I agree, but now that NASA is talking about a permanent presence on the moon, it implies that they will need some sort of local computing capacity to support their astronauts and possibly their robots, death rays and other machines, all of which they will want to control remotely due to the harsh environment. It's not very comfortable living inside of a space suit for a long time with a tube stuck in, well, you know where, so I expect them to work mostly in their specially constructed habitats.
But even if server farms are a bad idea, some NASA contractors who have their hand in the lobbyist equivalent of the cookie jar may demand them "just because."
"first you got to schlep all the stuff up there..."
Since none of this is happening until 2040, maybe 2080 (maybe, like *never* because we don't have the money...), just send up one self-replicating server or a self-replicating machine ecosystem and let that spread on the moon's surface on its own. Everything could be built from moon materials using solar energy if it's designed right. Of course it won't be designed right but that gives the astronauts something to do besides get there before somebody else.
"Cool it into what though? There's no atmosphere."
You don't need an atmosphere to radiate heat. Heat is just infrared and given enough time it'll happily bounce into the blackness of space. (If the moon couldn't radiate heat it would be a nasty molten mistress by now.) You just need a large enough heatsink, and there are lots of metals on the moon for that purpose as well as for being able to say "bite my shiny metal ass." Of course, you would want to place your heatsinks outside of direct sunlight, not to mention the server room itself. By 2040 one would hope that some really low wattage computers would be available, easing that problem. What TFA seems to be worried about is how several potentially competing moon communities would share the vast empty "real estate" on the lunar surface without conflict. Perhaps then each side should build the death ray enhancement first and only then worry about the server farm.
"when you're facing the sun good luck radiating all that heat away."
There are shady pockets in some craters that never receive sunlight. Or, at worst, you just have some rovers push moondirt to form a levee to provide shade. It's not like the wind will knock it down.
Of course, once the entire moon is covered in server farms peoples' jaws will drop at the video streaming capacity and the entire NSA will want to relocate there. Rather than build that many space suits, instead a lot of orbiting satellites with routers will be used to bring the capacity to earth, and these too will have to be built on the moon and launched from the moon. And the neat thing about all of it (possibly the *only* neat thing about server farms on the moon) will be that having all of that crap up there won't add to global warming here.
This one's easy all right, all that's missing are the leads with the signaling voltages to aim the turret.
Since this technology on the PC in TFA is supposed to identify individual faces, it might even be able to differentiate between friend or foe, though it wouldn't be a good idea to entrust anyone's life to its accuracy.
A previous slashdot story with an article from Rolling Stone revealed that the face recognition software sold to the Chinese by a U.S. security firm uses the distance between the eyes as one of the biometrics to identify people, and that this works even after people have aged for decades.
If so, once everyone on earth has their face in a database, all a leader has to do is push a button to send out a robot army against precisely specified targets, with all cameras also cooperating in the effort. Basically all of the robots can be hunting 24/7 for a person or group of people... and you hope they don't confuse you with someone else. After that, there is the possibility of a fully automated war where all of the enemy's faces are known in advance, and so are the friendlies, and it's basically a question of each side getting as many robots to a single location as they can. Of course, the winner of that will be the one with the smarter machines, and the next step of course is that the machines will become smarter than us and take over.
And all of that because we wanted to take better pictures. Tsk, tsk.
"if all six billion people on earth used hand calculators and performed calculations 24 hours a day and seven days a week, it would take them 46 years to do what the Roadrunner can in one day."
This contraption makes lots of people really, really, tired of punching on calculators?
Otherwise this law would shut down pro wrestling, or what passes for it. You know, the ones where each wrestler says what they will do to the other guy in gory detail.
And if someone says "I'm going to fucking kill Google" like some CEOs we've heard about, we will ignore that too.
And, well, do I need to go on? It seems like if it's done on cyber then its wrong, but if it's done live then its OK because its normal emotional behavior.
In this corner: AMD/ATI with the next generation of integrated video with their first pass at pairing up the CPU/GPU and claiming a 3x speedup (over previous integrated graphics solutions, most likely). Their second pass, if they stick the two onto a single die, should be much more interesting, as it should cost even less, use less power, and, if they do enough of a redesign should actually go even faster. I'm keeping my fingers crossed that this combination will result in a fully open system.
In the next corner: VIA and Nvidia, who claimed that they have a big can of "whoop-ass" for intel, but which, for the moment sounds like is empty except for a scrap of paper with the observation that they can put together a system with a small processor and a big video card that is usable enough for most things and can be sold cheaply. Alas, Nvidia can't open source their drivers and this can only hold them back in the long term compared to AMD/ATI and intel. As far as laptops, I'm less optimistic about this team producing something with long battery life.
In the other corner is Intel, who can afford to sit on its perch and see what the market does before reacting. According to the benchmark results in the article, the Atom is falling short in system performance compared to VIA's new offering. It could fix that by allowing the atom to be paired with a faster graphics chip, but if it did that, then Atom would surely eat into their profits in another market segment. Intel has opened its graphics drivers, but their graphics solutions being low-performance by comparison, these don't reveal any tricks that nVidia or ATI would want to use (and maybe that's by design). However, they can't sit on their laurels for ever, eventually the lack of GPU performance may come to haunt them.
NVidia may be worried that if it opened its high-performance drivers, intel and ATI would learn something. On the other hand, intel must be worried about that can of "whoop-ass" if it comes in the form of low-end competition, the sort of competition it can't answer without throwing its segmenting scheme into disarray. If the first two partnerships are both successfully attacking the low end and undercutting intel's high end offerings (and people do seem to be buying a lot of subnotebooks lately), intel will have to cut prices and reduce profitability. In the long term, it'll have to try to take back the low end, but with what we don't know yet.
Actually it's not a PC with face recognition that brings this grisly thought to mind, it's those cameras that "lock on" to any face in the scene and hang on to it like a pit bull to its opponent's neck no matter how much they move around.
It seems to me that one would only need to add a rotating machine gun turret to one of these cameras to create an automatic firing system. One shot per face and then round-robin. You know, the kind that you don't even need to lift your head to look, it does all of the looking and all of the shooting with one click convenience? Well, sure, you may not think this will turn out so great in terms of friendly fire but one application that it would truly shine brightly at would be crowd control.
So, who wants to help beta test this technology?
Anyone?
How about if they offer to add your faceprint to the whitelist?
"Intel's Mini-ITX has no fundamental reason of being less open..."
Maybe what they're referring to as far as "open" (but they're using the word loosely) is that intel doesn't want people to use its cheap atom chip to build systems that can compete in the same market segment with their more pricey offerings. That's what would keep an OEM from outfitting a system any way they want to.
This VIA/nVidia system that does have enough graphics performance to beat the atom and (they claim) to play recent games is basically a way to use intel's own (stupid) rule to grab marketshare away. If intel suddenly lifts the restriction, then their obedient OEMs won't be able to react for a while, and if it doesn't, it may have to drop prices on the low end of its more profitable segment.
It's a smart move on nVidia's part. What isn't so smart is to try to confuse the word "open" for this tactic. Everyone knows that nVidia's drivers aren't open source and that they're not likely to open them any time soon due to all of patents they had to use. They may have nice performance, but closed source drivers and open systems don't mix. They may think they can ignore this due to customer ignorance but I think it's going to catch up with them eventually. Personally, I wouldn't add a closed source video card to my system even at half price - I just don't want to bother with something that works today but breaks with the next kernel upgrade.
"What is different about "us" is that when these things come to light(wiretapping..etc), they are considered as a scandal because they go against the values "we" build our nations on. When they happen in China, the state may or may not publish the information itself, usually with a smile."
Even though we have lived in what has traditionally been viewed as a more open society, that doesn't mean that we are still being adequately informed about everything being done under our noses or over our heads. If some poor devil is being tasered in Canada we might not hear about it unless someone took video footage. I think we have our own gaps which are not completely being filled in by FOIA requests and the like. Combine that with the decline in investigative journalism and it looks like we're headed into some sort of Dark Age.
