That's what I Was getting at -- there must be needs other than *raw* power that IBM sees an opportunity to provide.
I'm not much of a hardware junkie, but are there still processing capabilities that supercomputers have that PC's don't, or at least that supercomputers are much more efficient at?
I know the lines between the two have begun to blur, now that desktop CPU's have vector instructions (SSE/3DNow); is there something left that desktop PC's (even clusters of them) just don't offer?
I guess there are companies that have their sights set on a *single* drug (like ImClone). But most such companies are always researching *something*, and usually several things at once, so they almost always need such capability (companies like Glaxo).
The difference, I'm guessing, is that they're trying to make it easier/cheaper to get access to this sort of power if you only have a limited or periodic need for it.
I'm not sure how many companies out there only need "a little" time on a "supercomputer" though...
...but, as the article points out, only for corporate/supercomputer types of situations.
After all, the PC revolution demonstrated that individual users want unrestricted computer usage on their own terms, and were/are willing to pay a fairly generous amount for it.
I only see this project working out as long as companies see it as cheaper than building their own solutions. Linux-based clusters can provide a fairly low-cost solution for a lot of high-end computing needs (like rendering tanks) -- that's what IBM will be competing with.
I think it'll boil down to how greedy IBM gets on pricing. If it's too pricey, companies with a fairly regular need for lots of computing power will deploy their own internal computing clusters -- which is ironic, considering that IBM will probably stay very interested in supplying such solutions. It sounds like they're just trying to play all sides of the game: Sell the big/pricey hardware, sell time on the big pricey hardware, sell the lower-cost alternatives -- or at least the contract to deploy and maintain them.
I'm not an expert on Search Engine Backends (IANA...ahh screw that).
But, wouldn't most search engines also at least try to grab index.html on directories in which they've found other files?
Of course, I doubt that's what happened here. From what I can tell on the "victim" website, Reuters just guessed what the URL for the report would be. Who hasn't done that before, in some way or another (e.g. guessing what a broken URL was supposed to be)?
There's clearly NO access control here, except a shining example of how security through obscurity is NOT security at all.
I meant the part about what MAC address the server sees.
The server "sees" the MAC address of his gateway (presumably); but the MAC address that he gets as far as the client is concerned is what the software grabbed from his machine.
I don't think that's what they're planning on doing.
It sounded to me like they have a little app bundled with their software, which grabs the MAC address of the machine it's running on. Then they submit that as part of the login/authentication. When someone makes trouble, they ban the MAC address that person authenticated with.
Which won't do jack, in the long run, since most grief players won't be authenticating with their real MAC address, or even with the same one twice.
Given the description, it will send the one of the PC running this 3rd party program -- which means the PC you're playing/working from.
Basically, they know how easy it is to change or mask IP addresses, and how (particularly for dialup users), banning an IP can punish a lot more people than just the original offender.
So, in the mind of some idiot who failed his CSC networking class before he went to business school, he figured "Hey, MAC addresses are unique! Let's grab that, and ban based on that!"
Just like back then, he didn't do his homework. As others have pointed out: 1) These days, altering your MAC address at run-time is easy, either on your machine or at a router (which is a common component of broadband connections these days) 2) Hackers will have little trouble cracking this "closed source" program, so they can make it emit any or a random MAC address, rather than the machine's actual MAC address. This will not affect connectivity, since its use in this context has nothing to do with the actual connection to the server. 3) If all else fails, network cards are dirt cheap; cheaters/griefers that can't manage #1 or #2 will just buy another network card.
Basically, this "solution" will only keep out the stupidest and poorest grief players. Smart cheaters won't be affected; smart NON-cheaters will probably hack the thing just to show them what a bad idea it was.
I've yet to see an access control system that can't be broken or circumvented; this one doesn't even come close.
Well, I think a lot more people are likely to hold the "woman" factor against her more than an "out of state" factor, if she had both, which is sad if true, but probably not surprising.
The out-of-state factor sort of depends on how centralized you like your government...
I've lived in North Carolina pretty much my entire life, and I'm celebrating the fact that Helms will finally be out of office after this election. Great, he made the right call this time, but a broken clock is right twice a day, too.
Helms has stayed on office for this long for ONE very simple reason: He's pro-tobacco, and tobacco farmers in NC are almost "activists" when it comes to keeping their interests represented in the government, around here. If the tobacco market collapses, the ones who can't effectively change to another cash crop will be out of business, and most such farms are family affairs.
