We've used that product line for almost 15 years now (we started with the predecessor, the system 36, and upgraded to the 400 later). We've never had a single hardware problem. The computer runs 24/7/365 without a single glitch. It's expensive, but it does work.
You had crappy math teachers. Math, esp at the higher level, is a very difficult challenging field that requires a lot of "creative" insight. It's not what you do in HS at all.
Algorithms by definition ARE math. They are not numeric based math, but they absolutly are math. Math is fundamentallly about patterns. Algorithms are imperative math statements, equations are declarative. Just because it's a different type of math doesn't mean that it's not math.
The only reason people don't realize this more is because most of the really hard stuff is already worked out for them. If you were stuck coding in assembler with no libraries to help you out, you'd realize how much math there is under the hood.
I don't think you understand me, because all of the stuff you've mentioned is pure and near pure CS stuff.
I'm talking about using software to solve complicated theoretical problems. For example when designing controllers for industrial equipment the designer needs both controls engineering and computer science expertise. If you were designing electronics simulation software, you'd need electrical engineering and computer science. If you were designing nuclear physics simulations for the DOE you'd need physics and CS.
Most of the time software is used either as a tool or as part of a product, for these sorts of things you'll be much more competative in the job market with another degree and a minor in CS. The rest of the time it's used on a PC in the form of an application or a web app, for this sort of stuff a pure CS degree is fine. The problem is that MOST of the people who like software go into the pure-CS stuff leaving the job market flooded and salaries relatively low. On the other hand, the mixed field stuff, where you need expertise in two or more areas is in relatively high demand. Hence, the salaries are relatively higher.
Because it is substantially easier to teach youself advance (and even basic) CS then it is to teach yourself engineering or physics, new students are better off finantially by getting a major in one of these harder fields and speciallizing in software then getting a CS degree.
It might be better to say instead that the knowledge domains available to a CS major do not provide you with significant job advantages because they represent a very small portion of the types of problems software is used for.
Almost every CS major knows the basics of graphics, databases, compilers, and operating systems. Yes you can specialize in one of these areas (or another CS area) but the field is so small and the jobs so few that it doesn't provide you with any significant advantage. Most CS graduates get jobs in the same small set of fields.
As a CS you won't learn anything about the areas where writting software does pay: control systems, embeded design, simulation, etc. If, for example, you had a degree in physics or engineering there would be more things you could write software TO DO, hence more/better job opportunities.
Software is primarily used as a tool to solve a problem. Very little of the software that gets written can be written by someone with just CS knowledge.
The problem with a CS degree and the reason it has poor job prospects is that it doesn't give you domain experience. When you graduate, you know the theory behind software and you've written a few programs.
When you get an engineering degree you hit the math and the theory hard at first and then you learn how to apply it to different situations. For example a electrical engineer might specialize in electromagnetic compatability and high frequency design. In addition to having a broad base of theory to use as a starting point for future learning, you have some specific domain of knowledge that you have deep understanding of and can immediatly bring to any team you join.
When you get a CS degree you don't have a knowledge domain, so tons of CS students are compeating for the relatively small number of software related fields where you don't need some specific knowledge.
People interested in software are much better off getting a degree in a field of knowledge that interests them and specializing in software related problems in that field. Almost all of the best CS-related jobs fall into this catagory --- simulation, controls, embeded systems, CAE, CAD, etc.
The reason that most schools don't offer software engineering (and, incidentally the reason that I don't recommend majoring in CS) is that the field of software problems where you don't need domain experience is very small, and for the most part un-theoretical. MOST people would be better off getting another degree and minoring in CS. At most schools the CS degree is not the best investment for your money.
For example, an electrical engineer specializing in embeded systems and with a CS minor is a much better candidate for many jobs then a guy with just a CS degree. Let's face it, it is much easier to teach the advanced CS stuff to yourself then it is to learn an entirely new field, and most of the good software jobs require some special field of knowledge.
