That's the point. Vista wasn't good enough. It may be true technically that software is never perfect, but your rationalization ignores the fact that there is a large range of quality in that imperfect gray area. A lot of software that hits the market is pretty good, with a few well-hidden bugs that need to be patched, and a range of upcoming features that its architects are planning to roll out over time.
There's a huge difference between that degree of market-readiness, and the turd called Vista.
Linux is still a work in progress
Yes. And when you download a linux distro, you know darn well whether you are getting a stable release, a beta release, or wild-crazy-brand-spankin-new alpha code. Vista was beta code in disguise, forcibly poured down the throats of > 90% of PC buyers. (Not the first time they pulled that little trick, either. Think WinME, first edition of '98, etc.)
When *isn't* something that is still "alive" and used a work in progress?
Ideally, the remaining work should be fine-tuning. Vista isn't yet to that stage, IMHO. People I know who make their living deploying and maintaining MS OS'es agree.
First outside the cuprate family? Maybe according to TFA, but what about the recently discussed silicon-hydrogen compounds and aluminum (or aluminium, if you prefer) nanoclusters which are a) outside the cuprate family, and b) showing superconductivity at elevated temperatures?
I read the summary and TFA, and still wasn't impressed. Sure, the new iron/arsenic compounds are somewhat interesting, and maybe in the future holds some kind of potential. It just seems a little underwhelming, since other research is producing compunds that seem more promising.
Meh... maybe. I think it is a different kind of survival tactic. Admins only tend to stay employed when they don't screw up. Working on a platform you aren't familiar with increases your chance of screwing something up. (new admin on a Solaris box: Hey, what does this 'sys-unconfig' command do? [enter]... )
The need to be error-free (hah, right!) breeds a dislike for the unfamiliar, and an affinity for systems that work EXACTLY as you expect.
Implementing change in an admin's world is often an unintentionally hostile thing. In my experience, change is often inflicted by uninformed people (the CEO!) for dubious business reasons (he had a great round of golf with the sales guy!). I'm not surprised that change is often greeted with suspicion and hostility.
Admins spend much time and effort acquiring the knowledge that makes them valuable. Change the platforms they work on, and you reduce the knowledge base they offer, and hence their (perceived) value.
Plus, it's fun having religious wars over who has the best hardware/os/app/db/etc. It gives admins a reason to comment on/.
So if we were crazy enough to bring up Penicillium roqueforti, (say as a composting agent for its ability to break down complex organic molecules in a bio-recycling process) and it got loose on the moon, you're saying that there's a chance it could survive and gradually spread over the whole surface?
Do you know what this means? We would FINALLY have a moon that really is made of green cheese! (Well, bleu-green, anyway.)
David Cross, a product unit manager at Microsoft, was the group program manager in charge of designing User Account Control (UAC).
There. Credentials established. He was in charge of designing it.
"The reason we put UAC into the (Vista) platform was to annoy users--I'm serious".
There. Intent established.
The media aren't exploiting it. They are reporting it. When the company with world's dominant desktop OS and dominant desktop productivity suite puts a group program manager on stage at a public event with press in attendance, and he specifically reveals that the reason for particular piece of so-called security software is to "annoy users"... THIS... IS... NEWSWORTHY.
The Gump quote is the only piece of your post I agree with. Speaking of which, his comments included the following stats:
- 80% of the warnings were generated by 10 apps
- Some undisclosed number of those 10 apps were from (... wait for it, wait for it... ) Microsoft. How the hell are they going to encourage 3rd party developers to clean up their act when they can't even build good code in house?
More:
- 66% of sessions now run without prompts. (means chance of annoying prompt = 34%)
- 88% of users have not turned off UAC. (means 12% are so fed up they switch it off)
- 7% of UAC permission dialog boxes get a "No" click. (means that 93% of sheeple^h^h^h^h^h users automagically click "Yes". Alternate explanation: Those 7% are too afraid of "Yes", and click "No" by default.)
Question about the robots in your sig: Are they iPhone killers in the sense that they stalk, attack, and kill iPhones? Or do they use iPhones to kill people? Kill people with iPhones?
Either way, please send two to my hotel room as soon as they are built.
IANAL, but here is the Mass law re: licensing a firm in the business of doing investigation work.
"Whoever violates any provision of this section shall be punished by a fine of not less than two hundred nor more than one thousand dollars or by imprisonment for not more than one year, or by both such fine and imprisonment."
Here is the law about impersonating a police officer.
