Aren't the Itanium and Hammer still Little-Endian architectures? If Intel&HP (and AMD) really had balls, they'd move to a Big-Endian architecture, like _real_ computers (Mainframes and RISC machines). The time saved by programmers (especially compiler writers) not having to bend their minds around byte, halfword, word, and doubleword inversions would yield a sudden surge in productivity. Understanding Little-Endian computing transforms makes my brain hurt. (But ASCII's OK, sort of.)
...and the consensus among the professors at NIU (one of the last mainframe-based universities) seems to be that all 16 of the processors combined add up to 120 MIPS.
OK, I get it. NIU (what's that, Northern Indiana University?) has an ancient IBM 16-way 308X (or older) machine with 12 MIPS per CPU running an even hoarier 1960s public-domain version of MVS. (Yes there is a public-domain version of MVS, but trust me, you _don't_ want to use it.) The CS professors just advise the administration, which hires computer operators at minimum wage to churn out student bills, and some CS students mess about with systems programming and tuning just before graduating and leaving everything mostly broken. Student development TSO runs below Long Batch work (billings for overdue library books are more important). What a nightmare!
The ongoing power and cooling costs alone for such a system would pay for a new mainframe (I assume it's not still under maintenance). Your school is _sorely_ in need of a consultant to run the numbers and explain how they can upgrade, likely at no additional cost. The administration (and CS faculty) are obviously not up to the task at hand.
...it is very disconcerting to have that big black IBM monitor on top of it, because it is running OS/2 on a Celeron board inside the mainframe to control the whole show.
Well, it's a very _stable_ variety of OS/2 (2.2, I believe) running the HCP (Host Control Processor in IBM-ese). All it does is configure I/O channels and memory to partitions, set up LPARS, etc. It's a configuration box like the notebook PC temporarily hooked up to a router to do configuration. (No one thinks twice that their router configuration notebook is running Win95 or suchlike.) Once the configuration is set, you IPL the mainframe and in most circumstances you could reboot the HCP and the mainframe wouldn't notice. However, on some very small mainframes (up to the size of the earlier Freeway machines), using onboard PC-class SCSI storage via I/O channel emulation was done through the HCP. Rebooting one of _those_ HCPs after IPL would ruin your day.
The average 16 processor mainframe is a 120 MIPS machine, whereas the average 1.5 GHz desktop system is a 3000 MIPS machine.
Mainframes run up to about 200 MIPS per processor and with multi-processor overhead a 16-way zSeries tops out somewhat below 3,000 MIPS. These are mainframe MIPS, not what you get as BogoMIPS out of Linux at boot (AFAIK, this is some quick integer timing loop calculation). There's a reason it's called BogoMIPS, troll.
IBM has successfully run over 40,000 Linux images on a mainframe (under VM). Try that on your 1.5Ghz desktop. Ever heard of Transactions Per Second (TPS) in four and five figures, I/O rates in GB/sec, multi-terabyte databases, 99.999% uptime for years? That's mainframe territory, and I sincerely doubt that you've ever seen it, or ever will.
Should we expect to see some !!!MAKE $$$ FAST!!! in the register then? Quite seriously. It's either all comments, or it's edited.
AFAIK, all public comments are published, by law (i.e., the Tunney Act), including stupid ones, ascii art (like goats.cx). But one might expect that only topical comments will be actually read. Especially by the DoJ and the Judge in this case. Important - she needs to see real opposition to Microsoft to counter their astroturf slaves. Wakeup!
and send an anti-Microsoft/DoJ settlement email to the DoJ. You don't have to be articulate or even polite. Numbers count here, sending just "NO to Microsoft" is enough.
And register it will. By law, all public comments submitted must be published in the Federal Register. And the judge in the case will read them, each and every one. Come on, get off your ass! What's two minutes cost for inscribing your opinion in governmental granite for all posterity? Send all those bastardos up in Redmond a real message!
There are reasonable (in my view, not sufficiently convincing, but credible) arguments to suggest that extensive bills of rights are unnecessary [...]
Three things about the US Constitution are interesting - and crucial to preserving liberty:
1) Government powers not explicitly granted to the Federal government are the responsibilities of the several States.
2) Recognition that powers not explicitly devolved to the government are enshrined in the hands of the People.
3) Provisions for Amendments to the Constitution.
The last procedure was employed early to establish some important Amendments identifying US citizens' Rights. They are not "extensive" (your key word), but instead basic. I won't list them here - you can pull them up with a simple search.
But I rather suspect you're an Australian who's quite comfortable with your Big-Brother as nanny socialist government. Sure, they handle everything for you, but you pay outrageous taxes and then pay through the nose for government sanctioned monopoly utilities on top of that. No worries mate, eh!
But Australia's far behind the US, and somewhat behind Britain, when it comes to questioning senior governmental appointments here.
