Now admittedly, Javascript as a language sucks rocks...
No argument there..
I cherish a hope that if this whole thing really does become popular, we'll start to see drop-in modules for other languages. Then people will be able to use Perl, Ruby, Java, Python, or whatever they already like best. Most of the popular interpreted languages today already have pretty good cross-platform support, and it wouldn't be too painful to find subsets that are safe, but replace Javascript.
platform independent - the same widget will run on a Mac or a PC.
application framework - Konfabulator (or Dashboard) does some of the work, so you don't have to slog through Mozilla-we've-created-our-own-version-of-every-dat a-class-in-the-universe hell.
one-size-fits-all constraints - you "download info and display it" when you play Halo, but nobody freaks out because they can't use Halo to read Slashdot or order books from Amazon. Halo can just concentrate on being Halo, rather than trying to be Halo, plus a web browser, plus Everquest, plus the Sims online, etc. Much the same way, a widget can concentrate on doing one thing well, without having to put up with the IE/Firefox/Mozilla/Opera/Safari windows, toolbars, menus, and other associated cruft.
doesn't have to be stateless - a widget can download and display information, then keep using that information while some of the other information changes.. as opposed to a web browser, which has to pass the same damn package of data back and forth every time the user goes to a new page, or play all sorts of stupid games with cookies, session keys, database storage, and so forth.
While I don't care much about Konfabulator per se, it's interesting to see how popular non-browser, web-based mini-apps are becoming. Looks like the first little frothy bit on the wave of the semantic web has arrived. It'll be interesting to see if this ends up generating as much buzz as the last few Next Big Things.
It certainly has the potential to. It's a platform-independent application framework, just like Java-on-the-browser was supposed to be, but this iteration is free from the one-size-fits-all constraints that browser-based development imposed (plus, widgets don't have to be stateless). In fact, this has all the earmarks of a disruptive technology, with the added advantage that it's based on well-deployed standard technologies.. i.e.: stuff Microsoft can't mess with as easily as Java.
The "magical property" that gold has is that governments can't print it up on a whim (or under the guise of loaning it out). Sure, you can mine it, but expanding the gold supply is still hard!
Right.. and "printing money up on a whim" is completely different from redefining the buying power of a gram of gold.
Of course, if you don't change the number of Big Macs you can buy with a gram of gold now and then, you end up having real problems whenever the population increases. But that leads to the problem with rich people being able to hoard gold, again forcing the rest of the population to make do with less gold per person. If you adjust the exchange rates to a point where the average person can survive, you reward the people who've hoarded gold by increasing the buying power of the stuff they've pulled out of the economy. Then they can turn around and acquire even more gold, forcing yet another adjustment, ad infinitum.
It's true that 'printing press money' has little or no intrinsic value, but that's a good thing. Currency is just an exchange medium. It's a convenient way of comparing the value of one hour of my work to the value of a cup of fluffy coffee at Starbucks. Issuing a currency with no value in and of itself just means that I compare the value of my time directly to the value of the coffee, rather than comparing both to some arbitrary amount of gold. The currency ends up being 'backed' by every single service and commodity in the entire economy. And as an added bonus, it does rich people absolutely no good to pack trunks full of hundred dollar bills away in the attic on the assumption that they'll become more valuable over time.
But I doubt any of this will change your mind, just as nothing you're likely to say will change mine.
Forgive me for not fitting a comprehensive model of economics into 25 words or less.
Keynes's work was a valuable step in our understanding of what makes economies tick. Among other things, it helped us move away from ideas like, "gold has some magical property that defines value," and "anyone who's unemployed must be so by choice." Later work has built on that foundation, and the Austrian model does give us some useful ways to look at the system. Of course, one could get snarky about that model too, since Gaillot himself said:
"Market process theory assumes that there is no force and fraud. Force includes force by individuals in the form of theft and so forth, and force by the government in the form of taxation, regulation, and so forth. Austrians recognize that force by both individuals and governments exists, and that market theory does not completely explain reality."
BTW - the notion that stagflation is impossible has more to do with misinterpretation of the Phillips curve than with Keynes's work per se.
The dotcom boom had less to do with the value of tech stocks than it did with a sudden prolferation of day-traders in the market.. basically untrained idiots whose only concept of stock valuation was to look at the ticker and see whether the numbers were still going up.
Throw enough numbskulls with money into a market, and you get what's called 'arbitrage'.. a situation where you can sell dollar bills for $1.50. The bubble ends when the people who think a $1.50 dollar bill is a good investment run out of money.
A negative frenzy would also lead to arbitrage, except that you'd have people selling dollar bills for fifty cents. Again, the cycle ends when the stupid people run out of money.
The dotcom boom, the Dutch Tulip frenzy, and all other such bubbles boil down to a sort of distributed ponzi scheme: even if the stock doesn't have any intrinsic value, it's still a safe investment while momentum is high because you can sell it to someone else who does think it has value. Money flows from the incoming investors to the outgoing investors with little or no relation to things like corporate earnings. The bubble inflates because people think they can not only resell the stock, but that they can resell it at a significant profit.
As soon as people stop believing they can resell the stuff at a significant profit, the whole thing collapses, and whoever ends up holding the last batch of shares wins the label of "idiot who threw money down a hole."
To make a frenzy go the other direction, you'd have to convince people that holding the stock would cause them significant loss, and there's only so much loss you can make someone expect when a company does $3.2B/year. Think of it this way: buying stock is a form of gambling. The lower a stock's price goes, the more attractive the bet becomes. In the ridiculous case, some stodge of a number-cruncher will eventually decide that paying twenty bucks a controlling interest in Google is worth the risk. And heck, at that price they could turn around and double their investment in no time..
So, assuming a major negative frenzy did occur, and somehow crashed through the barriers of sanity, all it would do is spawn another bubble.
In The General Theory of Employment, Interest, and Money, John Maynard Keynes laid out the big picture of 'value': When you look at an economy as a whole, you see that money flows from one place to another, then back again, making cycles. The question of value isn't whether one thing is intrinsically more worthwhile than another, but whether it has the power to pull a higher concentration of money into a given part of the system.
Along with that goes the idea that the overall value of money is based on people's willingness to use it. There are simple abberations.. put too much currency into an economy and you get inflation, restrict the flow of currency too much and you get a recession.. but even in a more or less balanced economy, people have to decide whether to spend their money on X today or Y tomorrow. Those decisions determine both where the cash goes, and how much buying power a unit of currency has.
Boil it all down, and you get the idea that 'wealth' is strongly tied to people's willingness to invest in the future.
Time Warner may be big, but is business model is old, well, understood, and frankly, not so healthy. Information on the internet competes with information managed by traditional media companies, and the differences in distribution models, usage models, and cost of entry are playing hell with prices in the traditional sector.
Google may be an unknown, but it's demonstrated its ability to make actual profits while distributing information on the internet. People are willing to invest in that model of the future, and that makes Google valuable.
Are people overestimating the potential value of the internet-information market? Possibly. But this time they're investing in a company that makes actual profits, as oppsed to the last dotcom boom, where people invested in the idea that 'branding' was synonymous with future profits.
If it turns out that Time Warner's $42B/year market dissolves into an internet information market worth, say, $8B/year, the investors who've bet that Google's cap will reflect current market conditions will lose money, and Google's stock price will go down. But it will happen slowly, as investors gradually work out the real value of the new information market. It won't be a cascade-failure like last time. At very least, there's no reason for Google's shares to drop below the reasonable value for a company that earns $3.2B/year.
Discussions like this would be less confusing if people referred to the documents in question as the Creative Commons Licenses. That's what they are: Licenses.
The breakdown runs like so:
Copyright law assigns certain rights to the creator of a work.
The CCLs explain how the creator wants to use those rights.
Licenses are important because we need some kind of legal document to fill the gap between, "no one is allowed to use this work but me," and "I hereby give up my rights to this work entirely." Licenses more or less boil down to, "I'm definitely holding onto the right to sue people, I'm just not going to sue you."
