market a candidate to the people in order to win a contested election. This can require large to massive amounts of money... and the strings that go along with that.
Media outlets get to further warp the funding/marketing issues by providing whatever balance and bias of coverage they want. Whether through carelessness, explicit bias, or even implicit biases, mass-media has come to have an astonishing effect on distorting our democratic processes. Not garnering media support can leave a campaign dead in the water. As a simple example, consider the viability of a candidate for any high-level office with a platform of serious media ownership reform. I have trouble imagining that getting very far.
I never got the (recent?) craze over using the latest SCM of the week myself. For those of us who really understand SCM systems, and are heavy users of same, the desire is to get a handle on the capabilities of the new generation systems. The reason is simple: CVS utterly sucks, and SVN wasn't a revolutionary improvement. Simple high-level tasks in a real working environment (e.g. tracking upstream branches against local changes) are frequently a time-consuming pain. Many collaboration use cases (e.g. developer/feature branches) are similarly a hassle to manage. SVN at least gave folks atomic commits, which was a huge step.
My guess is that SVN will turn out to be too little, too late with its merge tracking support. It'll be a boost for folks already using SVN who don't want to switch toolchains, but it's pretty easy to move from SVN to the new tools (beyond export, several newer SCMs have two-way commit support with SVN).
Generationally speaking, it feels like SVN is still trying to catch up to Perforce... but that ship has sailed. The teams working on the new round of decentralized SCMs[*] have done deep rethinking of source control problems and challenges, and the results are generally brilliant. These problems aren't esoteric -- administration and day-to-day usage really is easier with the new stuff. After a while using git, Bazaar, etc., the crufty old SCM tools seem like doing image editing in a hex editor instead of a GUI app.
[*] Includes: Bazaar, Darcs, git, and Mercurial (hg)
As a baseline, the best chair is the one that fits, and is adjusted properly. I've had nice, nice chairs before that simply didn't work for me very well due to sizing and/or adjustment issues. Chairs with more/better adjustments are all fine and well, but can produce a worse sitting experience than a "lesser" chair that is closer to what its user needs in the first place.
So take the specific recommendations here as good starting points, but make sure that your father has some quality time (i.e. more than five minutes) to try to sit in the chair in question and see if it's to his liking. Ask the staff to give a tutorial and recommendations on all adjustments, and go elsewhere if they aren't knowledgeable enough to do this. Have him bring a book, laptop, etc. to really get into working posture. Also, some places will let you try a demo chair for a period (usually for the nicer chair lines) -- if so, take 'em up on it. The last thing you want is to spend money on a good chair just to discover that the honeymoon ended on the showroom floor.
[Movie-based games] are the games most likely to suck! Yeah, a lot of movie-spawned games are pretty banal. On the other hand, one of my favorite console games was the PS2 Ghost In The Shell game -- jumping and skimming around as fast as you can in a fuchikoma. Sweet! The whole game felt well thought-out, from basic controls and game play, to level design, and niceties such as being able to replay any cut-scene you've already seen in the game(*). Even the practice mode was essentially a great little sub-game.
(*) It boggles the mind that publishers ever don't provide this feature.
My math is just fine, you need to revisit your basic English, logic, and social skills. I was trying to debunk the absurd prices cited by the parent. My "punch line" was to observe that back-of-the-napkin calculations using conservative bounds yielded (as you observe and agree with) completely unrealistic results.
Those prices are insanely high. I'm seeing current retail prices for high speed compact flash cards at 60 to 100 USD for 8GByte (varies with speed, rebates, etc.). Taking one of the lower-end prices for a top-tier part (since I want an upper bound on part manufacturing costs), that works out to a bit less than 0.001 USD per Mbit. Even if those numbers are per Gbit, that still leaves the cited current manufacturing cost at more than twice actual retail.
Slashdot Mirror
I hope the talented developers will favor [profitable] platforms over [unprofitable] ones, help create and improve [profitable] platforms, and help making the world more [profitable].
There, fixed that for ya. Really, when push comes to shove, developers want their proverbial bread on the table as much as anyone else. If openness coincides sufficiently well with developer self-interest, then openness may win out as well. If it doesn't, then there's not much hope for it; ignoring economic incentives (or disincentives) doesn't make them go away.