Maybe there's a business opportunity in providing "banned" content to people in Australia and other countries like Communist China and Iran which similarly attempt to censor the 'net. (Australia should be sooo proud to oppress its citizens just like the army satraps of Red China and the radical clerics of Iran, by the way.)
Charge a nominal amount, say AU $5 or so per month, and run an offshore proxy server. Compare search-engine TLD addresses reachable from outside against those reachable from within the customers' country, and mirror all the blocked domains. Give customers PGP (and tell them how to set it up) to protect the emailed proxy address from the censors. Keep a few spare domain names and proxy addresses to jump to whenever the censors catch on, and email customers with the new proxy address in response to inquiries ("Where'd you go?) in order to avoid conspicuous mass mailings. It might work, I think.
I realize there are other anonomizers and proxy-relay operations out there, but has anyone tried a secure subscription model proxy service to bypass oppressive censorship?
As others have mentioned, the Red Hat board is not obligated to do anything but act in the best interests of their stockholders (which, as with most public companies, includes themselves). Is the Red Hat board looking for a buyout? Do they want to sell the company - due to poor fundamentals, uncertain long-term prospects, or simple take-the-money-and-run greed? I don't know but I'd guess we'll find out soon enough. There are lots of variables that we have yet to learn. Here are relevant numbers:
RHAT Shares Outstanding, 180 million approximately
RHAT Closing Price last, $8.41
RHAT 52-week high, $10.14
RHAT Market Cap (see above), $1.43 billion
RHAT Last Quarter P/L, $0.01 per share (a profit)
Red Hat was down near $3.00 per share last October - if AOLTW wanted to acquire them, they should have bought then, either in the open market or through a buyout. Note that we don't know how much of Red Hat AOL has already acquired, but it can't be more than 20% as such a position would have triggered some financial news based upon SEC mandatory disclosure rules.
The question is how much AOLTW is willing to pay for Red Hat: $10/share? $15/share? $20/share? If I were a Red Hat board member, I'd laugh at $10/share, start to listen at $15, but (reluctantly) vote to sell out at $20. That is, unless the Red Hat board knows some things that the rest of us don't (which is certainly the case, but whether that information supports or opposes a buyout by AOLTW or anyone else is confidential). If I bought Red Hat today and got better than 100% profit on a buyout, I'd be rather content with it.
But Red Hat could do otherwise. They could enter into a partnership with AOLTW, or agree to let AOLTW acquire up to 49% of the shares with minority representation on the board.
If I were running Red Hat, I'd go with the latter choice, even if that meant that I might have to eventually go down fighting in the face of a successful hostile takeover by AOLTW. Red Hat might also adopt "poison pill" locks on their stock, of which there are many nasty variations that could effectively prevent any hostile takeover.
Suppose AOLTW bought Red Hat and took the software into Closed Source? Could they try this? Yes. Would they get away with it? One supposes that might depend upon what your definition of "get away with it" means. Who could afford to sue them back into compliance with the GPL? Would the GPL prevail? (It's never been tested in court.) Would tying a lawsuit up the courts for 5-10 years mean they "get away with it" win or lose?
If AOLTW took Red Hat closed source, Mandrake and other Red Hat based distributions would be up the creek. Mandrake (the slickest desktop Linux now) would have to change their base distribution, at great cost and delay. The resulting loss of momentum would surely hurt them and might even stagnate and kill Mandrake. This wouldn't be good.
Move to Europe and then come back and tell us about how great a government Post-Telecom utility was to deal with, in your experience. BT in the UK can't find its ass with both hands for broadband, their "watchdog" Oftel is asleep at the switch, and UK politicians are too busy passing electronic privacy invasion (internal security) laws to even notice. Do you want more of this?
...said Craig Barrett, CEO of Intel Corporation. "Regulatory policies should encourage all companies to deploy these expensive and risky facilities."
"Unless I'm reading something wrong here, is this guy encourging companies to hurry up and try like crazy to go belly up?!?"
Likely taken out of context or Mr. Barrett simply misspoke. What he probably wishes he'd said would be more along the lines of: "Regulatory policies should encourage companies to make the necessary investments in new facilities." But it does seem like somewhat of a Freudian slip though, admittedly ("Hey, we'd be crazy to underwrite this ourselves, but please encourage these other companies to build this expensive and risky infrastructure so we can profit from it if it happens to work to our advantage.")
The basic problem is the travesty of political campaign finance in this country, which basically creates incentives for politicians to waffle around trying to figure out who's going to pay them the most to sell their votes or, ideally, how they can get every opposing special interest to pay them a lot before they get around figuring out how they're going to screw the public with the most profitable (for the politicians) endgame playout. You can see this in the article, which mentions that the White House alone has held over a hundred meetings with interested parties (i.e., potential campaign contributors and their lobbyists) trying to figure out how best to structure public broadband rollout.
