It's entirely appropriate because otherwise he would go on to some other scam. Part of prison is to punish the felon, and part of prison is to protect us from them. Look at Sanford Wallace: after the Earthlink settlement, he moved on to other things. His nightclub went belly-up, so he resorted to selling a bogus spyware removal package which actually hijacked browsers and installed spyware.
Hijacking open relays, installing trojans on unprotected machines, using botnets, faking headers, stealing credit card information, these are not the acts of people with functioning consciences: look at the number of spammers with criminal records for fraud. Spammers are sociopaths, pure and simple, and society is better off without them.
Possibly 640 years in jail? It's a start. One down, 200 to go.
Why would people have broadband before the Net became useful? DSL et al came in because folks wanted a better experience using news and shopping sites, among others. There was no major market for broadband in the late 90s because the customer base did not see the Net as indispensible. Content-heavy sites such as Boo and all the other Flash-infested nightmares were barking up the wrong tree for their potential clients.
In any event, the VCs kept yammering at their various start-ups about fast growth to obtain "first-mover advantage" because Amazon had grown quickly and dominated the online book market. The real reason was because, post-Netscape, VC funds were massively subscribed and the managers had to do *something* with the money or they'd have to give it - and the management fees! - back to the investors. Since mgmt fees are somewhere on the order of 1% to 2% per annum, and $20 billion had sloshed in over the past couple of years, they'd have to return something like a half billion in fees as well as the principal. As it turns out, those fees had already been spent on fat bonuses, so returning the money was right out.
What then ensued was the opposite of a gold rush. Call it a crap rush. The funds were thrown at *anything* which even vaguely resembled a business. 20 pound bags of cat food by courier and you eat the shipping? Great, great, and the shakeout means we get first mover advantage. Here's 50 million. Selling toys over the internet? Fantastic, here's a hundred million, have a party.
At the end of the day, VC managers got to keep their fees, investors got completely reamed, and the rest of us bought Aereons at a discount. There's no way in hell, even now with broadband a fact of life, that most of the dot-bomb IPOs would have succeeded. The root cause of failure would still be the same: poor business plan means outgo > income.
The "New Economy" turned out to be the same old shuck, a game played by sharpies whose interests were only tangentially related to their investors'. Good news if you miss the dot-boom days, by the way: VCs are once again massively oversubscribed and they have to get rid of the money. The bad news: they're now investing in outsourcing. Will the last person to leave the tech industry turn out the lights?
You probably already read him, but Jorge Luis Borges comes to mind. The Library Of Babel fits all the definitions of SF I know, plus it's a mind-expanding read.
Great. Another theory consisting of pretty words strung in a vaguely plausible row. I'll believe this, ah, "model" when it can make quantifiable predictions from previously observed data.
Great. A bunch of pretty words strung up in a row. It means nothing to me unless this, ah, "model" 1) can make predictions from a set of facts 2) has some rigour in the form of numbers attached to it 3) and, as mentioned above for ID, can be falsified.
Until the above conditions are obtained, it's not likely anyone will give up the Standard Model.
The idea of Open Source is you're supposed to be able to look at or change the source if you need to, right? RIGHT? So why, by all the flaming sphincters in hell, did the coders use single letters for major variable names? I'm not talking about the standard i, j, k for index counters; just grep through the code for the data structure called "r". I'll wait.
When a cow-orker would NOT shut off the sound notification on his IM. Even better, he was the networking guy at a former startup, and the network had issues, so every five minutes: "uh-oh" "uh-oh" "uh-oh" "uh-oh" "uh-oh". Now, as y'all may guess from my moniker, I work on digital signal processing. This is fairly mathematical stuff, and when you combine it with machine language for speed, things get kinda intense, and there's nothing like being in the middle of an arcane inner loop only to have your concentration shattered by Yet Another IM Noise. HOLY LIVING MOTHER OF FUCK, WILL YOU TURN THAT GODDAMN THING OFF! Nope, he couldn't. He claimed he needed to know right away, even though the IM window would top itself over everything else anyway.
And they wondered why I wasn't as productive after they moved my desk within earshot of this schmuck. Christ, my fingers still twitch at the thought that it wasn't strangling distance.
