I prefer to look at it as 'With our own BIOS, we can now invoke Palladium on our terms'
If I'm running a server, I can enable Palladium, and require that all code be signed with my key, and thus that h4x0rz can't execute arbitrary code on my system. I could compile my kernels, sign them, move them over to the server, and install them when I want to upgrade. No one else can.
This is, of course, assuming that it would all work the way I think it will, but who knows? Maybe we'd have to do another step (flash a chip or something) to get it working.
Still, this is an important step in many many ways. Kudos to all those involved, good job guys.
Out of all the Xboxen sold, I'd guess it's a small percentage of gamers who go online with them
You obviously don't work at a gaming store. I work at a gaming store (a boutique, if you will) in a smallish city (~60k), where broadband is surprisingly uncommon (and sucks), and we haven't been able to keep our XBox Live kits on the shelf. For that matter, we never had any on the shelf, they were all preordered. Even our XBox rep was amazed at how many we've sold ('That's amazing. That's absolutely amazing.')
I'd say it's a fairly big percentage, considering how the Live kit just came out, and it'll only get bigger as time goes on.
I don't know what your university towns are like, but at UNB, students protesting Starbucks would be like transplant patients refusing invasive surgery.
When I moved to Fredericton to attend university, the guy I moved in with had a land line, which I tended to give to people when they asked for my number. Shortly after I got there though, I got a cellphone, and I'm glad I did. Why?
First of all, I'm almost never home, and when I am, it's to sleep, to eat, or to drop stuff off before I leave again. Also, there's no voicemail on the line, and I didn't feel like paying for it or checking it.
Secondly, I recently moved, and my old roommate stayed put. We both have cellphones, which means no sharing a land line, so we don't even have to get one (though we do have DSL, we're probably ditching it, since we're on promotion now).
Thirdly, I enjoy my privacy and solitude, and people know this. No one I know calls just to chat, except my mother (oy). They call for a reason, and once that reason is done, it's 'talk to you in class/tomorrow/at home/at work' time.
Fourth, if I don't feel like talking to someone, I can hit one of the volume buttons on the side to stop the ring/vibrate (so it only flashes), and then hit it again to send the call to voicemail - I alone decide when to accept a call and when not to.
And last, it's convenient. Closing shifts at work have to take the day's profits to the Bank of Montreal, and once we're done that, we go home - except I don't feel like walking for 45 minutes in blistering wind and a foot of snow and ice on the sidewalks. Wherever I go, I can call a cab and be home in ten minutes, as long as I can find some sort of shelter to wait in. This also keeps me from getting sick, which is good.
If you're on-call to everyone you know, maybe you need to set down guidelines. I'm a very relaxed person, but I know the feeling of 'What if someone called? Dear god, WHAT IF SOMEONE CALLED?!' and believe me, I didn't like it. Now, I can always know if someone's called, but I only answer it when I want to. I'm not resigned to a choice between freedom and connectivity - I have both.
Another way to do it is IFRAMEs. They're not entirely cross-browser, of course, but you can put non-IFRAME code into the ... area as a backup. That way, you can just call an external script or whatever, and rotate that separately.
You don't need an SUV, you just need good snow tires and to know how to drive. People around where I've lived drive F150s just as much as Geo Metros as far as weather's concerned.
On the other hand, you don't want to get a convertable because of the cold (unless you're in BC), or a sports car because of the low clearance (after a fresh snow you end up like a snowplow, then you end up with a snowbank on your hood; again, unless in BC).
On Shaw cable, there is a difference. The channels themselves aren't digital start to finish, but they are transmitted digitally, so there's no static, interference, etc. I was watching some television in Saskatoon (comparing it with our digital service in Prince Albert), and I couldn't believe how crappy the quality was.
As an aside, the Shaw packages are something like 60 is digital with a few exceptions.
If you want to see digital cable done right though, go to Jerusalem and take a look at what their cable companies have done. Not only is it high quality and has many good channels (damn you CRTC!), it looks very elegant and professional, not ass-ugly like everything I've seen in North America.
