Is it possible to go well beyond the current standard key sizes for encryption into something more vastly to decode (using "smart" techniques or brute hacking)?
Instead of using 512 bit encryption keys, is there a technical limitation (memory, computational time) that prevents, say, the use of 1Mbit keys? If so, at what point is a middle-of-the-road PC no longer able to produce and decode messages at a reasonable rate for real-time encryption and decryption?
That caught my eye as well, the statement about 2GB of RAM. That couldn't have been the deciding factor for using the Cray. A simple Sun E450 (workgroup server) with 4GB RAM is not an outrageous configuration and not all that expensive.
> You have suggested that one might enhance > security by 'playing games' with the order > and value of significant bytes of the message. > Unfortunatly, this does little to enhance > the security of the message. If one were to > crack the key for such a message, (not knowing > about the added measures) he would recieve > garbled answer, granted.
I suppose part of it is at what point you contort the message. If you confuse the message before encrypting (not encrypting plain text), and then contort it again once it is in encrypted form, a third party would have a much tougher time than if just the plain-text or the encypted-text version wasn't the norm. Even more so if you've played how with the logic of the DES routine itself... not just played with the output. (Stupid side questions... is the source for DES or RSA encryption available? Is there any open-source encryption methods?)
I'm probably barking up the wrong tree by giving these specific examples. Let me ask a more fundemental question. From an outsider's point of view, the race in encryption so far seems to be centered around the number of keys. It started out as a low number, and it has been creeping up and up ever since.
If this is the case, why is the battle being waged on the number of keys? Are people making significant progress in (strengthening) the encryption methods themselves?
Perhaps I'm going to make myself look even more stupid, but here goes. If you are going to do encryption for one-on-one communications (like not in a web browser, and not on a large scale), I can't imagine much benefit from following a well-known standard. It would seem to be quite the contrary.
Onto the key subject, if they are factoring the public key, part of the answer would seem to be to contort the public key format you are using for your private communications.
Mangled keys, mangled messages -- that adds up to a non-standard format and you can throw in a non-standard delivery method (like not through SSL port #s). Hey, you're not going to be automatically scanned, decoded, and archived. Someone would have to decide that you are worth enough attention to pour the man hours and the money into figuring out your own private encryption standard. And if you aren't the only one contorting the standard in weird ways, if you aren't big fish, you are going to have some decent privacy.
I guess what I'm trying to say here is that if someone is using a web browser and server with 'heavy encryption' in order to send sensitive information, it can be considered 'light encryption' because the message format and delivery method is very predictable and can automatically be intercepted, decoded, and stored. If you are taking PGP keys and messages, contorting those around, sprinking the bits here and there in a mis-mash of things like GIFs and WAVs, and putting it on your web site or sending them via IRC, it'll take a lot of effort to put things back together.
True, not many are going to go to that extreme. But if those mega-encryption breakers are out there, they're going to be looking for messages that follow the standard delivery methods, and they're going to be geared for cracking messages following the standard encryption method.
The battle shouldn't be made in the key size. It should be made in the algorithm itself and the flow of data.
I'll admit, I'm clueless on encryption. But I keep seeing something here that confuses me. It seems that they always have the roadmap of what is necessary to decrypt a message. It's just a matter of finding a clever approach to it, or finding massive horsepower.
What this points to, at least to me, is that when people are encrypting messages, they are using a pre-set published standard which describes how the message is going to be laid out. (Well, they would have to be, for any browser with encryption to seem to be able to send encrypted messages back and forth with various web servers.)
If someone really wanted to encrypt a message, instead of throwing more keys at it, why don't they change the order of some of the bytes, the values of others, and pad in a large number of random or semi-random digits?
If a massive key-cracker exists and works, it is going to plow right through someone running with a published encryption standard. But it'll stop dead cold on something like that. (But it might just be like waving a red flag, drawing attention.)
Is this generally correct, or am I way off-base here?
Thinking about this issue, Slashdot really isn't "Open Source Journalism". It is "Open Participation Journalism" which happens to run on open-source software and covers open-source issues. There is a big difference.
Take The Killer List of Videogames (arcade games information database) for example. Open participation? Very. Open source? No. Open participation works for databases and discussions, certainly. (KLOV owns the "open participation" database, but people still contribute. Interesting, no?)
Actually, I'm having a bit of trouble seperating some of the aspects of the two in some respects (aside that open SOURCE refers to source code, obviously). Perhaps these terms don't quite cover the full distinction of differences between, say, Slashdot, and the Linux kernel.