Usually, copyright holders hand out licenses to use the work in exchange for money. The document that lays out the exact details of that exchange is called a 'contract' because each party hands something over to the other.. one side supplies a big-bag-o-cash, the other supplies a license. Licenses can exist outside of a contract, though. As a copyright holder, I can exercise my rights any way I darn well please, and if I decide not to sue people who use my work, that's my decision.
Now, if I stand up and make a public declaration that I will not sue anyone who uses my work while wearing a green bowtie on a rainy Tuesday in Berlin, the courts will treat that statement as a binding license. If I try to sue somebody who can prove that they were wearing a green bowtie on a rainy Tuesday in Berlin, I'm gonna lose. If the courts ruled any other way, it would be way too easy for IP owners to trap people.
Thing is, writing a good IP license is tricky. It's generally the kind of thing you pay an IP lawyer a whole lot of money to do. The CCLs are just generic, check-the-appropriate-box licenses that have been written by people who know what they're talking about. Functionally, they're no different from the DIY legal kits you can get from www.nolo.com or the standard fill-in-the-blank contracts you'll find in any business.
The important thing the CCLs do is create nice, clean, well-defined boundaries between various kinds of IP use. As a creator, I can state cleanly and clearly whether I want to allow commercial or noncommercial use, attributed or unattributed use, and whether I do or don't want to allow other people to create derivative works.
And that's important, because if the license doesn't exist, or isn't written out clearly, the person who ultimately controls the rights will be the one with the most expensive IP lawyer.
People love to analyze big changes like this after they happen, but the analyses are almost always wrong. The problem is that the explanations carry an underlying assumption that only evolutionary changes make any sense.
Is there any real reason to assume that Steve Jobs, with his history of making big and frequently unpopular decisions, would refuse to call a switch like this just because he was unhappy with the price/performance numbers IBM was quoting him? Not really. Is there any reason to assume that his decision to call the switch took place in a vacuum? Not really.
One difference between leaders and followers is the willingness to make big changes on small justification. The people capable of thinking farthest outside the standard comfort zone are the ones who see the interesting possibilities first. You can build a retroactive chain of continuity after the fact, but those 'reasons' are *not* the reasons for taking the leap. They're merely reasons the person in charge (in this case Jobs) didn't reject that particular leap out of hand.
FOSS is doing the same thing to the entire software business model. People in the industry who Just Don't Get It (*cough*DarlMcBride*cough*) talk about how FOSS needs to be made 'acceptable to business' if it wants to 'succeed', because they can't think far enough out of their comfort zone to see how 'giving stuff away' works better for everyone.
One hallmark of genius is the ability to bridge the gap between "nobody ever thought of that before" and "totally obvious to anyone who sees it." I don't think this particular change rates as high as 'genius', but there's a similar gap between "unthinkable" and "justifiable after the fact".
Apple makes its money selling a vertically integrated solution stack.. hardware, OS, middleware, userland, and a collection of network services like.Mac and the iTMS.
Most of Apple's profits come from the hardware, not the software.
The software is basically just a value-added service that makes people think it's worthwhile to pay for the hardware.
Dropping into the world of commodity x86 boxes would kill Apple's main source of profits, so I doubt we'll see that happen.
It's also important to note that having 90% market share is not an unmixed blessing.
Just as a program spends 80% of its time looping through 20% of the code, 80% of the profit in any given market comes from 20% of the consumer base.
And 80% of the expense comes from having to support the 20% of consumers at the other end of the scale.
We've seen examples of that latter idea all over the place recently, with stores giving premuims to repeat customers, or refusing to provide service to 'problem' customers.
I doubt Apple really wants more than the 20% of the market that falls at the opposite end of the traditional Linux hacker space.. people who are willing to spend money to solve a problem rather than spending time and energy.
One, it's more profitable, and two, FOSS is doing such a good job of serving the high-maintenance, low-profit end of the market that nobody can really compete with it.
Both market groups are perfectly valid, they just approach the old "fast, good, cheap: pick two" formula in different ways.
But the simple fact is that PJ stepped up and made herself a public figure in an extremely controversial case.
I think if you check the definitions, the term "Limited Purpose Public Figure" is more appropriate.
A public figure is a famous person about whom the public has a broad, general interest.
A limited purpose public figure is someone who makes themselves visible in a specific issue, for purposes of influencing the outcome of that issue.
Both types have to prove actual malice in order to win a defamation case, but where a PF's whole life is up for grabs, a LPPF is only 'public' regarding things that are relevant to the issue in question.
Think of it as requirements for admissibility of information in court:
For a PF, any statement about them at all is admissible.
For a LPPF, only statements about issues where they've 'pushed themselves forward' are admissible.
So.. the questions we have to consider are: "How much of PJ's personal life is relevant to the SCO -v- IBM case?"
and
"How much of her personal life has PJ 'pushed into the spotlight' on Groklaw?"
Personally, I think the answers are, "damn little," and "damn little," respectively.
Yes, knowing whether PJ accepts money or other compensation from IBM is relevant to this issue.
No, her personal appearance, apartment number, the opinion of her landlord, and the contents of her car's back seat are not.
It's also important to note that PJ doesn't stonewall the press completely.
She does communicate with people, and she does do interviews.
She just prefers not to do them face to face.
It's not a question of whether the press can get information from her at all, it's a question of what means of communication they're allowed to use.
Even public figures have the right to refuse to give interviews to specific individuals, at specific times, and by specific means.
In this case, O'Gara published information that, to the best of my knowledge has never been 'pushed into the spotlight' by PJ.
In fact, it would be easy to make a case that PJ has made specific efforts to keep that information out of the spotlight.
The information does not strike me as relevant to SCO -v- IBM in any noticable way.
Nor do I see how approaching PJ face to face could convey any information about SCO -v- IBM that O'Gara couldn't get some other way.
The sole possible justification for face to face contact is to resolve the (IMO artificial) issue of whether PJ "actually exists."
And by that standard, O'Gara's approach failed miserably.
She didn't establish any meaningful connection between the woman whose privacy she invaded and the person who runs Groklaw.
She claimed to be looking for PJ, but pretty much abandoned that approach as soon as she got near the door.
Instead, she switched over to taking PJ's existence for granted, and spent the rest of the article listing potentially prejudicial details about the 'PJ' she 'found', and indulging in no small measure of 'we know where you live' style intimidation.
When you read Groklaw, you have to remember one basic fact about the western legal process: The only thing that matters is your evidence.
Bias is irellevant to the law. The attorneys on one side of a case will be firmly biased one way, and the attorneys on the other side will be firmly biased the other way.
Groklaw is 'open' in the sense that all its evidence is right out there for everyone to see, no matter which way their bias happens to point. This is different from SCO, which has still failed to identify the code it claims has been misappropriated.
And yes, Groklaw does have a clear bias against SCO. It began as an attempt to counter the FUD-storm that SCO was pumping into the court of public opinion. Remember the "millions of lines of code"? The "team of MIT researchers"? Darl's Open Letters too the Linux Community?
SCO went into this litigation playing a very hard flavor of hardball. They made large, bold, easy to understand claims, and left the tech community to respond with arguments that were logically correct, but too complicated to win any mindshare. And it worked. For probably the first year of the case, SCO's side of the story was the one that was heard in the mainstream media.
So PJ began reading documents and collecting evidence, and embarked on the long, slow process of chipping away at the FUD one little bit at a time. Over time, she developed a reputation for countering SCO's claims reliably, thoroughly, and on the basis of hard evidence. Eventually, the mainstream press started to pay attention, and started treating SCO's claims more critically than they had before.
At present, SCO has the FUD machine pointed at Groklaw and PJ herself. Within the last month, Darl was quoted in this article saying:
"if you look at the reality of the Pamela Jones situation, you have to conclude that all is not as appeared as it is in Groklaw land. We appreciate that many media sources disagree with us, but they're accountable. We think you need to know who's behind the news."
and:
"we're digging into who Pam Jones is, and we're close to the bottom."
because they'd really like people to see Groklaw as a haven for anti-SCO fanboys, and question (or better yet, ignore) all the evidence that Groklaw has accumulated. Then SCO could get an unfettered run at the media again.