My advice on what degree to get is to get the hardest degree you can survive in your field of interest. In the long run that will put you in a better possition in the job market, because there will be fewer people in the harder degrees compeating for the higher paying jobs, and if you can't find one of the higher paying jobs you can always out compeate someone who got an easier degree.
The reason it's hard to compile for them, simply put, is this: the chip reads 8 32-bit instructions each clock cycle and executes them simultaniously. Unlike a super scaler chip, there is no complicated scheduling hardware; it just executes what it sees. So the compiler has to do all of the instruction scheduling at compile time (I hope you have a good profiler).
To further complicate matters, registers can only be accessed so many times in each instruction group (unlike x86 there are no hidden internal registers), and each of the pipelines has preferential access to certain registers and can only execute some instructions. Then you have the possibility the instructions can be conditional and may or may not execute.
Put another way, the idea of VLIW is to move everything but the bare nessecities into the compiler. You give up a lot of non-performance advantages and significantly increase the difficulty of writing a compiler, but you free up a ton of chip real estate that can be used to make something REALLY fast.
The reason that other platforms have problems overcomming intel is that they make essentially the same trade offs (binary compatability, super-scaler, tune for good common case instead of best worst-case) and end up with essentially the same chip with a few minor differences.
The G4 is wide and shallow, but it's super scaler instead of very long instruction word. That adds a ton of chip realestate. If you think altivec is cool you should see some of the DSP math commands.
Asside from being a pain in the butt to do assembly and OS level stuff, it's really hard to write a compiler for a VLIW DSP. Maintaining binary compatability isn't something that's done for very long (a few years at most). Much of the performance comes by trading off the requirement for long-term binary compatability, something Intel can't afford to do.
The only thing it doesn't have is protected memory (which you could add with an external memory controller pretty trivially with minimal slowdown). The real strength of the chip is that it can do 8 instructions per clock cycle (it's wide and shallow) and has some really cool math instructions that save you a ton of time on common multimedia stuff.
Do you realise that you can buy a 300MHz DSP from TI that will run circles around an Athlon XP 3000 and only cost you about $30?
There is a ton of baggage in the x86 instruction set, the only reason people don't notice is that other PC archs also have similar crippling baggage. This junk slows you WAY down reguardless of how "clever" or "clean" the design.
It' POSSIBLE to do it. You need some expensive equipment and a three phase hook up but you COULD do it.
Were I going to attempt it I'd do something like the following (now there is a HOWTO on the net...):
*Buy very small perfect diamond *Buy LPCVD machine *Buy various carbon based gasses *Use DOE to tweak the tuning (my bet is that you need some combination of CH4 and CCl4 and a really high temp setting with the pressure at 1/1000000 ATM; even then it's probably going to be really slow) *Grow small diamond into larger diamond *Buy cutting tools *Separate the larger diamond into smaller ones *Repeate grow + separate until CVD machine is full *Harvest diamonds
A friend of mine, who is a deputy, has told me that the devices are very easy to manipulate into a false possitive. His advice is to always insist on going and taking a blood test immediately, stating that you don't trust the breathalizer device.
I can't say about everywhere, but I'm 120mi from the nearest big city and about 50mi from an interstate and Cingular has already swapped all of their towers to GSM (my old phone stopped working and I was forced to upgrade). As a bonus however I can now get 120kbps wireless internet way out here in the boonies(however this isn't really that impressive, the cable company already has 5Mbps cable available...):)
> This whole article is just a Red Hat employee tooting his company's horn.
I think you are wrong. Ulrich is just an idiot; one who likes to encourage bad programming habits. No I'm not trolling; he's managed to demonstrate that trait quite objectively.
In this case he doesn't realise that 9 times out of ten the minority platforms are exposing existing bugs that also effect the major platforms in less obvious ways. By giving up the minory platforms, you give up a the ability to catch bugs caused by bad programming. While not axing these bugs may be convinient now, in the long run it will bite you bad...