?? Help ??
Slashdot does have a "reputation" system.
on
Wired on Slashdot
·
· Score: 2
Slashdot does give a minor 'reputation' to posts. This post, for example, is starting out as a 2 instead of the normal 1. The reason is that in the past, I have posted replies that were often moderated upwards. So there is a starting point for my messages to begin at.
But there are advantages to NOT having a 'reputation tracker' running. For one, anyone can create a new ID at a whim. If your old one has a bad rap, come up with a new pen name. Second, the system seems to give the impression that every thought should be judged on its own merit. I've seen stuff that I would swear were written by Linus, but were cut down by a few Slashdotters.
Actually, that's a good thing. Just because a well reputable person has something to say, it doesn't make it interesting by default. Judge the ideas for themselves, I say.
Open Source Journalism w/compensation
on
Wired on Slashdot
·
· Score: 2
The term "open source journalism" is just a little odd... since it isn't quite journalism anymore. And I feel funny participating in a peer review of a singular review of a peer review system.
Anyhow, if journalists are afraid of an open-source journalism system, what they should get excited about is one w/compensation. The model is very close to Slashdot.
Open participation, with random readers ranking the results. The pariticpants (story providers) are compensated by the rankings of their readers. Imagine if Slashdot paid decent money for articles that rated a "5". That's incentive for you.
Albiet, there is a number of nagging flaws here and there, and such a system wouldn't have to be pure slashdot, but the basic concept seems to be relatively sound for providing an open-source journalism system that the journalists would buy into (or be bought into!).
It finally happened. I knew one day I'd browse on over to Slashdot and find a naughty story about the company I work for. Well, I can't say they weren't asking for it.
Yeah, the point about all the companies within MCI Worldcom rings true. Like the MCI local network, MCI long distance network, Worldcom long distnace network, and the Worldcom local network (which is actually the MFS local network and the Brooks fiber local network). Then you've got a layer on top of that, like UUNet. Not all that straightforward. They are working on these issues, though.
Responsibility issues? They are there. You should see the hoops I have to go through just to get someone in the same building to investigate a problem within our own LAN. MCI highly emphasises "the procedure".
But the story that Slashdot should be covering isn't the frame network outage. It is MCI's agreement to sell off their IT department to EDS. Most everyone assumed it has gone through. Guess what? They made the announcement 6+ months ago, and the deal (along with IT) is hanging in limbo.
Will we work for EDS? MCI? Do we get our current stock options? Benefits? Policies? Management is saying absolutely nothing. Rumors are flying that the negotiations are going bad and neither side wants to walk away because of a hefty $$ penalty. Even an arbitrator hasn't helped. (That's right, an arbitrator just to reach an agreement. Heaven help both sides over the next five years.)
The entire IT department of a major corporation that can't even say what conditions they will be working under next month. How smart is that?
The main problem is that EDS is a very bad fit for a company like MCI Worldcom. EDS is great for an organization that is low-growth, low-innovation, and reliability situations aren't critical. Say, a local government. But a phone company? Oh boy.
I can't say I know an employee that is enthusiastic about the merger. The only management that claims to love the idea is from the VP level and up.
"The guy's an asshole." What an image. But remember that the image you saw of Abe was carefully crafted by a team of MTV drama specialists. Thousands of hours of footage was condensed into a small package that was packed full of things to draw an emotional response. I can't say that Abe's an asshole. I can't say that he's a really cool guy, either. If I were put into the same situation, I'd question how c00l I would end up looking after MTV carefully crafts a dramatic spin upon my personality.
Slashdot in a written format is intelligent, and well thought-out. There is a large group of people participating, which leads to some very intelligent thoughts being expressed. But a call-in talk show may have a severe quality control problem. Ever watch that call in tech help show on ZDTV? The questions are often half-answered by someone who isn't really qualified to even understand the question. (And how many times have you said, "He wasn't asking about THAT. He was asking..." ?
Of course, Rob isn't an idiot. But I would think that a call-in would have to be more, forgive me, Donahue style focused than Joe Random calling and asking why he's having trouble with beta drivers connecting to a proprietary storage area network on a version of Debian Linux that is four revs back.
Admittedly, it looks rather bad to be flip-flopping on core aspects of a product. But this really isn't as negative as some have played it to be. This company wants to build something innovative -- and successful. Ask Apple... the two don't always go together.
They found that their success factor was unacceptably low, and they were forced to make a change. Heck, I'm glad they're carefully looking at these kind of issues. I'm even more happy they've chosen a Linux kernel.