Microsoft: Take this document, but beware! It's a proprietary format! Slashdot: That's bad. Microsoft: But it's XML compliant! Slashdot: That's good. Microsoft: The DTD is also proprietary. Slashdot: That's bad. Microsoft: But you can reverse-engineer and write your own parser. Slashdot: That's good. Microsoft: The parser will have to be written in perl. (blank stare) Microsoft: That's bad. Slashdot: Can I go now?
You can just set yourself to use a different channel than the access point in question. Voila, no interference. Besides, as mentioned in the article, the reason it's as powerful as it is is because it's directional - it beams the signal to you, not to everywhere, so you could be sitting beside it, but if you're not between it and a client, you probably won't pick anything up.
For example, one clause "By using the Site, you represent and warrant that you are 18 years of age or older and possess the legal right and ability to enter into this Agreement and to use the Site in accordance with all of the terms and conditions of this Agreement" could cause more than a number of problems. Say someone under the age of 18 purchases a ticket from them online...now what? are they legally obligated to actually provide a service to this person, even though said customer is technically not allowed to use the site?
What I remember from law class is that if someone is under 18, they (AA) are bound by the terms of the agreement, but the minor is not. This is a standard way of protecting oneself, and isn't nearly uncommon. That being said, I'd expect that this would only come into play in such a situation, and would prevent AA from being held to an unfair (to them) deal. It's basically a way to get out of a contract that the other side isn't necessarily being held to, or perhaps even to enforce the contract despite the minor's age.
I think the "burden of proof" is set on the buyer to verify that they are actually a student or teacher, but anyone can pick up the software off the shelf and buy it.
Anyone can pick it up off the shelf, but you have to show ID to buy it. Most places actually require you to leap through hoops to purchase stuff, and even then, it's not upgradable (the Office XP package at UNB's bookstore says this with a large sticker on the front). Thus, if you need to get a new version (i.e. if your professor upgrades when a new version comes out, or if the university upgrades them), you have to pay full student price again. Sure, it's only a hundred bucks or so, so it's not that bad, but since you don't really get anything new except compatibility, it's a pretty raw deal.
USB input is a standard, which input devices have to conform to. A device's USB firmware will report the manufacturer name, name of the product, and the type of device. Human Input Devices, or HID, use standard drivers to accomplish this, in the interests of usability. This is why you can take an Apple keyboard and plug it into a Windows PC and have it work, if Windows doesn't crash (it wanted me to reboot to install the mouse, then again for the keyboard). This is also why OS X will recognize any three button or wheel mouse you connect (assuming the mouse is properly designed). USB Storage works the same way - standard interface, standard drivers. You can use different drivers to access extended functionality, but you don't have to - example is the MS Intellimouse with the 87.2 buttons. With the Intellimouse drivers, they all work. Without, it's a standard wheel mouse.
FireWire is similar, in that it's very no-driver. Storage devices, cameras, and so on (ideally) Just Work, using standard interfaces and suchforth. Never used one though, so I can't tell you what the practical implementation is like.
The great thing about business or government or anywhere that isn't the users' home... "Use this computer. Or you're fired".
A friend of mine was telling me about how her university was trying to migrate the graphics arts and suchforth staff to IBM PCs from Macs. No one wanted to try, so when they got a new staff member in, they convinced them to try. After a few weeks, they gathered everyone together to hear her testimonial.
It basically went something like 'I hate my job, I hate IBM, I want a new computer or a new job.' Suffice to say, no one else switched.
Never underestimate the willingness of skilled professionals to tell you where to go. They are paid to do what they do best, and if they can't use the tools that let them do that, they'll go somewhere they can.
At my university, we're provided with e-mail addresses that can be used as webmail, IMAP, POP3, or can be forwarded, and the size limit on the account is 25 megs, but it's a soft limit.
My geology prof sent me, recently, a powerpoint presentation, her entire lecture for the whole year, including images, text, and embedded movies. The e-mail came to 63 megs (suffice to say, I was not expecting this). It came through fine, but any other mail was responded to with a temporary 'mailbox is full' message until I removed it from my account. This resulted, the second time she sent it (there was a problem with the first), in my saving to my attachments folder and then deleting the message, which in turn deleted it from my attachments folder, so she had to re-re-send it. Now she has discovered she left a lecture out, so she may have to send it again.