Another example of Ulrich's stupidity: He doesn't accept patches for glibc to support the BSD buffer-overflow-protected string functions(e.g. strlcpy() and friends) because programers are supposed to be smart enough not to write buffer overflows (his arguement). As a result of his policy, many open source apps have been forced to roll their own functions or else use the error prone posix ones. Every other major libc supports these functions (even Solaris), but GNU people won't because Ulrich says it's the programers' job to be smart. WTF?
An equally plausible and therefore equally valid explaination is that like Einstein, Newton and Maxwell saw the inherent order of the Universe as evidence of the existance of a transendant diety.
It's not an ad hominem. I'm serious. If you have a rational basis for what you are saying, you certainly arn't communicating it. Therefore it seems that your system is arbitrary, as further evidenced by your failure to acknoledge the relation between morality and ethics.
However, to please you, I'll rephrase: Please illustrate for everyone reading the logic behind the conclusion that the actions of these students were unethical. Please start from presupitions we can agree on.
I think however we might have found the real reason for the divergence of conclusions here. You seem to want to equate ethics with some abstract social standard based on implied consent. Whereas I see ethics as the rules used to ensure moral behavior. You seem to beleive that these rules are inherently obvious or otherwise need no rational justification. Whereas, I see that if we are to avoid chaos, any code of human conduct must obey reason. Is my interpretation of our dispute (that it lies at a deeper, metaphysical level) correct?
> Sure email does
No, it does not. Those technologies are things that arn't part of email, but can be used with it. That arn't part of the purpose of email. A web server exists for the purpose of distributing information to groups people efficently. Those people can assume that information it distributes to them is for them since by default it won't send the information unless the sender has told it to.
This is completely different from email, which is why different social conventions apply.
> Stanford did not provide them the URL and told them when they would get a decision - they chose to take another URL, redo it and attempt to access information Stanford clearly did not yet want released.
We keep going over this, ok I follow you this far. Now please tell me why it is that you see this as being morally wrong. Who was hurt? Does your moral system not require a victim? If so how does it separate right from wrong? Is you system even rationally based, or is it just something you hobbled together from junk lying around?
But they were released because people could get to them.
> Stanford et al did not want them to hev yet
As I have stated, desires to not translate into morality. Standford's wants, desires, and intentions, are compleately irrelevent to a discussion on the morality of accessing a web page.
> And while Stanford does not have to provide a rational reason, it's quite possible that decisons change at the last minute and they do not want soemone to get the wrong one before they finalize the acceptances.
That's a perfectly acceptable reason for not wanting to give the information out early, but it doesn't make the students accessing it wrong. It just means that they might get bad information.
> Your view means spam is ok because you make your email account accessable to the web, all I have to do is fiqure out an address.
No it doesn't, you can't take one conclusion and apply it to a situation with different premises. That would be irrational. The premise for why accessing a website is OK is because web sites exist for transmitting information to someone and the technology itself has the means to deny permission to individuals or groups. Therefore, if you can acess it, you can assume you are intended to.
Email OTOH, has no such mechanism, therefore, sending email to an address, unsolicited can be considered tresspass.
All of this is dancing around the core question though: What is the reasoning behind your assertion that the actions of the students are wrong?
You keep trying to avoid my question by indirection. Why is manipulating a URL to access something on a webserver that the server has granted them permission to access wrong? As I stated if they didn't want them to access it, they could have changed the permissions, or done what Stanford did, not posted it.
Quoting Stanford's Code of conduct doesn't add anything to this discussion. Because there is nothing there that can reasonably constitute a reasoned definition of ethics and morality.