Helms' approach wouldn't work, except most other people seem too apathetic to bother voting someone else in; so, while people complain about it, they don't DO anything about it, but that's the way the cookie crumbles.
I'm looking forward to this year's election: A woman from out of state on one side, and a Democrat on the other -- talk about a dilemma!;) I don't know what the conservative Republicans are making of this year's election, but I'm sure it'll be a lot more interesting than the last few have been...
The Helms Era is finally ending, and at least some folks are going to celebrate.
"You mention Enhanced CDs. As it happens, lots of consumers have had trouble with Enhanced CDs, because they may not play on all devices. So every time you mess with computer technology, there are unexpected effects."
Oh that cracks me up. The President of the RIAA trying to claim that he didn't know CDs with anti-copy technology "may not play on all devices".
What a crock.
It's a shame that "Jack" from the article didn't ask the MS guy, "But what about in a few years when the user *won't be allowed* to turn off copy protection?"
Plus, why would terrorists try to do something like that pilotless? It's higher-tech (meaning harder to acquire and costing more), and (for a while yet) less reliable than a suicidal but reasonably-well-trained human pilot.
I don't think we'll see a pilotless terrorist bomb/plane anytime soon -- it'll be crazy guys at the helm of a Cessna -- or a Jetliner:( -- for some time yet.
Yeah -- our numbers can definitely highlight where review time correlates with testing/support time. But, as you said, breaking that down into whether the code's robustness vs. support/maintainability is a little trickier.
And I agree about reusability -- hand-in-hand with that is using 3rd party solutions when appropriate. Forget solving the problem twice or even ONCE if someone else has done it for you (and you can legally and ethically use their solution, of course, e.g. we found using ACE on a C++ project to be well worth it).
Luckily, we have a leg up on most places for these issues -- we can generally use Java and C++, which lend themselves to maintainable, supportable code far better than C or assembly, and we make the effort for our code to be readable and understandable if at all possible.
I agree -- our own company suffers from giving less effort on code reviews than most of us know we should. People try to save time by under-planning for code reviews, but that saved time is always lost at least twofold in uncaught bugs, extra time for optimization, and so on -- all things that would be identified in a solid code review.
Identify the people in your company that have the best "critical eye" for each language you develop in -- and see to it that they get the time they need to really critique code, either during implementation or at least before starting integration testing (preferably before unit testing, actually). It may be hard to convince management the first time around, but if you account your hours precisely enough, the results WILL speak for themselves, in terms of hours spent overall and on testing/integration.
It's possible in my area at least -- I have a land line from Bell South but no long distance carrier.
I do pay a surcharge though, 5 USD, described on the bill thus: "FCC Charge -- A charge to recover costs associated with connecting to the interstate telecommunications service providers network. This includes the cost of equipment and facilities maintenance. Customers are billed one FCC charge per-line each month."
Xentax
Re:Discovery vs Invention
on
Fair IP Laws?
·
· Score: 2
Ahh. That seems reasonable. I'd make sure such a publisher had my lawyer's phone number and not mine, though;)
Xentax
Re:Discovery vs Invention
on
Fair IP Laws?
·
· Score: 2
Couple things:
Copyrights: Copyrights almost HAVE to be assignable -- not an _exclusive_ assignment necessarily; but without a non-exclusive copyright, a publisher can't distribute your Great American Novel (tm) legally. Even disallowing exclusive transfer of copyright is dangerous -- shouldn't the company that employs me have exclusive right to the code they're paying me to develop?
I agree with the general spirit though -- the goal is to prevent financial exploitation of your innovation by others without your express permission.
Patents: 3 -- I don't think this can work. If you can't cite relevant prior art that serves as... "inspiration" for lack of a better term... for your own work, you can't really patent ANYTHING.
If I just invented the helicopter, and I couldn't cite the airplane, someone else could say "look, he stole that idea from the prior art, the airplane" -- I have to be able to acknowledge existing technology that relates even though my idea IS new, non-obvious, and useful.
Obviously, the "portion" that's new in that example is identifiable; but I'm not sure it's ALWAYS true that the new parts can be identified so; the citation of relevant pre-existing work makes it easier for a review board (good idea) to decide if your work IS sufficiently new, non-obvious, and useful.
Xentax
Re:There ARE other ways
on
Fair IP Laws?
·
· Score: 2
The very reasons that the examples you cite work are the facts that they're both SO revolutionary and that the creators mentioned weren't really after making a lot of money off of them.
For one thing, creators like Van Gogh operated under sponsorship -- they were paid to "do their thing", not for specific results. It doesn't work like that anymore.