And if they get it, I invite you to consider just how 'open' SCO's comments to the media will be.
Because the company currently suing IBM, which calls itself "The SCO Group", used to be named "Caldera."
The company that used to be named "SCO" is now doing business under the name "Transmeta."
Caldera (aka: newSCO) bought the IP rights to the work done in Monterey from oldSCO (now Transmeta), and those rights are at the core of the present lawsuit.
So.. by showing that Caldera (now newSCO) knew Monterey was heading to POWER back before the Monterey project ended, PJ has produced evidence which directly contradicts the claims newSCO (then Caldera) has made in its third amended complaint.. namely that newSCO (then Caldera) thought:
.. that Monterey only gave IBM permission to build code for the Itanium platform
.. that by porting the code to POWER, IBM went beyond the rights it was granted by the Monterey contract
.. that newSCO (formerly Caldera) had the right to address that violation by revoking IBM's right to use or distribute the code
.. that anything IBM did with the code after that revocation was illegal
.. and that IBM owes newSCO (formerly Caldera) a whole bunch of money.
Interestingly enough, the company formely known as "Caldera" didn't change its name to "The SCO Group" until after it filed its lawsuit against IBM. Questions like the one you just asked show just how much confusion that bit of misdirection has caused.
For the record, Orlokowski is the one responsible for the whole "RMS 'can't believe' that Miguel de Icaza is working on.NET'" non-event a couple years back. (link)
IIRC, RMS actually said, "I don't believe he's doing that," when answering some specific question about de Icaza's work, in the, "I think that's incorrect," sense. Orlokowski got it secondhand from someone who attended, misinterpreted the out-of-context quote, and chucked a story onto the Reg's front page before doing any journalistic stuff like contacting the primary sources and checking his facts.
RMS wrote a response (link) which begins: "Your article about me, GNOME and.Net was inaccurate starting from the title."
The Reg has a habit of blurring the boundaries between 'reporting' and 'editorializing', which is part of its charm. Sometimes that means not letting the facts get in the way of a that will draw page views, but at least they don't do it too often.
This particular case happens to be pretty silly, though. For some reason Orlokowski, a general pundit who's never claimed any kind of in-depth legal knowledge, decided it would be a good idea to trade legal opinions with PJ, a paralegal who's read, discussed, archived, and cross-referenced every public document relating to this case that she can get her hands on.. not just the complaints, briefs, motions, responses, etc actually filed in court, but also a whole ton of external contextual information.
By and large, I think it's pretty safe to bet that PJ knows what she's doing.
He said right in the article that the problem he had was with the marketing. It doesn't change the way you use your computer. At least stick to the things he is wrong about.
Fair enough, but he does sort of miss the forest for the trees when he dismisses Spotlight as a kinda-neat-Windows-like-search-thingy. Being able to organize files by metadata (a smart folder with 'all the black&white photos I've worked on in the last couple of days') is 'a different way of using the computer' compared to building, maintaining, traversing (and forgetting) a series of folder hierarchies. He may as well have said the GUI is no big change from the command line because the underlying file trees are still the same.
And Automator at least has the potential to be as big a step as moving from 'single commands on the command line' to 'command line with pipes'. From a simple task-based analysis, the job: 'find all the black&white photos I've worked on in the last couple of days, convert them to JPEGs, dump those onto a CD, then burn it' can be done very differently in Tiger than it can in Panther.
Differently enough, in fact, that the phrase "change the way you use a computer" does have some merit, IMO.
1 - Vertical integration: Apple sells a (hardware, OS, software, network services) package where all the pieces are tweaked to work well "right out of the box." The fact that Apple controls the entire product stack gives them a lot of flexibility when it comes to tweaking the system. The applications people can push design requirements to the OS team, and the OS team can push requirements to the hardware people.
They'd lose that flexibility if they had to support machines designed by other companies with different agendas, and Apple would get most of the blame for anything that failed to work.
2 - Application base: Switching from PPC to x86 would force end-users to buy new software. No matter how you crunch the numbers, you can't save enough on performance-competitive x86 hardware to make up for having to buy a brand-new copy of Photoshop OSX-86.
Apple would also have a hell of a time bringing third-party developers on board, because even if compiling for the new platform were as easy as clicking a button (and it never is), it'd be one more SKU to worry about.
3 - Market position: Contrary to what most slashdotters think, it is *not* a good thing to own 95% of the market. The 80-20 rule applies to market profitability just as much as it applies to code use. 20% of the market generates 80% of the profit, and 20% of the market generates 80% of the expense.
Microsoft holds 95% of the market.. not because it wants to.. but because it has to. It continues to supply market segments that are a dead loss financially just to maintain the standards-and-protocols lock that keeps its profitable customers from wandering away.
There's no business reason for Apple ever to want more than the 15-20% of the market that generates 80% of the profit. Below that line, vertical integration becomes less important than price point, and the margins there are so thin that Dell (the big dog in that space) is branching out into camcorders and whiffle bats because it can't survive just by selling computers alone.
The thing slashdot needs to understand about the 80-20 rule is that 'small market share' can equal 'more pay for less work' if you hit the right part of the market. It's the choice between supporting two people who are willing to pay $4 each, or ten people (including two nuclear assholes) who'll net you a total of $10.
As an aside, Linux is actually helping Microsoft (temporarily) by taking over some of the market segments that cost Microsoft the most money. That will change in a few more years, though, when Linux becomes mature enough to compete for market share in segments Microsoft can't afford to lose. It's already threatening Microsoft's standards-and-protocols lock, and when 15% of the market embraces Open standards and protocols, things will get interesting.
4 - Brand support: Apple survives as a vertical integration vendor because it has a reputation for quality. Without its reputation for quality, Apple couldn't compete for the 20% segment that generates 80% of the profits. There's no way to embrace the x86 market without also embracing the crapfest at the low-20% end, though, and doing so would shoot Apple's reputation for quality straight through the head. That would knock Apple out of its high-margin target segment, and put Apple in head-to-head competition with Microsoft on Microsoft's home turf.
And history shows that competing with Microsoft in Microsoft's core markets, on Microsoft's target platform, is suicide.
---
Basically, Apple's core market has different priorities than the traditional/. Linux geek. Where the T/.LG resolves the 'soon/cheap/good - pick two' formula by picking 'cheap and good', Apple wants the chunk of the market that's willing to choose 'soon and good'. The T/.LG solves problems by throwing time and effort at them, while the ideal Apple customer throws money at them.
Well, my first reaction is that the leaker should have taken the time to go through the box before leaking the whole thing.
The court will weigh the interests on both sides, though. A judge would only release the leaker's name if Enron's actual injuries from the release of that contract outweighed the leaker's right to anonymity.. wiping out a billion-dollar deal to tell the world that Enron's CEO overstated his income tax deductions by $250, for instance. But unless it can demonstrate significant real injuries from the release of that one specific contract, Enron probably won't be able to force the court to release the leaker's name.
And even if Enron does get the name, it will have to be careful about what it does. The whistleblower protection laws specifically forbid companies from taking revenge.
If Enron does anything to the leaker, the leaker can sue on the grounds that it was revenge. And judges want to maintain a chilling effect that prevents companies from taking revenge, so Enron will have to turn itself inside out to prove its actions were absolutely squeaky clean and that it did everything but buy the leaker a pony and some ice cream to make the experience as pleasant as possible. If it can't, the judge will award the leaker a big 'ol pile of cash in order to remind all the other companies out there that beating up whistleblowers is A Really Bad Idea.
So basically, judges take care of the middle ground, and they work hard to make sure everyone gets fair treatment.
Apparently no one here has noticed the part of the article which reads:
"Larry is perfectly fine with somebody writing a free replacement."
This is not a question of whether McVoy is trying to crush free clients for his software. It's a question of how those free clients should be written. And Tridge should know the difference since he wrote this article on how Samba development is more about network and protocol analysis than reverse engineering.