You seem to believe that I'm saying what these students did was somehow "right". I'm not saying that. I'm just saying that this isn't, as you claim, morally wrong behavior. Of course Stanford has the right to reject anyone for any reason, no matter how petty, stupid, and short sited. They could reject all of the candidates that wore blue shirts to the interview, or all of the black candidates, or all of the white upper-middle class protestants. Instead they rejected everyone who followed a URL (by typing it in) to a dead end page with no information. Suddenly everyone wants to assert some sort of moral imperative one way or the other. All of the above scenarios are equally stupid and irrational; none of them can be justified morally.
I think you misunderstand my quip, I'm refering to the teaching of a "Business Ethics" course. The very name implies that "business ethics" are somehow different from say "moral ethics" or "social ethics" and that there isn't just ethics. Furthermore the PR surrounding these classes makes it pretty clear that these Universities feel that there is something different about being ethical while running a bussiness and about being ethical in general.
My theory is that the Universities are so over run by progressives, anti-moderns, and Marxists that running a business is generally considered unethical by the Liberal Arts faculty. As a result of internal politics, the Business program had to concede that business ethics is different from traditional ethics so that the LA faculty could go on their merry way preaching that business is by its very nature, unethical.
I accept the general rules of netiquette, which includes the assumption that information posted in a manner that can be read by someone by simply going to a URL is intended to be read by that someone. This doesn't let you crack passwords or break apps but these guys followed a link and were presented with a page. There is no sane reason what-so-ever to assume they shouldn't view it, because if someone felt that they shouldn't, they could have easily changed the file permisions to deny them the ability to do so.
Secondly, I assert that for something to be wrong there has to be a rational reason; to do otherwise is to invite insanity. As far as I currently know, there is no conceivable, rational basis for such a claim reguarding this situation.
P.S. Since you seem to think that reading something on the internet requires explicit permission:
I hearby grant permission to everyone everywhere to read this post.
Doesn't that just sound stupid?
P.P.S. In answer to you question, information is a mental abstration that is used in communication. It does not have physical properties, so for example it cannot be "accessed". Furtheremore for me to obtain information that only you have, you must give it to me. Either you give it to me freely ( by posting it on the web, or by telling me, by you leaving it written down and lying around were I might find it, etc.) or I force it out of you (by coercion, by fraud, etc.). The first is acceptable, the second is not.
> it was pretty clear that the school didn't want that info publicly available prior to its official release
I do not see it as clear. For all I know, the date could mearly be something like a final deadline for internal processing purposes.
Furthermore even positing that it is true, desire does not translate into morality. You may desire to have a consulting monopoly so that you can gouge prices, for example, but that doesn't make it wrong for me to prevent you from doing so.
that coupled with either an explicit or implicit agreement to follow the school's rules
No such implicit agreement exists until the applicant has accepted. Until then it is mearly an agreement for the applicant to provide information free of charge in exchange for consideration.
you obviously haven't been granted access.
This is a corporate mentality: If it's there in plain site you should ignore it unless you have permission. That is a compleately unjustified attitude. Publicly available information is by it's nature public, and information available to someone logged in on a server is by it's nature theirs to see.
This insn't like they used a buffer overflow or cracked a password, this is a page that the server admin had either granted them permission to see or had not denied them the ability to see. So by every rule of netique they were to assume that they had permission. Stanford, unlike Harvard, wisely chose to not post the information until closer to the deadline, so no students really learned anything, but even if they had, to claim these kids are unethical for looking at a web page is a far strech.
Slashdot Mirror
We've used that product line for almost 15 years now (we started with the predecessor, the system 36, and upgraded to the 400 later). We've never had a single hardware problem. The computer runs 24/7/365 without a single glitch. It's expensive, but it does work.
You had crappy math teachers. Math, esp at the higher level, is a very difficult challenging field that requires a lot of "creative" insight. It's not what you do in HS at all.
The only reason people don't realize this more is because most of the really hard stuff is already worked out for them. If you were stuck coding in assembler with no libraries to help you out, you'd realize how much math there is under the hood.