Many ideas created at the time were so revolutionary that no-one else could even claim to understand them; heck, half the ideas were considered, well, "crackpot" at the time they gained notice, and there's little danger in someone stealing your work if they think its madness, right?
These days, most "breakthroughs" are more evolutionary than revolutionary in nature, and/or are the solutions to known problems (like, say, a cure for cancer). The little guy who finds the answer has little defense against the big corporation that can more quickly capitalize upon it, UNLESS it's either so "breakthrough" that they can't even do it without his help, or he gets legal recognition of the fact that he did it *first*.
Xentax
Re:As far as patents are concerned
on
Fair IP Laws?
·
· Score: 2
We'd need to see something a lot more specific than that...
Would issues be required to be addressed? Would there have to be a certain minimum amount of feedback on EVERY patent (because let's face it, some things are sufficiently obscure that few people if any will bother or even be able to comment)? Would a popular "vote" against the patent's validity be sufficient to prevent it from being awarded?
I'd require at least prior art claims to be addressed by the applicant, if they meet some hard-and-fast criteria in terms of verifiability (especially if the prior art isn't itself patented).
That is, you're right -- we can define equations and systems of mathematics and the like to model or describe the universe we exist within.
I think (and no I haven't read the book, I'm extrapolating from the review) that Wolfram is claiming that 1) ANYTHING can be modeled with cellular automata; and therefore that 2) If 1 is true, then whether or not the universe really IS an "assemblage" of automata or not, we shouldn't care -- can you really do better (or do you CARE to do better) than a perfectly accurate model "Of Everything"?
Of course, I still would debate how you can tell when you have the RIGHT model -- presumably multiple models can predict all of the universe that we currently comprehend, but when there's a new area to understand, how do you tell which of those is right? It's the whole infinite-possibilities paradox, just one layer removed.
Hey, gcc is the worst compiler in the world -- except for all the other ones out there.
2.95 definitely made me happier than 2.71, though. We started with Sun's Forte/CC/we-can't-decide-what-to-call-it compiler, so it could have been worse;)
Slashdot Mirror
That's what I Was getting at -- there must be needs other than *raw* power that IBM sees an opportunity to provide.
I'm not much of a hardware junkie, but are there still processing capabilities that supercomputers have that PC's don't, or at least that supercomputers are much more efficient at?
I know the lines between the two have begun to blur, now that desktop CPU's have vector instructions (SSE/3DNow); is there something left that desktop PC's (even clusters of them) just don't offer?
Xentax
I guess there are companies that have their sights set on a *single* drug (like ImClone). But most such companies are always researching *something*, and usually several things at once, so they almost always need such capability (companies like Glaxo).
Xentax
The difference, I'm guessing, is that they're trying to make it easier/cheaper to get access to this sort of power if you only have a limited or periodic need for it.
I'm not sure how many companies out there only need "a little" time on a "supercomputer" though...
Xentax
...but, as the article points out, only for corporate/supercomputer types of situations.
After all, the PC revolution demonstrated that individual users want unrestricted computer usage on their own terms, and were/are willing to pay a fairly generous amount for it.
I only see this project working out as long as companies see it as cheaper than building their own solutions. Linux-based clusters can provide a fairly low-cost solution for a lot of high-end computing needs (like rendering tanks) -- that's what IBM will be competing with.
I think it'll boil down to how greedy IBM gets on pricing. If it's too pricey, companies with a fairly regular need for lots of computing power will deploy their own internal computing clusters -- which is ironic, considering that IBM will probably stay very interested in supplying such solutions. It sounds like they're just trying to play all sides of the game: Sell the big/pricey hardware, sell time on the big pricey hardware, sell the lower-cost alternatives -- or at least the contract to deploy and maintain them.
Xentax
I'm not an expert on Search Engine Backends (IANA...ahh screw that).
But, wouldn't most search engines also at least try to grab index.html on directories in which they've found other files?
Of course, I doubt that's what happened here. From what I can tell on the "victim" website, Reuters just guessed what the URL for the report would be. Who hasn't done that before, in some way or another (e.g. guessing what a broken URL was supposed to be)?
There's clearly NO access control here, except a shining example of how security through obscurity is NOT security at all.
Xentax
I meant the part about what MAC address the server sees.
;)
The server "sees" the MAC address of his gateway (presumably); but the MAC address that he gets as far as the client is concerned is what the software grabbed from his machine.
So I guess we're just vehemently agreeing
Xentax
I don't think that's what they're planning on doing.