Through all this, I've blasted past my quota three times so far, and likely will again, and no one has complained. Why? Because disk space is supercheap (and most students don't get large e-mails anyway, and the people who do are justified), bandwidth is cheap (my prof sends her e-mail from a large research lab outside of town, I download it at home on DSL, and the university is, well, the university), and proper administration and training goes a long way.
Yeah, but today, you don't just have one 500 MHz processor, you have several thousand (sometimes tens of thousands) of them. By the time IBM gets people to switch to this 'pay for cycles' method, desktop computers may be fast, but mainframes will be more reliable and clusters of them will still be faster.
Look at it this way: IBM can build a more powerful computer system than most corporations could buy. If IBM builds The Ultimate Cluster, then just rents time out, there's no long-term cost to people who use it. If you need to have access for a week, then rent CPU time for a week, use it, and then stop. You don't have to pay $50m for a system that can do the job you want in a reasonable time, but which won't be used often enough to justify the purchase, but you won't have to go without access to supercomputing power either.
There was an article a while ago on Canada.com (can't find it today) about some physicists in Toronto who are building a large 'net cluster' supercomputer spread over the largest universities in Canada. The problem they have is that they are about to produce so much data that they can't even store it at (I believe) UofT... Which is one of the largest, richest universities in the country, if not the largest, richest. If IBM could provide access to a powerful enough supercomputer (imagine another ASCI White for example), to which they could transfer the data and analyze it on, they might be interested. IBM would have to put it somewhere accessible to CA*Net III (i.e. a university, or the CRC), but it would be accessible for these huge experiments. I bet they'd like that. Even if it can't do the whole load, or costs too much to do the whole load, it'd be a good contributor to the system as a whole.
2. Dependency - you rely up on natural gas and electricity to be there, and yes they go down, but can they guarentee their utility won't have worse problems - especially if its Windows run and goes down once a week, cutting into your bought utility time.
I don't think IBM would use Windows on their supercomputing clusters. My main reasoning is that it doesn't boot on them, but there are political and logistic reasons as well.
Perhaps J2EE and.NET are irrelevant at this stage in the game, and a PHP vs. ASP review would be more relevant.
PHP doesn't offer the same things J2EE and.NET do though. PHP is a scripting language, J2EE and.NET are web application development platforms. The difference, in the easiest way I can think of it, is that PHP & ASP let you make pages, while J2EE &.NET let you make applications with web interfaces, basically. The distinction is a tricky one to make, but it's there nonetheless.
I think the issue with compiled languages on the server-side is that there's a larger problem than just RAD vs. speed. For example, writing your web apps in C could work fine, until some obscure pointer bug reared its ugly head and overflowed a buffer. This could hypothetically result in, for example, making an online store think you had x number of dollars in store credit, or think you'd already paid when you hadn't.
This problem arises because the deep magic is only run by one company (likely), so it may take longer to rear its head - and when it does, when an exploit is realized, it needs to be fixed immediately. This can also be a problem in that people don't necessarily understand the code someone else might've written, and they have to learn entire sections of code, or entire projects, just to fix.
PHP also takes time to learn about and so on, but you don't have pointer errors and buffer overflows (in that app specifically), since buffers are resized dynamically. You might pass bad data to something else, but you should be checking data once you have it all collected.
Then there's load times. On Apache, mod_php.so is loaded into memory, and then all that needs to be done is to read the page, compile, and execute, or (if pre-compiling/caching) just to execute. JSP is precompilied by default by Tomcat, and the JVM is always running. With a natively-compiled system, if you're not writing your own DSO or server software, it has to be executed for every page hit, and it's a lot harder to get someone who understands your setup than someone who understands JSP/ASP/PHP.
That all being said, I think the main factor is that the Internet moves at 'internet speed' - when you have an idea, it needs to get done now, with as little fuss, trouble, and debugging as possible. Scripting languages offer this by being highly abstracted and widely used (if a million people use PHP, it's more likely that people will find bugs. Slow but rapid beats fast and latent six days of the week.