The interesting thing is, if they both follow through on their announcement, you've got even more innovation and competition than before. QNX is going to continue with developing the operating system. Amiga is going to continue with developing their OS and with a Linux kernel.
If you're one to track win-loss comparisons, it seems like a win for the consumer, a win for Gateway, and a lesser win for QNX.
Of course, it isn't ALL roses. If you're porting or developing software, you've got an OS decision to make. And users will have one to make as well. One side to this to keep aware of is on the hardware end... the Linux version is going to take advantage of far more hardware than the QNX version. Might marginalize QNX if you've got the latest monster video card and QNX won't do it.
I'm trying to figure out if I like or I hate what you say. On one hand, you take the all too bitter "education" point of view that an evil high school teacher (Kimmery) of mine did. On the other hand, you write with the classy style that I find in Salon Magazine.
But we're engineer types. We focus on the meat and not so much the trimmings. It made sense even before reading your re-write. Actually, it made a little more sense. It didn't use "ancillery" and "postulation" or "expository".
But you are quite correct... it could use some touch-ups.
I have a about 25 servers with 4GB+ of ECC memory. Guess what? Every month or two, one of them needs a DIMM replaced due to persistent ECC memory errors. Memory is not absolutely perfect. In especially high quantities (50+ gb), there are going to be flaws on the chips.
So you've got a ~90gb solid state drive on a single chip. What's going to be my bit error rate? And it seems rather expensive to replace a single $900 chip when it goes bad.
Yet another reason why this article is bogus. (That, and it may have low access times... but one a single chip, what's going to be my throughput in mb/sec?)
You've got a vlarge company with about 15 significant competitors. You have developed an in-house piece of software "Y" to run all aspects of a new (and highly competitve) line of business. The software has extreme use value, but no sale value.
What I don't see Eric's model capturing is the fact that you would want to keep "Y" closed-source to prevent competitors from gaining benefit from the technology, code fragments, business models, etc of "Y". You're not afraid of your competitor making and selling "Z" from it, but from using it to gain insight into your business or to enhance their business in a way that causes revenue loss not directly related to software.
The other side of this coin is that most vendors buying this factory model will also fail in the longer run. Funding indefinitely-continuing support expenses from a fixed price is only viable in a market that is expanding fast enough to cover the support and life-cycle costs entailed in yesterday's sales with tomorrow's revenues. Once a market matures and sales slow down, most vendors will have no choice but to cut expenses by orphaning the product.
This is interesting, because it makes me think about Apple. Remember when they orphaned the IIe? What if, instead, they continued support on a pay-per model for the outdated product?
This middleman is a poor interface layer.
on
Feature:Geek Jobs
·
· Score: 3
Ironic, isn't it, that the Internet is supposed to get rid of the middleman which provides marginal services (such as the RIAA), yet this particular group is thriving. And I can say that they provide little value, other than advertising, to the candidate or the company.
From the side of the company, no matter how many ways we tell them we want someone competent at the enterprise level, they throw all sorts of near-entry level resumes at us. Misfits, rejects. People who you might want to change tapes for you. Our own technical interviews are far more enlightening.
From the side of the prospective employee, the recruiters get in the way. I recently visited a large jobs site to find some interesting jobs, but I couldn't get even a SIMPLE description of the job or the details without handing over the resume -- which means getting forever hounded.
I could do a lot better job of selling myself than these headhunters could. And even with them, it is difficult to sell yourself as a "signal" in a sea of "noise". They just don't know how to evaluate people. All they are interested in is pushing bodies through to get a commision... and I doubt that they have my true best interests at heart.
As for the company, they need to step up to the plate and have their own HR departments be more active in advertising and evaluation. These recruiting firms only thrive because HR isn't doing their job, so lower level managers have to work around the system.
The funny thing is that the headhunter/recruiting firms actually work AGAINST the companies that they find employees for. They recruit individuals away from their existing jobs. And once you are in the system, they'll be giving you a call back a year or two later, trying to move you somewhere else. And while they bring a new employee into the shop, they're talking to your other IT guys!
The value of headhunters, in most cases, isn't marginal. It is negative. They thrive on churn. They cost money. They provide poor service. The tight labor market and a poor HR department are the only two things that keep them afloat.
I found very interesting the level of depth put into Jobs' character versus the level of depth put into Gates' character. Steve came across looking (aside from an emotional artist-genius) three-dimensional than the two-dimensional (yet very lucky) Bill Gates. But I am left with some respect for Bill, and some awe at Steve.