OK, let's dispose of this political corruption problem first. (There might be some increased interest in this soon as the Enron bankruptcy scandal unfurls, laying bare how pervasive their influence-buying was for energy deregulation.) The campaign finance answer is really very simple, since the major cost of political campaigns is media buys: require the media (internet, TV, radio, and print media) to provide 20% of their advertising to legitimate political candidates, allocated equitably without charge during the campaign seasons. To get the media to go along, give them a tax credit equal to their opportunity cost for the "free" political advertising, times their marginal corporate tax rate. Voila! Public campaign financing without the corruption of the current back-room, bag-man lobbyist practices. And it would cost no more than the current system, where political lobbying costs are passed on to public in the form of higher prices for everything from electricity and natural gas to phone bills and groceries. (The previous FCC Chairman had actually suggested this, but Colin Powell's boy Michael doesn't seem to have the balls to piss off the regulated media, er... his likely future employers.)
Once we get the political corruption inherent in the current lobbying/campaign-finance system cleared up, then and only then will the politicos get their heads on straight about how to move forward for the good of the public rather than their presently myopic and self-serving focus on getting reelected and amply lining their own vest pockets.
Broadband internet access is now at about the point basic telephone service was 75 years ago - moderately affluent people in cities have it, others in rural areas don't. As a parallel, the rollout of broadband service to everyone who wants it (which eventually will be almost everyone) will look a lot like the provision of basic telephone service. It's going to take some public policy to make this happen. And the mix of technologies and companies involved is much more complicated now than it was with telephone services. It's going be complicated, but I'm sure some fair public utility policies can be devised to make it happen. But we need to make it happen rather quickly, if we are to maintain our economic competitiveness.
...was an FBI informant during the McCarthy era, doing his best to rat on the "different" people in Hollywood. This gives a new twist to "Mickey Mouse" surely.
"Good old Walt" was a misogynistic, homophobic, racist, red-baiting SOB who probably checked under his bed every night for communists. He was also a control freak, as you say, whose legacy survives at Disneyland, etc. with prohibitions of facial hair on employees, common practices of false arrest and strip-searches of (usually ethnic) "suspected" shoplifters, and on and on.
Hey, MP3.com got sued by the RIAA and lost, IIRC. Then Universal comes along and makes an offer for the company. The Board of MP3.com decides discretion is the better part of valor, gets the shareholders some value, and gets the H*ll out of Dodge while the getting is good. You can't blame the CEO (and the Board) for making that business decision in the best interests of their shareholders.
And what Vivendi Universal has done with MP3.com after acquiring it is _their_ responsibility, not the previous owners'. Michael R's one of the good guys trying new things, standing up to media dinosaurs and convicted software monopolists. I'd say he deserves moral support instead of cynical flames. I hope he wins this time.
Not much meat on the bones here. It seemed the interviewer was lobbing softballs and accepting facile replies without followup or pressing any issues. Frankly, I was not at all impressed with either the interviewer or the RedHat CTO answers. He might be brighter than he sounded, but one could not tell it from these interview responses.
I mean, RedHat's not about the desktop, OK? Did this interviewer not know it going into this? Where were the deeper questions about RedHat's working with IBM, HP, Compaq-Alpha-et-al, even Sun in the server space? Where were the clustering, scaling, fault-tolerance, instrumentation (performance and capacity monitoring) questions? No question about RedHat's broken GCC 2.96 compiler and what they're doing to fix it in later releases? This was just a joke, a parody of a real interview. What a shame.
What about making a small and quite [sic] system using a shuttle sv24 barebone
Tom's Hardware (www.tomshardware.com) reviewed this recently, too. Interesting concept to reduce the size and noise. Apple's new iMac is a small footprint and quiet machine also, by the way, even though I don't care for it's "lampbase" case form.
But the Shuttle SV24 has some shortfalls - sure it is small, but perhaps just a little too small - it only has room for _one_ PCI card, and the graphics is a Savage S3 on the small FlexATX mainboard. Good idea, nice package for what it offers, but not the right mix of components and choices, yet. Needs more integration and expansion options.
However, this is the direction I believe PCs will be going as they evolve into pervasive computing (along with their digital convergence with media for both work and play - TV, DVD, VoIP telephony, instant messaging, netmeetings, voice recognition / audio interaction, immersive environments, etc.). Let me review some history to explain.
My grandfather was legally blind during the latter decades of his life (glaucoma) as well as crippled by a back injury. He didn't get out much at all. His only link to the outside world was a shortwave radio that he listened to using wooden headphones; it brought him the 50's US network radio and BBC - news, concerts, lectures - in Oregon. His radio was about the size of my PC tower except nice wood and somewhat larger. I regret not having gotten to know him better, but I was just a small child at the time.
Within about ten years we had transistor radios that would fit in a shirt pocket (just about when I was in high school learning to design and build a superheterodyne tube radio, getting a Basic amateur radio license, getting into college, etc).