You are right that this sort of oppression doesn't cause the same sort of shock in other countries that it causes the U.S. or the U.K., but notice that one of the things we're not being told about is of the coping mechanisms that the subjects of other countries have developed to deal with the lack of freedoms. That is, they not only don't expect certain freedoms, they have workarounds for those situations that you don't know about and are not likely to develop until after decades of living under a regime like that.
The article (a good piece of investigative journalism) makes a very, very important point: the security industry is building up its own momentum internationally, is worth 300 billion already (IIRC) and that means it has its own lobby in Washington, and that means it's self-sustaining and... it isn't likely to go away any time soon. That doesn't portend very nicely for our open society and its values of personal freedom.
They will be listened to on their phone calls, monitored by digital voice-recognition technologies.
Sorry bud, we don't do that. There are a few wiretaps flying around, some illegal; but they comprise an absolutely tiny percentage of calls made, generally those made by Very Bad Men.
What are the NSA secret rooms at phone companies for then? Coffee breaks? Why do the phone companies want immunity if they're not being asked to do blatantly illegal things?
Their Internet access will be aggressively limited through the country's notorious system of online controls known as the "Great Firewall."
Yeah, we don't do this, nor will we ever.
Their system is the most notorious for the moment, but don't assume that all of your search queries aren't being watched and recorded, and don't assume the search results are always going to show everything found. Failing that, there are also more extreme measures to watch you.
National ID cards certainly come up every once in a while, but they get shot down consistently.
Some states have already implemented fingerprint cards, after trying to place tracking devices in cars unsuccessfully. Note that we have satellites to track people now, making such cards less important. UK has ubiquitous street cams + databases, we have redlight cameras + satellites, China has black dome cameras on streetlights. I don't see the satellites getting shot down any time soon.
7 years ago, we didn't even have a centralized database for ENEMIES OF THE STATE. What makes you think we have one now for all citizens?
Try reading about about the plans for this unprecedented biometric database.
So, once again, how are we any different except in terms of the degree of policestaticity? (Or is that policestatishness? Perhaps policestatoriality?...Ugh.)
This is how this Golden Shield will work: Chinese citizens will be watched around the clock through networked CCTV cameras and remote monitoring of computers. They will be listened to on their phone calls, monitored by digital voice-recognition technologies. Their Internet access will be aggressively limited through the country's notorious system of online controls known as the "Great Firewall." Their movements will be tracked through national ID cards with scannable computer chips and photos that are instantly uploaded to police databases and linked to their holder's personal data. This is the most important element of all: linking all these tools together in a massive, searchable database of names, photos, residency information, work history and biometric data. When Golden Shield is finished, there will be a photo in those databases for every person in China: 1.3 billion faces.
Correct me if I'm wrong, but I see nothing in the above that we're not already doing here or have announced that we will be doing soon. And the amazing thing is this really big giant coincidence that it's also happening everywhere else. What gives? It's like a world government has been instituted or something.
"He's quite right in saying that OSS developers [...] aren't unified."
He can say it without proof and most people will just take a vague statement like that at face value. I wasn't saying he was right or wrong, just saying he's not quite an objective observer on this. If you see his assertion repeated constantly, they are likely trying to nudge perceptions of it closer to what they want people to believe. Eventually they can convince developers that their astroturf posts represent actual views within the developer community and that the community is divided, making threats, retaliating and all that sort of melodramatic claptrap that can be used to fool banana republics into self-destructing without firing a shot.
He is right in the sense that corporate boards are used to dealing with the relatively few people at the top of other chains of command who can make their decisions and commitments stick. That much they are comfortable with, and they have lots of money, patents and lobbying influence to use as leverage in those situations, but with this group there is no single overlord to strike deals with that will cause the entire group to click their heels and salute.
He is wrong in the sense that open source developers do share some common interests, and this is enough to bind them together on a lot of issues even if they may disagree and pick nits about others.
Either way, he cares less about the accuracy of his statements than he cares about how to benefit his stockholders, so take all such webcasts intended for public consumption with a sizeable crystal of sea-salt. Executives don't lie, they are just "occasionally misinformed," and if that involves repeating some convenient view over and over within earshot of the press then, well that's just coincidental.
"...someone from Microsoft is suddenly all concerned for the financial well-being of OSS developers that makes me concerned."
It kind of reminds me of the joke about the blind man with a seeing-eye dog who stops at a street corner and gets pissed on by his dog. Then, moving the leash to the hand holding the cane, he pulls a treat from his pocket and offers it to the dog. Someone standing nearby points out what the dog has just done and asks the blind man why he is rewarding the dog for being naughty. The blind man answers "that's just so I can find which end to whack with my cane."
Open source developers are like the dog, and the blind man is MS, still no wiser as to where to hit to damage the adversary, and unable to hit too hard or too obviously because the other companies who benefit from OSS would sue.
Well, the treat they're offering may just be Microsoft trying to figure out who's who, so they can find out if their own employees ratted them out on a blog in violation of a gag order, or to quantify and understand their #1 enemy, or just to see if they'll be able to pounce on specific market sectors easily. For most people, no doubt they would try to act really nice and give out free stuff like SDKs, free marketing offers, free spam messages about developer conferences, discounts, beta test freebies, that sort of thing. They might convince some people to switch to the "dark side" eventually, and that would be a big win for them.
They specifically mention in the article that the OSS community is not "unified," which is actually just an opinion pulled out of the air. They would like you to begin thinking that it isn't unified so that it can seem more plausible to apply the divide and conquer strategy (e.g. GPL vs. BSDL, Mac OSS vs. Linux OSS vs. Windows OSS). They might try to fund only BSDL projects or add Windows-only features to the biggest OSS projects, for instance, if they think this will cause a rift. Gates saying that GPL code "can't be modified" is basically a hint of this. He's not so stupid as to actually believe such a shocking untruth, but it would be really difficult to prove willful ignorance on his part. In reality, they're worried that either BSDL or GPL could eat their lunch because either one can promote free standards and give their competitors a leg up (sorry for the pun). GPL has been around practically forever but they really started to dislike it recently because they've figured out that it evolves to counter new loopholes.
I don't expect them to bribe anyone directly, but maybe their affiliates will be funding lots of interoperability features that make sure their developers have a bridge to jump across if their Windows platform crashes and burns in the marketplace.
If they get enough critical mass with OSS developers, they might even convince some of them to add lots of bloat to open source projects so that their own products don't look so bad in comparison. Of course, they won't call it adding bloat, it'll be called cool new features, cool new devices, and cool new protocols, and they'll just need lots of them. They could, for instance, add a bunch of unnecessary modules for nonexistent hardware and make Linux take longer to boot and take up a big chunk of RAM. Since most of us would be suspicious if Microsoft asked us to put a binary blob into Zimbra or Samba, I would guess that their affiliates would be willing to act as the "customer" for this kind of enhancement.
My advice: take the free stuff but watch out what you agree to on the EULAs as it may prevent you from contributing to some OSS projects in the future. Don't give them any identity information if you can help it, just in case they turn into a RIAA-type troll organization someday.
"The authors present economic arguments without evidence, "markets function best in the presence of clearly defined property rights" (p11). "
That's more of a basic assumption of market economics than an argument that needs evidence to support it. If you want to see why such a thing could be taken as an assumption, you can think in terms of the following example: Let's say that a popular book author writes a manuscript. Every publishing house in the country would kill to get their hands on that manuscript and print it without paying the author. However, the author retains the property rights to that physical object. It's crystal clear who owns that property - it's in the author's hand and the author doesn't have to release it until a favorable deal is struck. Because those rights are extremely well-defined, a book publishing house is willing to make a deal with the book author in exchange for the ability to read the manuscript and print it (and sell the books). In the absence of clearly defined property rights, such as those surrounding imaginary property like patents and copyright, there will always be controversy because every party can argue that they are the injured party and there's enough subjectivity and gray areas for lawyers to swing it either way. Thus, if you want markets to function predictably and uniformly, you generally will be better off with clearly defined property rights than to have to constantly handle issues on a case-by-case, subjective basis and which may be here today and gone tomorrow.