The delicious irony that the movie pointed out was how Bill despised Big Blue/Big Brother, and went to the belly of the beast to slay it. (Although I think they put a little too much foresight and gave too much credit to the demands of Gates at the conference table.)
Just like the revelation that Darth Vader was Luke's father, Big Brother has taken off the mask, and it is none other than Bill himself! And our young Luke Skywalker (Jobs) performes a marriage of convenience with his mortal enemy to save his empire. Is there another Skywalker?
It's going to be tough for them to pull off. "AOL Everywhere"? The slogan has got to be bolder than the reality of it. It probably translates more into, "AOL available on every reasonable platform and pipe." I guess I can't argue with that.
Sun's vision of fat servers and dumb clients? Maybe. There are certainly a few issues which are going to work against that -- privacy, and games. I might want to type, send, and store my email on a remote server, but I'm going to be a whole lot less trusting to put personal finances and information on an online "excel/word" application to be stored and managed for me.
Even WORSE, it leads to "metered computing", which nobody wants. Quake type games become impossible to run, and you've got vendor lock-in with their decision of what application you run. (Well, kind of like Microsoft, huh?)
Sun has done a great job of defending its turf ever since Microsoft jumped it on the workstation space, and made a feeble attempt at the datacenter. I'm rooting for Sun here. As far as AOL? If it floats someone's boat, good for them. Just don't ram it down my throat like MSN.
Well, the author has made an interesting point, which I hate to say that I've fallen into. The point is that the battle against Microsoft is going to change the landscape in ways I may not like, win or lose. Linux is looking better even more these days. I need to install it.
DIVX favored the licensor too much.
on
DIVX is dead
·
· Score: 2
The problem with DIVX is that it favored the licensor far beyond what it did the consumer. They got their royalties, and then some -- and had powerful and rigid controls. But the consumer got a system less powerful than DVD, and it combined the *worst* aspects of ownership, rental, pay-per-view, and licensing. In fact, DIVX was more anti-consumer than anything, which is why in the end it was a failure to the company that brought it to market.
I celebrate this as a victory, because if DIVX suceeded, this type of arrangement may have spread elsewhere. It is very closed-system, inconvenient, and money-hungry. It'd work great for a monopoly.
Microsoft has it wrong on the E10k. It sounds like they've been talking to people here and there and haven't actually played with the hardware. The major SPOF is not the SSP workstation, it is the control board. If the control board dies, all your domains will go down. The control board is what, among other things, gives the clock to the entire system. But most E10ks are equipped with two system boards so that you can swap and get up-and-running again quick.
If the SSP (Ultra 5) dies... well, wait. It really doesn't happen. Something like a hard drive crash might do the trick. When you are without and SSP, the domains (virtual hardware systems) on the E10k continue to operate. But you're not going to catch things like record stop dumps (hardware error and warnings... such as persistant ECC memory errors). However, most sites that have purchased E10ks have also purchased two SSPs. They're so cheap in comparison, it makes sense. We have YET to fail over onto the secondary SSP on any of our 10 E10Ks. Since when is an Ultra 5 an "unreliable device"?
Sun complaining that the OS needs to be temporarily quiesced in order to move the kernel from one bank of ram to another? Heck, it's a miracle that it can even happen at all. I'd like to see microsoft write the code to move the kernel on the fly. Not a project I would want to be on.
Poo-poo on the adaptors that don't do DR? Hardly even an issue. Look at them... token ring, ATM, third party. I wouldn't even run a third party SBUS card on my E10k. The translation is that "a minority of SBUS cards are not a good choice for the E10k." Big deal, Bill.
About the swap space issue... they might actually have an issue there. I'm sure Sun is working on a warning now, if it is a problem. BTW... at that point you haven't actually REMOVED the system board. You are doing an operation called a "DR Drain" which moves all the pages of memory from the RAM in that system board to another. Once successful, you are able to remove the system board from the configuration, or abort the change.
Mind you, this is the german -> babelfish -> english translation, so be warned. A yoda filter would be just as correct.
The competitor star Office: " those will not remain in the market. "
Linux: " one of the five problems, which employ me before falling asleep. But I sleep nevertheless still quite well.
" Apache is simply better " " do not offer if we with our servers enough features, which justify our prices, are we ourselves debt. Apache is simply better than we, if it concerns, several Sites on a server version to hosten. Windows 2000 will solve, says this problem " Ballmer. [ the free ] ApacheServer holds with a market share of 57 per cent. Microsoft is with scarce 23 % of the server market because of second place.
