This company is restricted from selling their software (Hmm, I wonder if they could give an NDA to their customers prohibiting the re-release of their source code?) by the GPL. Their "rights" are not being protected here.
Two things here:
The GPL does not put restrictions on selling software per se. It puts restrictions on the restrictions you can put on downstream customers (namely, they must be given the same rights and restrictions as the GPL itself). Which leads to
The GPL explicitly forbids putting further restrictions (such as nondisclosure) on the GPL'ed source (see section 2b)
I can't find what the GPL says about modifying a program and then not distributing it. The closest thing I can see is in section 2, but it does not state that you must distribute your modified work, only that the modifications must meet certain standards (the modified files must contain appropriate notices, and the command when run must print out a copyright notice). But it only says that works that you distribute or publish must be covered by the GPL. So I presume from that that creating a proprietary service based on a mix of GPL and proprietary code, as long as only the output is distributed (and the output itself isn't a "work based on the Program", such as bison output), is OK. But I'm not a lawyer.
What the article omitted is the issue of how quickly bugs get fixed. Certainly patching things on the fly is much easier with open source, and there are alternate ways to get fixes. This, of course, can be a weakness, if you're not careful who you get your fixes from. But overall I think it's a strength.
Re sendmail and the DEBUG (and WIZ) commands: those commands were quite well known in the 1980's, but one has to look at the times to understand the issue in context. Until the Morris worm, people never really paid too much attention to back doors; the Arpanet was much more of a friendly club than the Internet of today. Also, sendmail wasn't really open source back then anyway; the vendor supplied it as a binary (often with proprietary extensions). You could plug in your own version of sendmail if you were motivated to, but you'd often lose something in the process, such as YP username resolution or some such. I still remember binary patching sendmail to turn off the DEBUG and WIZ commands because they were so obviously braindead (no particular kudos for that; just find the DEBUG and WIZ strings and overwrite them with nulls).
I don't think that "more people checking source" = "more out there introducing bugs". Most people who hack the source are doing it for their local site. They may well be introducing bugs, but they're not spreading very far. Certainly they're not likely to propagate back to the master source tree, assuming that the people controlling the master source are anywhere on the ball. Most people also don't need to get open source packages via back channels; it's just too easy to get it officially (most of the major Linux distributions include a ridiculous number of packages). One can argue about where they get their source -- with some validity, at least in principle -- but for players such as Red Hat, SuSE, and the like they employ enough known clued people so that there's a good chance that they're getting their software from reasonably trustworthy sources. Along those lines, I'd be more worried about warez; that always comes from back channels, and there are too many places where it's easy to stick nasties on (just do a rogue installer that installs a virus along with the real package).
Aside from moral issues, the problem with vetting access to open source software is the usual chicken and egg problem. How do you verify someone's bona fides if they don't get a chance to do something in the first place? For the maintainer of a package deciding whether to accept a change it's reasonable that that maintainer use whatever standards he or she wants. For simply inspecting the code, or making local modifications, why go through all this?
You GPL zealots really need to read your precious license sometime. It is deliberately vague and open to many interpretations. The only way to get this settled is for some deep-pocketed company to take the FSF to court over it. Hopefully that will happen someday soon, it will make these pointless discussions a lot simpler...
I doubt that the GPL is deliberately vague. To my (non-lawyer) reading, it's actually quite clear and precise. Yes, it's true that a lot of people misunderstand it, but I think that that's due to people reading what they want to read (on both sides) into it.
What about all of the free software developers who are not high school/college students, then? Who are you to dictate what other people's incentives may be? Why must everyone be looking for a high paying job?
Where would software developers find jobs in a world of all free software? Companies still need software written for their in-house IT needs. People would need applications written and would pay for those to be written. The software world would be different, certainly, but it would still exist.
The idea that _companies_ *must* give software, with source code, away for free doesn't really make sense. If companies were not going to make any (or much) money off their products, then the product wouldn't even exist in the first place.
Seems to me that the very existence of Red Hat, SuSE, etc. pretty much puts paid to that idea, I should think?
Maybe the fact that you're giving such an outrageous salary requirement is part of the reason you're getting interview you don't want? Perhaps the going salary for what you really want to do is out of line with your expectations.
I was very pleased to see a fairly mainstream publication cover Mandrake's model, why this is considered acceptable -- even desirable -- in the open source world, and the difference between Mandrake and LinuxOne.
Enough coverage like this and more people might start to understand the rules and mores of the open source community, and why it's as effective as it is at getting things done. Rather than fight things out in court, things are fought -- and borrowed and exchanged -- in the market.
If only the hard core free market capitalist types would wake up and realize that open source is one of the closest approximations there is to a classic free market. But I suppose free markets are too unpredictable and not profitable enough...
And why should companies restrict themselves? Isn't capitalism all about dog eat dog, and who can screw the opposition out of the most money? So why should they care about who will be affected when they apply for a patent for using red text on their billing form instead of blue?
The next time anyone starts talking about "free market capitalism", ask them to clarify whether they mean "free market" or "capitalism". Patents may be a tool of capitalism, but they certainly aren't part of any free market.
I hate spam as much as anyone, and I report both the spammer and any drop boxes to the appropriate service provder. That said, though, this ruling is good.