Microchannel was a joke. You needed some serious glue logic to get onto the bus, no worse than PCI, but at the time there were no off the shelf interface chips so you ended up using a lot of card space, which was in short supply since the cards themselves were quite a bit smaller than the standard AT form factor. All in all you could count on having maybe 55 to 60% of AT card space to play with, all this before surface mount silicon was readily available.
This made expansion cards more expensive, and adding in the licence fees IBM charged manufacturers set the price of a system+cards beyond most markets, especially when competing AT-bus boxes had a much broader selection of accessories. One or the other extra expense would've been doable given IBM's decent reputation at the time, but both together passed people's squick threshold.
Itanium was a collaboration between Intel and HP, who had a workstation architecture (PA, I think it was called) running HP/UX, so as such x86 compatibility wasn't high on the agenda.
I'll note that HP is one of the very few companies still shipping Itanium servers; they'd might as well since they sacrificed two outstanding architectures (remember Alpha?) to Intel.
About ten years ago, I answered a job ad from **** for a DSP engineer. As y'all may know, once you have the first half-dozen machine languages under your belt, moving the bit is no longer a Big Deal. They were looking for someone with TMS320C52 experience, as I recall, and I'd worked with other TI DSPs from the same family, not to mention some Motorola DSPs, so picking that one up would have been a doddle.
In any event, the phone screen went swimmingly, the guy said he'd recommend me to come soonest as this was a priority project, and he would phone me back within the week.
You know what's next, right? Nothing. Nada. Bugger-all. FInally, after another week or so had passed, I contacted him and asked what was going on.
"Well, you don't have the experience we need to hit the ground running, and we decided we would hire someone who already knew the processor."
"It's not that tough. Just another set of mnemonics."
"Well, that's what we're doing. Good luck on your search."
That req was still open for another six consecutive months. I sure hope they enjoyed saving the two weeks it would've taken me to get up to speed on the C52.
And, oh yeah, since then I've picked up another six or seven machine languages. As for ****, they're still around. Barely.
You sure about that $1.2 Gig? The wiki only says that was the program cost for 1966, which includes a heap of engineering. 1969 was the peak year for S-V flights, with Apollo 9 through 12.
Astronautix quotes the flyaway cost of a Saturn V as $400 million in 1967 dollars, or about four times that in 2005 bucks. On the other hand, the Sov.., er, Russians can build Soyuz vehicles for about $20 million because they use economies of scale. The things are put together on an assembly line. If we had stuck with the S-V, perhaps it wouldn't be that cheap, but experience shows the price would drop once steady-rate production was achieved.
Seriously, compute farms will never be the same. Even if they use double precision we're still talking an order of magnitude more processing per CPU, and the single precision numbers are just ridiculous.
What IBM really needs to do is build some libraries like Intel's excellent Math Kernel and Integrated Performance Primitives, which cover a lot of the operations heavy math users are likely to need. At the end of the day, getting a one or two order of magnitude speedup usually is worth recoding CPU-intensive apps.
I stopped reading his crap fifteen years ago, and from the quotes in the comments I can see I haven't missed a thing. He was a content-free wanker then, and he is now.
As others have said, he gets paid for trolling. Move along, nothing to see.
I was working at a smallish outfit a few years ago when someone recommended me to a VoIP startup. The position came with enough stock options to choke a horse and a nice bump in pay, so I took the gig on the condition I could finish my current project at the original company. They accepted, I did, and when I resigned from Company A the boss told me I was welcome to return any time. The guy didn't take it personally at all, so the whole thing was quite civilised.
Well, the startup did what startups tend to do, and once the paycheques stopped I phoned up my former employer. It turned out he had a great pile of pending projects which needed an experienced DSP engineer, and here comes this ghost from the past who also happens to be familiar with the code base. It took us somewhat less than a minute to reach an agreement.
If he had freaked out and pulled the above crap, I would've never called him up again even if I were so poor I had to eat cat food, but because the guy was a consummate professional we're both ahead of the game.
Dude. you blew it. Next time see if you can get the job, get root access to their servers, make copies of their scams, keep the FTC and FDA in the loop.
