Why? Google already has a solid and profitable business model.
The additional projects are just R&D with a very public face. They offer great brand exposure. They give Google's software team lots of experience dealing with the realities of building and maintaining software that thousands to millions of people use every day. They keep the company from developing tunnel vision. And in some cases they've put Microsoft on the defensive, so it has to spend time and resources defending its own markets rather than trying to take over Google's market.
Those are all solid and reasonable benefits for the company, even if none of the projects can ever be turned into a direct-revenue cash cow.
And if one of those projects happens to hit the combination of factors that does allow it to turn into The Next Big Thing, the payoff from that one winner will more than cover the cost of all the other research along the way.
It's pretty much the same plan Apple had for defending the iMac: it's very hard to make a complex system work cleanly and easily.
The iMac would have been a classic example of style over substance if it hadn't been a nice, solid little computer. The candy coating was justified as a statement that the underlying machine was about as good as it could reasonably be. The Wintel market's response was, OTOH, generally halfhearted and lame.
It takes a lot of institutional discipline to make a smoothly-integrated system work. A few idiots in the wrong place can inject enough friction to turn the whole thing into a train wreck. Correction: an expensive train wreck.
Now, this is Microsoft we're talking about: the company whose new (yet much-delayed), flagship operating system is waiting to be released at some still-undisclosed time. The company that's gotten so horribly convoluted and interdependent that no one can even locate the project groups that are dragging everything down. The company with a history of trying to lock customers into proprietary file formats, and is in bed with the content distribution cartels on creating DRM that basically amounts to putting a, "we know you're a fucking thief" label on every device, file, and webpage. And the company that's starting a huge internal reorganization, atop everything else.
The biggest obstacle to Microsoft's success in this is Microsoft itself.
While I don't buy the simplistic "if OS X had as many users as Windows, OS X would have just as many viruses" argument, I do believe in the power of Metcalfe's law: the value of joining a network increases geometrically with the size of the network itself.
Personally, I think the best estimate for expected viruses should be: (installed base * attack surface)^2.
The (installed base * attack surface) value defines the number of potential network connections that malware writers can use, so that number should drive the expected value of the network in terms of attracting malware.
If OS X had the same attack surface as Windows, but still only 1/20th the installed base, I'd expect to see 400 times as many viruses for Windows as for OS X. If the two had equal installed bases, but the Windows attack surface was 20 times as large as the OS X attack surface, I'd still expect to see 400 times as much malware for Windows as for OS X.
The fact that we have something like 10,000 pieces of malware for Windows to essentially nothing for the Mac suggests that the (installed base * attack surface) value for OS X is somewhere around 1/100th of Windows's. Or possibly even less.
Nobody seems to have looked at the basic economics of this whole thing.
SETI@home exists because no company or government would fund computing resources on that scale for that project. If everyday people don't crunch the numbers in their spare time, nobody will. Therefore, the founders of the SETI@home project found a way to harness the power of the Long Tail efficiently.
Medical research, OTOH, has a high expected payoff. If everyday people won't decidate CPU cycles to protein folding problems, drug companies will build their own clusters to handle the load.
So on the one hand, we have a project that will either be done through the efficient aggregation of support from anyone who happens to feel like chipping in a few CPU cycles, or not at all. On the other hand, we have research that attracts both private and government funding, and will be done whether the general public decides to participate or not.
Now comes Lee Gomes -- noted astrophysicist and expert on the allocation of computing resources -- to tell us that the SETI program should be abandoned. It's worthless, and anyone who supports it is wasting precious resources while people die.
The entire article, from start to finish, is hard-packed bullshit, folks. It's only a small step removed from the 'Email/The Web/[Fill In The New Technology Here] Costs Business $N Billion In Lost Productivity Per Year' crap that comes out every 18 months or so. The methodology is exactly the same: point at something people time or energy doing, declare it 'nonproductive', then write a thoroughly unrealistic screed about how great the world could be if people devoted those resources to something 'useful'.
It would take only a small extension of his reasoning to argue that all the CPU cycles 'wasted' on computer games should be devoted to 'important' medical research. One could take the same basic template and argue that Linux and F/OSS are a waste of time and effort: if all those coding resources were channeled into Microsoft's Shared Source program, they could be doing something worthwhile for the vast majority of people who use computers every day.
The fact that the article was posted to Slashdot by a WSJ employee smacks of outright click-whoring. The article itself lacks any meaningful substance. It fails to raise any issue worthy of discussion. It merely defines millions of people as stupid and wasteful because they don't happen to share Lee Gomes's personal set of priorities. It's a long-winded example of hypothesis contrary to fact, with a disingenuous and insulting "not that I'm telling anyone what they should do" coda at the end.
You can't license the idea, but you can license the specific implementation of that idea.. in this case, the lyrics poseted above.
Since the whole argument that Wierd Al wouldn't touch the lyrics above for legal reasons, slapping a BSD license or a Creative Commons license that allows commercial reproduction (maybe with attribution) would indeed give Wierd Al safe-and-legal grounds for making the song.
Sorry, but your assumptions about economic self-determinism are equally naive.
When you participate in a modern economy, you become an indirect supporter of every business in the supply chain of every product you buy. And since the supply chains for most American companies are extremely complex, it's practically impossible to vet everything so it meets your ideological standards. Let's say McDonalds does decide to buy fair trade coffee.. what about the cup? or the plastic spoon? or the components in the coffee maker? or the cash register?
And just saying, "I don't buy McDonald's products," doesn't cut it. If you buy your fair trade coffee in a grocery store, can you say with certainty that the store in question doesn't use sweatshop-built components in its cash registers and back-office computers? Or the paper the receipt was printed on? Or the ribbon for the printer? Or the bag you carry it home in?
Just try to imagine the number of products necessary to keep a store running for one day, then trace those back to their suppliers. Now try to imagine the products necessary to keep those suppliers up and running for a day, and trace them back to their suppliers. The complexity is staggering.
You may as well try to be a FOSS purist by refusing to connect to any computer on the internet that belongs to a company that uses non-FOSS products.
Bruce Schneier once said, of his own work, that he and his colleagues started out with the idea that cryptography was a sort of magic security dust that you could sprinkle on the world and make problems go away. But the more they explored the subject, the more they came to realize that was a false idea.
Encryption has its uses, sure, but it isn't easy. How do you plan to distribute keys, for instance? How do you keep people following reliable security protocols and avoid social engineering attacks? It's such a difficult issue that Schneier wrote a whole book (_Secrets and Lies_) to discuss it.
More generally, the feds won't have to ban encryption, but they can use encryption as a trigger for algorithmic analysis of data. Ed Felten is discussing that very subject now in his _Freedom to Tinker_ blog. As he puts it, the idea of objecting to a computer scanning your communications is kind of like freaking out because your dog has seen you naked. What happens in a computer is fine as long as it stays in the computer. The things we object to revolve around human access to information that was selected by algorithmic means.