Anyone remember the case of the California resident who set up a web site that some sheriff in Tennessee (I think) decided was indecent? The sheriff entrapped the guy by sending him a package of kiddie porn. The California cops were tipped off, and the guy was arrested and extradited to Tennessee. Of course, the kiddie porn charges were dropped (there was obviously no case), but he was now in Tennessee and tried -- and convicted -- for indecency or obscenity based on his web site in California.
Distinguishing addresses in Washington from addresses anywhere else in the world is well-nigh impossible, a fact that we all recognize when it comes to porn, or the CDA. An electronic registry is really no better. Think about how many jurisdictions there are in the US, not to mention the world, each of which might have their own registry (think Australia, with its new censorship laws!), and it's clear that this is an impossible burden. Furthermore, the act of gathering such a list itself has its own privacy implications, and I could well imagine some spammer taking revenge on this registry.
Now, prohibiting the SENDING of spam from Washington would be quite another matter -- a state has legitimate jurisdiction over what takes place within its boundaries. A federal law prohibiting the sending of spam from anywhere in the US, or by any US citizen from anywhere, would also be another matter. Putting a burden on the sender to prove that a potential recipient or consumer of information is not a member of a certain class is just asking for trouble.
Economic systems derive from the fact that resources are limited. An economic system is a system that provides for the division of limited resources in the face of unlimited demand. Communism believes that resources should be centrally distributed based on need. Capitalism believes that treating resources as private property provides for the most efficient division of resources.
But if the cost of duplicating and distributing something is negligible (as is the case with a lot of information), then the assumption that that resource is limited is invalid. The concept of "intellectual property" is attempting to force a model of a limited resource onto something that isn't limited. But you note this below.
(I'd also quibble with the notion that demand is unlimited -- even if storage space was perfectly unlimited and free, there's a lot of stuff I'd have no interest in.)
Because capitalism is not concerned with who actually needs a resource, it can appear cold-hearted. Because it is efficient,however, all members of a capitalist society eventually end up richer.
Capitalism nominally operates on the presumption that a free market with perfect information is the most efficient way to distribute resources. Whatever the truth of the presumption (and the few markets I've seen that really are free with good information, such as the market for free software, are rather convincing on a microscale), producers (and others, such as governments) inevitably attempt to rig the market in some way or other. Everything from restrictive software licenses to government regulation to the sheer difficulty an individual has in evaluating a complex, unfamiliar product reduces some of the freedom of the market. I'm not arguing that government regulations should be scrapped en masse -- there are very good reasons for a lot of them -- but rather to recognize that there are very few markets that are truly "free" in the classical sense.
Nor would a truly efficient market ensure that all members of a free market society would eventually end up richer. Certainly many people on all sides believe that restrictions on the freedom of markets allow them to enrich themselves beyond what they would achieve in a truly free market economy, and surely many of them are right. I find it hard to see how record company executives, for example, would be richer in a truly free, unencumbered market than they are in the current situation, and it's evident that they agree most strongly.
My problem with "intellectual property" in a free market system is that it enacts artificial barriers to the spread and use of information. The issue isn't so much with the initial sale -- one could construct a transaction in which the purchasers of a piece of music explicitly enter into a contract with the seller not to further distribute the CD -- but that it imposes restrictions on third parties not party to the transaction. Patents do this in a very obvious way, but so does copyright and trademark (to a lesser extent).
Private property is thus a horrible analogy for intellectual property. To determine a new paradigm for intellectual property, we need to think about exactly what we want from intellectual property. I believe it is this: We want the free flow of ideas only encumbered in such a way as to maximumally encourage the creation of new ideas.
Exactly. If we're very clear what it is and what its purpose is -- the Constitution of the United States makes it clear that the purpose is "to promote the progress of useful arts and sciences" or some such -- then we can start discussing it on a more rational plane. But we have to stop using the term "intellectual property", since it carries the wrong connotations. If we choose terminology that makes it clear that the purpose of this law is to promote progress by giving potential creators of information extra incentive to create and distribute their work, then we start from a position of "the producer, consumer, and third parties who otherwise legally come across the information all have rights, let's work out how they should best be balanced" rather than "the producer inherently owns all rights, and let's find the minimum set that must be given up to allow things to function at all".
What would it look like? Well, I have some ideas:
Copyright would retain enough force to allow authors who wish to be paid to be paid reasonably for the specific expression of their work, while third parties who prepare significant derivatives of the work (such as Trekkies) have the same rights to their derivation. The practice of forcing music fans to buy a CD of trash to get one good song would go by the wayside. This is basically "tying", which is a big part of the Microsoft trial (forcing system vendors to install Internet Explorer if they wished to install Windows). It would also allow creative derivation from existing ideas (in other words, the Disney characters should not be copyrightable; specific movies featuring them would be).
Patents might become voluntary in some fashion, perhaps by means of an appropriate antitrust exemption. If not, patents would be granted as narrowly as possible, for as short of a term as possible, to encourage innovation, and each case would be reviewed carefully with the presumption against granting the patent.
Trademarks would serve to identify the producer to the consumer to ensure that the consumer isn't being fooled. Use of trademarks for any other purpose (such as McDonalds trying to lean on anyone who uses "Mc" for just about anything, or console manufacturers whose software looks for the trademark in the game cartridge and refuses to play anything without the trademark) would risk forfeiture.
openssh.org does not attempt to set a cookie. It does not contain any scripts or applets. Its HTML is perfectly vanilla, and it doesn't even have any meta tags to redirect search engines. It also contains a link to openssh.com.
