Opportunities have always been openining up, but this does not mean the music industry is going to die. This was an argument some
time back only it was with the film industry when many invested in dot com companies who were going create movies for the Internet
and about 99.999% of them all are on FuckedCompany.com.
The precision of that measurement suggests that you think there were at least 100,000 dot com's who were going to create movies for the Internet (whatever that means!).:)
There is a well-known theorem in programming language theory called the Curry-Howard(-de Bruijn-Tait-Feys-...) correspondence. It says, approximately, that every program P of type T written in a functional programming language is equivalent to a proof P' of a theorem T', in a particular logical system. This suggests (but does not prove) that any program written in any strongly-typed language is probably a "piece of mathematics", and may provide an argument that programs and mathematical theorems be treated analogously by the law.
This article starts off with a loaded question (Q:"do you really want to be completely anonymous?") and uses the no-brainer answer to draw unwarranted conclusions. Obviously no one wants to be a complete nobody; but that doesn't mean we want Joe Q. Citizen peeping into our bedrooms. This is not far from one of those abortion/meat-is-murder arguments. "Do you approve of murder?" "No." "Then why are you murdering babies/animals?!?"
The online environment provides more privacy protections, not fewer. Online, you can encrypt things and provide conditional access. Yes, you could, if it remained in your possession, or you could somehow retain control over the information you give out; but you can't. Once it's gone, the holder's incompetence is your privacy's demise. And these days even apparently competent companies are getting cracked. That would be OK if you could pick and choose who you give your information to, but nowadays people want all sorts of information on you which is irrelevant, even for the most piddling transactions, and is later used only to target you for marketing purposes.
I have agreed to let my car company, for instance, track my every move through GPS satellites. Whoop-dee-doo. McNealy already gave up most of his privacy when he decided to become a loud-mouthed celebrity CEO. Once you go that far, relinquishing the rest of your privacy is not a big step.
So far, the industry has done a pretty good job of regulating itself. Where has this guy been the last two years? The amount of spam in my mailbox every morning, the most recent credit card crack headline, the opt-out/opt-in controversy, the passing on of private information from companies that get bought out---they all tell me otherwise.
On the Internet, even more than in other areas of our lives, trust is the real currency. Squander what you have and you'll find out how hard it can be to get more. Yeah, that works for large, well-established corporations like Sun, but it sounds like it would put smaller start-ups in a "you need credit to get credit" situation.
What is truly ironic, though, is that someone on whom all the symbolism of 2001 was lost also completely missed the fact that the Heidi Klum post was not a serious conspiracy theory, but rather a parody of a conspiracy theory!
There is little or no difference between people who dismiss everything but their own (or their own community's) ideas as quackery, and quacks themselves. Look, the simple truth is that there is no simple truth; at least, if there is a simple truth here, we can't effectively decide what it is. It's just not black and white.
Only a fool would argue that there is no symbolism in 2001: the very first scene, where the bone thrown into the heavens (where gods live) morphs into a spaceship (missile, somebody said?), immediately makes the connection with the Zarathustra story, as if the music (Strauss!) wasn't enough. Other people have already clearly shown that Kubrick played with words and anagrams (CRM-114, Vivian Darkbloom). It's not hard to believe that Kubrick embedded tons of hard-to-reveal metaphors and brain teasers into 2001.
That said, it is to be expected that some of Wheat's suggestions will turn out wrong. For example, TMA-1=NO-MEAT is hard for me to swallow. It's possible Kubrick intended this; but if he did, he sure made it difficult to find---so difficult that the chance of him intending it is about as small as the chance of someone correctly guessing that he intended it. And that makes it uninteresting to me.
But some of Wheat's suggestions are bound to be on target, and seem plausible. For example, `Bowman' meaning Odysseus. (Another thread established, in my mind, that this is consistent with the chronology revealed by Clarke and Kubrick in their notes.)
The notion that Hal-Discovery is intended to visually resemble God in the image of man, having a head, mouth, spine and anus (sans the bathroom tile bit) is a borderline case, but unlike TMA-1, I find it interesting because I can imagine both that some production engineer designed it that way accidentally (after all, it looks like many other movie spacecraft), and that Kubrick intended it to evoke God, since it is steered by an AI, and the technology-ruling-man theme is well-established.
Aristotle is reported to have said that the mark of an educated mind is that it can entertain an idea without necessarily accepting it. I don't know whether Kubrick intended all this signification, but I don't need to accept it or dismiss it at this point, so I won't.