One paragraph from that article reads:
Classical reverse engineering techniques in software engineering revolve around the use of disassemblers, debuggers and other object code analysis tools to examine directly the executable code of an existing product in order to create a "clone" that behaves in the same way. While these venerable techniques have been successfully used by many groups, they are not what we use in the Samba project.
So clearly Tridge does make a distinction between ripping the binaries and doing protocol analysis.
This is not a debate about Openness. This is not a debate about Freedom. It's a debate about methodology, and Tridge knows the distinction between cloning a binary with a disassembler and re-implementing a piece of software from the protocols better than most people do.
1. Yeah, generally. 'Civil offense' or 'civil tort' would be more correct, but 'crime' is good enough for street usage.
2. No. Enron could put the words "As part of my job, I agree not to blow any whistles" on a piece of paper and require all its employees sign it, but the courts wouldn't recognize that as a legal contract.
Contracts only work because the courts agree to enforce them. There's no actual law that says, "employee X will obey the terms of document Y while employed at company Z." There can't be. There are too many deals taking place every day to have specific laws for each and every one. Instead, we have what you might call the general specs for agreements between two parties, and those specs are called 'contract law'.
If you and I make a contract, then I refuse to hold up my end of the deal, I haven't (necessarily) broken an actual law.. you can't take me to court saying, "he broke statute X.Y.Z" (s criminal offense). All you can do is take me to court and say, "here's the contract that explains what he agreed to do, and he hasn't done it" (a civil offense). If the judge decides that our contract lives up to the general specs, she will then order me to deliver what I promised. And there *are* laws that say I have to obey the orders of the court.
A contract that doesn't live up to the specs is called an 'illegal' contract, but again, you don't get thrown in jail for writing or signing one. It just means that the courts won't enforce it, and if the courts won't enforce it, it's just a piece of paper with a bunch of funny words on it.
And contract law says a legal contract CAN NOT require someone to do something illegal. One of the precedents involves a guy who 'sold' a car to a woman in exchage for a certain amount of sex.. 10 visits to his bed rated at $50 a pop, or something like that. She stopped somewhere around 5, so the guy sued her for the rest of the promised nookie. The judge said that sex for pay is prostitution, prostitution was illegal in that state, and therefore the 'contract' was not legal. The brief is actually quite an amusing read.
Any Enron that tried to sue a whistleblower for violating an NDA would get a brisk "fuck you" from the courts if the secret involved anything actually illegal. And there are laws that specficially protect whistleblowers who reveal secrets that don't involve illegal activity per se, but are still 'in the public interest'.
----
However, as the judge in this case clearly pointed out, 'an interested public' is not the same as 'public interest'. The dividing line has nothing to do with curiosity, and everything to do with injury. 'Public interest' stops at 'preventing the public from being injured' and does not extend to 'what the public might be willing to watch on TV.'
If you violate your NDA to tell the world that the safety tests for a new airplane were forged because the company found a major flaw in the design and couldn't fix it before the test results were due, you're working in the public interest. Keeping the secret exposes people to risks they don't know about, and the courts will rule that letting people know about the risks is more important than obeying your NDA.
But no matter how tortured the "companies shouldn't be allowed to keep secrets" arguments get here on/. (and let's just test that theory by cracking into the Slashdot servers, stealing the username/password database, and giving it to a blogger who will then publish it 'in good faith' shall we? Everyone has a 'right' to a low uid), no one has yet advanced a really good line of reasoning to suggest that the anyone was actually injured by Apple's decision to keep its product specs secret. Therefore, there is no legitimate claim of 'public interest'.
----
And while I'm ranting, I'd like to take a minute to say that trade secrets and NDAs aren't just evil tools of big Sauron-like corporations. They exist to protect the little guys like you and me.
In response to the first post above, contracts don't have to be written and signed. Dig through _West's Business Law_ for some discussion and precedent. The real proof is in the behavior of the people who made the agreement. If two parties act like they signed a contract, the courts will generally rule that a contract exists.
In response to the post directly above, a contract can be illegal and still exist. The court will refuse to enforce an illegal contract, and will often tell people to give back what they took under the terms of an illegal contract, but that doesn't stop the contract from existing. Yes, that's a fiddly distinction, but we are talking about law, here.
Besides, I agree that the sources did something illegal, that the ThinkSecret NDA doesn't provide a legal excuse to keep those identities secret, and that the judge made the correct ruling. But I've seen too many people denouncing Apple, the judge, the law, and society in general because they think NDAs shouldn't exist, all the while demanding that ThinkSecret's NDA be judged legal and enforcable.
Current DRM schemes are no more flawed than the locks on your home or your car. Just ask any burglar or car thief. And even if the locks were truly unbreakable, the windows of your home and your car aren't. On the computer side, 8-character user account passwords are almost trivially easy to crack these days.
The most basic rule of security is that no security system is, or ever will be, perfect. Eventually, we always have to say, "oh well, their hearts are in the right place, so let's just honor it." We say that about locks, user account passwords, credit card authorization systems, or any other technology which restricts people's actions.
We draw the line in the sand where it provides the best balance between various opposing interests, not based on the perfection of the system preventing misbehavior.
The kind of argument you've offered could be extended to the idea that anything not 'effectively' prohibited must be allowed.. in other words, "if I can do it, I should be allowed to do it." That may sound attractive at first blush, but nobody wants to be on the receiving end of it.
Ultimately, we say, "let's just honor it" in support of someone else's flawed technology because we want them to say the same thing in support of ours. None of us can ever achieve perfect security, most of us can't afford truly excellent security, and few of us are diligent enough to manage even consistently good security. We agree to pretend that flawed systems work because we don't want to put up with the hassles inherent in living with something more effective.
It's fine -- in fact it's necessary -- to discuss exactly where we want to draw the line in the sand. But in those discussions, we always have to remember that we're talking about balancing interests, because in the long run, your right to life, liberty, and the pursuit of happiness interferes with my right to beat the shit out of you with a brick, and vice versa.
Arguments on the order of "I'll do it because you can't stop me" aren't useful, nor are arguments on the order of "anything that limits my freedom is wrong." Anyone can demonstrate how both theories are flawed given five minutes.. and a brick.
In this case, Apple's DRM strikes a fairly good balance of interests between the people who listen to music and the record companies. It's a compromise, i.e.: an agreement that's equally unacceptable to all parties involved. There are lunatic fringe elements on both sides of the debate, but at the moment, the ones on the record label side represent a denser concentration of serious idiocy backed up by a whole lot of contract negotiating power.
IMO, Jon is losing the war by winning the wrong battles. Every time he exploits a hole in Apple's current technology, he hands ammo to the most determined idiots on the record label side. If Jon manages to convince the labels that FairPlay will never work, the labels won't give up on DRM. They'll just throw their weight behind Microsoft's DRM and start lobbying congress for unique ID chips in every electronic device.. again.
Yes, the labels will eventually learn that no amount of DRM will ever be perfect. But how many years of increasingly draconian measures do you want to put up with while Jon makes headlines convincing the labels that the current version is still too permissive?
Has anyone else noted the irony that this whole issue is really about two NDAs?
On the one hand, we have the NDA between Apple and whoever leaked the information. People go to great lengths to explain how breaking that NDA, and/or publishing information gained from someone who did break that NDA, is Freedom Of Speech Goodness Galore.
On the other hand, we have ThinkSecret's promise of anonymity to its sources.
Now, if you think about it for five seconds, that pretty much boils down to another NDA, aka: an Agreement Not to Disclose information. But this NDA has to be protected at all costs because, again, that's Freedom Of Speech.
If "All Secrets Limit My Freedom", as some people have argued, and "Any Judge Who Enforces An NDA Is Pissing On Freedom Of Speech", as has also been argued (repeatedly), what makes ThinkSecret's decision to withold information so good?
"If ThinkSecret gave up the names of its sources, it won't be able to attract sources in the future," you say? But doesn't that pretty much boil down to the statement: "ThinkSecret uses NDAs to protect its business"?
If the people who gave the information to ThinkSecret didn't break any NDA, then Apple would have no case against them.