Oh, honey, we can't even get rid of the dumb ones. How long did the HP board take to ditch Carly? Three years after the Compaq merger, and well after she screwed up everything else at HP too? Another example: I worked for a guy at [ConsumerProductCompany] who insisted we stick to the the original memory spec even though Moore's law told us we could count on twice the memory by release. Recoding to fit the tight space delayed ship long enough the original RAM was obsoleted, forcing us to redesign the memory controller, and all of which pushed the release late enough that Sony stomped us like a steamroller crushing baby chicks. That guy's now a CTO at another outfit. Hell, Apple alone endured a number of inept CEOs, including a guy who'd crawl under his desk when the going got rough.
And remember, the VCs funded all the dot-bombs five or six years ago and told them to spend like mad to get "first-mover advantage", so we're not necessarily talking the brightest bulbs on the tree here.
Long story short, if even the dumb CEOs get multiyear grace periods and chances to screw up again repeatedly, how do you expect to catch the sharks smart enough to cover up their schemes? Especially, I should note, when luminaries such as Gates and Ellison have both prospered and made their venture investors prosper while being accused of sharp trading. That'll tend to discourage scrutiny so long as the numbers remain rosy.
Long story short, whoever in the blogosphere wrote that fatuous thing you paraphrased is talking out his ass.
Count one here. I can switch hands for the ten minutes a month I use a co-worker's machine. How often do you use someone else's machine, anyway?
You can avoid double-clicking, quite RSI-intensive, by pointing and hitting the return key - trivial to do with a left-handed mouse, PITA with a right handed mouse.
Just as an FYI, the hydrogen in the Hindenburg went straight up. The video-friendly fire was actually the aluminized cellulose nitrate coated skin of the airship burning after it caught a spark; once that started the H2 was additional fuel.
In other words, the hydrogen was safer than the surface of the airship.
Slashdot Mirror
It's entirely appropriate because otherwise he would go on to some other scam. Part of prison is to punish the felon, and part of prison is to protect us from them. Look at Sanford Wallace: after the Earthlink settlement, he moved on to other things. His nightclub went belly-up, so he resorted to selling a bogus spyware removal package which actually hijacked browsers and installed spyware.
Hijacking open relays, installing trojans on unprotected machines, using botnets, faking headers, stealing credit card information, these are not the acts of people with functioning consciences: look at the number of spammers with criminal records for fraud. Spammers are sociopaths, pure and simple, and society is better off without them.
Possibly 640 years in jail? It's a start. One down, 200 to go.
Why would people have broadband before the Net became useful? DSL et al came in because folks wanted a better experience using news and shopping sites, among others. There was no major market for broadband in the late 90s because the customer base did not see the Net as indispensible. Content-heavy sites such as Boo and all the other Flash-infested nightmares were barking up the wrong tree for their potential clients.
In any event, the VCs kept yammering at their various start-ups about fast growth to obtain "first-mover advantage" because Amazon had grown quickly and dominated the online book market. The real reason was because, post-Netscape, VC funds were massively subscribed and the managers had to do *something* with the money or they'd have to give it - and the management fees! - back to the investors. Since mgmt fees are somewhere on the order of 1% to 2% per annum, and $20 billion had sloshed in over the past couple of years, they'd have to return something like a half billion in fees as well as the principal. As it turns out, those fees had already been spent on fat bonuses, so returning the money was right out.
What then ensued was the opposite of a gold rush. Call it a crap rush. The funds were thrown at *anything* which even vaguely resembled a business. 20 pound bags of cat food by courier and you eat the shipping? Great, great, and the shakeout means we get first mover advantage. Here's 50 million. Selling toys over the internet? Fantastic, here's a hundred million, have a party.
At the end of the day, VC managers got to keep their fees, investors got completely reamed, and the rest of us bought Aereons at a discount. There's no way in hell, even now with broadband a fact of life, that most of the dot-bomb IPOs would have succeeded. The root cause of failure would still be the same: poor business plan means outgo > income.
The "New Economy" turned out to be the same old shuck, a game played by sharpies whose interests were only tangentially related to their investors'. Good news if you miss the dot-boom days, by the way: VCs are once again massively oversubscribed and they have to get rid of the money. The bad news: they're now investing in outsourcing. Will the last person to leave the tech industry turn out the lights?
You probably already read him, but Jorge Luis Borges comes to mind. The Library Of Babel fits all the definitions of SF I know, plus it's a mind-expanding read.
Frankie-Bob says two thumbs up!
Great. Another theory consisting of pretty words strung in a vaguely plausible row. I'll believe this, ah, "model" when it can make quantifiable predictions from previously observed data.