BTW, I find the Trojan Horse analogy pretty tenuous. It seems to me that the most significant thing about the Trojan Horse is not the chronology of events in that bit of the Odyssey, but rather the notion that the Horse tricked the Trojans into accepting an armed party of Achaians into their midst. This aspect doesn't seem to be reflected in the film at all. Besides, you would expect that the link between the small and big monoliths is engendered by a link between some small and big Trojan Horse, but there is only the one Horse...
Lisp programs are not programs in the interpreter. They are data. You can call (eval) on the data, but
it's still just data that conforms to a standard.
Not "just data": data together with an execution semantics. XML lacks this. (XML+XSLT is another matter, and begins to resemble something like a programming language.)
Look at it this way: give me the source code of a well-formed LISP program. I can tell you at least two other reasonable ways (e.g., call-by-name and call-by-need) to execute that source code, neither of which are LISP's semantics.
Now give me a well-formed piece of XML. I can give you a hundred ways to "execute" it. All of them are equally valid, because the XML standard says nothing about execution.
BTW, the "programs are data" paradigm of LISP is overhyped. In fact you even pointed out the reason: any program is treated as data by its own compiler/interpreter. Whether you use a LISP-style syntax or not is irrelevant. The advantage of a LISP-style syntax is that the concrete syntax and abstract syntax of programs is nearly the same, so there is one less layer of indirection to go through when you are manipulating programs.
Also, you can easily add macros and an eval command to any dynamically typed programming language; it's largely an implementation issue, except for some subtleties to do with scoping. Adding similar things to a statically typed language is much more challenging, and goes beyond implementation issues to semantical ones. In other words, if you want to thank someone for LISP's eval, you should thank foremost your LISP compiler's implementor, not the LISP designers, since all they had to do was declare that LISP implementors need to provide eval, rather than grapple with any deep issues related to it. (But the idea to allow eval in the first place was of course a bright idea, and McCarthy rightly deserves some acclaim for this---though honestly I dunno if it was McCarthy's idea or somebody else, say Steele or Gosling.)
Personally, I find LISP-like languages' approach to reflection far too low-level. The existence of something like eval is, of course, essential for reflection, but LISP-style reflection only really lets you elide code generation code. There is still a lot of other compiler code you have to write yourself. For example, the least I would expect from a language with good reflection capabilities is that it ensures statically that dynamically generated code is actually well-formed; in a statically typed language, I would additionally expect that it statically ensures that dynamically generated code is well-typed. LISP does not ensure this. It is easy to feed nonsense to eval, and evoke a runtime code generation error or---even worse---silently produce a nonsensical program. Hygiene issues compound the problem since, I don't think anyone really uses hygienic macros in practice as they are too complicated and restrictive.
Comparing LISP to XML is like comparing a Corvette to the chassis of a Trans Am. One is a car; the other looks like a car, but is missing a few things some people tend to insist on: for example, an engine, transmission, wheels, tires, fuel tank, etc.
Did you read Greenspun's article before you decided to rant?
First, there were supposed to be 7 people on the board, but the VCs blocked appointments for the last 2 seats. That enabled them to get a majority when voting as a bloc.
Second, the VCs stopped holding board meetings, and instead started holding "investor's meetings", which excluded Greenspun, which made board-level decisions.
Third, they used company money to finance litigation in an attempt to subvert the shareholders' legitimate interests in the financial welfare of the company.
In conclusion, it does not sound to me like Greenspun surrendered control of the company, stepped back and just hoped for the best. It sounds more like an internal takeover by a gang of people who want to bleed the shareholders dry.
one of my reasons against adding Scheme-style
continuations to the language (this has seriously been
proposed by the Stackless folks) is that it can't be
implemented in a JVM. I find the existence of Jython very
useful because it reminds me to think in terms of more abstract
language semantics, not just implementation details.
You find that disallowing a feature because it can't be implemented on a certain platform is a sign of thinking in terms of abstract language semantics?
The 'control-alt-del' to login. Someone
should pay for that.
I used to think this was odd too. But then I learned the reason for it. As I understand it, in Windows Ctl-Alt-Del is always trapped by the OS, never by a user program. So pressing Ctl-Alt-Del at the login ensures that the login screen you are seeing is actually the real OS login, and not a user program written by a malicious attacker who wants to deceive you into revealing your password.
As far as I can see, this is an improvement over standard Unix login procedures where the only you can determine its authenticity is to reboot first.
BH
Re:Geeks and filesystems.
on
MUD Shell
·
· Score: 1
The question is why do Geeks like computers more than non-geeks?
I'm a Ph.D. student in computer science; I majored in computer science at Cornell. Your question has generated so much traffic that I'll probably get lost in the shuffle, but since I feel strongly about this I will speak up anyway.