BUT, the judge can't decide whether the people in question broke an NDA until it knows who they are, now can he?
This decision deals with a subpoena.. a document the courts use to gather information the judge needs in order to make a good decision. It doesn't say that a crime was committed, that ThinkSecret broke some kind of law, that ThinkSecret owes Apple some kind of damages, or anything of the kind. It just says ThinkSecret has information the judge needs.
We can recast this whole thing in terms of the judge wanting to give ThinkSecret as much benefit of the doubt as he can. Let's imagine he wants to say, "ThinkSecret didn't do anything wrong".. he can't just wave his hands and do that, though. He has to cite some kind of evidence.
As it stands, the only resonable way the judge can get that information is to ask ThinkSecret for the name of the source.
Bottom line, the judge can't declare ThinkSecret not guilty until he gets the information about the source.
Slashdot Mirror
Now admittedly, Javascript as a language sucks rocks ...
No argument there..
I cherish a hope that if this whole thing really does become popular, we'll start to see drop-in modules for other languages. Then people will be able to use Perl, Ruby, Java, Python, or whatever they already like best. Most of the popular interpreted languages today already have pretty good cross-platform support, and it wouldn't be too painful to find subsets that are safe, but replace Javascript.
While I don't care much about Konfabulator per se, it's interesting to see how popular non-browser, web-based mini-apps are becoming. Looks like the first little frothy bit on the wave of the semantic web has arrived. It'll be interesting to see if this ends up generating as much buzz as the last few Next Big Things.
It certainly has the potential to. It's a platform-independent application framework, just like Java-on-the-browser was supposed to be, but this iteration is free from the one-size-fits-all constraints that browser-based development imposed (plus, widgets don't have to be stateless). In fact, this has all the earmarks of a disruptive technology, with the added advantage that it's based on well-deployed standard technologies.. i.e.: stuff Microsoft can't mess with as easily as Java.
The "magical property" that gold has is that governments can't print it up on a whim (or under the guise of loaning it out). Sure, you can mine it, but expanding the gold supply is still hard!
Right.. and "printing money up on a whim" is completely different from redefining the buying power of a gram of gold.
Of course, if you don't change the number of Big Macs you can buy with a gram of gold now and then, you end up having real problems whenever the population increases. But that leads to the problem with rich people being able to hoard gold, again forcing the rest of the population to make do with less gold per person. If you adjust the exchange rates to a point where the average person can survive, you reward the people who've hoarded gold by increasing the buying power of the stuff they've pulled out of the economy. Then they can turn around and acquire even more gold, forcing yet another adjustment, ad infinitum.
It's true that 'printing press money' has little or no intrinsic value, but that's a good thing. Currency is just an exchange medium. It's a convenient way of comparing the value of one hour of my work to the value of a cup of fluffy coffee at Starbucks. Issuing a currency with no value in and of itself just means that I compare the value of my time directly to the value of the coffee, rather than comparing both to some arbitrary amount of gold. The currency ends up being 'backed' by every single service and commodity in the entire economy. And as an added bonus, it does rich people absolutely no good to pack trunks full of hundred dollar bills away in the attic on the assumption that they'll become more valuable over time.
But I doubt any of this will change your mind, just as nothing you're likely to say will change mine.
Forgive me for not fitting a comprehensive model of economics into 25 words or less.
Keynes's work was a valuable step in our understanding of what makes economies tick. Among other things, it helped us move away from ideas like, "gold has some magical property that defines value," and "anyone who's unemployed must be so by choice." Later work has built on that foundation, and the Austrian model does give us some useful ways to look at the system. Of course, one could get snarky about that model too, since Gaillot himself said:
BTW - the notion that stagflation is impossible has more to do with misinterpretation of the Phillips curve than with Keynes's work per se.
The dotcom boom had less to do with the value of tech stocks than it did with a sudden prolferation of day-traders in the market.. basically untrained idiots whose only concept of stock valuation was to look at the ticker and see whether the numbers were still going up.
Throw enough numbskulls with money into a market, and you get what's called 'arbitrage'.. a situation where you can sell dollar bills for $1.50. The bubble ends when the people who think a $1.50 dollar bill is a good investment run out of money.
A negative frenzy would also lead to arbitrage, except that you'd have people selling dollar bills for fifty cents. Again, the cycle ends when the stupid people run out of money.
The game doesn't work that way.
The dotcom boom, the Dutch Tulip frenzy, and all other such bubbles boil down to a sort of distributed ponzi scheme: even if the stock doesn't have any intrinsic value, it's still a safe investment while momentum is high because you can sell it to someone else who does think it has value. Money flows from the incoming investors to the outgoing investors with little or no relation to things like corporate earnings. The bubble inflates because people think they can not only resell the stock, but that they can resell it at a significant profit.
As soon as people stop believing they can resell the stuff at a significant profit, the whole thing collapses, and whoever ends up holding the last batch of shares wins the label of "idiot who threw money down a hole."
To make a frenzy go the other direction, you'd have to convince people that holding the stock would cause them significant loss, and there's only so much loss you can make someone expect when a company does $3.2B/year. Think of it this way: buying stock is a form of gambling. The lower a stock's price goes, the more attractive the bet becomes. In the ridiculous case, some stodge of a number-cruncher will eventually decide that paying twenty bucks a controlling interest in Google is worth the risk. And heck, at that price they could turn around and double their investment in no time..
So, assuming a major negative frenzy did occur, and somehow crashed through the barriers of sanity, all it would do is spawn another bubble.
In The General Theory of Employment, Interest, and Money, John Maynard Keynes laid out the big picture of 'value': When you look at an economy as a whole, you see that money flows from one place to another, then back again, making cycles. The question of value isn't whether one thing is intrinsically more worthwhile than another, but whether it has the power to pull a higher concentration of money into a given part of the system.
Along with that goes the idea that the overall value of money is based on people's willingness to use it. There are simple abberations.. put too much currency into an economy and you get inflation, restrict the flow of currency too much and you get a recession.. but even in a more or less balanced economy, people have to decide whether to spend their money on X today or Y tomorrow. Those decisions determine both where the cash goes, and how much buying power a unit of currency has.
Boil it all down, and you get the idea that 'wealth' is strongly tied to people's willingness to invest in the future.
Time Warner may be big, but is business model is old, well, understood, and frankly, not so healthy. Information on the internet competes with information managed by traditional media companies, and the differences in distribution models, usage models, and cost of entry are playing hell with prices in the traditional sector.
Google may be an unknown, but it's demonstrated its ability to make actual profits while distributing information on the internet. People are willing to invest in that model of the future, and that makes Google valuable.
Are people overestimating the potential value of the internet-information market? Possibly. But this time they're investing in a company that makes actual profits, as oppsed to the last dotcom boom, where people invested in the idea that 'branding' was synonymous with future profits.
If it turns out that Time Warner's $42B/year market dissolves into an internet information market worth, say, $8B/year, the investors who've bet that Google's cap will reflect current market conditions will lose money, and Google's stock price will go down. But it will happen slowly, as investors gradually work out the real value of the new information market. It won't be a cascade-failure like last time. At very least, there's no reason for Google's shares to drop below the reasonable value for a company that earns $3.2B/year.
Discussions like this would be less confusing if people referred to the documents in question as the Creative Commons Licenses. That's what they are: Licenses.
The breakdown runs like so:
Licenses are important because we need some kind of legal document to fill the gap between, "no one is allowed to use this work but me," and "I hereby give up my rights to this work entirely." Licenses more or less boil down to, "I'm definitely holding onto the right to sue people, I'm just not going to sue you."
Usually, copyright holders hand out licenses to use the work in exchange for money. The document that lays out the exact details of that exchange is called a 'contract' because each party hands something over to the other.. one side supplies a big-bag-o-cash, the other supplies a license. Licenses can exist outside of a contract, though. As a copyright holder, I can exercise my rights any way I darn well please, and if I decide not to sue people who use my work, that's my decision.