Stated another way, if you can be reasonably certain that a CD is your own property after you buy it, you will be more likely to buy it than if you are told you are getting a license to use it with a bunch of conditions in fine print which you don't understand. Deals just flow more smoothly and quickly when each party knows what they're getting, that's all.
I haven't finished reading it, but I can cite a few things.
First of all, 20 of the 46 top selling drugs had no need for patents. These are things that doctors still prescribe or use like aspirin, insulin, penicillin, quinine, morphine, vaccines, vitamins, etc.
It claims 54% of new drug applications use active ingredients already in the market, the "innovation" being merely in dosage amounts or other incidentals.
The book says that 75% of drug R&D costs are for development of me-too drugs. The way I would explain this is this: a lot of otherwise unnecessary patents are filed on slightly modified drugs based on a successful earlier design. The new drugs are heavily promoted to doctors so that it can earn the company royalties while the old, perfectly good drug is ignored by the big name pharmas. Drugs are not developed as things that the whole society needs, but rather on the basis of marketability.
The book argues that clinical trials for new drugs could (and should) be paid completely by NIH grants, and this would remove the conflict of interest of having the drug company doing its own testing. Doing so would eliminate any need for drug patents to exist, because this is the cost that the companies are citing they would need to recover (the marketing cost being optional and really up to them to decide). The legal costs of fighting patents would of course also go away (too bad for the lawyers).
So yes, surprisingly they would still be invented. Then there's the question about whether the generics makers would take all of the profits. Think for a moment about how patents are filed, describing the drug in detail. Without patents, the first drug company to produce a successful drug would have a very large lead over the other companies. The others, once they see the drug is selling well would basically try to reverse engineer it, but in order to produce the drug in significant quantities they would have to retool their factories and that takes time. If typically takes the generics manufacturers about 4 years to get up to speed, and that's about half of the length of a drug patent period anyway. Overall, that means drug companies can still recover their costs, save hugely on legal costs and still make a tidy profit.
The upshot is that after 4 years any given drug of any importance would be very, very cheap everywhere and thus widely accessible by all economic strata.
"all in the name of corporate protectionism and profit"
That is indeed what is going on with "intellectual property." Proponents describe it as an enabler to unfettered innovation in free-market capitalism, but it's really an artificial monopoly designed to provide a form of protectionism for a few lucky companies, and any form of protectionism we all know is anathema to the ideas of a totally free market.
Just the same, the whole world would be better off completely without any form of it.
To save you 300 pages of reading "Against Intellectual Monopoly," basically patents don't spur innovation, not even the ones on concrete inventions. Case after case is presented where it is clear that the idea of spurring innovation through patents is flawed at best, and highly damaging at worst. They basically prove that steam engine development was slowed down by patents and only really began to chug when the patents expired. Inventors you thought were heroes finally come across in a more realistic light. They present lots of examples like this and you just basically see the light (at the end of the tunnel?). Now, if your goal was to slow down technological progress or science itself, then maybe patents would be a good idea.
Before reading some chapters from Boldrin & Levine I was somewhat convinced that copyright at least had some beneficial elements to it that should be respected and preserved, but they sure put the nail in that coffin too. They went through the origins of copyright as a *relaxation* to a censorship regime by the crown (IIRC), and it just went downhill from there. Now it just seems like copyright is extended to every damn little thing, and that wasn't the original purpose of it by far. While they don't prove that removing copyright would be beneficial to everyone, they take a shot at showing that it wouldn't be a total disaster to authors/artists. For everyone else, it wouldn't prevent new books from being written, new music from being produced, etc., and it would be a net gainer, by far.
If you have the time to read a 300 page book this summer, by all means at least read a few chapters of Boldrin & Levine. You will understand intellectual property much better and hopefully lose a few sacred cows in the process.
In that case I would have expected you to be much better acquainted with Microsoft's track record than what you've been demonstrating.
"Doesn't seem to have worked very well, and I still don't believe Microsoft are dumb enough to believe it would."
The fact is, they did send a takedown notice to this very site and it was viewed as censorship and duly caused an uproar. Does that make them stupid? No, it's not quite that. It's more at arrogant. It's what powerful people think they will get away with, and they just miscalculate. They love to take big gambles because when they win, it gives them an unbelievable high that they're smarter than all of those little people. When they lose, it's not quite as pleasant, but optimists like you just think they must be dumb to try it. Not much of a downside to their gambles then, is there?
"The second sounds like one of those changes made to support DRM. These I do disapprove of."
Whether or not we approve of it, they have foisted it on us, and I don't know about you but the control-freak nature of that concept gives me less of a reason to trust their upgrades. I've drawn the conclusion that closed source software is inherently less trustworthy as a result.
"[SP2] certainly wasn't hassle-free."
OK, we'll chalk it up to varying experiences of that then.
"both malware and anti-malware software is far more prevalent and more complex."
By the same token, MS had an opportunity to address these in SP3 but they not only went unaddressed, the upgrade itself fell victim to them. Don't say it wasn't in their power to survey the malware out there and prepare for it. This was precisely the sort of bugfix one would expect from a service pack.
One more thing. To the person who dies before the copyright expires it is certainly perpetual.
Since copyright extends beyond one lifetime then such perpetuity is guaranteed for the vast majority of works and the vast majority of people. Thus, the original words "for limited times" no longer really apply. That means that in practice, the current law is unconstitutional.
"I'm not sure I can agree that it is Microsoft's job to test
third-party products for this sort of problem..."
Stated that way, it sounds like an unlikely location to place blame,
but it can just as easily be stated another way: "who has the greatest
influence on final quality?" The fact is that Microsoft defines its
APIs any way it wants, produces whatever documentation and developer
kits it wants, and supports third party products any way it wants, and
if they don't want to support you and prefer your business to fail,
they can cause that and it wouldn't be the first instance of that.
"Hey, I'm at least as paranoid as the next guy, but it really does
seem you're assuming your conclusion and arguing backwards here."
Be serious. Everyone knows that for-profit companies have to keep
moving product to please investors with quarterly reports of high
growth rates. Investors love to hear about how a company will prevail
against its competition, and if the primary competition is their own
product - XP - it's not too "paranoid" to believe that some people are
going to pursue that competition diligently... even agressively.
Calling me paranoid about this makes you sound like you are really
trying hard to shine a positive light on this company - a company with
a decidedly spotty history. Once you know about their background of
monopoly abuse, expecting a scorched-earth legacy policy from them
would be the *least* of things to expect, but you apparently are so
new to the IT field that you don't know a bit of their history. If
I'm paranoid then you are uninformed or disingenuous.
"I don't believe Microsoft want this sort of PR."
Of course they don't, but they were counting on the large degree of
public ignorance surrounding computers, the effect of their well
funded deceptive marketing campaigns, and at last resort their
poweful legal arm to bury it with takedown notices to prevent the word
from getting out. It wouldn't be the first instance of that either.
"Are you honestly suggesting that Microsoft are deliberately
introducing new problems?"
Are you honestly suggesting that they're going to compete with
themselves and allow that to reduce their revenue? Some of their
investors may not like the idea of the public stuck on XP producing no
revenue and may try to switch out board members until somebody
listens. It's the short-term money they've been historically after,
not public acceptance. I wish I could think otherwise, but this sort
of breakage is the path of least resistance for MS. They can afford
to do it because they are a monopoly with such a gaping lead over
anyone else. Who else can the average person turn to?
"I don't see disk space or memory utilization as an inherent
problem. What sort of issue are you anticipating?"
It's simply Murphy's Law as applied over the quantity of code. The
number of bugs is thought to be proportional to the number of lines of
code, and the full installation of Vista weighs in about 10-100 times
larger than other modern operating systems. Vista is so big that even
the old rule of thumb of proportionality may not apply any more.
"...presumably most of the extra stuff is just candy - not likely to
have a significant impact on the underlying stability."