The parallels are obvious. My grandchildren will view my kickass self-built tower PC with much the same quiet acceptance as I had once regarded my grandfather's console shortwave radio. They will have the equivalent of transistor radio computers except they'll be things we don't know what, yet. I'm willing to tell what I want to see, though.
I want: a MicroATX mainboard with a good hardware modem and NIC (or 2) onboard, two ECC RAM slots, an AGP 8X/Pro slot, and 2-3 PCI (or next format) card slots for SCSI-160/320 and audio, room for two 3-1/2" SCSI HDs, a CD-R/RW drive, and a DVD drive. I want this all to fit under my coffee table, and a digital/TV display on the wall.
He's right. You don't know what you're talking about. True, AthlonXP CPUs have a built-in thermister diode. But they don't have circuitry to monitor that thermister and throttle the CPU if it exceeds temperature thresholds, unlike Intel's P4 CPUs, which do. AMD mainboards have to monitor the thermister and throttle the clockspeed, but so far only Siemens has marketed a board with such a feature (in Europe). Right now, you _will_ need a heatsink-fan on an AthlonXP CPU (if you don't want an expensive keyfob within seconds of power on).
OK, if you won't I will. My DSL-modem died about a week ago (cheapo Earthlink FlowPoint 2100 without any ventilation - finally the power-supply burned up), so I had to get a 56K modem. It took me four trips to Frys to find a real hardware modem! All the internal PCI modems were friggin WinModems, HSF or some such, even the ones claiming "Controller Based" and "Controller and DSP" and "Works with Linux" - Lucent chipsets, bleah. And these weren't cheap modems, they cost $50-$80. Finally I gave up on finding an internal PCI hardware modem and got a Diamond SupraExpress external 56K modem for $40. Works great, up to 13KB/sec PPP down using BSD compression.
What with mainboard vendors putting AC97 sound, a NIC, and (software) RAID onboard, I'd like to see them drop the poor AC97 sound and software RAID, and build in a solid hardware modem and a good, well supported NIC instead.
With an external hardware modem, SoundBlaster Live! 5.1, Matrox G400, and all SCSI2 storage, my 1.4Ghz AthlonXP box spends about 99% of its time crunching Distributed.net while surfing online and playing MP3s. So yes, PCs have gotten fast enough.
Well, although it seems you have some, er... unresolved issues with Nevada government as well as an unfortunate affinity for bold text, I will agree with you in principle. A mass driver to get this poisonous junk entirely off the planet seems like a reasonable proposition. Imagine mag-lev and an inclined railgun mass-driver able to accelerate a ton or more per shot to escape velocity. One problem is that it needs to face East (use the Earth's rotation rather than have to overcome it) and that makes most politicians based eastwards of wherever it's built nervous in case of a "partial" launch. A ton or more of highly radioactive nuclear waste landing anywhere can really ruin your next ten thousand years. And the US isn't a good place to site this - it's too far North - someplace near the Equator would be much better (that's why Ariane launches there). Maybe Colombia, Venezuela, Peru, even Panama or Puerto Rico? Their economies could sure use the massive cash infusion lasting decades until the job's done. And the system could be used to drive materials into orbit for space stations, Mars and beyond exploration (oops, different bureaucracy). But of course our vision impaired, cover-your-ass bureaucrats won't ever think of doing anything even remotely like this. They've spent $6 Billion on CYA for Yucca Mountain and they are determined to do it, whether or not it's the right answer or Nevada (or anyone else) objects to this. That said, it's a conservative choice - big hole in the ground under a stable desert mountain, way above the water table. Just hope they weld the doors shut when they leave, post warning signs, etc. Ten thousand years is a long, long time - about as long since people developed languages. Let's just hope global warming and geological changes don't turn Nevada into an inland sea in the interim, or something. Keeping it on-planet seems risky to me.
They're doing it to keep you from running servers from your home. If you want to do that, they want you to pay them a lot more.
Well yes, I reckon you're right - or it's for both reasons. But doesn't artificially limiting upload bandwidth basically suck? The ISPs and broadband providers costs track to aggregate bandwidth provided with no distinction between upstream vs downstream. We shouldn't let them get away with charging more for upstream (content serving) than downstream (content consuming). Where's the FCC, the FTC, the DoJ, when we need them? Looking the other way at direction of a bought-off Congress, I guess.
Slashdot Mirror
Aren't the Itanium and Hammer still Little-Endian architectures? If Intel&HP (and AMD) really had balls, they'd move to a Big-Endian architecture, like _real_ computers (Mainframes and RISC machines). The time saved by programmers (especially compiler writers) not having to bend their minds around byte, halfword, word, and doubleword inversions would yield a sudden surge in productivity. Understanding Little-Endian computing transforms makes my brain hurt. (But ASCII's OK, sort of.)