Uh... that was not a troll. I was quite serious, actually. I guess admitting a lack of space/time theory knowledge isn't popular? Oh well. No biggie...
Slashdot Mirror
Is it possible to go well beyond the current standard key sizes for encryption into something more vastly to decode (using "smart" techniques or brute hacking)?
Instead of using 512 bit encryption keys, is there a technical limitation (memory, computational time) that prevents, say, the use of 1Mbit keys? If so, at what point is a middle-of-the-road PC no longer able to produce and decode messages at a reasonable rate for real-time encryption and decryption?
That caught my eye as well, the statement about 2GB of RAM. That couldn't have been the deciding factor for using the Cray. A simple Sun E450 (workgroup server) with 4GB RAM is not an outrageous configuration and not all that expensive.
:)
How do you know so much about the Cray?
> You have suggested that one might enhance
> security by 'playing games' with the order
> and value of significant bytes of the message.
> Unfortunatly, this does little to enhance
> the security of the message. If one were to
> crack the key for such a message, (not knowing
> about the added measures) he would recieve > garbled answer, granted.
I suppose part of it is at what point you contort the message. If you confuse the message before encrypting (not encrypting plain text), and then contort it again once it is in encrypted form, a third party would have a much tougher time than if just the plain-text or the encypted-text version wasn't the norm. Even more so if you've played how with the logic of the DES routine itself... not just played with the output. (Stupid side questions... is the source for DES or RSA encryption available? Is there any open-source encryption methods?)
I'm probably barking up the wrong tree by giving these specific examples. Let me ask a more fundemental question. From an outsider's point of view, the race in encryption so far seems to be centered around the number of keys. It started out as a low number, and it has been creeping up and up ever since.
If this is the case, why is the battle being waged on the number of keys? Are people making significant progress in (strengthening) the encryption methods themselves?
Perhaps I'm going to make myself look even more stupid, but here goes. If you are going to do encryption for one-on-one communications (like not in a web browser, and not on a large scale), I can't imagine much benefit from following a well-known standard. It would seem to be quite the contrary.
Onto the key subject, if they are factoring the public key, part of the answer would seem to be to contort the public key format you are using for your private communications.
Mangled keys, mangled messages -- that adds up to a non-standard format and you can throw in a non-standard delivery method (like not through SSL port #s). Hey, you're not going to be automatically scanned, decoded, and archived. Someone would have to decide that you are worth enough attention to pour the man hours and the money into figuring out your own private encryption standard. And if you aren't the only one contorting the standard in weird ways, if you aren't big fish, you are going to have some decent privacy.
I guess what I'm trying to say here is that if someone is using a web browser and server with 'heavy encryption' in order to send sensitive information, it can be considered 'light encryption' because the message format and delivery method is very predictable and can automatically be intercepted, decoded, and stored. If you are taking PGP keys and messages, contorting those around, sprinking the bits here and there in a mis-mash of things like GIFs and WAVs, and putting it on your web site or sending them via IRC, it'll take a lot of effort to put things back together.
True, not many are going to go to that extreme. But if those mega-encryption breakers are out there, they're going to be looking for messages that follow the standard delivery methods, and they're going to be geared for cracking messages following the standard encryption method.
The battle shouldn't be made in the key size. It should be made in the algorithm itself and the flow of data.
Advice from an encryption idiot. Take it FWIW.
I'll admit, I'm clueless on encryption. But I keep seeing something here that confuses me. It seems that they always have the roadmap of what is necessary to decrypt a message. It's just a matter of finding a clever approach to it, or finding massive horsepower.
What this points to, at least to me, is that when people are encrypting messages, they are using a pre-set published standard which describes how the message is going to be laid out. (Well, they would have to be, for any browser with encryption to seem to be able to send encrypted messages back and forth with various web servers.)
If someone really wanted to encrypt a message, instead of throwing more keys at it, why don't they change the order of some of the bytes, the values of others, and pad in a large number of random or semi-random digits?
If a massive key-cracker exists and works, it is going to plow right through someone running with a published encryption standard. But it'll stop dead cold on something like that. (But it might just be like waving a red flag, drawing attention.)
Is this generally correct, or am I way off-base here?
Thinking about this issue, Slashdot really isn't "Open Source Journalism". It is "Open Participation Journalism" which happens to run on open-source software and covers open-source issues. There is a big difference.
Take The Killer List of Videogames (arcade games information database) for example. Open participation? Very. Open source? No. Open participation works for databases and discussions, certainly. (KLOV owns the "open participation" database, but people still contribute. Interesting, no?)