(btw, if FFT's aren't primarily used for image enhancement, I apologize. my pain point in bringing it back up is that in any complex process, all it takes is one single step that can be paralellized efficiently to make a beowulf worthwhile - and then you gain a huge money/performance ratio.)
That's exactly wrong. The point of Amdahl's Law is that all it takes is one sequential step to ruin the ability of an application to scale well. Amdahl was not a proponent of parallel computing; he liked big mainframes, and came up with his "law" to demonstrate that fast single processors were the way to go, not parallelism. I don't agree with his position (about parallel vs. sequential in general), but absent issues such as memory contention it's basically a pretty firm rule.
Also, Amdahl's Law sets a hard upper limit on parallelism. The ultimate speedup constraint imposed by the number of processors and the fraction of sequential work only applies in otherwise perfect conditions -- the communication cost is zero, and there is no inter-processor contention. These can be true in some circumstances (an embarrassingly parallel app on a pure MPP). A Beowulf cluster qualifies as a true MPP. Very few interesting apps truly qualify as embarrassingly parallel. Communication on a Beowulf cluster with anything less than a high quality gigabit switch (or better) is definitely not close to free.
Just because video consists of a stream of images doesn't mean that the images are independent images. I presume that good video (as opposed to photographic) image techniques rely heavily on correlation between frames
Well, an FFT's inner loop (butterfly) is pretty simple, something like 10 arithmetic ops if I remember. Higher radix helps, and there are other tricks to reduce the number of communication steps (address reordering), but it's still doing a lot of communication.
Also, answer my question about N-body problems.
I agree that the kinds of problems a supercomputing would be working on are usually very parallel. That doesn't mean that they're computationally easy to parallelize, particularly with limited communication bandwidth. If they require access to a shared resource (at least if it's update access), or the communication needs are heavy, it's going to be very hard to parallelize efficiently on a Beowulf-type cluster. That's where you need the massive bandwidth that a Starfire or SP or Connection Machine offers.
(And yes, I have programmed, from the inside, a Connection Machine. Worked for them for over 7 years, in fact, and spent most of that time tuning the FFT and suchlike. Even with a really good ratio of bandwidth to computation power, it still took a lot of work to get good performance even from a lowly FFT or matrix multiplication. And even though complete applications usually do FFT's and matrix multiplications in parallel, rather than one large one across the entire dataset, they usually have even more communication taking place than the core routines do.)
(this example uses 5 threads, accessing a common memory segment).
That's very nice, they're accessing read/write a common memory segment. Now, pray tell, how is this accomplished on a Beowulf efficiently?
As it happens, you've conveniently chosen something that's embarrassingly parallel anyway (no dependencies between elements not belonging to the same node, and no sequential component such as sequential I/O). I could put a node on Earth and a node on Mars and get perfect speedup in that situation. Think about something like finite element analysis with nearest neighbor interactions, or FFT, or n-body simulations (in order of increasing difficulty to scale well on most architectures).
And no, not every task can be broken down into multiple threads. As an example, consider decrypting a stream (where you know the key, you're not trying to break it or anything) that was encrypted with output feedback (simple electronic codebook encryption, where there's no dependence on prior state, is pretty easy to break with traditional cryptanalysis). You have to know the entire previous state of the message stream to decrypt the next block. That cannot be parallelized by any technique that I'm familiar with.
Not really; GoHip's ISP isn't publicly certifying anything about them; they're a common carrier. VeriSign is representing something about GoHip publicly, so I'd say they have a bit more responsibility. If VeriSign wants to certify something that weak, that's fine, but then perhaps VeriSign's not providing as strong a service as they should.
The Boston Globe actually does a fairly good job of covering technology issues. Hiawatha Bray and Simson Garfinkel do write columns about important technology issues, even if I don't always agree with them.
The whole issue of "timeliness" is overblown. So what if I don't find out that Bush beat McCain in South Carolina for a few more hours? Most of those hours I was asleep, anyhow. Likewise for the "graphically impaired" issue. Why do I need all that eye candy, anyhow?
There are a few things that newspapers do really well that won't happen digitally any time soon, if ever:
1) They're convenient to read over breakfast, or in bed on Sunday morning.
2) They're conveniently organized. Now, you're saying "WHAT!!!???" Yup, they're conveniently organized. The world/national/major local news is in the first section, the local stuff is in the second section, and so forth. No, it's not organized by subject or anything. That's just right. I get to see more than just headlines; if something catches my eye, I'll read it.
I don't know how many equivalent pixels a newspaper page is; I'd guesstimate 30Kx50K or thereabouts for a single broadsheet page. That's so much more visible information than I can fit on my screen that it's laughable. Plus, the screen forces me to be in a certain place.
Like it or not, the "information revolution" is still a relatively small part of life. Eating, sleeping, having shelter and transportation, are still more important to our lives than what happens on the net. The majority of people still don't own computers. The same vapidity that permeates popular culture permeates cyberspace.