In that context, using encryption would actually be more of a security risk than a benefit. Finding all data transmissions that originate from, or end with, an arab immigrant in the US is little more than a database lookup. Scanning those for transmissions whose entropy is high enough to make them look like line noise (a sign of good encryption) is almost trivial. If those two simple tools cough up a cluster of people communicating through encrypted channels, you devote more resources to investigating those specific individuals. If you find something that constitutes probable cause, you can then go to a judge and ask for a warrant or some other authority to investigate. If they turn out to be a cypherpunk play-by-email chess club, fine.. you drop the investigation. But even strict privacy advocates would have a hard time saying you were completely unjustified in checking them out.
In that scenario, you don't even need to know the contents of the messages to identify the cluster of people in question. All you need is a pattern of communication. The encryption itself doesn't amount to probable cause, any more than failing to signal a turn is probable cause for drunk driving.. but it's enough of an excuse for a cop to pull you over at 2:30 am and get a whiff of your breath.
Creative contacted Apple and spent months trying to negotiate a license. Even if they do win their infringement case against Apple, that behavior indicating intent to license is a matter of record. It will have to be considered as part of the four-part test recently endorsed by the supreme court in deciding whether an injunction is warranted, or whether something less will do.
There's no way in hell Creative, in this specific case, could get a blank check to shut down Apple's whole iPod business. You can haul generalities out of your own ass all day, but you can't make them correspond to actual legal practice. So let's just agree to disagree, and when Apple shuts down its entire iPod line because Creative got a ridiculously over-broad injunction, I'll be the first to say you have bragging rights on the "I told you so" circuit.
Advocacy to the contrary, this ruling doesn't say much. The two sides are still arranging their pieces on the board, and the major facts of the case are still in play.
This is really just an example of the adversarial legal system in action. Both sides state their ideas in the strongest terms they possibly can, then the other side gets a chance to chip away as much it can.
In the previous round, the bloggers floated the idea that anyone who puts information on the internet is a journalist, and that anyone who posts protected information should receive the same legal protection as a whistleblower. The court didn't buy that, nor should anyone have expected it to. But that's where the defense started, because it would have been the simplest, strongest win they could get. All they really lost was the right to claim blanket immunity from prosectution for anyone, anywhere, under pretty much any circumstances.
Now it's Apple's turn. Apple floated the idea that it should get a free pass for discovery since the information in question was vastly important, and that the bloggers had no possible interest in publishing it. The court didn't buy that, either. Had the bloggers posted the product's schematics, or a discussion of some new, patentable idea that Apple had been working on, the decision probably would have gone the other way.
So as things stand now, the bloggers can't make the case go away on the grounds of blanket immunity, and Apple can't ask the court to fast-track its subpoenas because of the massive-and-ongoing damage it received. Neither of those was really a viable claim in the first place, but that's how the game is played.
The courts still have to rule on whether Apple has done sufficient work trying to find the leak by other means, and the bloggers still have to face questions about whether they knew the information they posted was confidential, and put it online anyway.
And NONE of this has anything to do with the question of "whether bloggers are journalists."
You make a decent argument in the microscale, but you don't consider the macroscale. In epidemiological terms, one business equals one organism. A monoculture exists when all business/organisms are vulnerable to the same bug and each new infected organism becomes a vector for infection of its neighbors.
Your arguments about the cost of diversity fall flat when we assume that no more than 25% of businesses will run any given OS as an internal monoculture, though.. Business W is 100% FreeBSD; business X is 100% Windows; business Y is 100% Solaris; business Z is 100% OS X, etc.
A virus that wipes out all the Windows businesses will still leave 75% of the ecosystem up and running. No individual business sees any untoward cost from having to maintain internal diversity. Nor does the ecosystem between businesses suffer any increased cost form such diversity, because nobody is paying to keep everyone in synch.
In fact, it costs more to maintain a monoculture at that level than it does to maintain diversity. Imagine trying to put the entire Fortune 500 list of companies on exactly the same suite of computing resources. The synchronization costs would be horrific.
First of all, the whole "injunctively forc[ing] Apple to stop selling iPods, as we know them, until they completely redesign the interface," business only comes into play if Apple loses the case and refuses to pay Creative its licensing fees.
The worst reasonable scenario is that the courts force Apple to accept the license Creative offered in the first place and pay any back fees that apply to the iPods already sold.. in other words, pretty much what Creative wanted before taking the matter to court. The cash value of that downside is the $1-1.5B that Creative has been tossing around in its press releases.
Second, you apparently missed the Supreme Court's recent (unanimous) ruling WRT eBay, in which the justices said that plaintiffs are not automatically entitled to an injunction that shuts down a major product line simply because they've proved infringement.
Assuming Creative does manage to prove infringement, they'll still face a four-part test to determine whether shutting down Apple's iPod business is the only acceptable form of redress. Among other things, they'll have to prove that a cash settlement wouldn't be adequate, which will be difficult, because this whole thing started when Creative approached Apple with regard to a technology license.
Granted the ruling came down after Creative filed its suit, but the four-part test in question is already well established. eBay's argument to the Supremes was basically, "the lower court didn't use this test when it should have," and the majority opinion was pretty much, "yup."
So.. Creative's chance of winning its case still looks pretty low. Its chance of getting an injunction against the iPod, even if it does win, looks equally low. The chance of Apple actually shutting down its iPod business rather than just coughing up the cash is nil. The amount of cash a court would order Apple to cough up is pretty much the same amount Creative requested before filing its lawsuit.
Running those numbers, and using a standard strategy matrix, we have Apple settling with Creative and having a 100% chance of paying some amount.. call it $-X, since it's negative cashfllow for Apple. There's no winning or losing there, so we compare $-X to both of Apple's options for taking the case to court.
If Apple takes the matter to court and wins, it has a high probability of paying nothing. Call that probability P, and the cost to Apple of winning would be P * 0, or zero.. much better than $-X.
If Apple takes the matter to court and loses, it has a low probability of paying $-X, or thereabouts. Call that probability 1-P, and the cost to Apple of losing would be (1-P) * $-X. Since (1-P) is less than 1, that's also better than $-X.
Now in game theory, when every payout for one course of action is better than the corresponding payout for another course of action, that course of action is what we call a dominant strategy. Apple's taking the matter to court demonstrates that quality, so taking the matter to court is a dominant strategy for Apple.
The history of human interaction, illustrated with monkeys:
Phase 1: When two monkeys want the same banana, the big one gets it, and the little one writes a treatise on ethics.
Phase 2: Eventually, monkey-kind learns that a dozen little monkeys with nothing left to lose can kick the living shit out of a single, well-fed big monkey. Shortly thereafter, the big monkeys hold a press conference to announce their discovery of, and commitment to, the principle of enlightened self-interest.
Phase 3: From then on, all monkeys spend their time trying to invent a way to be: A) big enough to take bananas by force whenever they want, but B) 'small' enough, by some definition, to claim bananas by moral entitlement.