Now, if I stand up and make a public declaration that I will not sue anyone who uses my work while wearing a green bowtie on a rainy Tuesday in Berlin, the courts will treat that statement as a binding license. If I try to sue somebody who can prove that they were wearing a green bowtie on a rainy Tuesday in Berlin, I'm gonna lose. If the courts ruled any other way, it would be way too easy for IP owners to trap people.
Thing is, writing a good IP license is tricky. It's generally the kind of thing you pay an IP lawyer a whole lot of money to do. The CCLs are just generic, check-the-appropriate-box licenses that have been written by people who know what they're talking about. Functionally, they're no different from the DIY legal kits you can get from www.nolo.com or the standard fill-in-the-blank contracts you'll find in any business.
The important thing the CCLs do is create nice, clean, well-defined boundaries between various kinds of IP use. As a creator, I can state cleanly and clearly whether I want to allow commercial or noncommercial use, attributed or unattributed use, and whether I do or don't want to allow other people to create derivative works.
And that's important, because if the license doesn't exist, or isn't written out clearly, the person who ultimately controls the rights will be the one with the most expensive IP lawyer.
People love to analyze big changes like this after they happen, but the analyses are almost always wrong. The problem is that the explanations carry an underlying assumption that only evolutionary changes make any sense.
Is there any real reason to assume that Steve Jobs, with his history of making big and frequently unpopular decisions, would refuse to call a switch like this just because he was unhappy with the price/performance numbers IBM was quoting him? Not really. Is there any reason to assume that his decision to call the switch took place in a vacuum? Not really.
One difference between leaders and followers is the willingness to make big changes on small justification. The people capable of thinking farthest outside the standard comfort zone are the ones who see the interesting possibilities first. You can build a retroactive chain of continuity after the fact, but those 'reasons' are *not* the reasons for taking the leap. They're merely reasons the person in charge (in this case Jobs) didn't reject that particular leap out of hand.
FOSS is doing the same thing to the entire software business model. People in the industry who Just Don't Get It (*cough*DarlMcBride*cough*) talk about how FOSS needs to be made 'acceptable to business' if it wants to 'succeed', because they can't think far enough out of their comfort zone to see how 'giving stuff away' works better for everyone.
One hallmark of genius is the ability to bridge the gap between "nobody ever thought of that before" and "totally obvious to anyone who sees it." I don't think this particular change rates as high as 'genius', but there's a similar gap between "unthinkable" and "justifiable after the fact".
Apple makes its money selling a vertically integrated solution stack.. hardware, OS, middleware, userland, and a collection of network services like .Mac and the iTMS.
Most of Apple's profits come from the hardware, not the software.
The software is basically just a value-added service that makes people think it's worthwhile to pay for the hardware.
Dropping into the world of commodity x86 boxes would kill Apple's main source of profits, so I doubt we'll see that happen.
It's also important to note that having 90% market share is not an unmixed blessing. Just as a program spends 80% of its time looping through 20% of the code, 80% of the profit in any given market comes from 20% of the consumer base. And 80% of the expense comes from having to support the 20% of consumers at the other end of the scale. We've seen examples of that latter idea all over the place recently, with stores giving premuims to repeat customers, or refusing to provide service to 'problem' customers.
I doubt Apple really wants more than the 20% of the market that falls at the opposite end of the traditional Linux hacker space.. people who are willing to spend money to solve a problem rather than spending time and energy. One, it's more profitable, and two, FOSS is doing such a good job of serving the high-maintenance, low-profit end of the market that nobody can really compete with it. Both market groups are perfectly valid, they just approach the old "fast, good, cheap: pick two" formula in different ways.
But the simple fact is that PJ stepped up and made herself a public figure in an extremely controversial case.
I think if you check the definitions, the term "Limited Purpose Public Figure" is more appropriate. A public figure is a famous person about whom the public has a broad, general interest. A limited purpose public figure is someone who makes themselves visible in a specific issue, for purposes of influencing the outcome of that issue.
Both types have to prove actual malice in order to win a defamation case, but where a PF's whole life is up for grabs, a LPPF is only 'public' regarding things that are relevant to the issue in question. Think of it as requirements for admissibility of information in court: For a PF, any statement about them at all is admissible. For a LPPF, only statements about issues where they've 'pushed themselves forward' are admissible.
So.. the questions we have to consider are: "How much of PJ's personal life is relevant to the SCO -v- IBM case?" and "How much of her personal life has PJ 'pushed into the spotlight' on Groklaw?"
Personally, I think the answers are, "damn little," and "damn little," respectively.
Yes, knowing whether PJ accepts money or other compensation from IBM is relevant to this issue. No, her personal appearance, apartment number, the opinion of her landlord, and the contents of her car's back seat are not.
It's also important to note that PJ doesn't stonewall the press completely. She does communicate with people, and she does do interviews. She just prefers not to do them face to face. It's not a question of whether the press can get information from her at all, it's a question of what means of communication they're allowed to use.
Even public figures have the right to refuse to give interviews to specific individuals, at specific times, and by specific means.
In this case, O'Gara published information that, to the best of my knowledge has never been 'pushed into the spotlight' by PJ. In fact, it would be easy to make a case that PJ has made specific efforts to keep that information out of the spotlight. The information does not strike me as relevant to SCO -v- IBM in any noticable way. Nor do I see how approaching PJ face to face could convey any information about SCO -v- IBM that O'Gara couldn't get some other way.
The sole possible justification for face to face contact is to resolve the (IMO artificial) issue of whether PJ "actually exists." And by that standard, O'Gara's approach failed miserably. She didn't establish any meaningful connection between the woman whose privacy she invaded and the person who runs Groklaw. She claimed to be looking for PJ, but pretty much abandoned that approach as soon as she got near the door. Instead, she switched over to taking PJ's existence for granted, and spent the rest of the article listing potentially prejudicial details about the 'PJ' she 'found', and indulging in no small measure of 'we know where you live' style intimidation.
ARRRRRGH!
You're right. My typing got ahead of my fact-checker.
When you read Groklaw, you have to remember one basic fact about the western legal process: The only thing that matters is your evidence.
Bias is irellevant to the law. The attorneys on one side of a case will be firmly biased one way, and the attorneys on the other side will be firmly biased the other way.
Groklaw is 'open' in the sense that all its evidence is right out there for everyone to see, no matter which way their bias happens to point. This is different from SCO, which has still failed to identify the code it claims has been misappropriated.
And yes, Groklaw does have a clear bias against SCO. It began as an attempt to counter the FUD-storm that SCO was pumping into the court of public opinion. Remember the "millions of lines of code"? The "team of MIT researchers"? Darl's Open Letters too the Linux Community?
SCO went into this litigation playing a very hard flavor of hardball. They made large, bold, easy to understand claims, and left the tech community to respond with arguments that were logically correct, but too complicated to win any mindshare. And it worked. For probably the first year of the case, SCO's side of the story was the one that was heard in the mainstream media.
So PJ began reading documents and collecting evidence, and embarked on the long, slow process of chipping away at the FUD one little bit at a time. Over time, she developed a reputation for countering SCO's claims reliably, thoroughly, and on the basis of hard evidence. Eventually, the mainstream press started to pay attention, and started treating SCO's claims more critically than they had before.
At present, SCO has the FUD machine pointed at Groklaw and PJ herself. Within the last month, Darl was quoted in this article saying:
"if you look at the reality of the Pamela Jones situation, you have to conclude that all is not as appeared as it is in Groklaw land. We appreciate that many media sources disagree with us, but they're accountable. We think you need to know who's behind the news."
and:
"we're digging into who Pam Jones is, and we're close to the bottom."
because they'd really like people to see Groklaw as a haven for anti-SCO fanboys, and question (or better yet, ignore) all the evidence that Groklaw has accumulated. Then SCO could get an unfettered run at the media again.
And if they get it, I invite you to consider just how 'open' SCO's comments to the media will be.
The company that used to be named "SCO" is now doing business under the name "Transmeta."
Caldera (aka: newSCO) bought the IP rights to the work done in Monterey from oldSCO (now Transmeta), and those rights are at the core of the present lawsuit.