The IPod may have made Apple plenty of money, but the concept isn't revolutionary- its evolutionary.
The patentability of any particular innovation is a nuanced matter, but a blanket assessment that any product is "not patent-worthy" because it "isn't revolutionary- [it's] evolutionary" is utterly inane.
Here's a perspective: The iPod's design was the first digital music player that allowed quick and easy navigation of a large library. A collection of well-thought out design innovations made the iPod and its successors the smash hits they've been. Sure, Apple's had its marketing machine at work. But as Apple's varied market failures have well proven, even they can't sell a lemon.
By comparison, the contemporary players at the launch of the first iPod largely sucked. Many had UI so bad that you'd have had a hard time finding any of the music whether a few meg of flash or 20GB of music on a lurching laptop-sized drive. Others, the relatively successful ones, simply paled in comparison to the iPods relative simplicity and ease of use. This is the revolution that the iPod has ridden: that the user experience should kick ass, not just be a bunch of marketing bullet-points.
Anyone who thinks that stealing such things once laid off is a bright idea just does not have a criminal mind.
Actually, such a person DOES have a criminal mind: criminals are that because they make poor risk/reward assessments. This is part of what makes a lot of criminals seem fairly dumb, because poor judgement of risk is what led them to their crime in the first place. The myth/legend of the mastermind criminal falls down here (i.e. becomes unlikely) as brighter individuals tend to find lower risk ways to succeed in life.
Actually, the decoders don't often admit much quality optimization. Modern lossy codec formats for both audio and video typically allow for considerable implementation leeway (and computational expense) in the encoder, while the decoder's work is fairly cut-and-dried (and designed to be efficient). Consider that the encoder's job is to pick the reduced set of bits that best represents the original signal (within the format spec), but the decoder just has to handle reconstituting exactly one narrowly defined format.
This front-loading of the work has two benefits: one as seen here, where better encoders can come along and provide benefit to all decoders. The other is an efficiency concern: the media will be encoded just once, but decoded many times.
For those that are new here(tm), it wasn't that long ago that just decoding audio on a desktop computer or workstation was a fairly taxing operation. This set off this deliberate encoder/decoder work imbalance, but we continue to benefit wherever power draw is a concern.
I am under the belief that somewhere out there, black-hat organizations have some really scary databases of exploits that have never been reported to vendors.
No need for abstract belief; this is near certainty. Even "better", I've seen stuff that would curl your teeth that the vendor apparently knew about but remained quietly unpatched. That was in the toolset of a professional IT security testing company. Their stuff made Metasploit look like a Lego model of a battleship vs. the real thing. It's sobering knowing that tools exist that are the direct realization of the weakest link principle. With really well-thought out and easy to use UI, and backend code just as nice. Click, ownage, click, ownage... /shudder
Sounds like your program might not have gotten the curriculum sorted out yet. Brown and other programs had really clearly thought out the order of introduction for basic OO concepts (which came absolutely first) with control flow and basic algorithms, including making all of these things concrete through hands-on work. These schools were having immense success with their students using this approach.
You're right in that just mashing OO into the curriculum without a real rethink of the whole first year program is problematic at best. This is not what the top CS-edu schools were doing. Sadly, much of the 90's in many CS first-year programs was a nightmare of pseudo-OO concepts without solid pedagogy taught in C++. This, when in years prior CS departments shied away from C in the first year because its warts detracted from the essentials of algorithms and data structures. I spent a lot of time trying to un-break students back then... :-(
In fact, most girls that are into nerdstuff are quite resistant to all those jokes.
You are Legend..-arily clueless. Are you so blind that you don't realize this belief is self-fulfilling? I've known quite a number of women that really just don't want to put up with this disrespectful boys' locker-room crap... and they find other things to do. Spelling it out: insensitive bozos like you keep repelling bright creative minds from all manner of disciplines. STOP IT!