Scripting languages on the desktop would be cool though, but for mainstream apps, there has to be a lot of garbage collection, because variables in scripting languages (can) take up more memory than their C/C++ counterparts, which can get laggy.
I agree with you for the most part; that being said, I think there are two things to keep in mind. First of all, corporations exist solely for the purpose of making money. If the opportunity presents itself, then so shall it be done. (American) CEOs can be sued if they take an action that is not in the best interests of the company('s bottom line, presumably).
Also, of course you only hear about screwy corporations on the news, just like you only hear about terror attacks in Israel/Palestine, or hostages in Moscow. CNN wouldn't attract a lot of eyes if its headlines ran 'Corporation continues ethical business as usual; 30 people on bus in Jerusalem get where they're going without incident; 700 theatre patrons watch play without interruption'. Much though I wish that were all the news there was to report, people are attracted to news about death, corruption, greed, and so on, so when four Israelis die, they focus on that, not the five million that didn't; when corporations cheat, they focus on those, not the ones that did the math right, and so on. For that matter, every time people get together to protest globalization, we end up with riots, assaults, property damage, and so on. Can we really judge all anti-globalists based on those few? Can we judge all multi-nationals based on those few? I don't think so.
Priorities people, want to go after a price fixer? Then go after the Pharmacuetical Industry who definatly fixes prices! That sort of battle would benifit more people than this.
Pharmaceutical companies spend billions of dollars a year on research, development, and testing. Forcing them to lower their sources of income would mean they spend less on this, and thus would harm the people more than benefit them because of the unavailability of drugs.
As it is now, they can charge whatever they want, but after a while, other companies are allowed to sell the exact same drug - usually at a much lower (sometimes 50%) price. THAT benefits consumers more than 'solving' price fixing would.
While I know that prices in the UK are usually high, the figures quoted are irrelevant for a simple reason: if you're going to compare, compare against one country, and take things into account. The UK taxes cigarettes far more than most countries do (as I recall) - are taxes an issue? Germany is famous for its beer - is the beer in both situations made by the same company, so that you're having a fair comparison of products? If so, what about shipping costs and so on?
Even here in Canada, you can get a pre-cooked shrimp ring for about $9.99 in BC, $8.99 in Saskatchewan, and $4.99 in New Brunswick, all the same brand. You can also get $10 shrimp rings in New Brunswick from more widely known (i.e. larger, better) brands. Is this because people in BC get 'screwed', or because it costs a lot of money to ship refrigerated shrimp ten thousand kilometers?
If you're going to compare, you have to take more factors than just the price into account - local economy, shipping, VAT, local taxes, and so on. Is the US getting screwed because a BK Whopper costs more there than here? No, we just have cheaper beef, and a lower (but stronger) economy, so prices are less. It's good sense, and sensible economics.
Maybe they aren't necessarily being taught, but isn't this something that people should know about? The most popular, most common, and most versatile web server software in the world, the basis of IBM's WebSphere (as I recall), basically *the* implementation of the extra fun J2EE stuff (JSP, servlets, etc), and so on.
Isn't this sort of a big thing not to know about? For a web developer to not know what apache is seems like a database admin not knowing what Oracle is, or a network engineer not knowing what IPX is.
They're not taught, but people should be studying things they're interested in, and if you're interested in something, you don't restrict yourself to the classroom - you go to Chapters, find a book that looks neat on something you've never heard of in the computer section and then sit down in Starbucks until they kick you out for not buying the book. You check out some CodeNotes books, you read slashdot (even if only for amusement, as I do) or check out Wired online every now and then. You don't just learn it, you live it, and sure, you have to live other things too, but why are you spending $40-100k to learn about something you don't have a passion for?
People should open their eyes, and the submitter is complaining because they don't - and I agree.
Slashdot Mirror
I prefer to look at it as 'With our own BIOS, we can now invoke Palladium on our terms'
If I'm running a server, I can enable Palladium, and require that all code be signed with my key, and thus that h4x0rz can't execute arbitrary code on my system. I could compile my kernels, sign them, move them over to the server, and install them when I want to upgrade. No one else can.