It doesn't all smell like candy though. Some of the changes dig down
deep into the core of the kernel and change its inner workings. A new
facility for signed drivers and more stringently controlled and
prioritized video/audio paths come to mind. It basically threw away a
huge set of old, working drivers and replaced them with a set of new,
incomplete drivers. There are a host of potential problems just
lurking in this new driver code, simply because they all had to be
rewritten. All drivers can affect stability and security.
"...most of the problems seem to be the result of OEM or third-party-vendor defects, or malware"
It may be true that third party vendors and malware are immediately to
blame, but even so I find it difficult to believe that the third
parties in question would not have done any of their own testing for
critical bugs. I find it much easier to believe that it was MS who
skipped their part of the testing or failed to provide the appropriate
failure modes, developer training, support response, complete
documentation, etc. The company isn't exactly promoting XP anymore,
so if XP isn't left in a perfect state for posterity I doubt they'll
spend too many resources worrying about it.
In a case like this, if
the third parties pointed the finger back at MS I would tend to
believe them by default instead of believing MS. The third parties
have a much stronger motive for their product to actually work. They
are not monopolies after all, so they depend heavily on product
quality to get any repeat buyers. If MS lets their quality slip, they
are not similarly harmed by it - in fact some people will react to
these problems by buying a new computer with Vista on it. Thus, this
may just be a case where MS finds it convenient to blame someone else
and can make the accusation stick.
As to the malware, I see it as an implicit way of blaming the users
for either allowing the malware in, or of installing anti-malware
which also is blamed for problems. Thus, the hapless user faces a
no-win situation no matter which action is taken. You admit Windows'
design was suboptimal for security purposes, but on the other hand you
haven't fully deduced where the blame belongs. Malware is just a
force of nature, it cannot itself be blamed. Neither should the users
be expected to know about every part of an overwrought complex system
made by someone else and closed to inspection. Thus, that leaves only
Microsoft. No one else we know about has enough control and
visibility over the entire situation to take the steps to prevent
malware from messing with system internals. Blaming MS for actions
taken years ago or for a problem set that has grown too large may not
seem fair, but it also isn't fair that Microsoft can profitably use
the malware on XP as an inducement for people to upgrade to Vista.
"Some of the underlying design, in particular the way that updates and
service packs interact, looks to be superior. "
Vista's design features may seem like revolutionary advances right
now, but this sort of glowing promise is something that always appears
at the beginning of every upgrade cycle. Invariably, the next Windows
version is touted as having great solutions to the problems of the
previous version. In hindsight, we can see that this was done whether
or not it actually delivered the perfect solution as it was promised
(or hyped). It's not a coincidence that any *new* problems that the
current version is quietly introducing aren't emphasized until the
next upgrade cycle.
Specifically in the case of Vista you don't seem concerned that its
piggish size will be the source of future problems or excuses because
you are a "true believer" that its genial new features will be enough
to eventually trump most concerns. One only needs to have observed a
few upgrade cycles and remembered what was said to see how taking MS'
assertions at face value is risky.
And who is to say that Vista will fare any better than
XP? Past performance predicts that MS will have to find *something*
to use as an inducement for people to upgrade to Windows 7, so even if
Vista achieves perfection in its next service pack, I'm betting that
it will be undone in SP3 in order to generate sales. More likely, it
looks like Vista will never reach XP in terms of the overall user
experience.
Now, if you insist that Microsoft's software development methodology
isn't broken, I sense it must be because you are excluding the business
dec
"any major change to an existing Windows installation (such as, say, a
service pack) has associated risks."
If software engineering principles were adhered to across the
organization, these sorts of risks would be reduced. I suspect that
it must be something like office politics or featureitis that must be
working to undermine the methodology against the better judgment of
the developers, but there is really no excuse for SP3 being less
stable than SP2.
"If there's anything wrong with the existing installation, even if it
doesn't currently exhibit any symptoms, it may well fall over."
It's always easy to pass the blame onto the users doing the
installation, but they will resent being written off as the PEBKAC if
they upgraded to SP2 without running into problems this severe. Is it
too much to ask that MS test for critical bugs before releasing?
"anti-malware products are particularly risky because by design they
subvert the proper operation of the OS."
An OS that needs to be subverted in order to ensure correct operation
doesn't strike me as particularly well-designed. You seem to be
saying that it's not worth applying good software development
practices to a service pack, since these practices weren't used for
the OS itself, prone to malware as it was.
"Windows is just too big and complex for something like installing a
service pack to be a completely safe operation..."
"[but]...there are signs that Vista may be an improvement."
Windows is too big and complex, so the company decides to make Vista
an even *bigger* and *more* complex OS? If this problem is going to
be solved, they sure have a funny way of heading in the wrong
direction. Vista's own SP1 transition didn't inspire the sort of
confidence that they were turning the corner, new features and all.
"I'd really like to see Windows retired sooner rather than later,
preferably in favour of a brand new operating system built from the
ground up with today's security threats in mind. However,
this isn't likely to be feasible in the foreseeable future because of
the huge base of existing software."
If they can't transition, then they're the victims of their own
success. But, I think Microsoft will soon have no choice but to do
what you're suggesting. As Vista or its successor grows ever larger,
it will need a correspondingly large organization to maintain it, one
that itself will be fearsomely challenging to manage. Before
investors begin to suspect labor costs spiraling out of control,
they'll have to find some way to foist a redesign upon the world.
However, you would be correct to imply that a redesign that sacrifices
compatibility will be just as widely criticized as one with quality problems.
They could buy up IT companies left and right hoping to find miracle workers,
but there won't be any quick or easy solution for this no matter what.
Isn't the original purpose of a service pack to add reliability, rather than take it away?
One would think that by SP3 there would only the most minor bugs left to close, but instead giant new ones are opened. Machines that become unbootable? That's pre-alpha quality stuff.
Something is badly broken with their methodology... no wonder they were trying to do a people grab at Yahoo, the higher ups are probably pulling their hair out by now trying to figure out how to fix their organizational problem and maybe they thought a new project built on BSD (but independent from Apple code) with entirely new staff would bail them out.
"Another touted benefit is the ability to consolidate control- and data-plane functions on a single device, with "solid-wall" processor boundaries reinforcing security and licensing containment barrier. In this regard, the Tile64 chip resembles another heavily multicore MIPS64 chip, Cavium's 16-way Octeon."
Does anyone know what the heck a "licensing containment barrier" is? It definitely sounds like a performance hit if it's turned on. And if it's forced to have it on then this design just lost a lot of its sex-appeal.
"What you're talking about sounds like just mashing a CPU core and GPU core together on the same die. Which would be horrible for all kinds of reasons (heat, bus bottlenecks and yields!)."
One would hope that bus bottlenecks would not be a problem if the cores don't have to go off-chip to access a common bus. The bus could be a wider, dedicated video bus if kept on on-chip, and would enjoy vastly reduced latency. It would use less overall power than having an external mobo chip coordinate the handshaking, not to mention the signals that have to be sent to a physical video card. The die itself would be larger than a 1 CPU die, but that's not a problem anymore.
The problem now is what the hell to do with all of the transistors we can cram onto a chip. At some point people won't notice a difference between a 16 core unit and a 32 core unit connected to the same memory. Might as well stick a GPU in there. Heck, throw in two for good measure. If there's a defect in one, you can disable that and still sell the part as the economy version. If that means 2-4 fewer CPUs per die then so be it.
It's not all as horribly problematic as you make it sound. There are some significant savings that can be reaped by putting the two together. Right now they claim to be doing it as a way of producing a cheaper system with slightly slower video-card performance, but there are some opportunities to actually improve performance by doing this, and the benefit of lower power consumption for the overall system is definitely in the plus column.
Actually, if anybody can find the "right" way of doing this, it will probably be AMD/ATI... and then everybody else will just copy what they did, tweak it slightly and give it a different name.
"This case proves only one thing to me, and that is that we'd almost be better off with no system at all. Not quite, but almost."
If you consider Capitol vs. Thomas that is already the case. Paying $222,000 for 24 digital music files is completely vacuous, it lacks any sense of proportion. It's not like Thomas stole a house and has to give it back. It's not like they should become indentured servants to a lender for life due to a computing habit that they can easily quit. Did Capitol run out of digital copies for their songs? Finally, how many live performances could be purchased with that amount of money? It's just ridiculous. The lawyers must be foaming at the mouth.