I would be greatly obliged if you could point me to the IBM specs that bear out your MIPS figures - I haven't been able to find them,...
...and the consensus among the professors at NIU (one of the last mainframe-based universities) seems to be that all 16 of the processors combined add up to 120 MIPS.
Try this.
OK, I get it. NIU (what's that, Northern Indiana University?) has an ancient IBM 16-way 308X (or older) machine with 12 MIPS per CPU running an even hoarier 1960s public-domain version of MVS. (Yes there is a public-domain version of MVS, but trust me, you _don't_ want to use it.) The CS professors just advise the administration, which hires computer operators at minimum wage to churn out student bills, and some CS students mess about with systems programming and tuning just before graduating and leaving everything mostly broken. Student development TSO runs below Long Batch work (billings for overdue library books are more important). What a nightmare!
The ongoing power and cooling costs alone for such a system would pay for a new mainframe (I assume it's not still under maintenance). Your school is _sorely_ in need of a consultant to run the numbers and explain how they can upgrade, likely at no additional cost. The administration (and CS faculty) are obviously not up to the task at hand.
...it is very disconcerting to have that big black IBM monitor on top of it, because it is running OS/2 on a Celeron board inside the mainframe to control the whole show.
Well, it's a very _stable_ variety of OS/2 (2.2, I believe) running the HCP (Host Control Processor in IBM-ese). All it does is configure I/O channels and memory to partitions, set up LPARS, etc. It's a configuration box like the notebook PC temporarily hooked up to a router to do configuration. (No one thinks twice that their router configuration notebook is running Win95 or suchlike.) Once the configuration is set, you IPL the mainframe and in most circumstances you could reboot the HCP and the mainframe wouldn't notice. However, on some very small mainframes (up to the size of the earlier Freeway machines), using onboard PC-class SCSI storage via I/O channel emulation was done through the HCP. Rebooting one of _those_ HCPs after IPL would ruin your day.
The average 16 processor mainframe is a 120 MIPS machine, whereas the average 1.5 GHz desktop system is a 3000 MIPS machine.
Mainframes run up to about 200 MIPS per processor and with multi-processor overhead a 16-way zSeries tops out somewhat below 3,000 MIPS. These are mainframe MIPS, not what you get as BogoMIPS out of Linux at boot (AFAIK, this is some quick integer timing loop calculation). There's a reason it's called BogoMIPS, troll.
IBM has successfully run over 40,000 Linux images on a mainframe (under VM). Try that on your 1.5Ghz desktop. Ever heard of Transactions Per Second (TPS) in four and five figures, I/O rates in GB/sec, multi-terabyte databases, 99.999% uptime for years? That's mainframe territory, and I sincerely doubt that you've ever seen it, or ever will.
Even if the Athlon "XP" outperformed the intel (as I am told is often), it probably wouldn't be much, at least not enough worth talking about...
How much faster is "worth talking about" huh? The Athlon XP 2000+ beat the Intel Northwood 2.2Ghz by 45% on the integer benchmark in this comparison!
Should we expect to see some !!!MAKE $$$ FAST!!! in the register then? Quite seriously. It's either all comments, or it's edited.
AFAIK, all public comments are published, by law (i.e., the Tunney Act), including stupid ones, ascii art (like goats.cx). But one might expect that only topical comments will be actually read. Especially by the DoJ and the Judge in this case. Important - she needs to see real opposition to Microsoft to counter their astroturf slaves. Wakeup!
and send an anti-Microsoft/DoJ settlement email to the DoJ. You don't have to be articulate or even polite. Numbers count here, sending just "NO to Microsoft" is enough.
And register it will. By law, all public comments submitted must be published in the Federal Register. And the judge in the case will read them, each and every one. Come on, get off your ass! What's two minutes cost for inscribing your opinion in governmental granite for all posterity? Send all those bastardos up in Redmond a real message!
It's a joke, son. Operating Systems are viewed as religions in OS advocacy flame wars. Get it now?
Isn't there something in the Constitution about Separation of Church and State?
Three things about the US Constitution are interesting - and crucial to preserving liberty:
1) Government powers not explicitly granted to the Federal government are the responsibilities of the several States.
2) Recognition that powers not explicitly devolved to the government are enshrined in the hands of the People.
3) Provisions for Amendments to the Constitution.
The last procedure was employed early to establish some important Amendments identifying US citizens' Rights. They are not "extensive" (your key word), but instead basic. I won't list them here - you can pull them up with a simple search.
But I rather suspect you're an Australian who's quite comfortable with your Big-Brother as nanny socialist government. Sure, they handle everything for you, but you pay outrageous taxes and then pay through the nose for government sanctioned monopoly utilities on top of that. No worries mate, eh!
But Australia's far behind the US, and somewhat behind Britain, when it comes to questioning senior governmental appointments here.