This is, of course, assuming that it would all work the way I think it will, but who knows? Maybe we'd have to do another step (flash a chip or something) to get it working.
Still, this is an important step in many many ways. Kudos to all those involved, good job guys.
--Dan
Out of all the Xboxen sold, I'd guess it's a small percentage of gamers who go online with them
You obviously don't work at a gaming store. I work at a gaming store (a boutique, if you will) in a smallish city (~60k), where broadband is surprisingly uncommon (and sucks), and we haven't been able to keep our XBox Live kits on the shelf. For that matter, we never had any on the shelf, they were all preordered. Even our XBox rep was amazed at how many we've sold ('That's amazing. That's absolutely amazing.')
I'd say it's a fairly big percentage, considering how the Live kit just came out, and it'll only get bigger as time goes on.
--Dan
I don't know what your university towns are like, but at UNB, students protesting Starbucks would be like transplant patients refusing invasive surgery.
The effects would be as self-detrimental too.
--Dan
I've found something quite different, myself.
When I moved to Fredericton to attend university, the guy I moved in with had a land line, which I tended to give to people when they asked for my number. Shortly after I got there though, I got a cellphone, and I'm glad I did. Why?
First of all, I'm almost never home, and when I am, it's to sleep, to eat, or to drop stuff off before I leave again. Also, there's no voicemail on the line, and I didn't feel like paying for it or checking it.
Secondly, I recently moved, and my old roommate stayed put. We both have cellphones, which means no sharing a land line, so we don't even have to get one (though we do have DSL, we're probably ditching it, since we're on promotion now).
Thirdly, I enjoy my privacy and solitude, and people know this. No one I know calls just to chat, except my mother (oy). They call for a reason, and once that reason is done, it's 'talk to you in class/tomorrow/at home/at work' time.
Fourth, if I don't feel like talking to someone, I can hit one of the volume buttons on the side to stop the ring/vibrate (so it only flashes), and then hit it again to send the call to voicemail - I alone decide when to accept a call and when not to.
And last, it's convenient. Closing shifts at work have to take the day's profits to the Bank of Montreal, and once we're done that, we go home - except I don't feel like walking for 45 minutes in blistering wind and a foot of snow and ice on the sidewalks. Wherever I go, I can call a cab and be home in ten minutes, as long as I can find some sort of shelter to wait in. This also keeps me from getting sick, which is good.
If you're on-call to everyone you know, maybe you need to set down guidelines. I'm a very relaxed person, but I know the feeling of 'What if someone called? Dear god, WHAT IF SOMEONE CALLED?!' and believe me, I didn't like it. Now, I can always know if someone's called, but I only answer it when I want to. I'm not resigned to a choice between freedom and connectivity - I have both.
--Dan
Another way to do it is IFRAMEs. They're not entirely cross-browser, of course, but you can put non-IFRAME code into the ... area as a backup. That way, you can just call an external script or whatever, and rotate that separately.
I have no idea if this helps you at all, but hey.
--Dan
This is what my website does, and Google indexes it fine.
--Dan
Canada SUV because of the snow
You don't need an SUV, you just need good snow tires and to know how to drive. People around where I've lived drive F150s just as much as Geo Metros as far as weather's concerned.
On the other hand, you don't want to get a convertable because of the cold (unless you're in BC), or a sports car because of the low clearance (after a fresh snow you end up like a snowplow, then you end up with a snowbank on your hood; again, unless in BC).
--Dan
On Shaw cable, there is a difference. The channels themselves aren't digital start to finish, but they are transmitted digitally, so there's no static, interference, etc. I was watching some television in Saskatoon (comparing it with our digital service in Prince Albert), and I couldn't believe how crappy the quality was.
As an aside, the Shaw packages are something like 60 is digital with a few exceptions.
If you want to see digital cable done right though, go to Jerusalem and take a look at what their cable companies have done. Not only is it high quality and has many good channels (damn you CRTC!), it looks very elegant and professional, not ass-ugly like everything I've seen in North America.
--Dan
Microsoft: Take this document, but beware! It's a proprietary format!
Slashdot: That's bad.
Microsoft: But it's XML compliant!