Having swedish laws in mind, C. v. T. is also a regressive judgment. Clearly, justice here is heavily biased in favor of those who have lots of money, and against those who don't.
In Reiser's case, you get the nagging feeling that the whole story hasn't been told, and you *know* that the jury and the judge are more like the audience in a roman arena cheering and pointing their thumbs up or down according to their biases or preconceptions than they are the solemn, methodical guarantors of justice (I personally think that Nina might have had a motive to disappear to Russia or another place with bribeable authorities if the $50k salary she was offered for an OB-GYN seemed insufficient for the Bay Area, and not knowing what to do she (or her boyfriend) hatched a scheme that went terribly wrong). You get the feeling that a truly impartial investigation, using a real justice system would be able to deliver a fairer judgment, but that just isn't available. No such system exists. You wonder if it would be possible to create one from scratch out of smart, technologically current, impartial elements, but using the rapid rise of astroturfers on this site as an indicator, it seems that that route might not work permanently either. It's a tragedy because the justice system is probably the most crucial of the branches of government that needs to be kept impartial.
No doubt about it, we have a broken justice system. However, having no system at all would result in mob rule, which would actually the same thing: a justice system with uneven quality. The only difference would be a different set of gainers and losers. I guess we should just welcome our gavel-wielding overlords.
Slashdot Mirror
I agree, but now that NASA is talking about a permanent presence on the moon, it implies that they will need some sort of local computing capacity to support their astronauts and possibly their robots, death rays and other machines, all of which they will want to control remotely due to the harsh environment. It's not very comfortable living inside of a space suit for a long time with a tube stuck in, well, you know where, so I expect them to work mostly in their specially constructed habitats.
But even if server farms are a bad idea, some NASA contractors who have their hand in the lobbyist equivalent of the cookie jar may demand them "just because."
Since none of this is happening until 2040, maybe 2080 (maybe, like *never* because we don't have the money...), just send up one self-replicating server or a self-replicating machine ecosystem and let that spread on the moon's surface on its own. Everything could be built from moon materials using solar energy if it's designed right. Of course it won't be designed right but that gives the astronauts something to do besides get there before somebody else.
You don't need an atmosphere to radiate heat. Heat is just infrared and given enough time it'll happily bounce into the blackness of space. (If the moon couldn't radiate heat it would be a nasty molten mistress by now.) You just need a large enough heatsink, and there are lots of metals on the moon for that purpose as well as for being able to say "bite my shiny metal ass." Of course, you would want to place your heatsinks outside of direct sunlight, not to mention the server room itself. By 2040 one would hope that some really low wattage computers would be available, easing that problem. What TFA seems to be worried about is how several potentially competing moon communities would share the vast empty "real estate" on the lunar surface without conflict. Perhaps then each side should build the death ray enhancement first and only then worry about the server farm.
There are shady pockets in some craters that never receive sunlight. Or, at worst, you just have some rovers push moondirt to form a levee to provide shade. It's not like the wind will knock it down.
Of course, once the entire moon is covered in server farms peoples' jaws will drop at the video streaming capacity and the entire NSA will want to relocate there. Rather than build that many space suits, instead a lot of orbiting satellites with routers will be used to bring the capacity to earth, and these too will have to be built on the moon and launched from the moon. And the neat thing about all of it (possibly the *only* neat thing about server farms on the moon) will be that having all of that crap up there won't add to global warming here.
This one's easy all right, all that's missing are the leads with the signaling voltages to aim the turret.
Since this technology on the PC in TFA is supposed to identify individual faces, it might even be able to differentiate between friend or foe, though it wouldn't be a good idea to entrust anyone's life to its accuracy.
A previous slashdot story with an article from Rolling Stone revealed that the face recognition software sold to the Chinese by a U.S. security firm uses the distance between the eyes as one of the biometrics to identify people, and that this works even after people have aged for decades.
If so, once everyone on earth has their face in a database, all a leader has to do is push a button to send out a robot army against precisely specified targets, with all cameras also cooperating in the effort. Basically all of the robots can be hunting 24/7 for a person or group of people... and you hope they don't confuse you with someone else. After that, there is the possibility of a fully automated war where all of the enemy's faces are known in advance, and so are the friendlies, and it's basically a question of each side getting as many robots to a single location as they can. Of course, the winner of that will be the one with the smarter machines, and the next step of course is that the machines will become smarter than us and take over.
And all of that because we wanted to take better pictures. Tsk, tsk.
"if all six billion people on earth used hand calculators and performed calculations 24 hours a day and seven days a week, it would take them 46 years to do what the Roadrunner can in one day."
This contraption makes lots of people really, really, tired of punching on calculators?
Otherwise this law would shut down pro wrestling, or what passes for it. You know, the ones where each wrestler says what they will do to the other guy in gory detail.
And if someone says "I'm going to fucking kill Google" like some CEOs we've heard about, we will ignore that too.
And, well, do I need to go on? It seems like if it's done on cyber then its wrong, but if it's done live then its OK because its normal emotional behavior.
In this corner: AMD/ATI with the next generation of integrated video with their first pass at pairing up the CPU/GPU and claiming a 3x speedup (over previous integrated graphics solutions, most likely). Their second pass, if they stick the two onto a single die, should be much more interesting, as it should cost even less, use less power, and, if they do enough of a redesign should actually go even faster. I'm keeping my fingers crossed that this combination will result in a fully open system.
In the next corner: VIA and Nvidia, who claimed that they have a big can of "whoop-ass" for intel, but which, for the moment sounds like is empty except for a scrap of paper with the observation that they can put together a system with a small processor and a big video card that is usable enough for most things and can be sold cheaply. Alas, Nvidia can't open source their drivers and this can only hold them back in the long term compared to AMD/ATI and intel. As far as laptops, I'm less optimistic about this team producing something with long battery life.
In the other corner is Intel, who can afford to sit on its perch and see what the market does before reacting. According to the benchmark results in the article, the Atom is falling short in system performance compared to VIA's new offering. It could fix that by allowing the atom to be paired with a faster graphics chip, but if it did that, then Atom would surely eat into their profits in another market segment. Intel has opened its graphics drivers, but their graphics solutions being low-performance by comparison, these don't reveal any tricks that nVidia or ATI would want to use (and maybe that's by design). However, they can't sit on their laurels for ever, eventually the lack of GPU performance may come to haunt them.
NVidia may be worried that if it opened its high-performance drivers, intel and ATI would learn something. On the other hand, intel must be worried about that can of "whoop-ass" if it comes in the form of low-end competition, the sort of competition it can't answer without throwing its segmenting scheme into disarray. If the first two partnerships are both successfully attacking the low end and undercutting intel's high end offerings (and people do seem to be buying a lot of subnotebooks lately), intel will have to cut prices and reduce profitability. In the long term, it'll have to try to take back the low end, but with what we don't know yet.
Actually it's not a PC with face recognition that brings this grisly thought to mind, it's those cameras that "lock on" to any face in the scene and hang on to it like a pit bull to its opponent's neck no matter how much they move around.
It seems to me that one would only need to add a rotating machine gun turret to one of these cameras to create an automatic firing system. One shot per face and then round-robin. You know, the kind that you don't even need to lift your head to look, it does all of the looking and all of the shooting with one click convenience? Well, sure, you may not think this will turn out so great in terms of friendly fire but one application that it would truly shine brightly at would be crowd control.
So, who wants to help beta test this technology?
Anyone?
How about if they offer to add your faceprint to the whitelist?
"Intel's Mini-ITX has no fundamental reason of being less open..."
Maybe what they're referring to as far as "open" (but they're using the word loosely) is that intel doesn't want people to use its cheap atom chip to build systems that can compete in the same market segment with their more pricey offerings. That's what would keep an OEM from outfitting a system any way they want to.