Sheep!Maybe there's a business opportunity in providing "banned" content to people in Australia and other countries like Communist China and Iran which similarly attempt to censor the 'net. (Australia should be sooo proud to oppress its citizens just like the army satraps of Red China and the radical clerics of Iran, by the way.)
Charge a nominal amount, say AU $5 or so per month, and run an offshore proxy server. Compare search-engine TLD addresses reachable from outside against those reachable from within the customers' country, and mirror all the blocked domains. Give customers PGP (and tell them how to set it up) to protect the emailed proxy address from the censors. Keep a few spare domain names and proxy addresses to jump to whenever the censors catch on, and email customers with the new proxy address in response to inquiries ("Where'd you go?) in order to avoid conspicuous mass mailings. It might work, I think.
I realize there are other anonomizers and proxy-relay operations out there, but has anyone tried a secure subscription model proxy service to bypass oppressive censorship?
As others have mentioned, the Red Hat board is not obligated to do anything but act in the best interests of their stockholders (which, as with most public companies, includes themselves). Is the Red Hat board looking for a buyout? Do they want to sell the company - due to poor fundamentals, uncertain long-term prospects, or simple take-the-money-and-run greed? I don't know but I'd guess we'll find out soon enough. There are lots of variables that we have yet to learn. Here are relevant numbers:
RHAT Shares Outstanding, 180 million approximately
RHAT Closing Price last, $8.41
RHAT 52-week high, $10.14
RHAT Market Cap (see above), $1.43 billion
RHAT Last Quarter P/L, $0.01 per share (a profit)
Red Hat was down near $3.00 per share last October - if AOLTW wanted to acquire them, they should have bought then, either in the open market or through a buyout. Note that we don't know how much of Red Hat AOL has already acquired, but it can't be more than 20% as such a position would have triggered some financial news based upon SEC mandatory disclosure rules.
The question is how much AOLTW is willing to pay for Red Hat: $10/share? $15/share? $20/share? If I were a Red Hat board member, I'd laugh at $10/share, start to listen at $15, but (reluctantly) vote to sell out at $20. That is, unless the Red Hat board knows some things that the rest of us don't (which is certainly the case, but whether that information supports or opposes a buyout by AOLTW or anyone else is confidential). If I bought Red Hat today and got better than 100% profit on a buyout, I'd be rather content with it.
But Red Hat could do otherwise. They could enter into a partnership with AOLTW, or agree to let AOLTW acquire up to 49% of the shares with minority representation on the board.
If I were running Red Hat, I'd go with the latter choice, even if that meant that I might have to eventually go down fighting in the face of a successful hostile takeover by AOLTW. Red Hat might also adopt "poison pill" locks on their stock, of which there are many nasty variations that could effectively prevent any hostile takeover.
But, we will see what happens.
Suppose AOLTW bought Red Hat and took the software into Closed Source? Could they try this? Yes. Would they get away with it? One supposes that might depend upon what your definition of "get away with it" means. Who could afford to sue them back into compliance with the GPL? Would the GPL prevail? (It's never been tested in court.) Would tying a lawsuit up the courts for 5-10 years mean they "get away with it" win or lose?
If AOLTW took Red Hat closed source, Mandrake and other Red Hat based distributions would be up the creek. Mandrake (the slickest desktop Linux now) would have to change their base distribution, at great cost and delay. The resulting loss of momentum would surely hurt them and might even stagnate and kill Mandrake. This wouldn't be good.
Stiff Qwest on the bill. Sue in Small Claims court. Win.
Then move somewhere else.
Move to Europe and then come back and tell us about how great a government Post-Telecom utility was to deal with, in your experience. BT in the UK can't find its ass with both hands for broadband, their "watchdog" Oftel is asleep at the switch, and UK politicians are too busy passing electronic privacy invasion (internal security) laws to even notice. Do you want more of this?
...said Craig Barrett, CEO of Intel Corporation. "Regulatory policies should encourage all companies to deploy these expensive and risky facilities."
"Unless I'm reading something wrong here, is this guy encourging companies to hurry up and try like crazy to go belly up?!?"
Likely taken out of context or Mr. Barrett simply misspoke. What he probably wishes he'd said would be more along the lines of: "Regulatory policies should encourage companies to make the necessary investments in new facilities." But it does seem like somewhat of a Freudian slip though, admittedly ("Hey, we'd be crazy to underwrite this ourselves, but please encourage these other companies to build this expensive and risky infrastructure so we can profit from it if it happens to work to our advantage.")
The basic problem is the travesty of political campaign finance in this country, which basically creates incentives for politicians to waffle around trying to figure out who's going to pay them the most to sell their votes or, ideally, how they can get every opposing special interest to pay them a lot before they get around figuring out how they're going to screw the public with the most profitable (for the politicians) endgame playout. You can see this in the article, which mentions that the White House alone has held over a hundred meetings with interested parties (i.e., potential campaign contributors and their lobbyists) trying to figure out how best to structure public broadband rollout.