It sounded to me like they have a little app bundled with their software, which grabs the MAC address of the machine it's running on. Then they submit that as part of the login/authentication. When someone makes trouble, they ban the MAC address that person authenticated with.
Which won't do jack, in the long run, since most grief players won't be authenticating with their real MAC address, or even with the same one twice.
Xentax
Given the description, it will send the one of the PC running this 3rd party program -- which means the PC you're playing/working from.
Basically, they know how easy it is to change or mask IP addresses, and how (particularly for dialup users), banning an IP can punish a lot more people than just the original offender.
So, in the mind of some idiot who failed his CSC networking class before he went to business school, he figured "Hey, MAC addresses are unique! Let's grab that, and ban based on that!"
Just like back then, he didn't do his homework. As others have pointed out:
1) These days, altering your MAC address at run-time is easy, either on your machine or at a router (which is a common component of broadband connections these days)
2) Hackers will have little trouble cracking this "closed source" program, so they can make it emit any or a random MAC address, rather than the machine's actual MAC address. This will not affect connectivity, since its use in this context has nothing to do with the actual connection to the server.
3) If all else fails, network cards are dirt cheap; cheaters/griefers that can't manage #1 or #2 will just buy another network card.
Basically, this "solution" will only keep out the stupidest and poorest grief players. Smart cheaters won't be affected; smart NON-cheaters will probably hack the thing just to show them what a bad idea it was.
I've yet to see an access control system that can't be broken or circumvented; this one doesn't even come close.
Xentax
Well, I think a lot more people are likely to hold the "woman" factor against her more than an "out of state" factor, if she had both, which is sad if true, but probably not surprising.
The out-of-state factor sort of depends on how centralized you like your government...
Xentax
Boy, I ought to check my own facts first; Elizabeth Dole is a North Carolina native, so she's not pulling a Hillary on us, or anything.
Xentax
I've lived in North Carolina pretty much my entire life, and I'm celebrating the fact that Helms will finally be out of office after this election. Great, he made the right call this time, but a broken clock is right twice a day, too.
;) I don't know what the conservative Republicans are making of this year's election, but I'm sure it'll be a lot more interesting than the last few have been...
Helms has stayed on office for this long for ONE very simple reason: He's pro-tobacco, and tobacco farmers in NC are almost "activists" when it comes to keeping their interests represented in the government, around here. If the tobacco market collapses, the ones who can't effectively change to another cash crop will be out of business, and most such farms are family affairs.
Helms' approach wouldn't work, except most other people seem too apathetic to bother voting someone else in; so, while people complain about it, they don't DO anything about it, but that's the way the cookie crumbles.
I'm looking forward to this year's election: A woman from out of state on one side, and a Democrat on the other -- talk about a dilemma!
The Helms Era is finally ending, and at least some folks are going to celebrate.
Xentax
"You mention Enhanced CDs. As it happens, lots of consumers have had trouble with Enhanced CDs, because they may not play on all devices. So every time you mess with computer technology, there are unexpected effects."
Oh that cracks me up. The President of the RIAA trying to claim that he didn't know CDs with anti-copy technology "may not play on all devices".
What a crock.
It's a shame that "Jack" from the article didn't ask the MS guy, "But what about in a few years when the user *won't be allowed* to turn off copy protection?"
Xentax
Me? Nah...
It's not *my* ass that I want every Microsoft EULA shoved up...
Xentax
Plus, why would terrorists try to do something like that pilotless? It's higher-tech (meaning harder to acquire and costing more), and (for a while yet) less reliable than a suicidal but reasonably-well-trained human pilot.
:( -- for some time yet.
I don't think we'll see a pilotless terrorist bomb/plane anytime soon -- it'll be crazy guys at the helm of a Cessna -- or a Jetliner
Xentax
Agreed...I always figured he said "they were married" as a Milton roundabout way of describing what he watched them do.
.wav of that particular line of the movie...
Someone out there's gotta have a
Xentax
Yeah -- our numbers can definitely highlight where review time correlates with testing/support time. But, as you said, breaking that down into whether the code's robustness vs. support/maintainability is a little trickier.
And I agree about reusability -- hand-in-hand with that is using 3rd party solutions when appropriate. Forget solving the problem twice or even ONCE if someone else has done it for you (and you can legally and ethically use their solution, of course, e.g. we found using ACE on a C++ project to be well worth it).
Luckily, we have a leg up on most places for these issues -- we can generally use Java and C++, which lend themselves to maintainable, supportable code far better than C or assembly, and we make the effort for our code to be readable and understandable if at all possible.