So.. by showing that Caldera (now newSCO) knew Monterey was heading to POWER back before the Monterey project ended, PJ has produced evidence which directly contradicts the claims newSCO (then Caldera) has made in its third amended complaint.. namely that newSCO (then Caldera) thought:
Interestingly enough, the company formely known as "Caldera" didn't change its name to "The SCO Group" until after it filed its lawsuit against IBM. Questions like the one you just asked show just how much confusion that bit of misdirection has caused.
For the record, Orlokowski is the one responsible for the whole "RMS 'can't believe' that Miguel de Icaza is working on .NET'" non-event a couple years back. (link)
.Net was inaccurate starting from the title."
IIRC, RMS actually said, "I don't believe he's doing that," when answering some specific question about de Icaza's work, in the, "I think that's incorrect," sense. Orlokowski got it secondhand from someone who attended, misinterpreted the out-of-context quote, and chucked a story onto the Reg's front page before doing any journalistic stuff like contacting the primary sources and checking his facts.
RMS wrote a response (link) which begins: "Your article about me, GNOME and
The Reg has a habit of blurring the boundaries between 'reporting' and 'editorializing', which is part of its charm. Sometimes that means not letting the facts get in the way of a that will draw page views, but at least they don't do it too often.
This particular case happens to be pretty silly, though. For some reason Orlokowski, a general pundit who's never claimed any kind of in-depth legal knowledge, decided it would be a good idea to trade legal opinions with PJ, a paralegal who's read, discussed, archived, and cross-referenced every public document relating to this case that she can get her hands on.. not just the complaints, briefs, motions, responses, etc actually filed in court, but also a whole ton of external contextual information.
By and large, I think it's pretty safe to bet that PJ knows what she's doing.
He said right in the article that the problem he had was with the marketing. It doesn't change the way you use your computer. At least stick to the things he is wrong about.
Fair enough, but he does sort of miss the forest for the trees when he dismisses Spotlight as a kinda-neat-Windows-like-search-thingy. Being able to organize files by metadata (a smart folder with 'all the black&white photos I've worked on in the last couple of days') is 'a different way of using the computer' compared to building, maintaining, traversing (and forgetting) a series of folder hierarchies. He may as well have said the GUI is no big change from the command line because the underlying file trees are still the same.
And Automator at least has the potential to be as big a step as moving from 'single commands on the command line' to 'command line with pipes'. From a simple task-based analysis, the job: 'find all the black&white photos I've worked on in the last couple of days, convert them to JPEGs, dump those onto a CD, then burn it' can be done very differently in Tiger than it can in Panther.
Differently enough, in fact, that the phrase "change the way you use a computer" does have some merit, IMO.
1 - Vertical integration: Apple sells a (hardware, OS, software, network services) package where all the pieces are tweaked to work well "right out of the box." The fact that Apple controls the entire product stack gives them a lot of flexibility when it comes to tweaking the system. The applications people can push design requirements to the OS team, and the OS team can push requirements to the hardware people.
/. Linux geek. Where the T/.LG resolves the 'soon/cheap/good - pick two' formula by picking 'cheap and good', Apple wants the chunk of the market that's willing to choose 'soon and good'. The T/.LG solves problems by throwing time and effort at them, while the ideal Apple customer throws money at them.
They'd lose that flexibility if they had to support machines designed by other companies with different agendas, and Apple would get most of the blame for anything that failed to work.
2 - Application base: Switching from PPC to x86 would force end-users to buy new software. No matter how you crunch the numbers, you can't save enough on performance-competitive x86 hardware to make up for having to buy a brand-new copy of Photoshop OSX-86.
Apple would also have a hell of a time bringing third-party developers on board, because even if compiling for the new platform were as easy as clicking a button (and it never is), it'd be one more SKU to worry about.
3 - Market position: Contrary to what most slashdotters think, it is *not* a good thing to own 95% of the market. The 80-20 rule applies to market profitability just as much as it applies to code use. 20% of the market generates 80% of the profit, and 20% of the market generates 80% of the expense.
Microsoft holds 95% of the market.. not because it wants to.. but because it has to. It continues to supply market segments that are a dead loss financially just to maintain the standards-and-protocols lock that keeps its profitable customers from wandering away.
There's no business reason for Apple ever to want more than the 15-20% of the market that generates 80% of the profit. Below that line, vertical integration becomes less important than price point, and the margins there are so thin that Dell (the big dog in that space) is branching out into camcorders and whiffle bats because it can't survive just by selling computers alone.
The thing slashdot needs to understand about the 80-20 rule is that 'small market share' can equal 'more pay for less work' if you hit the right part of the market. It's the choice between supporting two people who are willing to pay $4 each, or ten people (including two nuclear assholes) who'll net you a total of $10.
As an aside, Linux is actually helping Microsoft (temporarily) by taking over some of the market segments that cost Microsoft the most money. That will change in a few more years, though, when Linux becomes mature enough to compete for market share in segments Microsoft can't afford to lose. It's already threatening Microsoft's standards-and-protocols lock, and when 15% of the market embraces Open standards and protocols, things will get interesting.
4 - Brand support: Apple survives as a vertical integration vendor because it has a reputation for quality. Without its reputation for quality, Apple couldn't compete for the 20% segment that generates 80% of the profits. There's no way to embrace the x86 market without also embracing the crapfest at the low-20% end, though, and doing so would shoot Apple's reputation for quality straight through the head. That would knock Apple out of its high-margin target segment, and put Apple in head-to-head competition with Microsoft on Microsoft's home turf.
And history shows that competing with Microsoft in Microsoft's core markets, on Microsoft's target platform, is suicide.
---
Basically, Apple's core market has different priorities than the traditional
Neither approach is inherently better or worse
Well, my first reaction is that the leaker should have taken the time to go through the box before leaking the whole thing.
The court will weigh the interests on both sides, though. A judge would only release the leaker's name if Enron's actual injuries from the release of that contract outweighed the leaker's right to anonymity.. wiping out a billion-dollar deal to tell the world that Enron's CEO overstated his income tax deductions by $250, for instance. But unless it can demonstrate significant real injuries from the release of that one specific contract, Enron probably won't be able to force the court to release the leaker's name.
And even if Enron does get the name, it will have to be careful about what it does. The whistleblower protection laws specifically forbid companies from taking revenge.
If Enron does anything to the leaker, the leaker can sue on the grounds that it was revenge. And judges want to maintain a chilling effect that prevents companies from taking revenge, so Enron will have to turn itself inside out to prove its actions were absolutely squeaky clean and that it did everything but buy the leaker a pony and some ice cream to make the experience as pleasant as possible. If it can't, the judge will award the leaker a big 'ol pile of cash in order to remind all the other companies out there that beating up whistleblowers is A Really Bad Idea.
So basically, judges take care of the middle ground, and they work hard to make sure everyone gets fair treatment.
Apparently no one here has noticed the part of the article which reads:
"Larry is perfectly fine with somebody writing a free replacement."
This is not a question of whether McVoy is trying to crush free clients for his software. It's a question of how those free clients should be written. And Tridge should know the difference since he wrote this article on how Samba development is more about network and protocol analysis than reverse engineering.
One paragraph from that article reads:
Classical reverse engineering techniques in software engineering revolve around the use of disassemblers, debuggers and other object code analysis tools to examine directly the executable code of an existing product in order to create a "clone" that behaves in the same way. While these venerable techniques have been successfully used by many groups, they are not what we use in the Samba project.
So clearly Tridge does make a distinction between ripping the binaries and doing protocol analysis.
This is not a debate about Openness. This is not a debate about Freedom. It's a debate about methodology, and Tridge knows the distinction between cloning a binary with a disassembler and re-implementing a piece of software from the protocols better than most people do.
1. Yeah, generally. 'Civil offense' or 'civil tort' would be more correct, but 'crime' is good enough for street usage.
/. (and let's just test that theory by cracking into the Slashdot servers, stealing the username/password database, and giving it to a blogger who will then publish it 'in good faith' shall we? Everyone has a 'right' to a low uid), no one has yet advanced a really good line of reasoning to suggest that the anyone was actually injured by Apple's decision to keep its product specs secret. Therefore, there is no legitimate claim of 'public interest'.