I know enough game theory to predict that your risk-aversive strategy will get you 5 years in Siberia in a Prisoner's Dilemma.;-)
I also think you haven't run the numbers. Game theory doesn't just say "avoid large risks." It says that you have to multiply the cost of losing by the chance of losing, then compare that to the cost of settling. Assuming your 80% chance of Apple winning is correct, and assuming Creative will get $1B if it wins, game theory says Apple should take the issue to court if settling with Creative will cost more than $200 million.
And that doesn't take Apple's four countersuits into consideration. If we give Apple the same 80% chance of winning those as in the Creative suit, the odds of Apple losing the whole $1B drop to about 3/100 of 1% (.2^5). The probability-adjusted expected loss on that is only about $320k. And if Apple wins any of its countersuits, those damages will cut into the damages awarded to Creative, so the $200 million settlement cost is an absolute maximum.
Beyond that, I find it highly unlikely that Creative would get the full billion even if Apple did lose. In practice, the court is more likely to set damages at a figure fairly close to what Creative wanted originally in licensing fees. To put it another way, Creative's lawyers would have to be utterly stupid to set their licensing fees significantly lower than what a court would give them. A bit lower, yes.. into the four-times-lower range, no.
Now, given the value of the portable music player market, it wouldn't take long for Creative to cost Apple $200 million in lost sales assuming Apple did choose to strike a cross-licensing deal.
Meanwhile, Apple has about a 30% chance of winning the Creative suit and all its countersuits, and even higher odds of winning Creative and at least some of its countersuits. In all of those scenarios, Apple comes out in the black on the litigation, and Creative suffers losses that will hurt its ability to compete with the iPod. That translates to some amount of additional revenue for Apple (figure 80% of Creative's lost sales, since the iPod has 80% market share), which translates to an even bigger win for Apple.
First, you only settle when you clearly expect to lose. I don't see Apple in that position, here. The Creative suit came out of a failed negotiation, so Apple already had the chance to surrender if they wanted to. Since they didn't, they probably don't expect to lose in court.
Second, settling only makes sense if the cost of losing the lawsuit outweighs the cost of splitting the market.
Given the size and value of the market, that cost would have to be huge. Personally, I don't see Creative swinging that much of a threat. Again, the fact that Apple didn't negotiate a license when they had the chance suggests that they didn't see Creative's claim as a major problem. At very least, I'd expect them to fight to limit the damages they pay Creative if they do lose.
Besides, as noted, Apple has already filed four countersuits. Even if they lose the Creative action, they have the potential to win back everything they lost (and more) by the time all the dust settles.
Dude.. everyone knows that the consumer electronics market shows seasonal variations. That's why analysts use year-over-year statistics rather than quarter-over-quarter.
The fact that the iPod has shown consistent Q/Q growth showed that the market was still in the process of discovering the product. The only 'saturation point' we've reached is in mindshare: there's nobody left who hasn't heard of an 'iPod' by now.
This does not signify the end of the iPod market. It marks the fact that the iPod market now, officially, exists. Now it's time to pay attention to Apple's Y/Y iPod sales growth, which, BTW, is phenomenal.
Rubbish. Why should Apple share a market it already dominates?
Look.. there's always a grey zone around patents. Most patent holders only sue over clear violations that hurt them in the market, because it takes a lot of very expensive time and effort to show just exactly if, and how much, product X infringes on roughly-similar patent Y. There's generally no payoff in being anal about marginal violations, and there are two strong benefits to taking a live-and-let-live approach:
First, it keeps business flowing smoothly. As long as everyone makes a general effort to do their own R&D, the climate stays friendly. The lawyers only come out when two companies follow the same line of grey-zone innovations to the same valuable result, and even then, they negotiate more than litigate.
Second, it's nice to have the option to drop the live-and-let-live attitude if someone gets difficult.
That's what's happening here. Apple was happy to let things slide as long as Creative was willing to play nice, too. But Creative knows that Apple has market dominance, first-mover advantage, economies of scale, phenomenal branding, and all that good stuff. There's no easy way for Creative to win that fight. Creative's management decided that their chance of making money by being anal over a possible marginal patent infringement was better than their chance of making money by maintaining the live-and-let-live mindset, so they sued.
Apple responded by showing what it could do if it chose to get anal, too.
There's nothing for Creative to win, here. Apple has no reason to settle, and plenty of resources to defend itself in court if it has to. The most likely result will be for Creative to quietly drop its suit in exchange for Apple doing the same, and both sides will go back to playing live-and-let-live, just like before.
Creative can't even win a crosss-licensing deal, because Apple has ponied up four patents to Creative's one. If anything, Creative will be lucky to get away without having to pay licensing fees to Apple.
'Is riding' is the imperfect tense of the verb 'to ride'. It indicates continuing action. It describes the boy's state between the time when he starts riding the bike and the time when he reaches the store.
'Rode' is the simple past tense of 'to ride'. It indicates a single completed action. It describes the boy's state after he reaches the store.
The passsive voice inverts the actor and the thing acted upon, and often ignores the actor entirely: The bike was ridden to the store.
The passive voice emphasizes the action over the actor. If we only care how the bike got to the store (this one was ridden, that one was shipped in a truck), using the passive is fine. Usually, though, readers want to know who did something.
In business communication, the passive voice is a great way of hiding high-level muddleheadedness. It gives managers a way to talk about what needs to be done without ever narrowing things down to a specific group of people who should do it, or assigning them the resources they'll need to do the job.
In general, using the passive suggests that you don't really know what you're saying, yet. You have a general idea, but haven't spent enough time thinking about it to get down to specifics. If you know for a fact that only the actions matter (one of the bike's went through Mrs. Grumpy's garden, there are only two bikes present, one was ridden, the other came in a truck) go ahead and use the passive. Otherwise, keep working on the idea until you can express it in 'X does Y' terms.
This is the internet. Around here, when something seems a little bit too stupid to be true, it's still on the leading edge of coherent thought.;-)
Besides, you Brits don't know sarcasm, you know satire. You have your lah-di-dah Jonathan Swift, Oscar Wilde, and all that lot. We Yanks produced Ambrose Bierce, who was one of the most unflinchingly sarcastic bastards ever to drip acid from the tip of a pen.
Subtlety has a place in satire, which is usually compared to a rapier. Sarcasm, on the other hand, is a saw. It derives its strength from stating the truth brutally, but with style.
Satire: Suggesting that Louderback might want to take a freshman-level creative writing class or two before trying to get any more of his "humorous" work published (and advising him not to give up his day job in the meantime, because editors can be tough cookies out there in the real world).
Sarcasm: Calling the article a piece of tripe that any self-respecting writer would have tossed into the garbage can along with all the other tissues coated with masturbatory residue.
Actually, it originally meant "screwed by your employer."
The original quote is from Hamlet: "For 'tis the sport to have the enginer / Hoist with his owne petar." There's nothing passive or accidental about that. It means, "once the guy has built the bomb and deployed it, it's funny to tie him up and hang him right next to it."