Actually, I'm having a bit of trouble seperating some of the aspects of the two in some respects (aside that open SOURCE refers to source code, obviously). Perhaps these terms don't quite cover the full distinction of differences between, say, Slashdot, and the Linux kernel.
?? Help ??
Slashdot does give a minor 'reputation' to posts. This post, for example, is starting out as a 2 instead of the normal 1. The reason is that in the past, I have posted replies that were often moderated upwards. So there is a starting point for my messages to begin at.
But there are advantages to NOT having a 'reputation tracker' running. For one, anyone can create a new ID at a whim. If your old one has a bad rap, come up with a new pen name. Second, the system seems to give the impression that every thought should be judged on its own merit. I've seen stuff that I would swear were written by Linus, but were cut down by a few Slashdotters.
Actually, that's a good thing. Just because a well reputable person has something to say, it doesn't make it interesting by default. Judge the ideas for themselves, I say.
The term "open source journalism" is just a little odd... since it isn't quite journalism anymore. And I feel funny participating in a peer review of a singular review of a peer review system.
Anyhow, if journalists are afraid of an open-source journalism system, what they should get excited about is one w/compensation. The model is very close to Slashdot.
Open participation, with random readers ranking the results. The pariticpants (story providers) are compensated by the rankings of their readers. Imagine if Slashdot paid decent money for articles that rated a "5". That's incentive for you.
Albiet, there is a number of nagging flaws here and there, and such a system wouldn't have to be pure slashdot, but the basic concept seems to be relatively sound for providing an open-source journalism system that the journalists would buy into (or be bought into!).
It finally happened. I knew one day I'd browse on over to Slashdot and find a naughty story about the company I work for. Well, I can't say they weren't asking for it.
Yeah, the point about all the companies within MCI Worldcom rings true. Like the MCI local network, MCI long distance network, Worldcom long distnace network, and the Worldcom local network (which is actually the MFS local network and the Brooks fiber local network). Then you've got a layer on top of that, like UUNet. Not all that straightforward. They are working on these issues, though.
Responsibility issues? They are there. You should see the hoops I have to go through just to get someone in the same building to investigate a problem within our own LAN. MCI highly emphasises "the procedure".
But the story that Slashdot should be covering isn't the frame network outage. It is MCI's agreement to sell off their IT department to EDS. Most everyone assumed it has gone through. Guess what? They made the announcement 6+ months ago, and the deal (along with IT) is hanging in limbo.
Will we work for EDS? MCI? Do we get our current stock options? Benefits? Policies? Management is saying absolutely nothing. Rumors are flying that the negotiations are going bad and neither side wants to walk away because of a hefty $$ penalty. Even an arbitrator hasn't helped. (That's right, an arbitrator just to reach an agreement. Heaven help both sides over the next five years.)
The entire IT department of a major corporation that can't even say what conditions they will be working under next month. How smart is that?
The main problem is that EDS is a very bad fit for a company like MCI Worldcom. EDS is great for an organization that is low-growth, low-innovation, and reliability situations aren't critical. Say, a local government. But a phone company? Oh boy.
I can't say I know an employee that is enthusiastic about the merger. The only management that claims to love the idea is from the VP level and up.
"The guy's an asshole." What an image. But remember that the image you saw of Abe was carefully crafted by a team of MTV drama specialists. Thousands of hours of footage was condensed into a small package that was packed full of things to draw an emotional response. I can't say that Abe's an asshole. I can't say that he's a really cool guy, either. If I were put into the same situation, I'd question how c00l I would end up looking after MTV carefully crafts a dramatic spin upon my personality.
Haven't you kept up with the news? The national speed limit has been abolished. Besides, not everyone drives on public roads. :)
Here's the link to the full article, not just the headline alert.
Slashdot in a written format is intelligent, and well thought-out. There is a large group of people participating, which leads to some very intelligent thoughts being expressed. But a call-in talk show may have a severe quality control problem. Ever watch that call in tech help show on ZDTV? The questions are often half-answered by someone who isn't really qualified to even understand the question. (And how many times have you said, "He wasn't asking about THAT. He was asking..." ?
Of course, Rob isn't an idiot. But I would think that a call-in would have to be more, forgive me, Donahue style focused than Joe Random calling and asking why he's having trouble with beta drivers connecting to a proprietary storage area network on a version of Debian Linux that is four revs back.