In my spare time I do hack. I do other things, too, and I need to do them more. I could use more exercise, which I've lost a lot of time on in the past few months due to a project (a print driver I'm working on). I haven't played my viola in a year, and I could use doing more of that. I've been on the net for upwards of 15 years (I'm a spry old 36). Yes, it's nice that due to the net there are people working on this print driver in Japan, Germany, and elsewhere in the US, but it's not my whole world either.
Software patents, DeCSS, and such are important (just ask my wife how angry I was when the injunctions were granted), but they really aren't the whole of everything. People aren't dying over them. Yes, it is important that sanity prevail here, and the current state of things isn't very sane, but big corporations have always been trying to take over everything and this is just the latest; we have to be just as vigilant as always. And we could use a little perspective beyond our little online world.
Don't get me wrong -- the net does bring important benefits. I've done a lot of research on various ailments and such that various members of my family have, and it's a lot easier to do than it would be in a traditional library, as one example. It can add a lot to life, if it's looked at as something more rather than a replacement for anything. But on the whole I think we're too wrapped up in our geek culture around here, and could use a little more getting out into the world of matter rather than just the world of information.
If a customer reports a bug, the presumption should be that it's of critical severity until the customer stipulates otherwise. If beta testers (and W2K "beta testers" are a fairly general population) are reporting things, the onus is on Microsoft to handle the situation, but calling it "niggling" is inappropriate unless the tester/customer says so (and customers will happily admit that something isn't important if that's what they really think; a lot of people like to report everything on principle but will happily characterize the importance of the issue because they understand that development time is limited and they want the software).
And yes, if a customer claims a misspelling or something is a serious issue, I'd be inclined to accept it as so on first glance -- note, for example, the infamous Chevy Nova ("no go" in Spanish). That doesn't mean that it absolutely must be fixed to ship, but if a customer says that something is important it should be treated as such.
Saying that the product shipped with no showstoppers doesn't mean very much, unless there's a really hard and fast rule about what constitutes a showstopper and the people with shipping responsibility have absolutely no control over the indicated severity). It's very common for something to be marked a showstopper, but then people start asking questions such as "is there *really* no workaround?"
While the 65000 bugs sounds outlandish (and is probably somewhat low but in the ballpark, given the number of lines of code), I agree that it probably overstates the severity of the situation. Nonetheless, saying that it shipped with no showstoppers (which is more or less true by definition) doesn't mean very much either.
There was one really nasty thing of this nature in 2.7.2 (I think it finally got fixed in 2.9 something). Compiling C++, a function
foo(int a) { }
was accepted. Notice that there is no declaration of return type -- a violation of every C++ standard -- and the compiler defaults to returning int, K&R-style. Even including -ansi -pedantic-errors would not trigger an error (it would trigger a warning, but not an error). I had a very hard time convincing the gcc folks at the time that this was a real, and serious, problem.
I'm glad that at least -ansi -pedantic-errors exists; for the most part, it's quite good. It's unfortunate that something as egregious as this (and in C++, where one of the major benefits is strong typing, this kind of thing is a real problem) slipped through for so long.
Officially released or not, W2K is widely available. They've found two holes in a layered service, and they're sending out patches in a fairly reasonable amount of time.
One can argue about the wisdom of turning on unnecessary services, but that problem is not unique to Microsoft. When I installed SuSE, I had to go and basically clean out inetd. Still nothing terribly new there. That's unfortunate, but it's an industry-wide problem.
There will be security holes in W2K. If Microsoft responds more quickly and openly, and the holes are in add-on services rather than appearing systematically in the core, then maybe they're finally learning their lesson. My guess is that they'll do better than NT4 (they've really been taking a beating over this) but not as good as the better Linux/Unix distributions. But that's just a guess, too. Time will tell.
Modulo the fact that we can't send Linus' lawyers after them, Linus has, after all, explicitly blessed binary modules (with the warning about what might happen when the kernel changes, but caveat hax0r). Linus's copyright only extends as far as the kernel, so he has no say over what happens in user space.
Most of the Linux distributions offer some binary-only stuff, even if it's just Netscape. Even Debian includes non-free options, I believe, although they're treated as an appendage rather than as part of the distribution per se. TurboLinux is quite explicit about offering proprietary value-add, and they're unquestionably offering something real.
As for whether Linus should withhold use of the trademark, it's his call, but LinuxOne does appear to be in the Linux business, however tenuously (as opposed to Serious Domains, which was strictly trying to capitalize on the name, with no pretense at actually offering Linux-related goods or services). Withholding the trademark from LinuxOne puts him in the business of judging particular uses of Linux, which is a rather slippery slope. I'd be very surprised if he goes after someone offering a distribution (however cheesy) with legal use of the Linux kernel (respecting the GPL on the kernel source proper) unless they seriously abuse the trademark, maybe by saying that Linus endorses their particular flavor.
But if LinuxOne's installation is as painful as it sounds, either they're not going to go very far or they'll quickly find out that it would actually help them to add real value.
check out iwin.nws.noaa.gov. I use the text interface; look under State Data, and then under most of the states you'll find Forecast Discussion. Depending upon who's on duty and how interesting the weather is, you'll get anything from "Will continue current forecast" to a long discussion of factors influencing the weather and local effects. Walt Drag of the Boston (actually Taunton), MA office is legendary for his discussions, which recently have sometimes filled two full screens. He's clearly a big-time snow buff.
http://iwin.nws.noaa.gov/iwin/ma/discussion.html is the current and last several discussions; they're usually updated about every 6 hours (again, except for Walt, who likes reissuing them for changing conditions). The current one (as of noon EST on Wednesday, January 19) is really juicy. Assuming he's still on duty this afternoon (I'm not certain how the shifts run), his afternoon discussion will be even better. It's interesting to read, to see how these folks interpret the data and the forecast models.