Speaking only to the Ivy schools listed, Brown is known for having a consistently outstanding CS program. Their undergraduate CS education in particular is reknowned. They were one of the pioneers in creating a first year program that taught using OO design from the ground up, and were great at engaging the students with interesting problems. I haven't tracked the evolution of their undergrad CS program for years, but I gather via the grapevine that it's still quite strong.
[Ob. Disclaimer: I have no personal association with Brown. I studied their CS education model back in the day, when I was in academia.]
Non free software does not get the love and attention it deserves, so it's almost always buggier an and more bloated than free alternatives.
This has got to be the most ignorant generalization I've read all week. Not to mention the "backdoors and spyware and TIGERS, OH MY!" FUD routine. Really, I'm all for successful free software... but spouting off completely unrealistic nonsense about the state of the software world helps no one.
Sure, there's a nice roster of free software that's wonderful; where non-free competitors are either non-existent, laughable, or (delightfully) moot. On the other hand, there's a LOT of non-free software that has no equivalent in terms of functionality, quality, or both in the free software world. Really, just look at free vs. non-free games, pro media authoring (audio, video, or still imaging), or scientific software markets for a plethora of examples. Yes, free software makes forays into all of these arenas, but major success stories are very thin on the ground. Some software simply requires massive domain expertise, software expertise, not to mention capital, to produce. In these areas, free software efforts to date have failed to muster the required resources to meet these needs. The open question remains: can and will this change?
[Sorry for the late response...]
Here's a link to the Hasselblad press release... which strangely doesn't actually appear on their own site (H's press releases page doesn't appear to have been updated since 2006). Third paragraph is the most relevant to your question.
http://dpnow.com/5010.html
How about a drive that advertises longevity instead of storage density. Seriously, I'd take half that storage if there was more assurance of my data integrity.
Nothing any storage manufacturer will EVER do will beat buying three drives and running regular backups of the primary drive to the spares. Remember to always keep one of the spares off-site. There's nothing like the feeling of non-panic when a drive buys the farm and you've got a recent (bootable!) backup. With cheap drives and external enclosures, keeping robust and up-to-date backups has never been easier.
How on earth do zero seek time and no moving parts not qualify under "significant and intrinsic advantage"? Zero seek time alone represents a sea change for mass storage -- access many orders of magnitude slower than the rest of the system has been a major assumption encoded in much software (and hardware) architecture for decades. We'll be feeling the repercussions of the end of rotating media for decades more. Yes, the price needs to come down for SSD's to annihilate traditional hard drives... but SSD's will steadily eat up HDD territory in the mean time.
Bing! Right on the heels of Hasselblad announcing their new H3DII-50 camera (to be released in October) which presumably uses this sensor. Hasselblad has also announced a future 645 format sensor (roughly 56mm x 45mm), more details to be revealed at Photokina 2008 (major bi-annual worldwide photography trade show) later this year.
Apologies for the reply-to-self, but I forgot to mention another important special-case use of SetDllDirectory(): if passed the empty string, it removes the current directory from the DLL search path. So even if your app doesn't require some particular directory to be 'blessed', this call can still mitigate a variety of DLL-related risks.
The above function is highly useful to force a single specified path (it takes ONE path, not a list) early into the DLL load search order. It's pretty much necessary to use this to avoid getting screwed by some fs!%!ing asshat installing a common third-party library into WINDOWS\System32.
If you're stuck wrestling the Windows alligator, this should pretty much always be used instead of trying to play stupid PATH games. But from other posters reports, it seems that MS was too stupid to use one of the very functions they created to tame various DLL hell problems.
In every organization I've worked in, developers have to manage source. Their own source, third-party source (whether OSS, vendor code, drops from another team, etc.) I.e. their work involves more than just blindly pounding out code. The new tools really do reduce the amount of time spent screwing around with the SCM system instead of doing other things. Even really basic stuff like just having more intelligent history-based merging algorithms can save staggering amounts of time.