I ain't giving up on the Standard Model just yet.
Francois.
Great. A bunch of pretty words strung up in a row. It means nothing to me unless this, ah, "model"
1) can make predictions from a set of facts
2) has some rigour in the form of numbers attached to it
3) and, as mentioned above for ID, can be falsified.
Until the above conditions are obtained, it's not likely anyone will give up the Standard Model.
Francois.
The idea of Open Source is you're supposed to be able to look at or change the source if you need to, right? RIGHT? So why, by all the flaming sphincters in hell, did the coders use single letters for major variable names? I'm not talking about the standard i, j, k for index counters; just grep through the code for the data structure called "r". I'll wait.
When a cow-orker would NOT shut off the sound notification on his IM. Even better, he was the networking guy at a former startup, and the network had issues, so every five minutes: "uh-oh" "uh-oh" "uh-oh" "uh-oh" "uh-oh". Now, as y'all may guess from my moniker, I work on digital signal processing. This is fairly mathematical stuff, and when you combine it with machine language for speed, things get kinda intense, and there's nothing like being in the middle of an arcane inner loop only to have your concentration shattered by Yet Another IM Noise. HOLY LIVING MOTHER OF FUCK, WILL YOU TURN THAT GODDAMN THING OFF! Nope, he couldn't. He claimed he needed to know right away, even though the IM window would top itself over everything else anyway.
And they wondered why I wasn't as productive after they moved my desk within earshot of this schmuck. Christ, my fingers still twitch at the thought that it wasn't strangling distance.
I work at home now.
Microchannel was a joke. You needed some serious glue logic to get onto the bus, no worse than PCI, but at the time there were no off the shelf interface chips so you ended up using a lot of card space, which was in short supply since the cards themselves were quite a bit smaller than the standard AT form factor. All in all you could count on having maybe 55 to 60% of AT card space to play with, all this before surface mount silicon was readily available.
This made expansion cards more expensive, and adding in the licence fees IBM charged manufacturers set the price of a system+cards beyond most markets, especially when competing AT-bus boxes had a much broader selection of accessories. One or the other extra expense would've been doable given IBM's decent reputation at the time, but both together passed people's squick threshold.
Itanium was a collaboration between Intel and HP, who had a workstation architecture (PA, I think it was called) running HP/UX, so as such x86 compatibility wasn't high on the agenda. I'll note that HP is one of the very few companies still shipping Itanium servers; they'd might as well since they sacrificed two outstanding architectures (remember Alpha?) to Intel.
About ten years ago, I answered a job ad from **** for a DSP engineer. As y'all may know, once you have the first half-dozen machine languages under your belt, moving the bit is no longer a Big Deal. They were looking for someone with TMS320C52 experience, as I recall, and I'd worked with other TI DSPs from the same family, not to mention some Motorola DSPs, so picking that one up would have been a doddle.
In any event, the phone screen went swimmingly, the guy said he'd recommend me to come soonest as this was a priority project, and he would phone me back within the week.
You know what's next, right? Nothing. Nada. Bugger-all. FInally, after another week or so had passed, I contacted him and asked what was going on.
"Well, you don't have the experience we need to hit the ground running, and we decided we would hire someone who already knew the processor."
"It's not that tough. Just another set of mnemonics."
"Well, that's what we're doing. Good luck on your search."
That req was still open for another six consecutive months. I sure hope they enjoyed saving the two weeks it would've taken me to get up to speed on the C52.
And, oh yeah, since then I've picked up another six or seven machine languages. As for ****, they're still around. Barely.
Francois.
You sure about that $1.2 Gig? The wiki only says that was the program cost for 1966, which includes a heap of engineering. 1969 was the peak year for S-V flights, with Apollo 9 through 12.
Astronautix quotes the flyaway cost of a Saturn V as $400 million in 1967 dollars, or about four times that in 2005 bucks. On the other hand, the Sov.., er, Russians can build Soyuz vehicles for about $20 million because they use economies of scale. The things are put together on an assembly line. If we had stuck with the S-V, perhaps it wouldn't be that cheap, but experience shows the price would drop once steady-rate production was achieved.
[Ed McMahon]
YOU, SIR, ARE CORRECT!