The upper air stuff is usually more important...
on
New Weather Computer
·
· Score: 1
in part because the upper air network is so sparse (and so infrequently sampled) compared to the surface network. There are also all sorts of boundary layer and topographical effects that cause the surface to be not representative of the entire atmospheric column. This, of course, cuts both ways -- local microclimates can be very important.
The classic example is a deep, bowl-shaped valley -- on clear, calm nights it will typically be sharply colder than on surrounding hills. On nights with a light wind, if the air in the valley decouples from the light breeze aloft, the difference might be very sharp indeed.
However, all in all I think it would be more cost effective to upgrade the upper air network, particularly over the oceans and Asia.
Most domain names, at $35/year, are priced below their market value. That's why there are so few good ones available. People register them as if they were free, because they nearly are. But that low cost is also what creates the shortage. Domain speculators help alleviate the shortage, by making names available to people who are willing to pay the most for them. In other words, by reselling the names at a profit, domain speculators help insure that the names will go to somebody who intends to use them, or at the very least, to somebody who will pay dearly for the privelege of not using it.
It's hard for me to see how speculators in tickets and domain names alleviate a shortage. To me, it appears that they create the shortage in the first place, and then magnanimously offer to resell them at a much higher price to alleviate the shortage. That strikes me as an excellent example of extortion.
Why is an auction any more efficient than a lottery? It's nicer for the person selling the good, but if that person doesn't want to capture the most value from it, why should someone else who has no interest in using the good jump in?
Slashdot Mirror
Two things here:
I can't find what the GPL says about modifying a program and then not distributing it. The closest thing I can see is in section 2, but it does not state that you must distribute your modified work, only that the modifications must meet certain standards (the modified files must contain appropriate notices, and the command when run must print out a copyright notice). But it only says that works that you distribute or publish must be covered by the GPL. So I presume from that that creating a proprietary service based on a mix of GPL and proprietary code, as long as only the output is distributed (and the output itself isn't a "work based on the Program", such as bison output), is OK. But I'm not a lawyer.
What the article omitted is the issue of how quickly bugs get fixed. Certainly patching things on the fly is much easier with open source, and there are alternate ways to get fixes. This, of course, can be a weakness, if you're not careful who you get your fixes from. But overall I think it's a strength.
Re sendmail and the DEBUG (and WIZ) commands: those commands were quite well known in the 1980's, but one has to look at the times to understand the issue in context. Until the Morris worm, people never really paid too much attention to back doors; the Arpanet was much more of a friendly club than the Internet of today. Also, sendmail wasn't really open source back then anyway; the vendor supplied it as a binary (often with proprietary extensions). You could plug in your own version of sendmail if you were motivated to, but you'd often lose something in the process, such as YP username resolution or some such. I still remember binary patching sendmail to turn off the DEBUG and WIZ commands because they were so obviously braindead (no particular kudos for that; just find the DEBUG and WIZ strings and overwrite them with nulls).
I don't think that "more people checking source" = "more out there introducing bugs". Most people who hack the source are doing it for their local site. They may well be introducing bugs, but they're not spreading very far. Certainly they're not likely to propagate back to the master source tree, assuming that the people controlling the master source are anywhere on the ball. Most people also don't need to get open source packages via back channels; it's just too easy to get it officially (most of the major Linux distributions include a ridiculous number of packages). One can argue about where they get their source -- with some validity, at least in principle -- but for players such as Red Hat, SuSE, and the like they employ enough known clued people so that there's a good chance that they're getting their software from reasonably trustworthy sources. Along those lines, I'd be more worried about warez; that always comes from back channels, and there are too many places where it's easy to stick nasties on (just do a rogue installer that installs a virus along with the real package).
Aside from moral issues, the problem with vetting access to open source software is the usual chicken and egg problem. How do you verify someone's bona fides if they don't get a chance to do something in the first place? For the maintainer of a package deciding whether to accept a change it's reasonable that that maintainer use whatever standards he or she wants. For simply inspecting the code, or making local modifications, why go through all this?
I doubt that the GPL is deliberately vague. To my (non-lawyer) reading, it's actually quite clear and precise. Yes, it's true that a lot of people misunderstand it, but I think that that's due to people reading what they want to read (on both sides) into it.
What about all of the free software developers who are not high school/college students, then? Who are you to dictate what other people's incentives may be? Why must everyone be looking for a high paying job?
Where would software developers find jobs in a world of all free software? Companies still need software written for their in-house IT needs. People would need applications written and would pay for those to be written. The software world would be different, certainly, but it would still exist.
Seems to me that the very existence of Red Hat, SuSE, etc. pretty much puts paid to that idea, I should think?
Maybe the fact that you're giving such an outrageous salary requirement is part of the reason you're getting interview you don't want?
Perhaps the going salary for what you really want to do is out of line with your expectations.
I was very pleased to see a fairly mainstream publication cover Mandrake's model, why this is considered acceptable -- even desirable -- in the open source world, and the difference between Mandrake and LinuxOne.