This VIA/nVidia system that does have enough graphics performance to beat the atom and (they claim) to play recent games is basically a way to use intel's own (stupid) rule to grab marketshare away. If intel suddenly lifts the restriction, then their obedient OEMs won't be able to react for a while, and if it doesn't, it may have to drop prices on the low end of its more profitable segment.
It's a smart move on nVidia's part. What isn't so smart is to try to confuse the word "open" for this tactic. Everyone knows that nVidia's drivers aren't open source and that they're not likely to open them any time soon due to all of patents they had to use. They may have nice performance, but closed source drivers and open systems don't mix. They may think they can ignore this due to customer ignorance but I think it's going to catch up with them eventually. Personally, I wouldn't add a closed source video card to my system even at half price - I just don't want to bother with something that works today but breaks with the next kernel upgrade.
Even though we have lived in what has traditionally been viewed as a more open society, that doesn't mean that we are still being adequately informed about everything being done under our noses or over our heads. If some poor devil is being tasered in Canada we might not hear about it unless someone took video footage. I think we have our own gaps which are not completely being filled in by FOIA requests and the like. Combine that with the decline in investigative journalism and it looks like we're headed into some sort of Dark Age.
You are right that this sort of oppression doesn't cause the same sort of shock in other countries that it causes the U.S. or the U.K., but notice that one of the things we're not being told about is of the coping mechanisms that the subjects of other countries have developed to deal with the lack of freedoms. That is, they not only don't expect certain freedoms, they have workarounds for those situations that you don't know about and are not likely to develop until after decades of living under a regime like that.
The article (a good piece of investigative journalism) makes a very, very important point: the security industry is building up its own momentum internationally, is worth 300 billion already (IIRC) and that means it has its own lobby in Washington, and that means it's self-sustaining and... it isn't likely to go away any time soon. That doesn't portend very nicely for our open society and its values of personal freedom.
What are the NSA secret rooms at phone companies for then? Coffee breaks? Why do the phone companies want immunity if they're not being asked to do blatantly illegal things?
Their system is the most notorious for the moment, but don't assume that all of your search queries aren't being watched and recorded, and don't assume the search results are always going to show everything found. Failing that, there are also more extreme measures to watch you.
Some states have already implemented fingerprint cards, after trying to place tracking devices in cars unsuccessfully. Note that we have satellites to track people now, making such cards less important. UK has ubiquitous street cams + databases, we have redlight cameras + satellites, China has black dome cameras on streetlights. I don't see the satellites getting shot down any time soon.
Try reading about about the plans for this unprecedented biometric database.
So, once again, how are we any different except in terms of the degree of policestaticity? (Or is that policestatishness? Perhaps policestatoriality? ...Ugh.)
Correct me if I'm wrong, but I see nothing in the above that we're not already doing here or have announced that we will be doing soon. And the amazing thing is this really big giant coincidence that it's also happening everywhere else. What gives? It's like a world government has been instituted or something.
"He's quite right in saying that OSS developers [...] aren't unified."
He can say it without proof and most people will just take a vague statement like that at face value. I wasn't saying he was right or wrong, just saying he's not quite an objective observer on this. If you see his assertion repeated constantly, they are likely trying to nudge perceptions of it closer to what they want people to believe. Eventually they can convince developers that their astroturf posts represent actual views within the developer community and that the community is divided, making threats, retaliating and all that sort of melodramatic claptrap that can be used to fool banana republics into self-destructing without firing a shot.
He is right in the sense that corporate boards are used to dealing with the relatively few people at the top of other chains of command who can make their decisions and commitments stick. That much they are comfortable with, and they have lots of money, patents and lobbying influence to use as leverage in those situations, but with this group there is no single overlord to strike deals with that will cause the entire group to click their heels and salute.
He is wrong in the sense that open source developers do share some common interests, and this is enough to bind them together on a lot of issues even if they may disagree and pick nits about others.
Either way, he cares less about the accuracy of his statements than he cares about how to benefit his stockholders, so take all such webcasts intended for public consumption with a sizeable crystal of sea-salt. Executives don't lie, they are just "occasionally misinformed," and if that involves repeating some convenient view over and over within earshot of the press then, well that's just coincidental.
"...someone from Microsoft is suddenly all concerned for the financial well-being of OSS developers that makes me concerned."
It kind of reminds me of the joke about the blind man with a seeing-eye dog who stops at a street corner and gets pissed on by his dog. Then, moving the leash to the hand holding the cane, he pulls a treat from his pocket and offers it to the dog. Someone standing nearby points out what the dog has just done and asks the blind man why he is rewarding the dog for being naughty. The blind man answers "that's just so I can find which end to whack with my cane."
Open source developers are like the dog, and the blind man is MS, still no wiser as to where to hit to damage the adversary, and unable to hit too hard or too obviously because the other companies who benefit from OSS would sue.
Well, the treat they're offering may just be Microsoft trying to figure out who's who, so they can find out if their own employees ratted them out on a blog in violation of a gag order, or to quantify and understand their #1 enemy, or just to see if they'll be able to pounce on specific market sectors easily. For most people, no doubt they would try to act really nice and give out free stuff like SDKs, free marketing offers, free spam messages about developer conferences, discounts, beta test freebies, that sort of thing. They might convince some people to switch to the "dark side" eventually, and that would be a big win for them.
They specifically mention in the article that the OSS community is not "unified," which is actually just an opinion pulled out of the air. They would like you to begin thinking that it isn't unified so that it can seem more plausible to apply the divide and conquer strategy (e.g. GPL vs. BSDL, Mac OSS vs. Linux OSS vs. Windows OSS). They might try to fund only BSDL projects or add Windows-only features to the biggest OSS projects, for instance, if they think this will cause a rift. Gates saying that GPL code "can't be modified" is basically a hint of this. He's not so stupid as to actually believe such a shocking untruth, but it would be really difficult to prove willful ignorance on his part. In reality, they're worried that either BSDL or GPL could eat their lunch because either one can promote free standards and give their competitors a leg up (sorry for the pun). GPL has been around practically forever but they really started to dislike it recently because they've figured out that it evolves to counter new loopholes.
I don't expect them to bribe anyone directly, but maybe their affiliates will be funding lots of interoperability features that make sure their developers have a bridge to jump across if their Windows platform crashes and burns in the marketplace.
If they get enough critical mass with OSS developers, they might even convince some of them to add lots of bloat to open source projects so that their own products don't look so bad in comparison. Of course, they won't call it adding bloat, it'll be called cool new features, cool new devices, and cool new protocols, and they'll just need lots of them. They could, for instance, add a bunch of unnecessary modules for nonexistent hardware and make Linux take longer to boot and take up a big chunk of RAM. Since most of us would be suspicious if Microsoft asked us to put a binary blob into Zimbra or Samba, I would guess that their affiliates would be willing to act as the "customer" for this kind of enhancement.
My advice: take the free stuff but watch out what you agree to on the EULAs as it may prevent you from contributing to some OSS projects in the future. Don't give them any identity information if you can help it, just in case they turn into a RIAA-type troll organization someday.
Ssssssssss...
"The authors present economic arguments without evidence, "markets function best in the presence of clearly defined property rights" (p11). "
That's more of a basic assumption of market economics than an argument that needs evidence to support it. If you want to see why such a thing could be taken as an assumption, you can think in terms of the following example: Let's say that a popular book author writes a manuscript. Every publishing house in the country would kill to get their hands on that manuscript and print it without paying the author. However, the author retains the property rights to that physical object. It's crystal clear who owns that property - it's in the author's hand and the author doesn't have to release it until a favorable deal is struck. Because those rights are extremely well-defined, a book publishing house is willing to make a deal with the book author in exchange for the ability to read the manuscript and print it (and sell the books). In the absence of clearly defined property rights, such as those surrounding imaginary property like patents and copyright, there will always be controversy because every party can argue that they are the injured party and there's enough subjectivity and gray areas for lawyers to swing it either way. Thus, if you want markets to function predictably and uniformly, you generally will be better off with clearly defined property rights than to have to constantly handle issues on a case-by-case, subjective basis and which may be here today and gone tomorrow.