Admittedly, it looks rather bad to be flip-flopping on core aspects of a product. But this really isn't as negative as some have played it to be. This company wants to build something innovative -- and successful. Ask Apple... the two don't always go together.
They found that their success factor was unacceptably low, and they were forced to make a change. Heck, I'm glad they're carefully looking at these kind of issues. I'm even more happy they've chosen a Linux kernel.
The interesting thing is, if they both follow through on their announcement, you've got even more innovation and competition than before. QNX is going to continue with developing the operating system. Amiga is going to continue with developing their OS and with a Linux kernel.
If you're one to track win-loss comparisons, it seems like a win for the consumer, a win for Gateway, and a lesser win for QNX.
Of course, it isn't ALL roses. If you're porting or developing software, you've got an OS decision to make. And users will have one to make as well. One side to this to keep aware of is on the hardware end... the Linux version is going to take advantage of far more hardware than the QNX version. Might marginalize QNX if you've got the latest monster video card and QNX won't do it.
I'm trying to figure out if I like or I hate what you say. On one hand, you take the all too bitter "education" point of view that an evil high school teacher (Kimmery) of mine did. On the other hand, you write with the classy style that I find in Salon Magazine.
But we're engineer types. We focus on the meat and not so much the trimmings. It made sense even before reading your re-write. Actually, it made a little more sense. It didn't use "ancillery" and "postulation" or "expository".
But you are quite correct... it could use some touch-ups.
I have a about 25 servers with 4GB+ of ECC memory. Guess what? Every month or two, one of them needs a DIMM replaced due to persistent ECC memory errors. Memory is not absolutely perfect. In especially high quantities (50+ gb), there are going to be flaws on the chips.
So you've got a ~90gb solid state drive on a single chip. What's going to be my bit error rate? And it seems rather expensive to replace a single $900 chip when it goes bad.
Yet another reason why this article is bogus. (That, and it may have low access times... but one a single chip, what's going to be my throughput in mb/sec?)
You've got a vlarge company with about 15 significant competitors. You have developed an in-house piece of software "Y" to run all aspects of a new (and highly competitve) line of business. The software has extreme use value, but no sale value.
What I don't see Eric's model capturing is the fact that you would want to keep "Y" closed-source to prevent competitors from gaining benefit from the technology, code fragments, business models, etc of "Y". You're not afraid of your competitor making and selling "Z" from it, but from using it to gain insight into your business or to enhance their business in a way that causes revenue loss not directly related to software.
Does this connect with anyone?
This is interesting, because it makes me think about Apple. Remember when they orphaned the IIe? What if, instead, they continued support on a pay-per model for the outdated product?
Ironic, isn't it, that the Internet is supposed to get rid of the middleman which provides marginal services (such as the RIAA), yet this particular group is thriving. And I can say that they provide little value, other than advertising, to the candidate or the company.
From the side of the company, no matter how many ways we tell them we want someone competent at the enterprise level, they throw all sorts of near-entry level resumes at us. Misfits, rejects. People who you might want to change tapes for you. Our own technical interviews are far more enlightening.
From the side of the prospective employee, the recruiters get in the way. I recently visited a large jobs site to find some interesting jobs, but I couldn't get even a SIMPLE description of the job or the details without handing over the resume -- which means getting forever hounded.
I could do a lot better job of selling myself than these headhunters could. And even with them, it is difficult to sell yourself as a "signal" in a sea of "noise". They just don't know how to evaluate people. All they are interested in is pushing bodies through to get a commision... and I doubt that they have my true best interests at heart.
As for the company, they need to step up to the plate and have their own HR departments be more active in advertising and evaluation. These recruiting firms only thrive because HR isn't doing their job, so lower level managers have to work around the system.
The funny thing is that the headhunter/recruiting firms actually work AGAINST the companies that they find employees for. They recruit individuals away from their existing jobs. And once you are in the system, they'll be giving you a call back a year or two later, trying to move you somewhere else. And while they bring a new employee into the shop, they're talking to your other IT guys!
The value of headhunters, in most cases, isn't marginal. It is negative. They thrive on churn. They cost money. They provide poor service. The tight labor market and a poor HR department are the only two things that keep them afloat.
I found very interesting the level of depth put into Jobs' character versus the level of depth put into Gates' character. Steve came across looking (aside from an emotional artist-genius) three-dimensional than the two-dimensional (yet very lucky) Bill Gates. But I am left with some respect for Bill, and some awe at Steve.