Third parties face huge barriers to entry due to a collection of factors. The first cut: a lot of intelligent folks simply stay away from politics in the first place due to these and other issues. The second cut: those interested in pursuing all but the most minor political positions really must be career-minded the way things work today. To have a viable career, this pretty much means running with one of the two major parties. (Head of snake, meet tail.) The third cut: unfortunately, many of the folks left after the first two cuts seem to be wholly unelectable. As in, if you read their platforms in detail, you realize that they're fscking nuts. At best, they are well-meaning but lack requisite insight into human nature and/or the real-world ramifications of their lofty ideas. At worst, they're really nuts: spouting off about eliminating UFO influence on our toaster ovens and the like. (My state's voter pre-election voter pamphlets occasionally offer excellent comic-relief while researching candidate backgrounds.)
The more interesting question is why candidates in the "second cut" above don't rally around a third party. Part of the reason is simple: virtually none of the USA voting districts use a ranking-based system of election, such as IRV or Condorcet. This provides a barrier to entry most notably seen in the 2004 US Presidential election as the "Nader effect". Nader was never a viable candidate to win, so many voters felt they were forced to choose between "voting their conscience" and "voting for the lesser evil." The real effects are much deeper, however. An excellent third-party or independent candidate might win an election in such a system by garnering a lot of first and second place votes from voters across the spectrum. The effect could be rather de-polarizing, and would allow a foot in the door for new parties.
Now all of this neglects other serious issues, such as campaign funding and media influence and coverage. Third-parties have an additional barrier in the form of achieving sufficient fund-raising to win a campaign, and achieving media backing. As sad as it sounds, it is absolutely necessary today to get the word out and successfully market a candidate to the people in order to win a contested election. This can require large to massive amounts of money... and the strings that go along with that.
Media outlets get to further warp the funding/marketing issues by providing whatever balance and bias of coverage they want. Whether through carelessness, explicit bias, or even implicit biases, mass-media has come to have an astonishing effect on distorting our democratic processes. Not garnering media support can leave a campaign dead in the water. As a simple example, consider the viability of a candidate for any high-level office with a platform of serious media ownership reform. I have trouble imagining that getting very far.
My guess is that SVN will turn out to be too little, too late with its merge tracking support. It'll be a boost for folks already using SVN who don't want to switch toolchains, but it's pretty easy to move from SVN to the new tools (beyond export, several newer SCMs have two-way commit support with SVN).
Generationally speaking, it feels like SVN is still trying to catch up to Perforce... but that ship has sailed. The teams working on the new round of decentralized SCMs[*] have done deep rethinking of source control problems and challenges, and the results are generally brilliant. These problems aren't esoteric -- administration and day-to-day usage really is easier with the new stuff. After a while using git, Bazaar, etc., the crufty old SCM tools seem like doing image editing in a hex editor instead of a GUI app.
[*] Includes: Bazaar, Darcs, git, and Mercurial (hg)
As a baseline, the best chair is the one that fits, and is adjusted properly. I've had nice, nice chairs before that simply didn't work for me very well due to sizing and/or adjustment issues. Chairs with more/better adjustments are all fine and well, but can produce a worse sitting experience than a "lesser" chair that is closer to what its user needs in the first place.
So take the specific recommendations here as good starting points, but make sure that your father has some quality time (i.e. more than five minutes) to try to sit in the chair in question and see if it's to his liking. Ask the staff to give a tutorial and recommendations on all adjustments, and go elsewhere if they aren't knowledgeable enough to do this. Have him bring a book, laptop, etc. to really get into working posture. Also, some places will let you try a demo chair for a period (usually for the nicer chair lines) -- if so, take 'em up on it. The last thing you want is to spend money on a good chair just to discover that the honeymoon ended on the showroom floor.
(*) It boggles the mind that publishers ever don't provide this feature.
My math is just fine, you need to revisit your basic English, logic, and social skills. I was trying to debunk the absurd prices cited by the parent. My "punch line" was to observe that back-of-the-napkin calculations using conservative bounds yielded (as you observe and agree with) completely unrealistic results.
So your issue is... what?
Those prices are insanely high. I'm seeing current retail prices for high speed compact flash cards at 60 to 100 USD for 8GByte (varies with speed, rebates, etc.). Taking one of the lower-end prices for a top-tier part (since I want an upper bound on part manufacturing costs), that works out to a bit less than 0.001 USD per Mbit. Even if those numbers are per Gbit, that still leaves the cited current manufacturing cost at more than twice actual retail.