On one hand, I think that, as long as you pick a major which is generally related to what you will want to pursue as a career, you'll be alright, provided you take enough classes in peripheral subjects. For example, if you major in CS but take a few more advanced classes in computer architecture than normal, you will still make a respectable computer engineer, and vice versa. So your choice of major is not so critical for general job skills.
On the other hand, I do think that the choice of major strongly influences the way in which you approach problems, and how big your `box' is. By `box' here, I mean the same `box' as in the phrase `thinking outside the box.' Typically you find that people in more theoretical departments, like physics and math and literature and philosophy, develop very robust and flexible ways of problem-solving, whereas those in more applied fields, like engineering and architecture, develop more specialized skills.
The flipside is that theory is only useful when it is applied, and to apply theory you need to develop a skill in formal reasoning, which means some kind of mathematics. From this point of view, then, physics is the ideal major.
But it all depends on what you want to do. If you want to build new things, you will need to develop those formal reasoning skills, and a goodly amount of patience. If you want to build things which are useful but not really new, you may be satisfied with an engineering degree. If you want to build things which are elegant but not immediately useful, mathematics or philosophy are good choices. If you don't want to build things, then you are probably better off in a non-technical field.
Personally, I think you should play it safe until you know what you want. Take a mix of courses which introduces you to a wide range of subjects. Don't bias yourself towards theoretical or applied classes until you know what you want. (For example, I regret not taking more advanced theoretical classes in college.) If you are required to declare a major, then make your best guess: you can always switch later. (I did.)
I have looked at the TeX code, and in fact I own the book produced from the WEB (literate Pascal) sources. I have not seen the CWEB (literate Pascal) sources, so I cannot comment on those, although I suspect they are virtually identical modulo C/Pascal differences.
Now, I can tell you that TeX, as a program in the abstract, is designed very well, i.e., the specification is quite well thought out and not too difficult to follow. Furthermore, it handles nearly all of the things you would want from a typesetter, so it is also very complete. And, when you consider how old TeX is, this is even more amazing.
But the code itself -- I am talking now about the implementation, not the specification -- is not particularly spectacular. If it weren't in literate form, it would be difficult to grasp, that's for sure! With the literate documentation, it isn't too bad, although it's not what I'd call modular, doesn't take any advantage of strong typing, and you will find lots of places where Knuth makes what will seem to us now some very odd design decisions; the latter are mostly things to do with dynamic memory allocation (hint: there is none!:) and I/O, and are partly due to Pascal's peculiarities, and partly due to the age of the code, since it was originally written to be run on what we would now consider to be a very puny computer.
My conclusion is, if you want to see how to design a program well, look at the TeX WEB; but if you want to see how to implement a program well, you should look at some programs written in more modern programming languages, as they can more precisely capture formally (i.e., also in machine-readable form rather than just human-readable form) some of the higher-level abstractions which appear in the specification.
I don't believe in copy control, patents
of any kind, or intellectual property,
but anybody should have the right to
market anything they want. That includes
media companies---if they want to sell
content on media which is difficult to
duplicate, then that's fine with me. On
the other hand, they should not expect
the government to establish laws which
prevent tampering with their
devices. That is like trying to have
your cake and eat it too: they want to
sell a physical device or object, but
not the rights to use or modify it.
But look: if you try to take away the
right of these companies to sell what
they want to sell in the way they want
to sell it, then you may also be taking
away your own right in the future of
ever selling something in the way
you want to sell it. Maybe
someday it will be the case that the
public wants to tell you that
you should behave in a certain way. It
doesn't have to be copy control; it
could be something else: the point is
that you have the right to sell anything
you own.
The trouble with this particular
instance, namely mass media like films,
music and software, is that we think of
it in the same way as we think of
utilities like water and power, and
resources like food. The fact is that
you don't need to see Hollywood films,
or listen to big-name bands, or use
Microsoft software to live. Maybe you
grew up taking it for granted, and now
it seems indispensible, but that's your
fault: you put your faith in the makers
of these things, and now they are
stabbing you in the back. You trusted
the wrong people.
Now live with it. If everybody just
accepted that fact, then no one would
buy these copy-controlled media, and the
media companies would soon abandon the
idea. Otherwise, you are being a
hypocrite.
Now, certainly I would like to continue
to enjoy good films and good music, but
if I just boycott the industry, the mass
of sheep that inhabits the US will just
go along with whatever the industry
decides for them, and I will never get
to enjoy my favorite media again. So
what can I do?
1. I can create my own content. Instead of
merely benefitting from others'
producing it, I can start contributing,
and ensure that it remains in the public
domain. Nearly everyone here is either a
programmer or an artist, so I don't
think I'm preaching to the wrong
audience. Maybe you're a hardware designer; start making free hardware. (You know what I mean by "free"...!)