I'm talking about using software to solve complicated theoretical problems. For example when designing controllers for industrial equipment the designer needs both controls engineering and computer science expertise. If you were designing electronics simulation software, you'd need electrical engineering and computer science. If you were designing nuclear physics simulations for the DOE you'd need physics and CS.
Most of the time software is used either as a tool or as part of a product, for these sorts of things you'll be much more competative in the job market with another degree and a minor in CS. The rest of the time it's used on a PC in the form of an application or a web app, for this sort of stuff a pure CS degree is fine. The problem is that MOST of the people who like software go into the pure-CS stuff leaving the job market flooded and salaries relatively low. On the other hand, the mixed field stuff, where you need expertise in two or more areas is in relatively high demand. Hence, the salaries are relatively higher.
Because it is substantially easier to teach youself advance (and even basic) CS then it is to teach yourself engineering or physics, new students are better off finantially by getting a major in one of these harder fields and speciallizing in software then getting a CS degree.
Almost every CS major knows the basics of graphics, databases, compilers, and operating systems. Yes you can specialize in one of these areas (or another CS area) but the field is so small and the jobs so few that it doesn't provide you with any significant advantage. Most CS graduates get jobs in the same small set of fields.
As a CS you won't learn anything about the areas where writting software does pay: control systems, embeded design, simulation, etc. If, for example, you had a degree in physics or engineering there would be more things you could write software TO DO, hence more/better job opportunities.
Software is primarily used as a tool to solve a problem. Very little of the software that gets written can be written by someone with just CS knowledge.
When you get an engineering degree you hit the math and the theory hard at first and then you learn how to apply it to different situations. For example a electrical engineer might specialize in electromagnetic compatability and high frequency design. In addition to having a broad base of theory to use as a starting point for future learning, you have some specific domain of knowledge that you have deep understanding of and can immediatly bring to any team you join.
When you get a CS degree you don't have a knowledge domain, so tons of CS students are compeating for the relatively small number of software related fields where you don't need some specific knowledge.
People interested in software are much better off getting a degree in a field of knowledge that interests them and specializing in software related problems in that field. Almost all of the best CS-related jobs fall into this catagory --- simulation, controls, embeded systems, CAE, CAD, etc.
For example, an electrical engineer specializing in embeded systems and with a CS minor is a much better candidate for many jobs then a guy with just a CS degree. Let's face it, it is much easier to teach the advanced CS stuff to yourself then it is to learn an entirely new field, and most of the good software jobs require some special field of knowledge.
My advice on what degree to get is to get the hardest degree you can survive in your field of interest. In the long run that will put you in a better possition in the job market, because there will be fewer people in the harder degrees compeating for the higher paying jobs, and if you can't find one of the higher paying jobs you can always out compeate someone who got an easier degree.
To further complicate matters, registers can only be accessed so many times in each instruction group (unlike x86 there are no hidden internal registers), and each of the pipelines has preferential access to certain registers and can only execute some instructions. Then you have the possibility the instructions can be conditional and may or may not execute.
Put another way, the idea of VLIW is to move everything but the bare nessecities into the compiler. You give up a lot of non-performance advantages and significantly increase the difficulty of writing a compiler, but you free up a ton of chip real estate that can be used to make something REALLY fast.
The reason that other platforms have problems overcomming intel is that they make essentially the same trade offs (binary compatability, super-scaler, tune for good common case instead of best worst-case) and end up with essentially the same chip with a few minor differences.
Asside from being a pain in the butt to do assembly and OS level stuff, it's really hard to write a compiler for a VLIW DSP. Maintaining binary compatability isn't something that's done for very long (a few years at most). Much of the performance comes by trading off the requirement for long-term binary compatability, something Intel can't afford to do.
The only thing it doesn't have is protected memory (which you could add with an external memory controller pretty trivially with minimal slowdown). The real strength of the chip is that it can do 8 instructions per clock cycle (it's wide and shallow) and has some really cool math instructions that save you a ton of time on common multimedia stuff.