OK, let's dispose of this political corruption problem first. (There might be some increased interest in this soon as the Enron bankruptcy scandal unfurls, laying bare how pervasive their influence-buying was for energy deregulation.) The campaign finance answer is really very simple, since the major cost of political campaigns is media buys: require the media (internet, TV, radio, and print media) to provide 20% of their advertising to legitimate political candidates, allocated equitably without charge during the campaign seasons. To get the media to go along, give them a tax credit equal to their opportunity cost for the "free" political advertising, times their marginal corporate tax rate. Voila! Public campaign financing without the corruption of the current back-room, bag-man lobbyist practices. And it would cost no more than the current system, where political lobbying costs are passed on to public in the form of higher prices for everything from electricity and natural gas to phone bills and groceries. (The previous FCC Chairman had actually suggested this, but Colin Powell's boy Michael doesn't seem to have the balls to piss off the regulated media, er... his likely future employers.)
Once we get the political corruption inherent in the current lobbying/campaign-finance system cleared up, then and only then will the politicos get their heads on straight about how to move forward for the good of the public rather than their presently myopic and self-serving focus on getting reelected and amply lining their own vest pockets.
Broadband internet access is now at about the point basic telephone service was 75 years ago - moderately affluent people in cities have it, others in rural areas don't. As a parallel, the rollout of broadband service to everyone who wants it (which eventually will be almost everyone) will look a lot like the provision of basic telephone service. It's going to take some public policy to make this happen. And the mix of technologies and companies involved is much more complicated now than it was with telephone services. It's going be complicated, but I'm sure some fair public utility policies can be devised to make it happen. But we need to make it happen rather quickly, if we are to maintain our economic competitiveness.
...was an FBI informant during the McCarthy era, doing his best to rat on the "different" people in Hollywood. This gives a new twist to "Mickey Mouse" surely.
"Good old Walt" was a misogynistic, homophobic, racist, red-baiting SOB who probably checked under his bed every night for communists. He was also a control freak, as you say, whose legacy survives at Disneyland, etc. with prohibitions of facial hair on employees, common practices of false arrest and strip-searches of (usually ethnic) "suspected" shoplifters, and on and on.
Hey, MP3.com got sued by the RIAA and lost, IIRC. Then Universal comes along and makes an offer for the company. The Board of MP3.com decides discretion is the better part of valor, gets the shareholders some value, and gets the H*ll out of Dodge while the getting is good. You can't blame the CEO (and the Board) for making that business decision in the best interests of their shareholders.
And what Vivendi Universal has done with MP3.com after acquiring it is _their_ responsibility, not the previous owners'. Michael R's one of the good guys trying new things, standing up to media dinosaurs and convicted software monopolists. I'd say he deserves moral support instead of cynical flames. I hope he wins this time.
Not much meat on the bones here. It seemed the interviewer was lobbing softballs and accepting facile replies without followup or pressing any issues. Frankly, I was not at all impressed with either the interviewer or the RedHat CTO answers. He might be brighter than he sounded, but one could not tell it from these interview responses.
I mean, RedHat's not about the desktop, OK? Did this interviewer not know it going into this? Where were the deeper questions about RedHat's working with IBM, HP, Compaq-Alpha-et-al, even Sun in the server space? Where were the clustering, scaling, fault-tolerance, instrumentation (performance and capacity monitoring) questions? No question about RedHat's broken GCC 2.96 compiler and what they're doing to fix it in later releases? This was just a joke, a parody of a real interview. What a shame.
What about making a small and quite [sic] system using a shuttle sv24 barebone
Tom's Hardware (www.tomshardware.com) reviewed this recently, too. Interesting concept to reduce the size and noise. Apple's new iMac is a small footprint and quiet machine also, by the way, even though I don't care for it's "lampbase" case form.
But the Shuttle SV24 has some shortfalls - sure it is small, but perhaps just a little too small - it only has room for _one_ PCI card, and the graphics is a Savage S3 on the small FlexATX mainboard. Good idea, nice package for what it offers, but not the right mix of components and choices, yet. Needs more integration and expansion options.
However, this is the direction I believe PCs will be going as they evolve into pervasive computing (along with their digital convergence with media for both work and play - TV, DVD, VoIP telephony, instant messaging, netmeetings, voice recognition / audio interaction, immersive environments, etc.). Let me review some history to explain.
My grandfather was legally blind during the latter decades of his life (glaucoma) as well as crippled by a back injury. He didn't get out much at all. His only link to the outside world was a shortwave radio that he listened to using wooden headphones; it brought him the 50's US network radio and BBC - news, concerts, lectures - in Oregon. His radio was about the size of my PC tower except nice wood and somewhat larger. I regret not having gotten to know him better, but I was just a small child at the time.