"... shall be punished by a fine of not more than four hundred dollars or by imprisonment for not more than one year."
The investigators might actually have it easier if they posed as officers. Hell, the fines imposed are a lot less than the RIAA would like to hit the John Does with for far slighter infractions. No indication if either qualifies as a felony, though.
I think the author's point was that to squeeze more performance out of a piece of silicon, you can either crank the speed up or pile on more cores + threads. IBM is (mostly) doing the former, while Sun is (primarily) doing the latter.
IBM's multithreading is a baby step compared to Sun's T1 and T2. Power6 = 2 cores*2threads each. T1 = (4 to 8 cores)*4 threads each. T2 = (up to 8 cores)*8 threads each.
Of course, this simply underlines the fact that Sun and IBM are coming at this from 2 different directions. Sun set out to build a economical cpu for entry-level systems that would perform well with lots of everyday tasks (emphasis on low cost of ownership, easily fit into any environment, server consolidation roles), with minimal need to optimise code. If your code supports multithreading on SPARC, you are good to go. Their higher-end systems will get more cores per processor in the future.
IBM started at the top end of the spectrum with a datacenter-centric platform, and will require heavier modification / optimisation of apps to see performance improvements. Their aim is to have screaming sequential execution in a big-iron box. They are planning to merge the technology down into more commoditized products in the future.
Each of the threads on a T2 only sucks up about 1.5 watts. Does anyone out there have any idea what the thermal envelope on a Power6 is? I haven't seen anything more accurate than
educated guesses. Assuming 100 watts for a 4.7Ghz Power6 with 4 threads, you're looking at about 25 watts per thread. Okay if you want to run a few things at max speed (batch processing), bad if you want to run lots of threads (ie, server consolidation & virtualization).
A liquid cooling system is good for establishing uber-geek status among peers, but it's bad for cost of ownership, simplicity of operation, etc.
Overall, Sun offers way more threads at half the speed, while IBM gives you top-end clock speed. Ultimately, while the Power6 is terribly sexy with its ginormous clock speed and liquid-cooling system, you can get +/- similar performance out of less expensive hardware.
Never confuse processor speed with processing power.
Great. Let me know when fission reactors reach car-engine size.
(Cue up plutonium-powered DeLorean jokes...)
No, really I get it, and I expect we'll see nuclear power expand, although we are limited in how fast we can bring new plants online. But what the hell are we going to do with all the damn waste? Yucca Mountain is stupid idea. Are we going to start reprocessing our nuclear waste like France? Move to a more progressive reactor design like MSRs, for which reprocessing is an integral part?
Personally, I like the MSR model for a number of reasons. It can burn old, nasty, long-half-life waste into less-nasty stuff, little/no chance of an out-of-control reaction, and relatively efficient burnup of the fuel it is fed.
... if I operate a search engine based solely in the US, why would a EU visitor expect that I have to conform to EU regs?
AFAIK, Google's situation is different because it has a physical presence in Ireland and Belgium, and so they probably will have to eventually conform to the regulations in question. In principle, I like the EU personal data protection laws, but in general I believe that people need to protect themselves by controlling their own behavior rather than the actions of others.
[setenv RANT on; export RANT]
Want to protect your privacy? Read privacy policies on websites. You don't like Google's policies? Search elsewhere. Use EU-based search engines. Use search engines whose privacy statements conform to EU standards. Invent your own damn search engine that is based in the EU and conforms to EU laws. Who knows, maybe US surfers would prefer to use your engine in relative privacy rather than Google's. Maybe you'll be wildly successful at Google's expense.
[setenv JOKING on; export JOKING]
We already fought one war because of onnerous regulations from that side of the ocean. Don't piss us off again.
Agreed.
I just couldn't resist commenting. Your post gave me a hilarious metal image of hundreds of CoS drones wandering around a CoS campus somewhere, each mindlessly pumping away at their standard-issue CoS brand penis pump as they go about their business.
Slashdot Mirror
Dammit ... I wasn't fast enough.
releases tend to be when it's "good enough"
That's the point. Vista wasn't good enough. It may be true technically that software is never perfect, but your rationalization ignores the fact that there is a large range of quality in that imperfect gray area. A lot of software that hits the market is pretty good, with a few well-hidden bugs that need to be patched, and a range of upcoming features that its architects are planning to roll out over time.
There's a huge difference between that degree of market-readiness, and the turd called Vista.