Xentax
I agree -- our own company suffers from giving less effort on code reviews than most of us know we should. People try to save time by under-planning for code reviews, but that saved time is always lost at least twofold in uncaught bugs, extra time for optimization, and so on -- all things that would be identified in a solid code review.
Identify the people in your company that have the best "critical eye" for each language you develop in -- and see to it that they get the time they need to really critique code, either during implementation or at least before starting integration testing (preferably before unit testing, actually). It may be hard to convince management the first time around, but if you account your hours precisely enough, the results WILL speak for themselves, in terms of hours spent overall and on testing/integration.
Xentax
It's possible in my area at least -- I have a land line from Bell South but no long distance carrier.
I do pay a surcharge though, 5 USD, described on the bill thus: "FCC Charge -- A charge to recover costs associated with connecting to the interstate telecommunications service providers network. This includes the cost of equipment and facilities maintenance. Customers are billed one FCC charge per-line each month."
Xentax
Ahh. That seems reasonable. I'd make sure such a publisher had my lawyer's phone number and not mine, though ;)
Xentax
Couple things:
... "inspiration" for lack of a better term ... for your own work, you can't really patent ANYTHING.
Copyrights:
Copyrights almost HAVE to be assignable -- not an _exclusive_ assignment necessarily; but without a non-exclusive copyright, a publisher can't distribute your Great American Novel (tm) legally. Even disallowing exclusive transfer of copyright is dangerous -- shouldn't the company that employs me have exclusive right to the code they're paying me to develop?
I agree with the general spirit though -- the goal is to prevent financial exploitation of your innovation by others without your express permission.
Patents:
3 -- I don't think this can work. If you can't cite relevant prior art that serves as
If I just invented the helicopter, and I couldn't cite the airplane, someone else could say "look, he stole that idea from the prior art, the airplane" -- I have to be able to acknowledge existing technology that relates even though my idea IS new, non-obvious, and useful.
Obviously, the "portion" that's new in that example is identifiable; but I'm not sure it's ALWAYS true that the new parts can be identified so; the citation of relevant pre-existing work makes it easier for a review board (good idea) to decide if your work IS sufficiently new, non-obvious, and useful.
Xentax
The very reasons that the examples you cite work are the facts that they're both SO revolutionary and that the creators mentioned weren't really after making a lot of money off of them.
For one thing, creators like Van Gogh operated under sponsorship -- they were paid to "do their thing", not for specific results. It doesn't work like that anymore.
Many ideas created at the time were so revolutionary that no-one else could even claim to understand them; heck, half the ideas were considered, well, "crackpot" at the time they gained notice, and there's little danger in someone stealing your work if they think its madness, right?
These days, most "breakthroughs" are more evolutionary than revolutionary in nature, and/or are the solutions to known problems (like, say, a cure for cancer). The little guy who finds the answer has little defense against the big corporation that can more quickly capitalize upon it, UNLESS it's either so "breakthrough" that they can't even do it without his help, or he gets legal recognition of the fact that he did it *first*.
Xentax
We'd need to see something a lot more specific than that...
Would issues be required to be addressed? Would there have to be a certain minimum amount of feedback on EVERY patent (because let's face it, some things are sufficiently obscure that few people if any will bother or even be able to comment)? Would a popular "vote" against the patent's validity be sufficient to prevent it from being awarded?
I'd require at least prior art claims to be addressed by the applicant, if they meet some hard-and-fast criteria in terms of verifiability (especially if the prior art isn't itself patented).
Xentax
I think it's the issue of universality.
That is, you're right -- we can define equations and systems of mathematics and the like to model or describe the universe we exist within.
I think (and no I haven't read the book, I'm extrapolating from the review) that Wolfram is claiming that 1) ANYTHING can be modeled with cellular automata; and therefore that 2) If 1 is true, then whether or not the universe really IS an "assemblage" of automata or not, we shouldn't care -- can you really do better (or do you CARE to do better) than a perfectly accurate model "Of Everything"?
Of course, I still would debate how you can tell when you have the RIGHT model -- presumably multiple models can predict all of the universe that we currently comprehend, but when there's a new area to understand, how do you tell which of those is right? It's the whole infinite-possibilities paradox, just one layer removed.
Xentax
Sounds interesting...got some linkage you could share to elaborate?
Xentax
Hey, gcc is the worst compiler in the world -- except for all the other ones out there.
;)
2.95 definitely made me happier than 2.71, though. We started with Sun's Forte/CC/we-can't-decide-what-to-call-it compiler, so it could have been worse
Xentax