That reading fits the line's context extremely well: Hamlet has just stolen the letter Rosencrantz and Guildenstern are carrying to the King of England (which says: "Kill Hamlet") and replaced it with his own (which says: "Kill the men who give you this letter").
One little-known fact about Shakespeare is that he spiced his work up with controversies about military technology. This line demonstrates the love-hate relationship kings had with the engineers who handled explosives: it's damned useful to be able to blow down a door during a seige, but you don't want the guy who did it hanging around and possibly blowing down your own door next time. Besides, the royals of that day found it offensive to think that a commoner with a barrel full of smelly chemicals could wield that kind of power. It leads to the idea of commoners having the capacity to blow the bejeezus out of royalty, and that was just an affront to what they considered the natural order of things.
(and Romeo and Juliet -- if you know your historical weaponry -- revolves around the introduction of rapier fighting. Tybalt, "the captain of courageous complements" uses a parry-and-thrust fighting style, while the Montagues use the older block-and-slash style. The newer style could be faster and more lethal, but only as long as you did it right. All the fighting between the two families is based on the two gangs with different styles working out who has the power to take the other down)
Folks... back in the old days, we had the right to share music files with our friends, make backup copies, and copy files to different media.
No we didn't. We were just breaking the law in a context where nobody cared.
I'd say I've made fewer than 250 mix tapes and CDs over the past 30 years. I have friends who may have made a thousand, and that's fairly active for old-school swapping.
A P2P file sharer can beat those numbers in a single day.
On top of that, the tapes and CDs I made all stayed within my social network. I didn't go around my college putting "Get a bootleg tape of the latest Metallica album" flyers on every phone pole and bulletin board I could find. Can those of you who've filled your 80 Gig drives with P2P music pull a random track out of the collection and say you know the person who gave it to you?
Probably not.
Internet file sharing supports vastly larger numbers and vastly wider peer networks than good old-fashioned, "hey, I made you a tape," copyright violation. And like it or not, that huge increase in scale has changed the landscape. File sharing is no longer a trivial nuisance. The jury is still out on whether it's a beneficial manifestation of word-of-mouth advertising or a parasitic drain on legitimate revenue channels, but either way, it's big enough to be worth noticing these days.
1. I should not be allowed to use a car unless I can prove that I have a driver's license. Mag strip swipe by the ignition maybe?
If you get arrested for drunk driving and your BAC is above a certain limit, you may have to get your ignition lock fitted with a breathalyzer before you can get your license back. So yes, the state could enforce that precise limit if it wanted to, and thought that the benefit outweighed the cost.
Greymail is a fairly common legal tactic. You try to force your opponent to choose between 'losing the case' and 'giving up an important secret'. In criminal cases, it tends to be used as an attempt to make the prosecutors drop a charge. In civil cases, like this one, it's usually an attempt to get the defendant to settle.
From what I can see, and what I recall of previous pageRank suits against Google, this one looks like a non-starter. Google has never promised anyone that their pageRank score will remain the same.. in fact, they pretty much guarantee that pageRank will change whenever new pages about a subject appear, old pages about a subject disappear, or the content of existing pages changes. Neither has Google ever promised, or even suggested, that their pageRank algorithms will remain stable and predictable. They're pretty open about the fact that they're constantly tuning and adjusting their system.
The plaintiff has no contractual claim against Google. They didn't pay for a Sponsored Link, and even those positions are subject to change based on what other people are willing to pay for the same keywords. Neither can the plaintiff claim that Google specifically altered their rankings to injure that company (and even then, Google has won suits against pageRank spammers who sued because Google broke their system).
When we get right down to it, we have a plaintiff that took opportunistic advantage of a good pageRank, and then suffered a reversal when the weather changed. Now they've brought a lawsuit against Google claiming specific and prejudicial injury, but they can't prove that unless they see Google's pageRank algorithm. Thus, they hope they can force Google will choose to settle rather than being forced to defend their right to keep the pageRank algorithms secret.
Like I said, it looks to me like a non-starter. Google's argument to the judge will probably run along these lines:
1 - The plaintiff doesn't have standing to sue Google on contractual grounds. The only way plaintiff can bring suit at all is to claim that Google is a public service.
2.1 - Assuming for the moment that Google is a public service, adhering to the plaintiff's concept of fairness would force Google to do one of two things: publish its pageRank algorithm, or provide sufficient notice of change that its pageRank algorithm could easily be reverse-engineered by a third party.
2.2 - If the pageRank algorithm was public knowledge, third parties (aka: spammers) would abuse it to assign pageRanks based on a client's willingness to pay, not on any standard of relevance or utility to the person doing the search.
2.3 - Allowing spammers to hijack the pageRank system would destroy the 'public service' the plaintiff claims to exist.
3 - But Google isn't a public service.
3.1 - Google never promised the plaintiff (or anyone else) a consistently high pageRank.
3.2 - Google doesn't give anyone 'warning' that their pageRank is about to change.
3.3 - The change in the client's pageRank happened for one of the following 'reasons':
3.3.1 - Someone else's pageRank got better.
3.3.2 - The plaintiff's page passed its freshness date, and the fact that new pages get a higher rank than old, unchanged pages is public knowledge.
3.3.3 - Google adjusted its pageRank algorithms to make searches even better, and this is only one of millions of resulting changes.
That's great, except that the ACID2 test is about CSS compliance which does indeed specify a particular page rendering.
Point taken, but the original question didn't mention ACID2 or DTD specification. It was a general complaint against the (W3C recommended) policy of being forgiving about what you're willing to parse.
ACID2 is a set of standard rendering conventions (and very valuable), but it doesn't trump the need to be generous with badly- or poorly-formed HTML. If an XHTML browser renders:
<ol> <li> item 1 <li> item 2 </ol>
the same way it renders:
<ol> <li> item 1 </li> <li> item 2 </li> </ol>
and both meet the ACID2 criteria for unordered lists, is the browser well-written or not?
The first version is clearly malformed XHTML, so should the browser just forgive the missing close tags, or should it barf out a badly-rendered page?
We have 30-odd years of data that show a strong correlation between ocean surface temperature and the force of hurricanes. Fine. We also have a very long history of periodic fluctuation in the force of hurricanes over time.
So why does this correlation only work today, as a 'result' of global warming? Why can't we use it to infer the existence of a periodic variation in ocean surface temperatures that corresponds to the observed history of strong hurricanes? More specifically, why can't we use this correlation and the history of strong hurricanes to show that ocean surface temperatures were rising and falling as much as they are today, long before the 150-year period associated with global warming?
If this round of strong hurricanes are the result of global warming, but none of the others in history were, what 'missing mechanism' produced all those previous cycles of strong hurricanes, and why isn't that mechanism working today?
Slashdot Mirror
Why? Google already has a solid and profitable business model.