Linux is still a work in progress
Yes. And when you download a linux distro, you know darn well whether you are getting a stable release, a beta release, or wild-crazy-brand-spankin-new alpha code. Vista was beta code in disguise, forcibly poured down the throats of > 90% of PC buyers. (Not the first time they pulled that little trick, either. Think WinME, first edition of '98, etc.)
When *isn't* something that is still "alive" and used a work in progress?
Ideally, the remaining work should be fine-tuning. Vista isn't yet to that stage, IMHO. People I know who make their living deploying and maintaining MS OS'es agree.
First outside the cuprate family? Maybe according to TFA, but what about the recently discussed silicon-hydrogen compounds and aluminum (or aluminium, if you prefer) nanoclusters which are a) outside the cuprate family, and b) showing superconductivity at elevated temperatures?
I read the summary and TFA, and still wasn't impressed. Sure, the new iron/arsenic compounds are somewhat interesting, and maybe in the future holds some kind of potential. It just seems a little underwhelming, since other research is producing compunds that seem more promising.
RAIB? (Redundant Array of Independent Birds?) A quad-core 3ghz pigeon?
My electric bill will be down to nothing, but the birdseed bill will be killing me.
Didn't we just see a far warmer superconductor just a little while ago?
Not to mention this one operating at 200 kelvin.
I feel kind of bad for these guys doing their research and coming in 150 kelvin behind everyone else.
Meh ... maybe. I think it is a different kind of survival tactic. Admins only tend to stay employed when they don't screw up. Working on a platform you aren't familiar with increases your chance of screwing something up. (new admin on a Solaris box: Hey, what does this 'sys-unconfig' command do? [enter]... )
/.
The need to be error-free (hah, right!) breeds a dislike for the unfamiliar, and an affinity for systems that work EXACTLY as you expect.
Implementing change in an admin's world is often an unintentionally hostile thing. In my experience, change is often inflicted by uninformed people (the CEO!) for dubious business reasons (he had a great round of golf with the sales guy!). I'm not surprised that change is often greeted with suspicion and hostility.
Admins spend much time and effort acquiring the knowledge that makes them valuable. Change the platforms they work on, and you reduce the knowledge base they offer, and hence their (perceived) value.
Plus, it's fun having religious wars over who has the best hardware/os/app/db/etc. It gives admins a reason to comment on
So if we were crazy enough to bring up Penicillium roqueforti, (say as a composting agent for its ability to break down complex organic molecules in a bio-recycling process) and it got loose on the moon, you're saying that there's a chance it could survive and gradually spread over the whole surface?
Do you know what this means? We would FINALLY have a moon that really is made of green cheese! (Well, bleu-green, anyway.)
Mod parent down: Ignorant. From here:
... THIS ... IS ... NEWSWORTHY.
... wait for it, wait for it ... ) Microsoft. How the hell are they going to encourage 3rd party developers to clean up their act when they can't even build good code in house?
David Cross, a product unit manager at Microsoft, was the group program manager in charge of designing User Account Control (UAC).
There. Credentials established. He was in charge of designing it.
"The reason we put UAC into the (Vista) platform was to annoy users--I'm serious".
There. Intent established.
The media aren't exploiting it. They are reporting it. When the company with world's dominant desktop OS and dominant desktop productivity suite puts a group program manager on stage at a public event with press in attendance, and he specifically reveals that the reason for particular piece of so-called security software is to "annoy users"
The Gump quote is the only piece of your post I agree with. Speaking of which, his comments included the following stats:
- 80% of the warnings were generated by 10 apps
- Some undisclosed number of those 10 apps were from (
More:
- 66% of sessions now run without prompts. (means chance of annoying prompt = 34%)
- 88% of users have not turned off UAC. (means 12% are so fed up they switch it off)
- 7% of UAC permission dialog boxes get a "No" click. (means that 93% of sheeple^h^h^h^h^h users automagically click "Yes". Alternate explanation: Those 7% are too afraid of "Yes", and click "No" by default.)
Lucid, interesting, and well written.
Question about the robots in your sig: Are they iPhone killers in the sense that they stalk, attack, and kill iPhones? Or do they use iPhones to kill people? Kill people with iPhones?
Either way, please send two to my hotel room as soon as they are built.
All the excuses I make for my poor choices are based on cognitive dissonance! What the hell am I supposed to do now?
Egad! You may be right.
IANAL, but here is the Mass law re: licensing a firm in the business of doing investigation work.
"Whoever violates any provision of this section shall be punished by a fine of not less than two hundred nor more than one thousand dollars or by imprisonment for not more than one year, or by both such fine and imprisonment."