2. I can support people who
create free content. Donate to the FSF
and to free software projects, or to
specific bands or independent filmmakers
who want to keep their films free. I can
start buying goods from companies that
don't participate in the CPRM agenda,
and write drivers for their hardware. If
an appropriate organization doesn't
exist, I might think about founding one.
3. I can tell other people why they
should not buy non-free goods, or goods
which the government will not let me own
completely. Then I can tell them why
they need to tell other people ("pay it
forward"?).
Maybe there other ways. The point is to
give the market and our democracy a
chance to sort the situation out itself
using its own rules and mechanisms,
rather than imposing more artificial
rules which restrict anybody
else's freedoms, including those of the
media companies.
This is not a perfect solution maybe,
but it is the only one I see wherein I
can protect both my freedoms and my
integrity.
I was fortunate enough to go to a school with many great teachers, but one of my favorites was Dr. Holmes, who taught English. He was one of the few teachers at my school who actually had a Ph.D. To look at him, you would not think he was very likeable. He always wore a suit and a vest, and occasionally a stern expression as well. But in fact he was not stuffy at all, but quite congenial, and I think he was one of the best-loved teachers at the school.
I think English is one of those topics which many students feel is extremely wishy-washy, that for example you can make an argument for whatever thesis you feel like supporting that day. Dr. Holmes showed me, at least, that that was false. Whenever he asked the class to interpret a passage in a book or a verse of poetry, and somebody came out with some outlandish and abstract perspective, Holmes would ask him to support that with specific evidence from the text. When that student tried to answer again, sure enough, his interpretation would usually end up unconvincing, and Dr. Holmes could always point out specific and convincing evidence for his own conclusions.
I was always good in English, but I also had this wishy-washy perspective on the subject before I took Holmes' class. But not afterwards. I think this was the first time I really learned that critical thinking and reasoning is something that transcends formal systems like mathematics and programming.
One thing I remember specifically which he said, and which surprised me at the time, was that students should major either in physics or philosophy, because these are the only two curricula at the undergraduate level which build critical reasoning skills. His perspective was that the undergraduate education is too short to study anything in depth, and so it should be considered as only a primer for graduate-level studies, which is when you should start specializing.
His advice had a great effect on me. I had planned to major in computer science, but when I got to college I couldn't bring myself to do it, and chose physics instead. Unfortunately, I only pursued it for two years, and then switched to CS in the end anyway, but I did benefit from it, because I caught a glimpse of how theory and practice can be made to work together harmoniously, and how they inform each other.
Now I am back in school, getting a Ph.D. in CS, and I can really appreciate what Dr. Holmes said. I think he was right about undergraduate education: it is too short to waste on studying specifics. The best way to spend your undergraduate years is to study a field which gives you a broad and solid grounding in fundamental reasoning skills. Physics and philosophy satisfy that condition, and I would also add mathematics. But CS is too specific and biased.
Anyway, I am grateful for the opportunity to share this story with./ readers. Maybe someone else reading this will know the Dr. Holmes I am talking about?:)
I think that this whole incident shows how well the hacker norms that ESR outlined in his
"Homesteading the Noosphere" can work to reduce conflict in ways that the corporate "let's call in the lawyers" world will never understand.
This reminds me of what Karl Popper wrote about Freud and Marx: their claims are so broad and vague that virtually anything can be marshalled into "further evidence" for them. Would Popper feel the same way about ESR and "hacker culture"?
Of all the stories I have read on Slashdot in the last few years, this story about CPRM has incensed me the most. I am not the sort of person who gets up in arms about personal privacy issues, or entertains paranoid illusions about corporate superstates, but this is the straw that broke the camel's back!
If the movie companies want to put copy protection on their DVDs, fine; I don't really care. If the music companies want to copy protect their music, fine; I can live with that. But when IBM, Toshiba, Intel and Matsushita start putting restrictions on how I can use my own data, and when and where, just because the RIAA and other media producers want to protect their own data, which I don't use anyway, I have a big problem with it! This is not an improvement for the consumer, even if it stimulates the development and distribution of digitally-converted analog media; this is blatant cow-towing, and these companies want to force consumers to follow their agenda.
The beauty of capitalism was that it was supposed to give both freedom of choice and high-quality goods to the consumer, by encouraging and rewarding suppliers who successfully compete on those standards. But if I can't choose what I want to buy because some industrial oligarchy has already decided that only certain things are allowed to reach the average consumer, I can't make an effective vote for what I consider the best product.