The additional projects are just R&D with a very public face. They offer great brand exposure. They give Google's software team lots of experience dealing with the realities of building and maintaining software that thousands to millions of people use every day. They keep the company from developing tunnel vision. And in some cases they've put Microsoft on the defensive, so it has to spend time and resources defending its own markets rather than trying to take over Google's market.
Those are all solid and reasonable benefits for the company, even if none of the projects can ever be turned into a direct-revenue cash cow.
And if one of those projects happens to hit the combination of factors that does allow it to turn into The Next Big Thing, the payoff from that one winner will more than cover the cost of all the other research along the way.
It's pretty much the same plan Apple had for defending the iMac: it's very hard to make a complex system work cleanly and easily.
The iMac would have been a classic example of style over substance if it hadn't been a nice, solid little computer. The candy coating was justified as a statement that the underlying machine was about as good as it could reasonably be. The Wintel market's response was, OTOH, generally halfhearted and lame.
It takes a lot of institutional discipline to make a smoothly-integrated system work. A few idiots in the wrong place can inject enough friction to turn the whole thing into a train wreck. Correction: an expensive train wreck.
Now, this is Microsoft we're talking about: the company whose new (yet much-delayed), flagship operating system is waiting to be released at some still-undisclosed time. The company that's gotten so horribly convoluted and interdependent that no one can even locate the project groups that are dragging everything down. The company with a history of trying to lock customers into proprietary file formats, and is in bed with the content distribution cartels on creating DRM that basically amounts to putting a, "we know you're a fucking thief" label on every device, file, and webpage. And the company that's starting a huge internal reorganization, atop everything else.
The biggest obstacle to Microsoft's success in this is Microsoft itself.
Well, we are talking about a company that markets a product called 'WinCE'.
While I don't buy the simplistic "if OS X had as many users as Windows, OS X would have just as many viruses" argument, I do believe in the power of Metcalfe's law: the value of joining a network increases geometrically with the size of the network itself.
Personally, I think the best estimate for expected viruses should be: (installed base * attack surface)^2.
The (installed base * attack surface) value defines the number of potential network connections that malware writers can use, so that number should drive the expected value of the network in terms of attracting malware.
If OS X had the same attack surface as Windows, but still only 1/20th the installed base, I'd expect to see 400 times as many viruses for Windows as for OS X. If the two had equal installed bases, but the Windows attack surface was 20 times as large as the OS X attack surface, I'd still expect to see 400 times as much malware for Windows as for OS X.
The fact that we have something like 10,000 pieces of malware for Windows to essentially nothing for the Mac suggests that the (installed base * attack surface) value for OS X is somewhere around 1/100th of Windows's. Or possibly even less.
Nobody seems to have looked at the basic economics of this whole thing.
SETI@home exists because no company or government would fund computing resources on that scale for that project. If everyday people don't crunch the numbers in their spare time, nobody will. Therefore, the founders of the SETI@home project found a way to harness the power of the Long Tail efficiently.
Medical research, OTOH, has a high expected payoff. If everyday people won't decidate CPU cycles to protein folding problems, drug companies will build their own clusters to handle the load.
So on the one hand, we have a project that will either be done through the efficient aggregation of support from anyone who happens to feel like chipping in a few CPU cycles, or not at all. On the other hand, we have research that attracts both private and government funding, and will be done whether the general public decides to participate or not.
Now comes Lee Gomes -- noted astrophysicist and expert on the allocation of computing resources -- to tell us that the SETI program should be abandoned. It's worthless, and anyone who supports it is wasting precious resources while people die.
The entire article, from start to finish, is hard-packed bullshit, folks. It's only a small step removed from the 'Email/The Web/[Fill In The New Technology Here] Costs Business $N Billion In Lost Productivity Per Year' crap that comes out every 18 months or so. The methodology is exactly the same: point at something people time or energy doing, declare it 'nonproductive', then write a thoroughly unrealistic screed about how great the world could be if people devoted those resources to something 'useful'.
It would take only a small extension of his reasoning to argue that all the CPU cycles 'wasted' on computer games should be devoted to 'important' medical research. One could take the same basic template and argue that Linux and F/OSS are a waste of time and effort: if all those coding resources were channeled into Microsoft's Shared Source program, they could be doing something worthwhile for the vast majority of people who use computers every day.
The fact that the article was posted to Slashdot by a WSJ employee smacks of outright click-whoring. The article itself lacks any meaningful substance. It fails to raise any issue worthy of discussion. It merely defines millions of people as stupid and wasteful because they don't happen to share Lee Gomes's personal set of priorities. It's a long-winded example of hypothesis contrary to fact, with a disingenuous and insulting "not that I'm telling anyone what they should do" coda at the end.
You can't license the idea, but you can license the specific implementation of that idea.. in this case, the lyrics poseted above.
Since the whole argument that Wierd Al wouldn't touch the lyrics above for legal reasons, slapping a BSD license or a Creative Commons license that allows commercial reproduction (maybe with attribution) would indeed give Wierd Al safe-and-legal grounds for making the song.
Whether he wants to is another question.
Sorry, but your assumptions about economic self-determinism are equally naive.
When you participate in a modern economy, you become an indirect supporter of every business in the supply chain of every product you buy. And since the supply chains for most American companies are extremely complex, it's practically impossible to vet everything so it meets your ideological standards. Let's say McDonalds does decide to buy fair trade coffee.. what about the cup? or the plastic spoon? or the components in the coffee maker? or the cash register?
And just saying, "I don't buy McDonald's products," doesn't cut it. If you buy your fair trade coffee in a grocery store, can you say with certainty that the store in question doesn't use sweatshop-built components in its cash registers and back-office computers? Or the paper the receipt was printed on? Or the ribbon for the printer? Or the bag you carry it home in?
Just try to imagine the number of products necessary to keep a store running for one day, then trace those back to their suppliers. Now try to imagine the products necessary to keep those suppliers up and running for a day, and trace them back to their suppliers. The complexity is staggering.
You may as well try to be a FOSS purist by refusing to connect to any computer on the internet that belongs to a company that uses non-FOSS products.
Bruce Schneier once said, of his own work, that he and his colleagues started out with the idea that cryptography was a sort of magic security dust that you could sprinkle on the world and make problems go away. But the more they explored the subject, the more they came to realize that was a false idea.
Encryption has its uses, sure, but it isn't easy. How do you plan to distribute keys, for instance? How do you keep people following reliable security protocols and avoid social engineering attacks? It's such a difficult issue that Schneier wrote a whole book (_Secrets and Lies_) to discuss it.
More generally, the feds won't have to ban encryption, but they can use encryption as a trigger for algorithmic analysis of data. Ed Felten is discussing that very subject now in his _Freedom to Tinker_ blog. As he puts it, the idea of objecting to a computer scanning your communications is kind of like freaking out because your dog has seen you naked. What happens in a computer is fine as long as it stays in the computer. The things we object to revolve around human access to information that was selected by algorithmic means.