Here is the law about impersonating a police officer.
"... shall be punished by a fine of not more than four hundred dollars or by imprisonment for not more than one year."
The investigators might actually have it easier if they posed as officers. Hell, the fines imposed are a lot less than the RIAA would like to hit the John Does with for far slighter infractions. No indication if either qualifies as a felony, though.
This is a civil matter, not a criminal one.
Laws only apply to little people like you, not big people like us. Now give us some money and go away. Sincerely, RIAA
Hey, psst. Can you get me the phone numbers for those sluts? Thanks.
... big numbers sound impressive, but they don't necessarily mean better performance.
I think the author's point was that to squeeze more performance out of a piece of silicon, you can either crank the speed up or pile on more cores + threads. IBM is (mostly) doing the former, while Sun is (primarily) doing the latter.
IBM's multithreading is a baby step compared to Sun's T1 and T2. Power6 = 2 cores*2threads each. T1 = (4 to 8 cores)*4 threads each. T2 = (up to 8 cores)*8 threads each.
Of course, this simply underlines the fact that Sun and IBM are coming at this from 2 different directions. Sun set out to build a economical cpu for entry-level systems that would perform well with lots of everyday tasks (emphasis on low cost of ownership, easily fit into any environment, server consolidation roles), with minimal need to optimise code. If your code supports multithreading on SPARC, you are good to go. Their higher-end systems will get more cores per processor in the future.
IBM started at the top end of the spectrum with a datacenter-centric platform, and will require heavier modification / optimisation of apps to see performance improvements. Their aim is to have screaming sequential execution in a big-iron box. They are planning to merge the technology down into more commoditized products in the future.
Each of the threads on a T2 only sucks up about 1.5 watts. Does anyone out there have any idea what the thermal envelope on a Power6 is? I haven't seen anything more accurate than educated guesses. Assuming 100 watts for a 4.7Ghz Power6 with 4 threads, you're looking at about 25 watts per thread. Okay if you want to run a few things at max speed (batch processing), bad if you want to run lots of threads (ie, server consolidation & virtualization).
A liquid cooling system is good for establishing uber-geek status among peers, but it's bad for cost of ownership, simplicity of operation, etc.
Overall, Sun offers way more threads at half the speed, while IBM gives you top-end clock speed. Ultimately, while the Power6 is terribly sexy with its ginormous clock speed and liquid-cooling system, you can get +/- similar performance out of less expensive hardware.
Never confuse processor speed with processing power.
Great. Let me know when fission reactors reach car-engine size. ...)
(Cue up plutonium-powered DeLorean jokes
No, really I get it, and I expect we'll see nuclear power expand, although we are limited in how fast we can bring new plants online. But what the hell are we going to do with all the damn waste? Yucca Mountain is stupid idea. Are we going to start reprocessing our nuclear waste like France? Move to a more progressive reactor design like MSRs, for which reprocessing is an integral part?
Personally, I like the MSR model for a number of reasons. It can burn old, nasty, long-half-life waste into less-nasty stuff, little/no chance of an out-of-control reaction, and relatively efficient burnup of the fuel it is fed.
----
Imitation is the sincerest form of mockery
Make the damn cell phone addicts pay extra to fly cell-class!
... if I operate a search engine based solely in the US, why would a EU visitor expect that I have to conform to EU regs?
AFAIK, Google's situation is different because it has a physical presence in Ireland and Belgium, and so they probably will have to eventually conform to the regulations in question. In principle, I like the EU personal data protection laws, but in general I believe that people need to protect themselves by controlling their own behavior rather than the actions of others.
[setenv RANT on; export RANT]
Want to protect your privacy? Read privacy policies on websites. You don't like Google's policies? Search elsewhere. Use EU-based search engines. Use search engines whose privacy statements conform to EU standards. Invent your own damn search engine that is based in the EU and conforms to EU laws. Who knows, maybe US surfers would prefer to use your engine in relative privacy rather than Google's. Maybe you'll be wildly successful at Google's expense.
[setenv JOKING on; export JOKING]
We already fought one war because of onnerous regulations from that side of the ocean. Don't piss us off again.
[setenv JOKING off; export JOKING]
[setenv RANT off; export RANT]
Agreed. I just couldn't resist commenting. Your post gave me a hilarious metal image of hundreds of CoS drones wandering around a CoS campus somewhere, each mindlessly pumping away at their standard-issue CoS brand penis pump as they go about their business.
Are you implying that Scientologists have small penises?