Slashdot: That's good.
Microsoft: The DTD is also proprietary.
Slashdot: That's bad.
Microsoft: But you can reverse-engineer and write your own parser.
Slashdot: That's good.
Microsoft: The parser will have to be written in perl.
(blank stare)
Microsoft: That's bad.
Slashdot: Can I go now?
--Dan
You can just set yourself to use a different channel than the access point in question. Voila, no interference. Besides, as mentioned in the article, the reason it's as powerful as it is is because it's directional - it beams the signal to you, not to everywhere, so you could be sitting beside it, but if you're not between it and a client, you probably won't pick anything up.
--Dan
For example, one clause "By using the Site, you represent and warrant that you are 18 years of age or older and possess the legal right and ability to enter into this Agreement and to use the Site in accordance with all of the terms and conditions of this Agreement" could cause more than a number of problems. Say someone under the age of 18 purchases a ticket from them online...now what? are they legally obligated to actually provide a service to this person, even though said customer is technically not allowed to use the site?
What I remember from law class is that if someone is under 18, they (AA) are bound by the terms of the agreement, but the minor is not. This is a standard way of protecting oneself, and isn't nearly uncommon. That being said, I'd expect that this would only come into play in such a situation, and would prevent AA from being held to an unfair (to them) deal. It's basically a way to get out of a contract that the other side isn't necessarily being held to, or perhaps even to enforce the contract despite the minor's age.
--Dan
I think the "burden of proof" is set on the buyer to verify that they are actually a student or teacher, but anyone can pick up the software off the shelf and buy it.
Anyone can pick it up off the shelf, but you have to show ID to buy it. Most places actually require you to leap through hoops to purchase stuff, and even then, it's not upgradable (the Office XP package at UNB's bookstore says this with a large sticker on the front). Thus, if you need to get a new version (i.e. if your professor upgrades when a new version comes out, or if the university upgrades them), you have to pay full student price again. Sure, it's only a hundred bucks or so, so it's not that bad, but since you don't really get anything new except compatibility, it's a pretty raw deal.
--Dan
USB input is a standard, which input devices have to conform to. A device's USB firmware will report the manufacturer name, name of the product, and the type of device. Human Input Devices, or HID, use standard drivers to accomplish this, in the interests of usability. This is why you can take an Apple keyboard and plug it into a Windows PC and have it work, if Windows doesn't crash (it wanted me to reboot to install the mouse, then again for the keyboard). This is also why OS X will recognize any three button or wheel mouse you connect (assuming the mouse is properly designed). USB Storage works the same way - standard interface, standard drivers. You can use different drivers to access extended functionality, but you don't have to - example is the MS Intellimouse with the 87.2 buttons. With the Intellimouse drivers, they all work. Without, it's a standard wheel mouse.
FireWire is similar, in that it's very no-driver. Storage devices, cameras, and so on (ideally) Just Work, using standard interfaces and suchforth. Never used one though, so I can't tell you what the practical implementation is like.
--Dan
The great thing about business or government or anywhere that isn't the users' home... "Use this computer. Or you're fired".
A friend of mine was telling me about how her university was trying to migrate the graphics arts and suchforth staff to IBM PCs from Macs. No one wanted to try, so when they got a new staff member in, they convinced them to try. After a few weeks, they gathered everyone together to hear her testimonial.
It basically went something like 'I hate my job, I hate IBM, I want a new computer or a new job.' Suffice to say, no one else switched.
Never underestimate the willingness of skilled professionals to tell you where to go. They are paid to do what they do best, and if they can't use the tools that let them do that, they'll go somewhere they can.
--Dan
I'd tend to agree in this case.
At my university, we're provided with e-mail addresses that can be used as webmail, IMAP, POP3, or can be forwarded, and the size limit on the account is 25 megs, but it's a soft limit.
My geology prof sent me, recently, a powerpoint presentation, her entire lecture for the whole year, including images, text, and embedded movies. The e-mail came to 63 megs (suffice to say, I was not expecting this). It came through fine, but any other mail was responded to with a temporary 'mailbox is full' message until I removed it from my account. This resulted, the second time she sent it (there was a problem with the first), in my saving to my attachments folder and then deleting the message, which in turn deleted it from my attachments folder, so she had to re-re-send it. Now she has discovered she left a lecture out, so she may have to send it again.