I am sick and tired of this irresponsible, condescending behavior and collusional tactics. If I have to pay three times as much, I am going to get my storage devices from a source that does not try to control the manner in which I opt to use it. I hold IBM, Toshiba, Intel and Matushita responsible for trying curtail my freedom as a consumer, and I will not spend a cent on a hard drive, processor, CD player or any other goods from any of these corporations until they recant and this outrageous CPRM idea is dead and buried never to return.
Slashdot Mirror
Open source is a one-time gift.
GPL source is the gift that keeps on giving.
BH
The precision of that measurement suggests that you think there were at least 100,000 dot com's who were going to create movies for the Internet (whatever that means!). :)
BH
BH
This article starts off with a loaded question (Q:"do you really want to be completely anonymous?") and uses the no-brainer answer to draw unwarranted conclusions. Obviously no one wants to be a complete nobody; but that doesn't mean we want Joe Q. Citizen peeping into our bedrooms. This is not far from one of those abortion/meat-is-murder arguments. "Do you approve of murder?" "No." "Then why are you murdering babies/animals?!?"
The online environment provides more privacy protections, not fewer. Online, you can encrypt things and provide conditional access. Yes, you could, if it remained in your possession, or you could somehow retain control over the information you give out; but you can't. Once it's gone, the holder's incompetence is your privacy's demise. And these days even apparently competent companies are getting cracked. That would be OK if you could pick and choose who you give your information to, but nowadays people want all sorts of information on you which is irrelevant, even for the most piddling transactions, and is later used only to target you for marketing purposes.
I have agreed to let my car company, for instance, track my every move through GPS satellites. Whoop-dee-doo. McNealy already gave up most of his privacy when he decided to become a loud-mouthed celebrity CEO. Once you go that far, relinquishing the rest of your privacy is not a big step.
So far, the industry has done a pretty good job of regulating itself. Where has this guy been the last two years? The amount of spam in my mailbox every morning, the most recent credit card crack headline, the opt-out/opt-in controversy, the passing on of private information from companies that get bought out---they all tell me otherwise.
On the Internet, even more than in other areas of our lives, trust is the real currency. Squander what you have and you'll find out how hard it can be to get more. Yeah, that works for large, well-established corporations like Sun, but it sounds like it would put smaller start-ups in a "you need credit to get credit" situation.
BH
That post is hilarious, and recommended reading. Here is a more direct link:
What is truly ironic, though, is that someone on whom all the symbolism of 2001 was lost also completely missed the fact that the Heidi Klum post was not a serious conspiracy theory, but rather a parody of a conspiracy theory!
There is little or no difference between people who dismiss everything but their own (or their own community's) ideas as quackery, and quacks themselves. Look, the simple truth is that there is no simple truth; at least, if there is a simple truth here, we can't effectively decide what it is. It's just not black and white.
Only a fool would argue that there is no symbolism in 2001: the very first scene, where the bone thrown into the heavens (where gods live) morphs into a spaceship (missile, somebody said?), immediately makes the connection with the Zarathustra story, as if the music (Strauss!) wasn't enough. Other people have already clearly shown that Kubrick played with words and anagrams (CRM-114, Vivian Darkbloom). It's not hard to believe that Kubrick embedded tons of hard-to-reveal metaphors and brain teasers into 2001.
That said, it is to be expected that some of Wheat's suggestions will turn out wrong. For example, TMA-1=NO-MEAT is hard for me to swallow. It's possible Kubrick intended this; but if he did, he sure made it difficult to find---so difficult that the chance of him intending it is about as small as the chance of someone correctly guessing that he intended it. And that makes it uninteresting to me.
But some of Wheat's suggestions are bound to be on target, and seem plausible. For example, `Bowman' meaning Odysseus. (Another thread established, in my mind, that this is consistent with the chronology revealed by Clarke and Kubrick in their notes.)
The notion that Hal-Discovery is intended to visually resemble God in the image of man, having a head, mouth, spine and anus (sans the bathroom tile bit) is a borderline case, but unlike TMA-1, I find it interesting because I can imagine both that some production engineer designed it that way accidentally (after all, it looks like many other movie spacecraft), and that Kubrick intended it to evoke God, since it is steered by an AI, and the technology-ruling-man theme is well-established.
Aristotle is reported to have said that the mark of an educated mind is that it can entertain an idea without necessarily accepting it. I don't know whether Kubrick intended all this signification, but I don't need to accept it or dismiss it at this point, so I won't.
BTW, I find the Trojan Horse analogy pretty tenuous. It seems to me that the most significant thing about the Trojan Horse is not the chronology of events in that bit of the Odyssey, but rather the notion that the Horse tricked the Trojans into accepting an armed party of Achaians into their midst. This aspect doesn't seem to be reflected in the film at all. Besides, you would expect that the link between the small and big monoliths is engendered by a link between some small and big Trojan Horse, but there is only the one Horse...