Stated another way, if you can be reasonably certain that a CD is your own property after you buy it, you will be more likely to buy it than if you are told you are getting a license to use it with a bunch of conditions in fine print which you don't understand. Deals just flow more smoothly and quickly when each party knows what they're getting, that's all.
I haven't finished reading it, but I can cite a few things.
First of all, 20 of the 46 top selling drugs had no need for patents. These are things that doctors still prescribe or use like aspirin, insulin, penicillin, quinine, morphine, vaccines, vitamins, etc.
It claims 54% of new drug applications use active ingredients already in the market, the "innovation" being merely in dosage amounts or other incidentals.
The book says that 75% of drug R&D costs are for development of me-too drugs. The way I would explain this is this: a lot of otherwise unnecessary patents are filed on slightly modified drugs based on a successful earlier design. The new drugs are heavily promoted to doctors so that it can earn the company royalties while the old, perfectly good drug is ignored by the big name pharmas. Drugs are not developed as things that the whole society needs, but rather on the basis of marketability.
The book argues that clinical trials for new drugs could (and should) be paid completely by NIH grants, and this would remove the conflict of interest of having the drug company doing its own testing. Doing so would eliminate any need for drug patents to exist, because this is the cost that the companies are citing they would need to recover (the marketing cost being optional and really up to them to decide). The legal costs of fighting patents would of course also go away (too bad for the lawyers).
So yes, surprisingly they would still be invented. Then there's the question about whether the generics makers would take all of the profits. Think for a moment about how patents are filed, describing the drug in detail. Without patents, the first drug company to produce a successful drug would have a very large lead over the other companies. The others, once they see the drug is selling well would basically try to reverse engineer it, but in order to produce the drug in significant quantities they would have to retool their factories and that takes time. If typically takes the generics manufacturers about 4 years to get up to speed, and that's about half of the length of a drug patent period anyway. Overall, that means drug companies can still recover their costs, save hugely on legal costs and still make a tidy profit.
The upshot is that after 4 years any given drug of any importance would be very, very cheap everywhere and thus widely accessible by all economic strata.
"all in the name of corporate protectionism and profit"
That is indeed what is going on with "intellectual property." Proponents describe it as an enabler to unfettered innovation in free-market capitalism, but it's really an artificial monopoly designed to provide a form of protectionism for a few lucky companies, and any form of protectionism we all know is anathema to the ideas of a totally free market.
Just the same, the whole world would be better off completely without any form of it.
To save you 300 pages of reading "Against Intellectual Monopoly," basically patents don't spur innovation, not even the ones on concrete inventions. Case after case is presented where it is clear that the idea of spurring innovation through patents is flawed at best, and highly damaging at worst. They basically prove that steam engine development was slowed down by patents and only really began to chug when the patents expired. Inventors you thought were heroes finally come across in a more realistic light. They present lots of examples like this and you just basically see the light (at the end of the tunnel?). Now, if your goal was to slow down technological progress or science itself, then maybe patents would be a good idea.
Before reading some chapters from Boldrin & Levine I was somewhat convinced that copyright at least had some beneficial elements to it that should be respected and preserved, but they sure put the nail in that coffin too. They went through the origins of copyright as a *relaxation* to a censorship regime by the crown (IIRC), and it just went downhill from there. Now it just seems like copyright is extended to every damn little thing, and that wasn't the original purpose of it by far. While they don't prove that removing copyright would be beneficial to everyone, they take a shot at showing that it wouldn't be a total disaster to authors/artists. For everyone else, it wouldn't prevent new books from being written, new music from being produced, etc., and it would be a net gainer, by far.
If you have the time to read a 300 page book this summer, by all means at least read a few chapters of Boldrin & Levine. You will understand intellectual property much better and hopefully lose a few sacred cows in the process.
You can select what you may want to read from this landing page:
http://levine.sscnet.ucla.edu/general/intellectual/againstfinal.htm
In that case I would have expected you to be much better acquainted with Microsoft's track record than what you've been demonstrating.
The fact is, they did send a takedown notice to this very site and it was viewed as censorship and duly caused an uproar. Does that make them stupid? No, it's not quite that. It's more at arrogant. It's what powerful people think they will get away with, and they just miscalculate. They love to take big gambles because when they win, it gives them an unbelievable high that they're smarter than all of those little people. When they lose, it's not quite as pleasant, but optimists like you just think they must be dumb to try it. Not much of a downside to their gambles then, is there?
Whether or not we approve of it, they have foisted it on us, and I don't know about you but the control-freak nature of that concept gives me less of a reason to trust their upgrades. I've drawn the conclusion that closed source software is inherently less trustworthy as a result.
OK, we'll chalk it up to varying experiences of that then.
By the same token, MS had an opportunity to address these in SP3 but they not only went unaddressed, the upgrade itself fell victim to them. Don't say it wasn't in their power to survey the malware out there and prepare for it. This was precisely the sort of bugfix one would expect from a service pack.
One more thing. To the person who dies before the copyright expires it is certainly perpetual.
Since copyright extends beyond one lifetime then such perpetuity is guaranteed for the vast majority of works and the vast majority of people. Thus, the original words "for limited times" no longer really apply. That means that in practice, the current law is unconstitutional.
Everyone (with time to read a 300-page book) needs to read this life-changing book containing the most comprehensive treatment of intellectual property out there: http://levine.sscnet.ucla.edu/papers/imbookfinalall.pdf
I used to think copyrights were beneficial before I read it. Now it's clear they are obsolete.
Stated that way, it sounds like an unlikely location to place blame, but it can just as easily be stated another way: "who has the greatest influence on final quality?" The fact is that Microsoft defines its APIs any way it wants, produces whatever documentation and developer kits it wants, and supports third party products any way it wants, and if they don't want to support you and prefer your business to fail, they can cause that and it wouldn't be the first instance of that.
Be serious. Everyone knows that for-profit companies have to keep moving product to please investors with quarterly reports of high growth rates. Investors love to hear about how a company will prevail against its competition, and if the primary competition is their own product - XP - it's not too "paranoid" to believe that some people are going to pursue that competition diligently... even agressively. Calling me paranoid about this makes you sound like you are really trying hard to shine a positive light on this company - a company with a decidedly spotty history. Once you know about their background of monopoly abuse, expecting a scorched-earth legacy policy from them would be the *least* of things to expect, but you apparently are so new to the IT field that you don't know a bit of their history. If I'm paranoid then you are uninformed or disingenuous.
Of course they don't, but they were counting on the large degree of public ignorance surrounding computers, the effect of their well funded deceptive marketing campaigns, and at last resort their poweful legal arm to bury it with takedown notices to prevent the word from getting out. It wouldn't be the first instance of that either.
Are you honestly suggesting that they're going to compete with themselves and allow that to reduce their revenue? Some of their investors may not like the idea of the public stuck on XP producing no revenue and may try to switch out board members until somebody listens. It's the short-term money they've been historically after, not public acceptance. I wish I could think otherwise, but this sort of breakage is the path of least resistance for MS. They can afford to do it because they are a monopoly with such a gaping lead over anyone else. Who else can the average person turn to?
It's simply Murphy's Law as applied over the quantity of code. The number of bugs is thought to be proportional to the number of lines of code, and the full installation of Vista weighs in about 10-100 times larger than other modern operating systems. Vista is so big that even the old rule of thumb of proportionality may not apply any more.
It doesn't all smell like candy though. Some of the changes dig down deep into the core of the kernel and change its inner workings. A new facility for signed drivers and more stringently controlled and prioritized video/audio paths come to mind. It basically threw away a huge set of old, working drivers and replaced them with a set of new, incomplete drivers. There are a host of potential problems just lurking in this new driver code, simply because they all had to be rewritten. All drivers can affect stability and security.
It may be true that third party vendors and malware are immediately to blame, but even so I find it difficult to believe that the third parties in question would not have done any of their own testing for critical bugs. I find it much easier to believe that it was MS who skipped their part of the testing or failed to provide the appropriate failure modes, developer training, support response, complete documentation, etc. The company isn't exactly promoting XP anymore, so if XP isn't left in a perfect state for posterity I doubt they'll spend too many resources worrying about it.