In that context, using encryption would actually be more of a security risk than a benefit. Finding all data transmissions that originate from, or end with, an arab immigrant in the US is little more than a database lookup. Scanning those for transmissions whose entropy is high enough to make them look like line noise (a sign of good encryption) is almost trivial. If those two simple tools cough up a cluster of people communicating through encrypted channels, you devote more resources to investigating those specific individuals. If you find something that constitutes probable cause, you can then go to a judge and ask for a warrant or some other authority to investigate. If they turn out to be a cypherpunk play-by-email chess club, fine.. you drop the investigation. But even strict privacy advocates would have a hard time saying you were completely unjustified in checking them out.
In that scenario, you don't even need to know the contents of the messages to identify the cluster of people in question. All you need is a pattern of communication. The encryption itself doesn't amount to probable cause, any more than failing to signal a turn is probable cause for drunk driving.. but it's enough of an excuse for a cop to pull you over at 2:30 am and get a whiff of your breath.
Don't confuse generalities with specific cases.
Creative contacted Apple and spent months trying to negotiate a license. Even if they do win their infringement case against Apple, that behavior indicating intent to license is a matter of record. It will have to be considered as part of the four-part test recently endorsed by the supreme court in deciding whether an injunction is warranted, or whether something less will do.
There's no way in hell Creative, in this specific case, could get a blank check to shut down Apple's whole iPod business. You can haul generalities out of your own ass all day, but you can't make them correspond to actual legal practice. So let's just agree to disagree, and when Apple shuts down its entire iPod line because Creative got a ridiculously over-broad injunction, I'll be the first to say you have bragging rights on the "I told you so" circuit.
Advocacy to the contrary, this ruling doesn't say much. The two sides are still arranging their pieces on the board, and the major facts of the case are still in play.
This is really just an example of the adversarial legal system in action. Both sides state their ideas in the strongest terms they possibly can, then the other side gets a chance to chip away as much it can.
In the previous round, the bloggers floated the idea that anyone who puts information on the internet is a journalist, and that anyone who posts protected information should receive the same legal protection as a whistleblower. The court didn't buy that, nor should anyone have expected it to. But that's where the defense started, because it would have been the simplest, strongest win they could get. All they really lost was the right to claim blanket immunity from prosectution for anyone, anywhere, under pretty much any circumstances.
Now it's Apple's turn. Apple floated the idea that it should get a free pass for discovery since the information in question was vastly important, and that the bloggers had no possible interest in publishing it. The court didn't buy that, either. Had the bloggers posted the product's schematics, or a discussion of some new, patentable idea that Apple had been working on, the decision probably would have gone the other way.
So as things stand now, the bloggers can't make the case go away on the grounds of blanket immunity, and Apple can't ask the court to fast-track its subpoenas because of the massive-and-ongoing damage it received. Neither of those was really a viable claim in the first place, but that's how the game is played.
The courts still have to rule on whether Apple has done sufficient work trying to find the leak by other means, and the bloggers still have to face questions about whether they knew the information they posted was confidential, and put it online anyway.
And NONE of this has anything to do with the question of "whether bloggers are journalists."
You make a decent argument in the microscale, but you don't consider the macroscale. In epidemiological terms, one business equals one organism. A monoculture exists when all business/organisms are vulnerable to the same bug and each new infected organism becomes a vector for infection of its neighbors.
Your arguments about the cost of diversity fall flat when we assume that no more than 25% of businesses will run any given OS as an internal monoculture, though.. Business W is 100% FreeBSD; business X is 100% Windows; business Y is 100% Solaris; business Z is 100% OS X, etc.
A virus that wipes out all the Windows businesses will still leave 75% of the ecosystem up and running. No individual business sees any untoward cost from having to maintain internal diversity. Nor does the ecosystem between businesses suffer any increased cost form such diversity, because nobody is paying to keep everyone in synch.
In fact, it costs more to maintain a monoculture at that level than it does to maintain diversity. Imagine trying to put the entire Fortune 500 list of companies on exactly the same suite of computing resources. The synchronization costs would be horrific.
First of all, the whole "injunctively forc[ing] Apple to stop selling iPods, as we know them, until they completely redesign the interface," business only comes into play if Apple loses the case and refuses to pay Creative its licensing fees.
The worst reasonable scenario is that the courts force Apple to accept the license Creative offered in the first place and pay any back fees that apply to the iPods already sold.. in other words, pretty much what Creative wanted before taking the matter to court. The cash value of that downside is the $1-1.5B that Creative has been tossing around in its press releases.
Second, you apparently missed the Supreme Court's recent (unanimous) ruling WRT eBay, in which the justices said that plaintiffs are not automatically entitled to an injunction that shuts down a major product line simply because they've proved infringement.
Assuming Creative does manage to prove infringement, they'll still face a four-part test to determine whether shutting down Apple's iPod business is the only acceptable form of redress. Among other things, they'll have to prove that a cash settlement wouldn't be adequate, which will be difficult, because this whole thing started when Creative approached Apple with regard to a technology license.
Granted the ruling came down after Creative filed its suit, but the four-part test in question is already well established. eBay's argument to the Supremes was basically, "the lower court didn't use this test when it should have," and the majority opinion was pretty much, "yup."
So.. Creative's chance of winning its case still looks pretty low. Its chance of getting an injunction against the iPod, even if it does win, looks equally low. The chance of Apple actually shutting down its iPod business rather than just coughing up the cash is nil. The amount of cash a court would order Apple to cough up is pretty much the same amount Creative requested before filing its lawsuit.
Running those numbers, and using a standard strategy matrix, we have Apple settling with Creative and having a 100% chance of paying some amount.. call it $-X, since it's negative cashfllow for Apple. There's no winning or losing there, so we compare $-X to both of Apple's options for taking the case to court.
If Apple takes the matter to court and wins, it has a high probability of paying nothing. Call that probability P, and the cost to Apple of winning would be P * 0, or zero.. much better than $-X.
If Apple takes the matter to court and loses, it has a low probability of paying $-X, or thereabouts. Call that probability 1-P, and the cost to Apple of losing would be (1-P) * $-X. Since (1-P) is less than 1, that's also better than $-X.
Now in game theory, when every payout for one course of action is better than the corresponding payout for another course of action, that course of action is what we call a dominant strategy. Apple's taking the matter to court demonstrates that quality, so taking the matter to court is a dominant strategy for Apple.
The history of human interaction, illustrated with monkeys:
Phase 1: When two monkeys want the same banana, the big one gets it, and the little one writes a treatise on ethics.
Phase 2: Eventually, monkey-kind learns that a dozen little monkeys with nothing left to lose can kick the living shit out of a single, well-fed big monkey. Shortly thereafter, the big monkeys hold a press conference to announce their discovery of, and commitment to, the principle of enlightened self-interest.