2. No. Enron could put the words "As part of my job, I agree not to blow any whistles" on a piece of paper and require all its employees sign it, but the courts wouldn't recognize that as a legal contract.
Contracts only work because the courts agree to enforce them. There's no actual law that says, "employee X will obey the terms of document Y while employed at company Z." There can't be. There are too many deals taking place every day to have specific laws for each and every one. Instead, we have what you might call the general specs for agreements between two parties, and those specs are called 'contract law'.
If you and I make a contract, then I refuse to hold up my end of the deal, I haven't (necessarily) broken an actual law.. you can't take me to court saying, "he broke statute X.Y.Z" (s criminal offense). All you can do is take me to court and say, "here's the contract that explains what he agreed to do, and he hasn't done it" (a civil offense). If the judge decides that our contract lives up to the general specs, she will then order me to deliver what I promised. And there *are* laws that say I have to obey the orders of the court.
A contract that doesn't live up to the specs is called an 'illegal' contract, but again, you don't get thrown in jail for writing or signing one. It just means that the courts won't enforce it, and if the courts won't enforce it, it's just a piece of paper with a bunch of funny words on it.
And contract law says a legal contract CAN NOT require someone to do something illegal. One of the precedents involves a guy who 'sold' a car to a woman in exchage for a certain amount of sex.. 10 visits to his bed rated at $50 a pop, or something like that. She stopped somewhere around 5, so the guy sued her for the rest of the promised nookie. The judge said that sex for pay is prostitution, prostitution was illegal in that state, and therefore the 'contract' was not legal. The brief is actually quite an amusing read.
Any Enron that tried to sue a whistleblower for violating an NDA would get a brisk "fuck you" from the courts if the secret involved anything actually illegal. And there are laws that specficially protect whistleblowers who reveal secrets that don't involve illegal activity per se, but are still 'in the public interest'.
----
However, as the judge in this case clearly pointed out, 'an interested public' is not the same as 'public interest'. The dividing line has nothing to do with curiosity, and everything to do with injury. 'Public interest' stops at 'preventing the public from being injured' and does not extend to 'what the public might be willing to watch on TV.'
If you violate your NDA to tell the world that the safety tests for a new airplane were forged because the company found a major flaw in the design and couldn't fix it before the test results were due, you're working in the public interest. Keeping the secret exposes people to risks they don't know about, and the courts will rule that letting people know about the risks is more important than obeying your NDA.
But no matter how tortured the "companies shouldn't be allowed to keep secrets" arguments get here on
----
And while I'm ranting, I'd like to take a minute to say that trade secrets and NDAs aren't just evil tools of big Sauron-like corporations. They exist to protect the little guys like you and me.
Leaks hurt cor
In response to the first post above, contracts don't have to be written and signed. Dig through _West's Business Law_ for some discussion and precedent. The real proof is in the behavior of the people who made the agreement. If two parties act like they signed a contract, the courts will generally rule that a contract exists.
In response to the post directly above, a contract can be illegal and still exist. The court will refuse to enforce an illegal contract, and will often tell people to give back what they took under the terms of an illegal contract, but that doesn't stop the contract from existing. Yes, that's a fiddly distinction, but we are talking about law, here.
Besides, I agree that the sources did something illegal, that the ThinkSecret NDA doesn't provide a legal excuse to keep those identities secret, and that the judge made the correct ruling. But I've seen too many people denouncing Apple, the judge, the law, and society in general because they think NDAs shouldn't exist, all the while demanding that ThinkSecret's NDA be judged legal and enforcable.
Current DRM schemes are no more flawed than the locks on your home or your car. Just ask any burglar or car thief. And even if the locks were truly unbreakable, the windows of your home and your car aren't. On the computer side, 8-character user account passwords are almost trivially easy to crack these days.
The most basic rule of security is that no security system is, or ever will be, perfect. Eventually, we always have to say, "oh well, their hearts are in the right place, so let's just honor it." We say that about locks, user account passwords, credit card authorization systems, or any other technology which restricts people's actions.
We draw the line in the sand where it provides the best balance between various opposing interests, not based on the perfection of the system preventing misbehavior.
The kind of argument you've offered could be extended to the idea that anything not 'effectively' prohibited must be allowed.. in other words, "if I can do it, I should be allowed to do it." That may sound attractive at first blush, but nobody wants to be on the receiving end of it.
Ultimately, we say, "let's just honor it" in support of someone else's flawed technology because we want them to say the same thing in support of ours. None of us can ever achieve perfect security, most of us can't afford truly excellent security, and few of us are diligent enough to manage even consistently good security. We agree to pretend that flawed systems work because we don't want to put up with the hassles inherent in living with something more effective.
It's fine -- in fact it's necessary -- to discuss exactly where we want to draw the line in the sand. But in those discussions, we always have to remember that we're talking about balancing interests, because in the long run, your right to life, liberty, and the pursuit of happiness interferes with my right to beat the shit out of you with a brick, and vice versa.
Arguments on the order of "I'll do it because you can't stop me" aren't useful, nor are arguments on the order of "anything that limits my freedom is wrong." Anyone can demonstrate how both theories are flawed given five minutes.. and a brick.
In this case, Apple's DRM strikes a fairly good balance of interests between the people who listen to music and the record companies. It's a compromise, i.e.: an agreement that's equally unacceptable to all parties involved. There are lunatic fringe elements on both sides of the debate, but at the moment, the ones on the record label side represent a denser concentration of serious idiocy backed up by a whole lot of contract negotiating power.
IMO, Jon is losing the war by winning the wrong battles. Every time he exploits a hole in Apple's current technology, he hands ammo to the most determined idiots on the record label side. If Jon manages to convince the labels that FairPlay will never work, the labels won't give up on DRM. They'll just throw their weight behind Microsoft's DRM and start lobbying congress for unique ID chips in every electronic device.. again.
Yes, the labels will eventually learn that no amount of DRM will ever be perfect. But how many years of increasingly draconian measures do you want to put up with while Jon makes headlines convincing the labels that the current version is still too permissive?
Has anyone else noted the irony that this whole issue is really about two NDAs?
On the one hand, we have the NDA between Apple and whoever leaked the information. People go to great lengths to explain how breaking that NDA, and/or publishing information gained from someone who did break that NDA, is Freedom Of Speech Goodness Galore.
On the other hand, we have ThinkSecret's promise of anonymity to its sources.
Now, if you think about it for five seconds, that pretty much boils down to another NDA, aka: an Agreement Not to Disclose information. But this NDA has to be protected at all costs because, again, that's Freedom Of Speech.
If "All Secrets Limit My Freedom", as some people have argued, and "Any Judge Who Enforces An NDA Is Pissing On Freedom Of Speech", as has also been argued (repeatedly), what makes ThinkSecret's decision to withold information so good?
"If ThinkSecret gave up the names of its sources, it won't be able to attract sources in the future," you say? But doesn't that pretty much boil down to the statement: "ThinkSecret uses NDAs to protect its business"?
And this is different from Apple's NDA.. how?
If the people who gave the information to ThinkSecret didn't break any NDA, then Apple would have no case against them.
BUT, the judge can't decide whether the people in question broke an NDA until it knows who they are, now can he?
This decision deals with a subpoena.. a document the courts use to gather information the judge needs in order to make a good decision. It doesn't say that a crime was committed, that ThinkSecret broke some kind of law, that ThinkSecret owes Apple some kind of damages, or anything of the kind. It just says ThinkSecret has information the judge needs.
We can recast this whole thing in terms of the judge wanting to give ThinkSecret as much benefit of the doubt as he can. Let's imagine he wants to say, "ThinkSecret didn't do anything wrong".. he can't just wave his hands and do that, though. He has to cite some kind of evidence.
As it stands, the only resonable way the judge can get that information is to ask ThinkSecret for the name of the source.
Bottom line, the judge can't declare ThinkSecret not guilty until he gets the information about the source.