Enough coverage like this and more people might start to understand the rules and mores of the open source community, and why it's as effective as it is at getting things done. Rather than fight things out in court, things are fought -- and borrowed and exchanged -- in the market.
If only the hard core free market capitalist types would wake up and realize that open source is one of the closest approximations there is to a classic free market. But I suppose free markets are too unpredictable and not profitable enough...
I hate spam as much as anyone, and I report both the spammer and any drop boxes to the appropriate service provder. That said, though, this ruling is good.
Anyone remember the case of the California resident who set up a web site that some sheriff in Tennessee (I think) decided was indecent? The sheriff entrapped the guy by sending him a package of kiddie porn. The California cops were tipped off, and the guy was arrested and extradited to Tennessee. Of course, the kiddie porn charges were dropped (there was obviously no case), but he was now in Tennessee and tried -- and convicted -- for indecency or obscenity based on his web site in California.
Distinguishing addresses in Washington from addresses anywhere else in the world is well-nigh impossible, a fact that we all recognize when it comes to porn, or the CDA. An electronic registry is really no better. Think about how many jurisdictions there are in the US, not to mention the world, each of which might have their own registry (think Australia, with its new censorship laws!), and it's clear that this is an impossible burden. Furthermore, the act of gathering such a list itself has its own privacy implications, and I could well imagine some spammer taking revenge on this registry.
Now, prohibiting the SENDING of spam from Washington would be quite another matter -- a state has legitimate jurisdiction over what takes place within its boundaries. A federal law prohibiting the sending of spam from anywhere in the US, or by any US citizen from anywhere, would also be another matter. Putting a burden on the sender to prove that a potential recipient or consumer of information is not a member of a certain class is just asking for trouble.
But if the cost of duplicating and distributing something is negligible (as is the case with a lot of information), then the assumption that that resource is limited is invalid. The concept of "intellectual property" is attempting to force a model of a limited resource onto something that isn't limited. But you note this below.
(I'd also quibble with the notion that demand is unlimited -- even if storage space was perfectly unlimited and free, there's a lot of stuff I'd have no interest in.)
Capitalism nominally operates on the presumption that a free market with perfect information is the most efficient way to distribute resources. Whatever the truth of the presumption (and the few markets I've seen that really are free with good information, such as the market for free software, are rather convincing on a microscale), producers (and others, such as governments) inevitably attempt to rig the market in some way or other. Everything from restrictive software licenses to government regulation to the sheer difficulty an individual has in evaluating a complex, unfamiliar product reduces some of the freedom of the market. I'm not arguing that government regulations should be scrapped en masse -- there are very good reasons for a lot of them -- but rather to recognize that there are very few markets that are truly "free" in the classical sense.
Nor would a truly efficient market ensure that all members of a free market society would eventually end up richer. Certainly many people on all sides believe that restrictions on the freedom of markets allow them to enrich themselves beyond what they would achieve in a truly free market economy, and surely many of them are right. I find it hard to see how record company executives, for example, would be richer in a truly free, unencumbered market than they are in the current situation, and it's evident that they agree most strongly.
My problem with "intellectual property" in a free market system is that it enacts artificial barriers to the spread and use of information. The issue isn't so much with the initial sale -- one could construct a transaction in which the purchasers of a piece of music explicitly enter into a contract with the seller not to further distribute the CD -- but that it imposes restrictions on third parties not party to the transaction. Patents do this in a very obvious way, but so does copyright and trademark (to a lesser extent).
Exactly. If we're very clear what it is and what its purpose is -- the Constitution of the United States makes it clear that the purpose is "to promote the progress of useful arts and sciences" or some such -- then we can start discussing it on a more rational plane. But we have to stop using the term "intellectual property", since it carries the wrong connotations. If we choose terminology that makes it clear that the purpose of this law is to promote progress by giving potential creators of information extra incentive to create and distribute their work, then we start from a position of "the producer, consumer, and third parties who otherwise legally come across the information all have rights, let's work out how they should best be balanced" rather than "the producer inherently owns all rights, and let's find the minimum set that must be given up to allow things to function at all".
What would it look like? Well, I have some ideas:
openssh.org does not attempt to set a cookie. It does not contain any scripts or applets. Its HTML is perfectly vanilla, and it doesn't even have any meta tags to redirect search engines. It also contains a link to openssh.com.
Certainly looks harmless enough to me.
That's exactly wrong. The point of Amdahl's Law is that all it takes is one sequential step to ruin the ability of an application to scale well. Amdahl was not a proponent of parallel computing; he liked big mainframes, and came up with his "law" to demonstrate that fast single processors were the way to go, not parallelism. I don't agree with his position (about parallel vs. sequential in general), but absent issues such as memory contention it's basically a pretty firm rule.
Also, Amdahl's Law sets a hard upper limit on parallelism. The ultimate speedup constraint imposed by the number of processors and the fraction of sequential work only applies in otherwise perfect conditions -- the communication cost is zero, and there is no inter-processor contention. These can be true in some circumstances (an embarrassingly parallel app on a pure MPP). A Beowulf cluster qualifies as a true MPP. Very few interesting apps truly qualify as embarrassingly parallel. Communication on a Beowulf cluster with anything less than a high quality gigabit switch (or better) is definitely not close to free.