[/Ed McMahon]
Seriously, compute farms will never be the same. Even if they use double precision we're still talking an order of magnitude more processing per CPU, and the single precision numbers are just ridiculous.
What IBM really needs to do is build some libraries like Intel's excellent Math Kernel and Integrated Performance Primitives, which cover a lot of the operations heavy math users are likely to need. At the end of the day, getting a one or two order of magnitude speedup usually is worth recoding CPU-intensive apps.
Francois.
Oh no. How terrible. ONLY 25 GFlops of double precision on a processor = whatever will we do with such a slug?
I know! Imagine a beo... what? ok, I'll come quietly.
Francois.
Funny you should mention that. You can now get a WinAmp plugin to make your MP3s sound like old vinyl. http://www.pcsoftland.com/audio/plug-in/iZotope-Vi nyl-for-Winamp-2.htm
Dude, most of that 18 years was software. They had backups. The rest was bolting parts together. "Big-ass planet destroying beam goes *here*...."
I stopped reading his crap fifteen years ago, and from the quotes in the comments I can see I haven't missed a thing. He was a content-free wanker then, and he is now.
As others have said, he gets paid for trolling. Move along, nothing to see.
Francois.
I was working at a smallish outfit a few years ago when someone recommended me to a VoIP startup. The position came with enough stock options to choke a horse and a nice bump in pay, so I took the gig on the condition I could finish my current project at the original company. They accepted, I did, and when I resigned from Company A the boss told me I was welcome to return any time. The guy didn't take it personally at all, so the whole thing was quite civilised.
Well, the startup did what startups tend to do, and once the paycheques stopped I phoned up my former employer. It turned out he had a great pile of pending projects which needed an experienced DSP engineer, and here comes this ghost from the past who also happens to be familiar with the code base. It took us somewhat less than a minute to reach an agreement.
If he had freaked out and pulled the above crap, I would've never called him up again even if I were so poor I had to eat cat food, but because the guy was a consummate professional we're both ahead of the game.
Francois.
Dude. you blew it. Next time see if you can get the job, get root access to their servers, make copies of their scams, keep the FTC and FDA in the loop.
http://www.fuckthesouth.com
http://www.machinadynamica.com/machina27.htmp ://www.shakti-innovations.com/audiovideo.htm
htt
Pseudo-science at its best.
Oh, honey, we can't even get rid of the dumb ones. How long did the HP board take to ditch Carly? Three years after the Compaq merger, and well after she screwed up everything else at HP too? Another example: I worked for a guy at [ConsumerProductCompany] who insisted we stick to the the original memory spec even though Moore's law told us we could count on twice the memory by release. Recoding to fit the tight space delayed ship long enough the original RAM was obsoleted, forcing us to redesign the memory controller, and all of which pushed the release late enough that Sony stomped us like a steamroller crushing baby chicks. That guy's now a CTO at another outfit. Hell, Apple alone endured a number of inept CEOs, including a guy who'd crawl under his desk when the going got rough.
And remember, the VCs funded all the dot-bombs five or six years ago and told them to spend like mad to get "first-mover advantage", so we're not necessarily talking the brightest bulbs on the tree here.
Long story short, if even the dumb CEOs get multiyear grace periods and chances to screw up again repeatedly, how do you expect to catch the sharks smart enough to cover up their schemes? Especially, I should note, when luminaries such as Gates and Ellison have both prospered and made their venture investors prosper while being accused of sharp trading. That'll tend to discourage scrutiny so long as the numbers remain rosy.
Long story short, whoever in the blogosphere wrote that fatuous thing you paraphrased is talking out his ass.
Count one here. I can switch hands for the ten minutes a month I use a co-worker's machine. How often do you use someone else's machine, anyway?
You can avoid double-clicking, quite RSI-intensive, by pointing and hitting the return key - trivial to do with a left-handed mouse, PITA with a right handed mouse.
"People don't read the morning newspaper, they slip into it like a warm bath."
Switching power supplies can have end-to-end efficiencies of 90%, which means you lose 10%. No huge loss there.
Just as an FYI, the hydrogen in the Hindenburg went straight up. The video-friendly fire was actually the aluminized cellulose nitrate coated skin of the airship burning after it caught a spark; once that started the H2 was additional fuel.
In other words, the hydrogen was safer than the surface of the airship.