Through all this, I've blasted past my quota three times so far, and likely will again, and no one has complained. Why? Because disk space is supercheap (and most students don't get large e-mails anyway, and the people who do are justified), bandwidth is cheap (my prof sends her e-mail from a large research lab outside of town, I download it at home on DSL, and the university is, well, the university), and proper administration and training goes a long way.
--Dan
Yeah, but today, you don't just have one 500 MHz processor, you have several thousand (sometimes tens of thousands) of them. By the time IBM gets people to switch to this 'pay for cycles' method, desktop computers may be fast, but mainframes will be more reliable and clusters of them will still be faster.
Look at it this way: IBM can build a more powerful computer system than most corporations could buy. If IBM builds The Ultimate Cluster, then just rents time out, there's no long-term cost to people who use it. If you need to have access for a week, then rent CPU time for a week, use it, and then stop. You don't have to pay $50m for a system that can do the job you want in a reasonable time, but which won't be used often enough to justify the purchase, but you won't have to go without access to supercomputing power either.
--Dan
There was an article a while ago on Canada.com (can't find it today) about some physicists in Toronto who are building a large 'net cluster' supercomputer spread over the largest universities in Canada. The problem they have is that they are about to produce so much data that they can't even store it at (I believe) UofT... Which is one of the largest, richest universities in the country, if not the largest, richest. If IBM could provide access to a powerful enough supercomputer (imagine another ASCI White for example), to which they could transfer the data and analyze it on, they might be interested. IBM would have to put it somewhere accessible to CA*Net III (i.e. a university, or the CRC), but it would be accessible for these huge experiments. I bet they'd like that. Even if it can't do the whole load, or costs too much to do the whole load, it'd be a good contributor to the system as a whole.
--Dan
2. Dependency - you rely up on natural gas and electricity to be there, and yes they go down, but can they guarentee their utility won't have worse problems - especially if its Windows run and goes down once a week, cutting into your bought utility time.
I don't think IBM would use Windows on their supercomputing clusters. My main reasoning is that it doesn't boot on them, but there are political and logistic reasons as well.
--Dan
Perhaps J2EE and .NET are irrelevant at this stage in the game, and a PHP vs. ASP review would be more relevant.
.NET do though. PHP is a scripting language, J2EE and .NET are web application development platforms. The difference, in the easiest way I can think of it, is that PHP & ASP let you make pages, while J2EE & .NET let you make applications with web interfaces, basically. The distinction is a tricky one to make, but it's there nonetheless.
PHP doesn't offer the same things J2EE and
--Dan
I think the issue with compiled languages on the server-side is that there's a larger problem than just RAD vs. speed. For example, writing your web apps in C could work fine, until some obscure pointer bug reared its ugly head and overflowed a buffer. This could hypothetically result in, for example, making an online store think you had x number of dollars in store credit, or think you'd already paid when you hadn't.
This problem arises because the deep magic is only run by one company (likely), so it may take longer to rear its head - and when it does, when an exploit is realized, it needs to be fixed immediately. This can also be a problem in that people don't necessarily understand the code someone else might've written, and they have to learn entire sections of code, or entire projects, just to fix.
PHP also takes time to learn about and so on, but you don't have pointer errors and buffer overflows (in that app specifically), since buffers are resized dynamically. You might pass bad data to something else, but you should be checking data once you have it all collected.
Then there's load times. On Apache, mod_php.so is loaded into memory, and then all that needs to be done is to read the page, compile, and execute, or (if pre-compiling/caching) just to execute. JSP is precompilied by default by Tomcat, and the JVM is always running. With a natively-compiled system, if you're not writing your own DSO or server software, it has to be executed for every page hit, and it's a lot harder to get someone who understands your setup than someone who understands JSP/ASP/PHP.
That all being said, I think the main factor is that the Internet moves at 'internet speed' - when you have an idea, it needs to get done now, with as little fuss, trouble, and debugging as possible. Scripting languages offer this by being highly abstracted and widely used (if a million people use PHP, it's more likely that people will find bugs. Slow but rapid beats fast and latent six days of the week.