Just because video consists of a stream of images doesn't mean that the images are independent images. I presume that good video (as opposed to photographic) image techniques rely heavily on correlation between frames
Well, an FFT's inner loop (butterfly) is pretty simple, something like 10 arithmetic ops if I remember. Higher radix helps, and there are other tricks to reduce the number of communication steps (address reordering), but it's still doing a lot of communication.
Also, answer my question about N-body problems.
I agree that the kinds of problems a supercomputing would be working on are usually very parallel. That doesn't mean that they're computationally easy to parallelize, particularly with limited communication bandwidth. If they require access to a shared resource (at least if it's update access), or the communication needs are heavy, it's going to be very hard to parallelize efficiently on a Beowulf-type cluster. That's where you need the massive bandwidth that a Starfire or SP or Connection Machine offers.
(And yes, I have programmed, from the inside, a Connection Machine. Worked for them for over 7 years, in fact, and spent most of that time tuning the FFT and suchlike. Even with a really good ratio of bandwidth to computation power, it still took a lot of work to get good performance even from a lowly FFT or matrix multiplication. And even though complete applications usually do FFT's and matrix multiplications in parallel, rather than one large one across the entire dataset, they usually have even more communication taking place than the core routines do.)
That's very nice, they're accessing read/write a common memory segment. Now, pray tell, how is this accomplished on a Beowulf efficiently?
As it happens, you've conveniently chosen something that's embarrassingly parallel anyway (no dependencies between elements not belonging to the same node, and no sequential component such as sequential I/O). I could put a node on Earth and a node on Mars and get perfect speedup in that situation. Think about something like finite element analysis with nearest neighbor interactions, or FFT, or n-body simulations (in order of increasing difficulty to scale well on most architectures).
And no, not every task can be broken down into multiple threads. As an example, consider decrypting a stream (where you know the key, you're not trying to break it or anything) that was encrypted with output feedback (simple electronic codebook encryption, where there's no dependence on prior state, is pretty easy to break with traditional cryptanalysis). You have to know the entire previous state of the message stream to decrypt the next block. That cannot be parallelized by any technique that I'm familiar with.
Not really; GoHip's ISP isn't publicly certifying anything about them; they're a common carrier. VeriSign is representing something about GoHip publicly, so I'd say they have a bit more responsibility. If VeriSign wants to certify something that weak, that's fine, but then perhaps VeriSign's not providing as strong a service as they should.
The Boston Globe actually does a fairly good job of covering technology issues. Hiawatha Bray and Simson Garfinkel do write columns about important technology issues, even if I don't always agree with them.
The whole issue of "timeliness" is overblown. So what if I don't find out that Bush beat McCain in South Carolina for a few more hours? Most of those hours I was asleep, anyhow. Likewise for the "graphically impaired" issue. Why do I need all that eye candy, anyhow?
There are a few things that newspapers do really well that won't happen digitally any time soon, if ever:
1) They're convenient to read over breakfast, or in bed on Sunday morning.
2) They're conveniently organized. Now, you're saying "WHAT!!!???" Yup, they're conveniently organized. The world/national/major local news is in the first section, the local stuff is in the second section, and so forth. No, it's not organized by subject or anything. That's just right. I get to see more than just headlines; if something catches my eye, I'll read it.
I don't know how many equivalent pixels a newspaper page is; I'd guesstimate 30Kx50K or thereabouts for a single broadsheet page. That's so much more visible information than I can fit on my screen that it's laughable. Plus, the screen forces me to be in a certain place.
Like it or not, the "information revolution" is still a relatively small part of life. Eating, sleeping, having shelter and transportation, are still more important to our lives than what happens on the net. The majority of people still don't own computers. The same vapidity that permeates popular culture permeates cyberspace.
In my spare time I do hack. I do other things, too, and I need to do them more. I could use more exercise, which I've lost a lot of time on in the past few months due to a project (a print driver I'm working on). I haven't played my viola in a year, and I could use doing more of that. I've been on the net for upwards of 15 years (I'm a spry old 36). Yes, it's nice that due to the net there are people working on this print driver in Japan, Germany, and elsewhere in the US, but it's not my whole world either.
Software patents, DeCSS, and such are important (just ask my wife how angry I was when the injunctions were granted), but they really aren't the whole of everything. People aren't dying over them. Yes, it is important that sanity prevail here, and the current state of things isn't very sane, but big corporations have always been trying to take over everything and this is just the latest; we have to be just as vigilant as always. And we could use a little perspective beyond our little online world.
Don't get me wrong -- the net does bring important benefits. I've done a lot of research on various ailments and such that various members of my family have, and it's a lot easier to do than it would be in a traditional library, as one example. It can add a lot to life, if it's looked at as something more rather than a replacement for anything. But on the whole I think we're too wrapped up in our geek culture around here, and could use a little more getting out into the world of matter rather than just the world of information.
If a customer reports a bug, the presumption should be that it's of critical severity until the customer stipulates otherwise. If beta testers (and W2K "beta testers" are a fairly general population) are reporting things, the onus is on Microsoft to handle the situation, but calling it "niggling" is inappropriate unless the tester/customer says so (and customers will happily admit that something isn't important if that's what they really think; a lot of people like to report everything on principle but will happily characterize the importance of the issue because they understand that development time is limited and they want the software).