There is a ton of baggage in the x86 instruction set, the only reason people don't notice is that other PC archs also have similar crippling baggage. This junk slows you WAY down reguardless of how "clever" or "clean" the design.
It' POSSIBLE to do it. You need some expensive equipment and a three phase hook up but you COULD do it.
Were I going to attempt it I'd do something like the following (now there is a HOWTO on the net...):
*Buy very small perfect diamond
*Buy LPCVD machine
*Buy various carbon based gasses
*Use DOE to tweak the tuning (my bet is that you need some combination of CH4 and CCl4 and a really high temp setting with the pressure at 1/1000000 ATM; even then it's probably going to be really slow)
*Grow small diamond into larger diamond
*Buy cutting tools
*Separate the larger diamond into smaller ones
*Repeate grow + separate until CVD machine is full
*Harvest diamonds
Est Cost: $85,000
A friend of mine, who is a deputy, has told me that the devices are very easy to manipulate into a false possitive. His advice is to always insist on going and taking a blood test immediately, stating that you don't trust the breathalizer device.
Certainly true; however it is still really silly to have versions of the same function compiled into 20 different apps.
I can't say about everywhere, but I'm 120mi from the nearest big city and about 50mi from an interstate and Cingular has already swapped all of their towers to GSM (my old phone stopped working and I was forced to upgrade). As a bonus however I can now get 120kbps wireless internet way out here in the boonies(however this isn't really that impressive, the cable company already has 5Mbps cable available...) :)
I think you are wrong. Ulrich is just an idiot; one who likes to encourage bad programming habits. No I'm not trolling; he's managed to demonstrate that trait quite objectively.
In this case he doesn't realise that 9 times out of ten the minority platforms are exposing existing bugs that also effect the major platforms in less obvious ways. By giving up the minory platforms, you give up a the ability to catch bugs caused by bad programming. While not axing these bugs may be convinient now, in the long run it will bite you bad...
Another example of Ulrich's stupidity: He doesn't accept patches for glibc to support the BSD buffer-overflow-protected string functions(e.g. strlcpy() and friends) because programers are supposed to be smart enough not to write buffer overflows (his arguement). As a result of his policy, many open source apps have been forced to roll their own functions or else use the error prone posix ones. Every other major libc supports these functions (even Solaris), but GNU people won't because Ulrich says it's the programers' job to be smart. WTF?
An equally plausible and therefore equally valid explaination is that like Einstein, Newton and Maxwell saw the inherent order of the Universe as evidence of the existance of a transendant diety.
It's not an ad hominem. I'm serious. If you have a rational basis for what you are saying, you certainly arn't communicating it. Therefore it seems that your system is arbitrary, as further evidenced by your failure to acknoledge the relation between morality and ethics.
However, to please you, I'll rephrase: Please illustrate for everyone reading the logic behind the conclusion that the actions of these students were unethical. Please start from presupitions we can agree on.
I think however we might have found the real reason for the divergence of conclusions here. You seem to want to equate ethics with some abstract social standard based on implied consent. Whereas I see ethics as the rules used to ensure moral behavior. You seem to beleive that these rules are inherently obvious or otherwise need no rational justification. Whereas, I see that if we are to avoid chaos, any code of human conduct must obey reason. Is my interpretation of our dispute (that it lies at a deeper, metaphysical level) correct?
> Stanford did not provide them the URL and told them when they would get a decision - they chose to take another URL, redo it and attempt to access information Stanford clearly did not yet want released.
We keep going over this, ok I follow you this far. Now please tell me why it is that you see this as being morally wrong. Who was hurt? Does your moral system not require a victim? If so how does it separate right from wrong? Is you system even rationally based, or is it just something you hobbled together from junk lying around?
But they were released because people could get to them.