BH
BH
BH
Not "just data": data together with an execution semantics. XML lacks this. (XML+XSLT is another matter, and begins to resemble something like a programming language.)
Look at it this way: give me the source code of a well-formed LISP program. I can tell you at least two other reasonable ways (e.g., call-by-name and call-by-need) to execute that source code, neither of which are LISP's semantics.
Now give me a well-formed piece of XML. I can give you a hundred ways to "execute" it. All of them are equally valid, because the XML standard says nothing about execution.
BTW, the "programs are data" paradigm of LISP is overhyped. In fact you even pointed out the reason: any program is treated as data by its own compiler/interpreter. Whether you use a LISP-style syntax or not is irrelevant. The advantage of a LISP-style syntax is that the concrete syntax and abstract syntax of programs is nearly the same, so there is one less layer of indirection to go through when you are manipulating programs.
Also, you can easily add macros and an eval command to any dynamically typed programming language; it's largely an implementation issue, except for some subtleties to do with scoping. Adding similar things to a statically typed language is much more challenging, and goes beyond implementation issues to semantical ones. In other words, if you want to thank someone for LISP's eval, you should thank foremost your LISP compiler's implementor, not the LISP designers, since all they had to do was declare that LISP implementors need to provide eval, rather than grapple with any deep issues related to it. (But the idea to allow eval in the first place was of course a bright idea, and McCarthy rightly deserves some acclaim for this---though honestly I dunno if it was McCarthy's idea or somebody else, say Steele or Gosling.)
Personally, I find LISP-like languages' approach to reflection far too low-level. The existence of something like eval is, of course, essential for reflection, but LISP-style reflection only really lets you elide code generation code. There is still a lot of other compiler code you have to write yourself. For example, the least I would expect from a language with good reflection capabilities is that it ensures statically that dynamically generated code is actually well-formed; in a statically typed language, I would additionally expect that it statically ensures that dynamically generated code is well-typed. LISP does not ensure this. It is easy to feed nonsense to eval, and evoke a runtime code generation error or---even worse---silently produce a nonsensical program. Hygiene issues compound the problem since, I don't think anyone really uses hygienic macros in practice as they are too complicated and restrictive.
BH
Comparing LISP to XML is like comparing a Corvette to the chassis of a Trans Am. One is a car; the other looks like a car, but is missing a few things some people tend to insist on: for example, an engine, transmission, wheels, tires, fuel tank, etc.
LISP is a programming language. XML is a syntax.
'nuff said.
BH
Did you read Greenspun's article before you decided to rant?
First, there were supposed to be 7 people on the board, but the VCs blocked appointments for the last 2 seats. That enabled them to get a majority when voting as a bloc.
Second, the VCs stopped holding board meetings, and instead started holding "investor's meetings", which excluded Greenspun, which made board-level decisions.
Third, they used company money to finance litigation in an attempt to subvert the shareholders' legitimate interests in the financial welfare of the company.
In conclusion, it does not sound to me like Greenspun surrendered control of the company, stepped back and just hoped for the best. It sounds more like an internal takeover by a gang of people who want to bleed the shareholders dry.
BH
You find that disallowing a feature because it can't be implemented on a certain platform is a sign of thinking in terms of abstract language semantics?
Your second sentence contradicts the first.
BH
I used to think this was odd too. But then I learned the reason for it. As I understand it, in Windows Ctl-Alt-Del is always trapped by the OS, never by a user program. So pressing Ctl-Alt-Del at the login ensures that the login screen you are seeing is actually the real OS login, and not a user program written by a malicious attacker who wants to deceive you into revealing your password.
As far as I can see, this is an improvement over standard Unix login procedures where the only you can determine its authenticity is to reboot first.
BH
Yeah, and why is the Left so goddamn liberal?
BH
On one hand, I think that, as long as you pick a major which is generally related to what you will want to pursue as a career, you'll be alright, provided you take enough classes in peripheral subjects. For example, if you major in CS but take a few more advanced classes in computer architecture than normal, you will still make a respectable computer engineer, and vice versa. So your choice of major is not so critical for general job skills.
On the other hand, I do think that the choice of major strongly influences the way in which you approach problems, and how big your `box' is. By `box' here, I mean the same `box' as in the phrase `thinking outside the box.' Typically you find that people in more theoretical departments, like physics and math and literature and philosophy, develop very robust and flexible ways of problem-solving, whereas those in more applied fields, like engineering and architecture, develop more specialized skills.
The flipside is that theory is only useful when it is applied, and to apply theory you need to develop a skill in formal reasoning, which means some kind of mathematics. From this point of view, then, physics is the ideal major.