And yes, if a customer claims a misspelling or something is a serious issue, I'd be inclined to accept it as so on first glance -- note, for example, the infamous Chevy Nova ("no go" in Spanish). That doesn't mean that it absolutely must be fixed to ship, but if a customer says that something is important it should be treated as such.
Saying that the product shipped with no showstoppers doesn't mean very much, unless there's a really hard and fast rule about what constitutes a showstopper and the people with shipping responsibility have absolutely no control over the indicated severity). It's very common for something to be marked a showstopper, but then people start asking questions such as "is there *really* no workaround?"
While the 65000 bugs sounds outlandish (and is probably somewhat low but in the ballpark, given the number of lines of code), I agree that it probably overstates the severity of the situation. Nonetheless, saying that it shipped with no showstoppers (which is more or less true by definition) doesn't mean very much either.
There was one really nasty thing of this nature in 2.7.2 (I think it finally got fixed in 2.9 something). Compiling C++, a function
foo(int a)
{
}
was accepted. Notice that there is no declaration of return type -- a violation of every C++ standard -- and the compiler defaults to returning int, K&R-style. Even including -ansi -pedantic-errors would not trigger an error (it would trigger a warning, but not an error). I had a very hard time convincing the gcc folks at the time that this was a real, and serious, problem.
I'm glad that at least -ansi -pedantic-errors exists; for the most part, it's quite good. It's unfortunate that something as egregious as this (and in C++, where one of the major benefits is strong typing, this kind of thing is a real problem) slipped through for so long.
Officially released or not, W2K is widely available. They've found two holes in a layered service, and they're sending out patches in a fairly reasonable amount of time.
One can argue about the wisdom of turning on unnecessary services, but that problem is not unique to Microsoft. When I installed SuSE, I had to go and basically clean out inetd. Still nothing terribly new there. That's unfortunate, but it's an industry-wide problem.
There will be security holes in W2K. If Microsoft responds more quickly and openly, and the holes are in add-on services rather than appearing systematically in the core, then maybe they're finally learning their lesson. My guess is that they'll do better than NT4 (they've really been taking a beating over this) but not as good as the better Linux/Unix distributions. But that's just a guess, too. Time will tell.
Modulo the fact that we can't send Linus' lawyers after them, Linus has, after all, explicitly blessed binary modules (with the warning about what might happen when the kernel changes, but caveat hax0r). Linus's copyright only extends as far as the kernel, so he has no say over what happens in user space.
Most of the Linux distributions offer some binary-only stuff, even if it's just Netscape. Even Debian includes non-free options, I believe, although they're treated as an appendage rather than as part of the distribution per se. TurboLinux is quite explicit about offering proprietary value-add, and they're unquestionably offering something real.
As for whether Linus should withhold use of the trademark, it's his call, but LinuxOne does appear to be in the Linux business, however tenuously (as opposed to Serious Domains, which was strictly trying to capitalize on the name, with no pretense at actually offering Linux-related goods or services). Withholding the trademark from LinuxOne puts him in the business of judging particular uses of Linux, which is a rather slippery slope. I'd be very surprised if he goes after someone offering a distribution (however cheesy) with legal use of the Linux kernel (respecting the GPL on the kernel source proper) unless they seriously abuse the trademark, maybe by saying that Linus endorses their particular flavor.
But if LinuxOne's installation is as painful as it sounds, either they're not going to go very far or they'll quickly find out that it would actually help them to add real value.
It's not the military-industrial complex we need to be afraid of any more...
check out iwin.nws.noaa.gov. I use the text interface; look under State Data, and then under most of the states you'll find Forecast Discussion. Depending upon who's on duty and how interesting the weather is, you'll get anything from "Will continue current forecast" to a long discussion of factors influencing the weather and local effects. Walt Drag of the Boston (actually Taunton), MA office is legendary for his discussions, which recently have sometimes filled two full screens. He's clearly a big-time snow buff.
l is the current and last several discussions; they're usually updated about every 6 hours (again, except for Walt, who likes reissuing them for changing conditions). The current one (as of noon EST on Wednesday, January 19) is really juicy. Assuming he's still on duty this afternoon (I'm not certain how the shifts run), his afternoon discussion will be even better. It's interesting to read, to see how these folks interpret the data and the forecast models.
http://iwin.nws.noaa.gov/iwin/ma/discussion.htm
in part because the upper air network is so sparse (and so infrequently sampled) compared to the surface network. There are also all sorts of boundary layer and topographical effects that cause the surface to be not representative of the entire atmospheric column. This, of course, cuts both ways -- local microclimates can be very important.
The classic example is a deep, bowl-shaped valley -- on clear, calm nights it will typically be sharply colder than on surrounding hills. On nights with a light wind, if the air in the valley decouples from the light breeze aloft, the difference might be very sharp indeed.
However, all in all I think it would be more cost effective to upgrade the upper air network, particularly over the oceans and Asia.
It's hard for me to see how speculators in tickets and domain names alleviate a shortage. To me, it appears that they create the shortage in the first place, and then magnanimously offer to resell them at a much higher price to alleviate the shortage. That strikes me as an excellent example of extortion.
Why is an auction any more efficient than a lottery? It's nicer for the person selling the good, but if that person doesn't want to capture the most value from it, why should someone else who has no interest in using the good jump in?