Phase 3: From then on, all monkeys spend their time trying to invent a way to be: A) big enough to take bananas by force whenever they want, but B) 'small' enough, by some definition, to claim bananas by moral entitlement.
I know enough game theory to predict that your risk-aversive strategy will get you 5 years in Siberia in a Prisoner's Dilemma. ;-)
I also think you haven't run the numbers. Game theory doesn't just say "avoid large risks." It says that you have to multiply the cost of losing by the chance of losing, then compare that to the cost of settling. Assuming your 80% chance of Apple winning is correct, and assuming Creative will get $1B if it wins, game theory says Apple should take the issue to court if settling with Creative will cost more than $200 million.
And that doesn't take Apple's four countersuits into consideration. If we give Apple the same 80% chance of winning those as in the Creative suit, the odds of Apple losing the whole $1B drop to about 3/100 of 1% (.2^5). The probability-adjusted expected loss on that is only about $320k. And if Apple wins any of its countersuits, those damages will cut into the damages awarded to Creative, so the $200 million settlement cost is an absolute maximum.
Beyond that, I find it highly unlikely that Creative would get the full billion even if Apple did lose. In practice, the court is more likely to set damages at a figure fairly close to what Creative wanted originally in licensing fees. To put it another way, Creative's lawyers would have to be utterly stupid to set their licensing fees significantly lower than what a court would give them. A bit lower, yes.. into the four-times-lower range, no.
Now, given the value of the portable music player market, it wouldn't take long for Creative to cost Apple $200 million in lost sales assuming Apple did choose to strike a cross-licensing deal.
Meanwhile, Apple has about a 30% chance of winning the Creative suit and all its countersuits, and even higher odds of winning Creative and at least some of its countersuits. In all of those scenarios, Apple comes out in the black on the litigation, and Creative suffers losses that will hurt its ability to compete with the iPod. That translates to some amount of additional revenue for Apple (figure 80% of Creative's lost sales, since the iPod has 80% market share), which translates to an even bigger win for Apple.
But hey, what do I know about game theory?
First, you only settle when you clearly expect to lose. I don't see Apple in that position, here. The Creative suit came out of a failed negotiation, so Apple already had the chance to surrender if they wanted to. Since they didn't, they probably don't expect to lose in court.
Second, settling only makes sense if the cost of losing the lawsuit outweighs the cost of splitting the market.
Given the size and value of the market, that cost would have to be huge. Personally, I don't see Creative swinging that much of a threat. Again, the fact that Apple didn't negotiate a license when they had the chance suggests that they didn't see Creative's claim as a major problem. At very least, I'd expect them to fight to limit the damages they pay Creative if they do lose.
Besides, as noted, Apple has already filed four countersuits. Even if they lose the Creative action, they have the potential to win back everything they lost (and more) by the time all the dust settles.
Dude.. everyone knows that the consumer electronics market shows seasonal variations. That's why analysts use year-over-year statistics rather than quarter-over-quarter.
The fact that the iPod has shown consistent Q/Q growth showed that the market was still in the process of discovering the product. The only 'saturation point' we've reached is in mindshare: there's nobody left who hasn't heard of an 'iPod' by now.
This does not signify the end of the iPod market. It marks the fact that the iPod market now, officially, exists. Now it's time to pay attention to Apple's Y/Y iPod sales growth, which, BTW, is phenomenal.
Rubbish. Why should Apple share a market it already dominates?
Look.. there's always a grey zone around patents. Most patent holders only sue over clear violations that hurt them in the market, because it takes a lot of very expensive time and effort to show just exactly if, and how much, product X infringes on roughly-similar patent Y. There's generally no payoff in being anal about marginal violations, and there are two strong benefits to taking a live-and-let-live approach:
First, it keeps business flowing smoothly. As long as everyone makes a general effort to do their own R&D, the climate stays friendly. The lawyers only come out when two companies follow the same line of grey-zone innovations to the same valuable result, and even then, they negotiate more than litigate.
Second, it's nice to have the option to drop the live-and-let-live attitude if someone gets difficult.
That's what's happening here. Apple was happy to let things slide as long as Creative was willing to play nice, too. But Creative knows that Apple has market dominance, first-mover advantage, economies of scale, phenomenal branding, and all that good stuff. There's no easy way for Creative to win that fight. Creative's management decided that their chance of making money by being anal over a possible marginal patent infringement was better than their chance of making money by maintaining the live-and-let-live mindset, so they sued.
Apple responded by showing what it could do if it chose to get anal, too.
There's nothing for Creative to win, here. Apple has no reason to settle, and plenty of resources to defend itself in court if it has to. The most likely result will be for Creative to quietly drop its suit in exchange for Apple doing the same, and both sides will go back to playing live-and-let-live, just like before.
Creative can't even win a crosss-licensing deal, because Apple has ponied up four patents to Creative's one. If anything, Creative will be lucky to get away without having to pay licensing fees to Apple.
Both of those sentences are in the active voice.
'Is riding' is the imperfect tense of the verb 'to ride'. It indicates continuing action. It describes the boy's state between the time when he starts riding the bike and the time when he reaches the store.
'Rode' is the simple past tense of 'to ride'. It indicates a single completed action. It describes the boy's state after he reaches the store.
The passsive voice inverts the actor and the thing acted upon, and often ignores the actor entirely: The bike was ridden to the store.
The passive voice emphasizes the action over the actor. If we only care how the bike got to the store (this one was ridden, that one was shipped in a truck), using the passive is fine. Usually, though, readers want to know who did something.
In business communication, the passive voice is a great way of hiding high-level muddleheadedness. It gives managers a way to talk about what needs to be done without ever narrowing things down to a specific group of people who should do it, or assigning them the resources they'll need to do the job.
In general, using the passive suggests that you don't really know what you're saying, yet. You have a general idea, but haven't spent enough time thinking about it to get down to specifics. If you know for a fact that only the actions matter (one of the bike's went through Mrs. Grumpy's garden, there are only two bikes present, one was ridden, the other came in a truck) go ahead and use the passive. Otherwise, keep working on the idea until you can express it in 'X does Y' terms.
This is the internet. Around here, when something seems a little bit too stupid to be true, it's still on the leading edge of coherent thought. ;-)
Besides, you Brits don't know sarcasm, you know satire. You have your lah-di-dah Jonathan Swift, Oscar Wilde, and all that lot. We Yanks produced Ambrose Bierce, who was one of the most unflinchingly sarcastic bastards ever to drip acid from the tip of a pen.
Subtlety has a place in satire, which is usually compared to a rapier. Sarcasm, on the other hand, is a saw. It derives its strength from stating the truth brutally, but with style.
Satire: Suggesting that Louderback might want to take a freshman-level creative writing class or two before trying to get any more of his "humorous" work published (and advising him not to give up his day job in the meantime, because editors can be tough cookies out there in the real world).
Sarcasm: Calling the article a piece of tripe that any self-respecting writer would have tossed into the garbage can along with all the other tissues coated with masturbatory residue.