> Stanford et al did not want them to hev yet
As I have stated, desires to not translate into morality. Standford's wants, desires, and intentions, are compleately irrelevent to a discussion on the morality of accessing a web page.
> And while Stanford does not have to provide a rational reason, it's quite possible that decisons change at the last minute and they do not want soemone to get the wrong one before they finalize the acceptances.
That's a perfectly acceptable reason for not wanting to give the information out early, but it doesn't make the students accessing it wrong. It just means that they might get bad information.
> Your view means spam is ok because you make your email account accessable to the web, all I have to do is fiqure out an address.
No it doesn't, you can't take one conclusion and apply it to a situation with different premises. That would be irrational. The premise for why accessing a website is OK is because web sites exist for transmitting information to someone and the technology itself has the means to deny permission to individuals or groups. Therefore, if you can acess it, you can assume you are intended to.
Email OTOH, has no such mechanism, therefore, sending email to an address, unsolicited can be considered tresspass.
All of this is dancing around the core question though: What is the reasoning behind your assertion that the actions of the students are wrong?
Quoting Stanford's Code of conduct doesn't add anything to this discussion. Because there is nothing there that can reasonably constitute a reasoned definition of ethics and morality.
You seem to believe that I'm saying what these students did was somehow "right". I'm not saying that. I'm just saying that this isn't, as you claim, morally wrong behavior. Of course Stanford has the right to reject anyone for any reason, no matter how petty, stupid, and short sited. They could reject all of the candidates that wore blue shirts to the interview, or all of the black candidates, or all of the white upper-middle class protestants. Instead they rejected everyone who followed a URL (by typing it in) to a dead end page with no information. Suddenly everyone wants to assert some sort of moral imperative one way or the other. All of the above scenarios are equally stupid and irrational; none of them can be justified morally.
My theory is that the Universities are so over run by progressives, anti-moderns, and Marxists that running a business is generally considered unethical by the Liberal Arts faculty. As a result of internal politics, the Business program had to concede that business ethics is different from traditional ethics so that the LA faculty could go on their merry way preaching that business is by its very nature, unethical.
Agreed
Secondly, I assert that for something to be wrong there has to be a rational reason; to do otherwise is to invite insanity. As far as I currently know, there is no conceivable, rational basis for such a claim reguarding this situation.
P.S. Since you seem to think that reading something on the internet requires explicit permission:
I hearby grant permission to everyone everywhere to read this post.
Doesn't that just sound stupid?
P.P.S. In answer to you question, information is a mental abstration that is used in communication. It does not have physical properties, so for example it cannot be "accessed". Furtheremore for me to obtain information that only you have, you must give it to me. Either you give it to me freely ( by posting it on the web, or by telling me, by you leaving it written down and lying around were I might find it, etc.) or I force it out of you (by coercion, by fraud, etc.). The first is acceptable, the second is not.
I do not see it as clear. For all I know, the date could mearly be something like a final deadline for internal processing purposes.
Furthermore even positing that it is true, desire does not translate into morality. You may desire to have a consulting monopoly so that you can gouge prices, for example, but that doesn't make it wrong for me to prevent you from doing so.
that coupled with either an explicit or implicit agreement to follow the school's rules
No such implicit agreement exists until the applicant has accepted. Until then it is mearly an agreement for the applicant to provide information free of charge in exchange for consideration.
you obviously haven't been granted access.
This is a corporate mentality: If it's there in plain site you should ignore it unless you have permission. That is a compleately unjustified attitude. Publicly available information is by it's nature public, and information available to someone logged in on a server is by it's nature theirs to see.
This insn't like they used a buffer overflow or cracked a password, this is a page that the server admin had either granted them permission to see or had not denied them the ability to see. So by every rule of netique they were to assume that they had permission. Stanford, unlike Harvard, wisely chose to not post the information until closer to the deadline, so no students really learned anything, but even if they had, to claim these kids are unethical for looking at a web page is a far strech.