The delicious irony that the movie pointed out was how Bill despised Big Blue/Big Brother, and went to the belly of the beast to slay it. (Although I think they put a little too much foresight and gave too much credit to the demands of Gates at the conference table.)
Just like the revelation that Darth Vader was Luke's father, Big Brother has taken off the mask, and it is none other than Bill himself! And our young Luke Skywalker (Jobs) performes a marriage of convenience with his mortal enemy to save his empire. Is there another Skywalker?
It's going to be tough for them to pull off. "AOL Everywhere"? The slogan has got to be bolder than the reality of it. It probably translates more into, "AOL available on every reasonable platform and pipe." I guess I can't argue with that.
Sun's vision of fat servers and dumb clients? Maybe. There are certainly a few issues which are going to work against that -- privacy, and games. I might want to type, send, and store my email on a remote server, but I'm going to be a whole lot less trusting to put personal finances and information on an online "excel/word" application to be stored and managed for me.
Even WORSE, it leads to "metered computing", which nobody wants. Quake type games become impossible to run, and you've got vendor lock-in with their decision of what application you run. (Well, kind of like Microsoft, huh?)
Sun has done a great job of defending its turf ever since Microsoft jumped it on the workstation space, and made a feeble attempt at the datacenter. I'm rooting for Sun here. As far as AOL? If it floats someone's boat, good for them. Just don't ram it down my throat like MSN.
Well, the author has made an interesting point, which I hate to say that I've fallen into. The point is that the battle against Microsoft is going to change the landscape in ways I may not like, win or lose. Linux is looking better even more these days. I need to install it.
The problem with DIVX is that it favored the licensor far beyond what it did the consumer. They got their royalties, and then some -- and had powerful and rigid controls. But the consumer got a system less powerful than DVD, and it combined the *worst* aspects of ownership, rental, pay-per-view, and licensing. In fact, DIVX was more anti-consumer than anything, which is why in the end it was a failure to the company that brought it to market.
I celebrate this as a victory, because if DIVX suceeded, this type of arrangement may have spread elsewhere. It is very closed-system, inconvenient, and money-hungry. It'd work great for a monopoly.
Microsoft has it wrong on the E10k. It sounds like they've been talking to people here and there and haven't actually played with the hardware. The major SPOF is not the SSP workstation, it is the control board. If the control board dies, all your domains will go down. The control board is what, among other things, gives the clock to the entire system. But most E10ks are equipped with two system boards so that you can swap and get up-and-running again quick.
If the SSP (Ultra 5) dies... well, wait. It really doesn't happen. Something like a hard drive crash might do the trick. When you are without and SSP, the domains (virtual hardware systems) on the E10k continue to operate. But you're not going to catch things like record stop dumps (hardware error and warnings... such as persistant ECC memory errors). However, most sites that have purchased E10ks have also purchased two SSPs. They're so cheap in comparison, it makes sense. We have YET to fail over onto the secondary SSP on any of our 10 E10Ks. Since when is an Ultra 5 an "unreliable device"?
Sun complaining that the OS needs to be temporarily quiesced in order to move the kernel from one bank of ram to another? Heck, it's a miracle that it can even happen at all. I'd like to see microsoft write the code to move the kernel on the fly. Not a project I would want to be on.
Poo-poo on the adaptors that don't do DR? Hardly even an issue. Look at them... token ring, ATM, third party. I wouldn't even run a third party SBUS card on my E10k. The translation is that "a minority of SBUS cards are not a good choice for the E10k." Big deal, Bill.
About the swap space issue... they might actually have an issue there. I'm sure Sun is working on a warning now, if it is a problem. BTW... at that point you haven't actually REMOVED the system board. You are doing an operation called a "DR Drain" which moves all the pages of memory from the RAM in that system board to another. Once successful, you are able to remove the system board from the configuration, or abort the change.
Mind you, this is the german -> babelfish -> english translation, so be warned. A yoda filter would be just as correct.
The competitor star Office: " those will not remain in the market. "
Linux: " one of the five problems, which employ me before falling asleep. But I sleep nevertheless still quite well.
" Apache is simply better "
" do not offer if we with our servers enough features, which justify our prices, are we ourselves debt. Apache is simply better than we, if it concerns, several Sites on a server version to hosten. Windows 2000 will solve, says this problem " Ballmer. [ the free ] ApacheServer holds with a market share of 57 per cent. Microsoft is with scarce 23 % of the server market because of second place.
Uh... that was not a troll. I was quite serious, actually. I guess admitting a lack of space/time theory knowledge isn't popular? Oh well. No biggie...
DDuck