But it all depends on what you want to do. If you want to build new things, you will need to develop those formal reasoning skills, and a goodly amount of patience. If you want to build things which are useful but not really new, you may be satisfied with an engineering degree. If you want to build things which are elegant but not immediately useful, mathematics or philosophy are good choices. If you don't want to build things, then you are probably better off in a non-technical field.
Personally, I think you should play it safe until you know what you want. Take a mix of courses which introduces you to a wide range of subjects. Don't bias yourself towards theoretical or applied classes until you know what you want. (For example, I regret not taking more advanced theoretical classes in college.) If you are required to declare a major, then make your best guess: you can always switch later. (I did.)
Good luck!
BH
I have looked at the TeX code, and in fact I own the book produced from the WEB (literate Pascal) sources. I have not seen the CWEB (literate Pascal) sources, so I cannot comment on those, although I suspect they are virtually identical modulo C/Pascal differences.
Now, I can tell you that TeX, as a program in the abstract, is designed very well, i.e., the specification is quite well thought out and not too difficult to follow. Furthermore, it handles nearly all of the things you would want from a typesetter, so it is also very complete. And, when you consider how old TeX is, this is even more amazing.
But the code itself -- I am talking now about the implementation, not the specification -- is not particularly spectacular. If it weren't in literate form, it would be difficult to grasp, that's for sure! With the literate documentation, it isn't too bad, although it's not what I'd call modular, doesn't take any advantage of strong typing, and you will find lots of places where Knuth makes what will seem to us now some very odd design decisions; the latter are mostly things to do with dynamic memory allocation (hint: there is none! :) and I/O, and are partly due to Pascal's peculiarities, and partly due to the age of the code, since it was originally written to be run on what we would now consider to be a very puny computer.
My conclusion is, if you want to see how to design a program well, look at the TeX WEB; but if you want to see how to implement a program well, you should look at some programs written in more modern programming languages, as they can more precisely capture formally (i.e., also in machine-readable form rather than just human-readable form) some of the higher-level abstractions which appear in the specification.
I don't believe in copy control, patents of any kind, or intellectual property, but anybody should have the right to market anything they want. That includes media companies---if they want to sell content on media which is difficult to duplicate, then that's fine with me. On the other hand, they should not expect the government to establish laws which prevent tampering with their devices. That is like trying to have your cake and eat it too: they want to sell a physical device or object, but not the rights to use or modify it.
But look: if you try to take away the right of these companies to sell what they want to sell in the way they want to sell it, then you may also be taking away your own right in the future of ever selling something in the way you want to sell it. Maybe someday it will be the case that the public wants to tell you that you should behave in a certain way. It doesn't have to be copy control; it could be something else: the point is that you have the right to sell anything you own.
The trouble with this particular instance, namely mass media like films, music and software, is that we think of it in the same way as we think of utilities like water and power, and resources like food. The fact is that you don't need to see Hollywood films, or listen to big-name bands, or use Microsoft software to live. Maybe you grew up taking it for granted, and now it seems indispensible, but that's your fault: you put your faith in the makers of these things, and now they are stabbing you in the back. You trusted the wrong people.
Now live with it. If everybody just accepted that fact, then no one would buy these copy-controlled media, and the media companies would soon abandon the idea. Otherwise, you are being a hypocrite.
Now, certainly I would like to continue to enjoy good films and good music, but if I just boycott the industry, the mass of sheep that inhabits the US will just go along with whatever the industry decides for them, and I will never get to enjoy my favorite media again. So what can I do?
1. I can create my own content. Instead of merely benefitting from others' producing it, I can start contributing, and ensure that it remains in the public domain. Nearly everyone here is either a programmer or an artist, so I don't think I'm preaching to the wrong audience. Maybe you're a hardware designer; start making free hardware. (You know what I mean by "free"...!)
2. I can support people who create free content. Donate to the FSF and to free software projects, or to specific bands or independent filmmakers who want to keep their films free. I can start buying goods from companies that don't participate in the CPRM agenda, and write drivers for their hardware. If an appropriate organization doesn't exist, I might think about founding one.
3. I can tell other people why they should not buy non-free goods, or goods which the government will not let me own completely. Then I can tell them why they need to tell other people ("pay it forward"?).
Maybe there other ways. The point is to give the market and our democracy a chance to sort the situation out itself using its own rules and mechanisms, rather than imposing more artificial rules which restrict anybody else's freedoms, including those of the media companies.
This is not a perfect solution maybe, but it is the only one I see wherein I can protect both my freedoms and my integrity.
Are you sure that's a keyboard and not a couch?