Scripting languages on the desktop would be cool though, but for mainstream apps, there has to be a lot of garbage collection, because variables in scripting languages (can) take up more memory than their C/C++ counterparts, which can get laggy.
--Dan
I agree with you for the most part; that being said, I think there are two things to keep in mind. First of all, corporations exist solely for the purpose of making money. If the opportunity presents itself, then so shall it be done. (American) CEOs can be sued if they take an action that is not in the best interests of the company('s bottom line, presumably).
Also, of course you only hear about screwy corporations on the news, just like you only hear about terror attacks in Israel/Palestine, or hostages in Moscow. CNN wouldn't attract a lot of eyes if its headlines ran 'Corporation continues ethical business as usual; 30 people on bus in Jerusalem get where they're going without incident; 700 theatre patrons watch play without interruption'. Much though I wish that were all the news there was to report, people are attracted to news about death, corruption, greed, and so on, so when four Israelis die, they focus on that, not the five million that didn't; when corporations cheat, they focus on those, not the ones that did the math right, and so on. For that matter, every time people get together to protest globalization, we end up with riots, assaults, property damage, and so on. Can we really judge all anti-globalists based on those few? Can we judge all multi-nationals based on those few? I don't think so.
--Dan
Priorities people, want to go after a price fixer? Then go after the Pharmacuetical Industry who definatly fixes prices! That sort of battle would benifit more people than this.
Pharmaceutical companies spend billions of dollars a year on research, development, and testing. Forcing them to lower their sources of income would mean they spend less on this, and thus would harm the people more than benefit them because of the unavailability of drugs.
As it is now, they can charge whatever they want, but after a while, other companies are allowed to sell the exact same drug - usually at a much lower (sometimes 50%) price. THAT benefits consumers more than 'solving' price fixing would.
--Dan
While I know that prices in the UK are usually high, the figures quoted are irrelevant for a simple reason: if you're going to compare, compare against one country, and take things into account. The UK taxes cigarettes far more than most countries do (as I recall) - are taxes an issue? Germany is famous for its beer - is the beer in both situations made by the same company, so that you're having a fair comparison of products? If so, what about shipping costs and so on?
Even here in Canada, you can get a pre-cooked shrimp ring for about $9.99 in BC, $8.99 in Saskatchewan, and $4.99 in New Brunswick, all the same brand. You can also get $10 shrimp rings in New Brunswick from more widely known (i.e. larger, better) brands. Is this because people in BC get 'screwed', or because it costs a lot of money to ship refrigerated shrimp ten thousand kilometers?
If you're going to compare, you have to take more factors than just the price into account - local economy, shipping, VAT, local taxes, and so on. Is the US getting screwed because a BK Whopper costs more there than here? No, we just have cheaper beef, and a lower (but stronger) economy, so prices are less. It's good sense, and sensible economics.
--Dan
Maybe they aren't necessarily being taught, but isn't this something that people should know about? The most popular, most common, and most versatile web server software in the world, the basis of IBM's WebSphere (as I recall), basically *the* implementation of the extra fun J2EE stuff (JSP, servlets, etc), and so on.
Isn't this sort of a big thing not to know about? For a web developer to not know what apache is seems like a database admin not knowing what Oracle is, or a network engineer not knowing what IPX is.
They're not taught, but people should be studying things they're interested in, and if you're interested in something, you don't restrict yourself to the classroom - you go to Chapters, find a book that looks neat on something you've never heard of in the computer section and then sit down in Starbucks until they kick you out for not buying the book. You check out some CodeNotes books, you read slashdot (even if only for amusement, as I do) or check out Wired online every now and then. You don't just learn it, you live it, and sure, you have to live other things too, but why are you spending $40-100k to learn about something you don't have a passion for?
People should open their eyes, and the submitter is complaining because they don't - and I agree.
--Dan
I played Descent 1-3 and Tomb Raider 1-5, and now I keep getting these urges to explore caves and mines. Why does no one ever talk about my plight?
--Dan