In a case like this, if the third parties pointed the finger back at MS I would tend to believe them by default instead of believing MS. The third parties have a much stronger motive for their product to actually work. They are not monopolies after all, so they depend heavily on product quality to get any repeat buyers. If MS lets their quality slip, they are not similarly harmed by it - in fact some people will react to these problems by buying a new computer with Vista on it. Thus, this may just be a case where MS finds it convenient to blame someone else and can make the accusation stick.
As to the malware, I see it as an implicit way of blaming the users for either allowing the malware in, or of installing anti-malware which also is blamed for problems. Thus, the hapless user faces a no-win situation no matter which action is taken. You admit Windows' design was suboptimal for security purposes, but on the other hand you haven't fully deduced where the blame belongs. Malware is just a force of nature, it cannot itself be blamed. Neither should the users be expected to know about every part of an overwrought complex system made by someone else and closed to inspection. Thus, that leaves only Microsoft. No one else we know about has enough control and visibility over the entire situation to take the steps to prevent malware from messing with system internals. Blaming MS for actions taken years ago or for a problem set that has grown too large may not seem fair, but it also isn't fair that Microsoft can profitably use the malware on XP as an inducement for people to upgrade to Vista.
Vista's design features may seem like revolutionary advances right now, but this sort of glowing promise is something that always appears at the beginning of every upgrade cycle. Invariably, the next Windows version is touted as having great solutions to the problems of the previous version. In hindsight, we can see that this was done whether or not it actually delivered the perfect solution as it was promised (or hyped). It's not a coincidence that any *new* problems that the current version is quietly introducing aren't emphasized until the next upgrade cycle.
Specifically in the case of Vista you don't seem concerned that its piggish size will be the source of future problems or excuses because you are a "true believer" that its genial new features will be enough to eventually trump most concerns. One only needs to have observed a few upgrade cycles and remembered what was said to see how taking MS' assertions at face value is risky.
And who is to say that Vista will fare any better than XP? Past performance predicts that MS will have to find *something* to use as an inducement for people to upgrade to Windows 7, so even if Vista achieves perfection in its next service pack, I'm betting that it will be undone in SP3 in order to generate sales. More likely, it looks like Vista will never reach XP in terms of the overall user experience.
Now, if you insist that Microsoft's software development methodology isn't broken, I sense it must be because you are excluding the business dec
If software engineering principles were adhered to across the organization, these sorts of risks would be reduced. I suspect that it must be something like office politics or featureitis that must be working to undermine the methodology against the better judgment of the developers, but there is really no excuse for SP3 being less stable than SP2.
It's always easy to pass the blame onto the users doing the installation, but they will resent being written off as the PEBKAC if they upgraded to SP2 without running into problems this severe. Is it too much to ask that MS test for critical bugs before releasing?
An OS that needs to be subverted in order to ensure correct operation doesn't strike me as particularly well-designed. You seem to be saying that it's not worth applying good software development practices to a service pack, since these practices weren't used for the OS itself, prone to malware as it was.
Windows is too big and complex, so the company decides to make Vista an even *bigger* and *more* complex OS? If this problem is going to be solved, they sure have a funny way of heading in the wrong direction. Vista's own SP1 transition didn't inspire the sort of confidence that they were turning the corner, new features and all.
If they can't transition, then they're the victims of their own success. But, I think Microsoft will soon have no choice but to do what you're suggesting. As Vista or its successor grows ever larger, it will need a correspondingly large organization to maintain it, one that itself will be fearsomely challenging to manage. Before investors begin to suspect labor costs spiraling out of control, they'll have to find some way to foist a redesign upon the world.
However, you would be correct to imply that a redesign that sacrifices compatibility will be just as widely criticized as one with quality problems. They could buy up IT companies left and right hoping to find miracle workers, but there won't be any quick or easy solution for this no matter what.
Isn't the original purpose of a service pack to add reliability, rather than take it away?
One would think that by SP3 there would only the most minor bugs left to close, but instead giant new ones are opened. Machines that become unbootable? That's pre-alpha quality stuff.
Something is badly broken with their methodology... no wonder they were trying to do a people grab at Yahoo, the higher ups are probably pulling their hair out by now trying to figure out how to fix their organizational problem and maybe they thought a new project built on BSD (but independent from Apple code) with entirely new staff would bail them out.
I went on to read their in-depth article (linked to the main article) at http://linuxdevices.com/news/NS8981295285.html and I found this:
"Another touted benefit is the ability to consolidate control- and data-plane functions on a single device, with "solid-wall" processor boundaries reinforcing security and licensing containment barrier. In this regard, the Tile64 chip resembles another heavily multicore MIPS64 chip, Cavium's 16-way Octeon."
Does anyone know what the heck a "licensing containment barrier" is? It definitely sounds like a performance hit if it's turned on. And if it's forced to have it on then this design just lost a lot of its sex-appeal.
"What you're talking about sounds like just mashing a CPU core and GPU core together on the same die. Which would be horrible for all kinds of reasons (heat, bus bottlenecks and yields!)."
One would hope that bus bottlenecks would not be a problem if the cores don't have to go off-chip to access a common bus. The bus could be a wider, dedicated video bus if kept on on-chip, and would enjoy vastly reduced latency. It would use less overall power than having an external mobo chip coordinate the handshaking, not to mention the signals that have to be sent to a physical video card. The die itself would be larger than a 1 CPU die, but that's not a problem anymore.
The problem now is what the hell to do with all of the transistors we can cram onto a chip. At some point people won't notice a difference between a 16 core unit and a 32 core unit connected to the same memory. Might as well stick a GPU in there. Heck, throw in two for good measure. If there's a defect in one, you can disable that and still sell the part as the economy version. If that means 2-4 fewer CPUs per die then so be it.
It's not all as horribly problematic as you make it sound. There are some significant savings that can be reaped by putting the two together. Right now they claim to be doing it as a way of producing a cheaper system with slightly slower video-card performance, but there are some opportunities to actually improve performance by doing this, and the benefit of lower power consumption for the overall system is definitely in the plus column.
Actually, if anybody can find the "right" way of doing this, it will probably be AMD/ATI... and then everybody else will just copy what they did, tweak it slightly and give it a different name.
"This case proves only one thing to me, and that is that we'd almost be better off with no system at all. Not quite, but almost."
If you consider Capitol vs. Thomas that is already the case. Paying $222,000 for 24 digital music files is completely vacuous, it lacks any sense of proportion. It's not like Thomas stole a house and has to give it back. It's not like they should become indentured servants to a lender for life due to a computing habit that they can easily quit. Did Capitol run out of digital copies for their songs? Finally, how many live performances could be purchased with that amount of money? It's just ridiculous. The lawyers must be foaming at the mouth.
Having swedish laws in mind, C. v. T. is also a regressive judgment. Clearly, justice here is heavily biased in favor of those who have lots of money, and against those who don't.
In Reiser's case, you get the nagging feeling that the whole story hasn't been told, and you *know* that the jury and the judge are more like the audience in a roman arena cheering and pointing their thumbs up or down according to their biases or preconceptions than they are the solemn, methodical guarantors of justice (I personally think that Nina might have had a motive to disappear to Russia or another place with bribeable authorities if the $50k salary she was offered for an OB-GYN seemed insufficient for the Bay Area, and not knowing what to do she (or her boyfriend) hatched a scheme that went terribly wrong). You get the feeling that a truly impartial investigation, using a real justice system would be able to deliver a fairer judgment, but that just isn't available. No such system exists. You wonder if it would be possible to create one from scratch out of smart, technologically current, impartial elements, but using the rapid rise of astroturfers on this site as an indicator, it seems that that route might not work permanently either. It's a tragedy because the justice system is probably the most crucial of the branches of government that needs to be kept impartial.
No doubt about it, we have a broken justice system. However, having no system at all would result in mob rule, which would actually the same thing: a justice system with uneven quality. The only difference would be a different set of gainers and losers. I guess we should just welcome our gavel-wielding overlords.