Actually, it originally meant "screwed by your employer."
The original quote is from Hamlet: "For 'tis the sport to have the enginer / Hoist with his owne petar." There's nothing passive or accidental about that. It means, "once the guy has built the bomb and deployed it, it's funny to tie him up and hang him right next to it."
That reading fits the line's context extremely well: Hamlet has just stolen the letter Rosencrantz and Guildenstern are carrying to the King of England (which says: "Kill Hamlet") and replaced it with his own (which says: "Kill the men who give you this letter").
One little-known fact about Shakespeare is that he spiced his work up with controversies about military technology. This line demonstrates the love-hate relationship kings had with the engineers who handled explosives: it's damned useful to be able to blow down a door during a seige, but you don't want the guy who did it hanging around and possibly blowing down your own door next time. Besides, the royals of that day found it offensive to think that a commoner with a barrel full of smelly chemicals could wield that kind of power. It leads to the idea of commoners having the capacity to blow the bejeezus out of royalty, and that was just an affront to what they considered the natural order of things.
(and Romeo and Juliet -- if you know your historical weaponry -- revolves around the introduction of rapier fighting. Tybalt, "the captain of courageous complements" uses a parry-and-thrust fighting style, while the Montagues use the older block-and-slash style. The newer style could be faster and more lethal, but only as long as you did it right. All the fighting between the two families is based on the two gangs with different styles working out who has the power to take the other down)
Folks... back in the old days, we had the right to share music files with our friends, make backup copies, and copy files to different media.
No we didn't. We were just breaking the law in a context where nobody cared.
I'd say I've made fewer than 250 mix tapes and CDs over the past 30 years. I have friends who may have made a thousand, and that's fairly active for old-school swapping.
A P2P file sharer can beat those numbers in a single day.
On top of that, the tapes and CDs I made all stayed within my social network. I didn't go around my college putting "Get a bootleg tape of the latest Metallica album" flyers on every phone pole and bulletin board I could find. Can those of you who've filled your 80 Gig drives with P2P music pull a random track out of the collection and say you know the person who gave it to you?
Probably not.
Internet file sharing supports vastly larger numbers and vastly wider peer networks than good old-fashioned, "hey, I made you a tape," copyright violation. And like it or not, that huge increase in scale has changed the landscape. File sharing is no longer a trivial nuisance. The jury is still out on whether it's a beneficial manifestation of word-of-mouth advertising or a parasitic drain on legitimate revenue channels, but either way, it's big enough to be worth noticing these days.
1. I should not be allowed to use a car unless I can prove that I have a driver's license. Mag strip swipe by the ignition maybe?
If you get arrested for drunk driving and your BAC is above a certain limit, you may have to get your ignition lock fitted with a breathalyzer before you can get your license back. So yes, the state could enforce that precise limit if it wanted to, and thought that the benefit outweighed the cost.
It's called 'greymail'.
Greymail is a fairly common legal tactic. You try to force your opponent to choose between 'losing the case' and 'giving up an important secret'. In criminal cases, it tends to be used as an attempt to make the prosecutors drop a charge. In civil cases, like this one, it's usually an attempt to get the defendant to settle.
From what I can see, and what I recall of previous pageRank suits against Google, this one looks like a non-starter. Google has never promised anyone that their pageRank score will remain the same.. in fact, they pretty much guarantee that pageRank will change whenever new pages about a subject appear, old pages about a subject disappear, or the content of existing pages changes. Neither has Google ever promised, or even suggested, that their pageRank algorithms will remain stable and predictable. They're pretty open about the fact that they're constantly tuning and adjusting their system.
The plaintiff has no contractual claim against Google. They didn't pay for a Sponsored Link, and even those positions are subject to change based on what other people are willing to pay for the same keywords. Neither can the plaintiff claim that Google specifically altered their rankings to injure that company (and even then, Google has won suits against pageRank spammers who sued because Google broke their system).
When we get right down to it, we have a plaintiff that took opportunistic advantage of a good pageRank, and then suffered a reversal when the weather changed. Now they've brought a lawsuit against Google claiming specific and prejudicial injury, but they can't prove that unless they see Google's pageRank algorithm. Thus, they hope they can force Google will choose to settle rather than being forced to defend their right to keep the pageRank algorithms secret.
Like I said, it looks to me like a non-starter. Google's argument to the judge will probably run along these lines:
1 - The plaintiff doesn't have standing to sue Google on contractual grounds. The only way plaintiff can bring suit at all is to claim that Google is a public service.
2.1 - Assuming for the moment that Google is a public service, adhering to the plaintiff's concept of fairness would force Google to do one of two things: publish its pageRank algorithm, or provide sufficient notice of change that its pageRank algorithm could easily be reverse-engineered by a third party.
2.2 - If the pageRank algorithm was public knowledge, third parties (aka: spammers) would abuse it to assign pageRanks based on a client's willingness to pay, not on any standard of relevance or utility to the person doing the search.
2.3 - Allowing spammers to hijack the pageRank system would destroy the 'public service' the plaintiff claims to exist.
3 - But Google isn't a public service.
3.1 - Google never promised the plaintiff (or anyone else) a consistently high pageRank.
3.2 - Google doesn't give anyone 'warning' that their pageRank is about to change.
3.3 - The change in the client's pageRank happened for one of the following 'reasons':
3.3.1 - Someone else's pageRank got better.
3.3.2 - The plaintiff's page passed its freshness date, and the fact that new pages get a higher rank than old, unchanged pages is public knowledge.
3.3.3 - Google adjusted its pageRank algorithms to make searches even better, and this is only one of millions of resulting changes.
That's great, except that the ACID2 test is about CSS compliance which does indeed specify a particular page rendering.
Point taken, but the original question didn't mention ACID2 or DTD specification. It was a general complaint against the (W3C recommended) policy of being forgiving about what you're willing to parse.
ACID2 is a set of standard rendering conventions (and very valuable), but it doesn't trump the need to be generous with badly- or poorly-formed HTML. If an XHTML browser renders:
the same way it renders:
and both meet the ACID2 criteria for unordered lists, is the browser well-written or not?
The first version is clearly malformed XHTML, so should the browser just forgive the missing close tags, or should it barf out a badly-rendered page?
My question is: "Where's the missing mechanism?"
We have 30-odd years of data that show a strong correlation between ocean surface temperature and the force of hurricanes. Fine. We also have a very long history of periodic fluctuation in the force of hurricanes over time.
So why does this correlation only work today, as a 'result' of global warming? Why can't we use it to infer the existence of a periodic variation in ocean surface temperatures that corresponds to the observed history of strong hurricanes? More specifically, why can't we use this correlation and the history of strong hurricanes to show that ocean surface temperatures were rising and falling as much as they are today, long before the 150-year period associated with global warming?
If this round of strong hurricanes are the result of global warming, but none of the others in history were, what 'missing mechanism' produced all those previous cycles of strong hurricanes, and why isn't that mechanism working today?