Within about ten years we had transistor radios that would fit in a shirt pocket (just about when I was in high school learning to design and build a superheterodyne tube radio, getting a Basic amateur radio license, getting into college, etc).
The parallels are obvious. My grandchildren will view my kickass self-built tower PC with much the same quiet acceptance as I had once regarded my grandfather's console shortwave radio. They will have the equivalent of transistor radio computers except they'll be things we don't know what, yet. I'm willing to tell what I want to see, though.
I want: a MicroATX mainboard with a good hardware modem and NIC (or 2) onboard, two ECC RAM slots, an AGP 8X/Pro slot, and 2-3 PCI (or next format) card slots for SCSI-160/320 and audio, room for two 3-1/2" SCSI HDs, a CD-R/RW drive, and a DVD drive. I want this all to fit under my coffee table, and a digital/TV display on the wall.
He's right. You don't know what you're talking about. True, AthlonXP CPUs have a built-in thermister diode. But they don't have circuitry to monitor that thermister and throttle the CPU if it exceeds temperature thresholds, unlike Intel's P4 CPUs, which do. AMD mainboards have to monitor the thermister and throttle the clockspeed, but so far only Siemens has marketed a board with such a feature (in Europe). Right now, you _will_ need a heatsink-fan on an AthlonXP CPU (if you don't want an expensive keyfob within seconds of power on).
Don't even get me started on modems.
OK, if you won't I will. My DSL-modem died about a week ago (cheapo Earthlink FlowPoint 2100 without any ventilation - finally the power-supply burned up), so I had to get a 56K modem. It took me four trips to Frys to find a real hardware modem! All the internal PCI modems were friggin WinModems, HSF or some such, even the ones claiming "Controller Based" and "Controller and DSP" and "Works with Linux" - Lucent chipsets, bleah. And these weren't cheap modems, they cost $50-$80. Finally I gave up on finding an internal PCI hardware modem and got a Diamond SupraExpress external 56K modem for $40. Works great, up to 13KB/sec PPP down using BSD compression.
What with mainboard vendors putting AC97 sound, a NIC, and (software) RAID onboard, I'd like to see them drop the poor AC97 sound and software RAID, and build in a solid hardware modem and a good, well supported NIC instead.
With an external hardware modem, SoundBlaster Live! 5.1, Matrox G400, and all SCSI2 storage, my 1.4Ghz AthlonXP box spends about 99% of its time crunching Distributed.net while surfing online and playing MP3s. So yes, PCs have gotten fast enough.
On the dual processor theme, the X Window System probably takes as much CPU scrolling your terminal window during the make as the compiler...
nohup make [etc] & # background compile
Wait a bit, depending on what you're compiling...
tail -f nohup.out # see the output
Well, although it seems you have some, er... unresolved issues with Nevada government as well as an unfortunate affinity for bold text, I will agree with you in principle. A mass driver to get this poisonous junk entirely off the planet seems like a reasonable proposition. Imagine mag-lev and an inclined railgun mass-driver able to accelerate a ton or more per shot to escape velocity. One problem is that it needs to face East (use the Earth's rotation rather than have to overcome it) and that makes most politicians based eastwards of wherever it's built nervous in case of a "partial" launch. A ton or more of highly radioactive nuclear waste landing anywhere can really ruin your next ten thousand years. And the US isn't a good place to site this - it's too far North - someplace near the Equator would be much better (that's why Ariane launches there). Maybe Colombia, Venezuela, Peru, even Panama or Puerto Rico? Their economies could sure use the massive cash infusion lasting decades until the job's done. And the system could be used to drive materials into orbit for space stations, Mars and beyond exploration (oops, different bureaucracy). But of course our vision impaired, cover-your-ass bureaucrats won't ever think of doing anything even remotely like this. They've spent $6 Billion on CYA for Yucca Mountain and they are determined to do it, whether or not it's the right answer or Nevada (or anyone else) objects to this. That said, it's a conservative choice - big hole in the ground under a stable desert mountain, way above the water table. Just hope they weld the doors shut when they leave, post warning signs, etc. Ten thousand years is a long, long time - about as long since people developed languages. Let's just hope global warming and geological changes don't turn Nevada into an inland sea in the interim, or something. Keeping it on-planet seems risky to me.
They're doing it to keep you from running servers from your home. If you want to do that, they want you to pay them a lot more.
Well yes, I reckon you're right - or it's for both reasons. But doesn't artificially limiting upload bandwidth basically suck? The ISPs and broadband providers costs track to aggregate bandwidth provided with no distinction between upstream vs downstream. We shouldn't let them get away with charging more for upstream (content serving) than downstream (content consuming). Where's the FCC, the FTC, the DoJ, when we need them? Looking the other way at direction of a bought-off Congress, I guess.