I was fortunate enough to go to a school with many great teachers, but one of my favorites was Dr. Holmes, who taught English. He was one of the few teachers at my school who actually had a Ph.D. To look at him, you would not think he was very likeable. He always wore a suit and a vest, and occasionally a stern expression as well. But in fact he was not stuffy at all, but quite congenial, and I think he was one of the best-loved teachers at the school.
I think English is one of those topics which many students feel is extremely wishy-washy, that for example you can make an argument for whatever thesis you feel like supporting that day. Dr. Holmes showed me, at least, that that was false. Whenever he asked the class to interpret a passage in a book or a verse of poetry, and somebody came out with some outlandish and abstract perspective, Holmes would ask him to support that with specific evidence from the text. When that student tried to answer again, sure enough, his interpretation would usually end up unconvincing, and Dr. Holmes could always point out specific and convincing evidence for his own conclusions.
I was always good in English, but I also had this wishy-washy perspective on the subject before I took Holmes' class. But not afterwards. I think this was the first time I really learned that critical thinking and reasoning is something that transcends formal systems like mathematics and programming.
One thing I remember specifically which he said, and which surprised me at the time, was that students should major either in physics or philosophy, because these are the only two curricula at the undergraduate level which build critical reasoning skills. His perspective was that the undergraduate education is too short to study anything in depth, and so it should be considered as only a primer for graduate-level studies, which is when you should start specializing.
His advice had a great effect on me. I had planned to major in computer science, but when I got to college I couldn't bring myself to do it, and chose physics instead. Unfortunately, I only pursued it for two years, and then switched to CS in the end anyway, but I did benefit from it, because I caught a glimpse of how theory and practice can be made to work together harmoniously, and how they inform each other.
Now I am back in school, getting a Ph.D. in CS, and I can really appreciate what Dr. Holmes said. I think he was right about undergraduate education: it is too short to waste on studying specifics. The best way to spend your undergraduate years is to study a field which gives you a broad and solid grounding in fundamental reasoning skills. Physics and philosophy satisfy that condition, and I would also add mathematics. But CS is too specific and biased.
Anyway, I am grateful for the opportunity to share this story with ./ readers. Maybe someone else reading this will know the Dr. Holmes I am talking about? :)
Thanks, Dr. Holmes!
This reminds me of what Karl Popper wrote about Freud and Marx: their claims are so broad and vague that virtually anything can be marshalled into "further evidence" for them. Would Popper feel the same way about ESR and "hacker culture"?
I am concerned over the recent fascination with death here on /. Look at these headlines, all of which appeared in the last 24 hours:
This is unhealthy. Maybe you should consider alternative, more PC and euphimistic phrasings?
Then I can let my little toddlers read /. again.
Would you rather be a non-geek, or is geek good?
Haven't heard this one before. Is this "mal-" as in "bad", as in "malicious", "malevolent" and "maladjusted"?
Of all the stories I have read on Slashdot in the last few years, this story about CPRM has incensed me the most. I am not the sort of person who gets up in arms about personal privacy issues, or entertains paranoid illusions about corporate superstates, but this is the straw that broke the camel's back!
If the movie companies want to put copy protection on their DVDs, fine; I don't really care. If the music companies want to copy protect their music, fine; I can live with that. But when IBM, Toshiba, Intel and Matsushita start putting restrictions on how I can use my own data, and when and where, just because the RIAA and other media producers want to protect their own data, which I don't use anyway, I have a big problem with it! This is not an improvement for the consumer, even if it stimulates the development and distribution of digitally-converted analog media; this is blatant cow-towing, and these companies want to force consumers to follow their agenda.
The beauty of capitalism was that it was supposed to give both freedom of choice and high-quality goods to the consumer, by encouraging and rewarding suppliers who successfully compete on those standards. But if I can't choose what I want to buy because some industrial oligarchy has already decided that only certain things are allowed to reach the average consumer, I can't make an effective vote for what I consider the best product.
I am sick and tired of this irresponsible, condescending behavior and collusional tactics. If I have to pay three times as much, I am going to get my storage devices from a source that does not try to control the manner in which I opt to use it. I hold IBM, Toshiba, Intel and Matushita responsible for trying curtail my freedom as a consumer, and I will not spend a cent on a hard drive, processor, CD player or any other goods from any of these corporations until they recant and this outrageous CPRM idea is dead and buried never to return.
Will start-ups taking advantage of this new technology be known henceforth as `quantum dot coms'?
<ducks>
oh, and have you ever noticed that our logo isn't the sun logo?
You mean that gray thing with the brown in it is the /. rendition of the Java coffee cup? I always thought it was a waste basket or a potted plant...!