A couple of years ago - when Acorn was still Acorn (well, actually Olivetti) - a representative hinted to open sourcing the OS in case they'd go bust. At that time the downward spiral was pretty evident, and he didn't seem to think the monetary value of the OS' IP was worth much anyway due to their small market share.
That's not the way things went, unfortunately. Far from it. In my opinion the IP has been the hostage victim of a number of quarreling dinosaurs - RISC OS Ltd. and Pace being the most prominent parties. For obvious budget reasons nothing worth mentioning has evolved on the RISCOS front for the last decade, while a lot of disapointed developers and users have turned to open source alternatives.
Free RISCOS now! Open *all* sources - and let's hope it's not too late...
It has a fantastic UI and was in its day a supurb platform [..] but it no longer suits my needs and the OS architecture is frankly antiquated despite various moves in the past to drag it up into the modern world (notably some attempts to introduce pre-emptive multitasking and the Hydra multiprocessor boards).
Couldn't agree more. My father is a RISCOS devotee as well. Recently bought an Iyonic and actually *uses* it as well on a dayly basis. Personally I think even Microsoft Windows has moved beyond any competitive advantage RISCOS has or might have had. Although IMHO archimedes and especially the ARM processor had great design value; I just can't imagine using ARM-based hardware for running RISCOS instead of Linux.
To me, RISCOS died when I found out RISCOS3 yet again depended on cooperative multitasking. (I guess, that's about the same time the Hydra thing turned out to be vapourware and BeOS became hype). With intel processors quickly catching up on ARM it became evident that only radical changes could keep Archimedes on rails. But the uninspiring A5000 lacked soft and hardware features to compete in any way with Apple or even Microsoft (even in early nineties).
I remember speaking to an Acorn representative a few years later, probing him as to whether their business was still viable. He told me they were still planning pre-emptive scheduling and multiprocessors. (Not a word about 26-bit issues.) I asked about opensourcing RISCOS in case they'd go bust. He frowned but admitted the'd played with the idea....If only they had done so - I guess it could have been a viable complement to Linux which IMHO lacked a decent GUI until at least 2000. The rest of the OS -though a bit entangled with the hardware - had great design quality as well. (snif)
As noted in other reply, the reason for Solaris' small set of supported platforms is not primarily technical.
However, your point does illustrate that "the lack of abstraction" in Linux isn't that much of an issue in portablity. Moreover, I think the paragraph exaggerates the whole issue. First, I guess all mentioned solaris modules *are* to some extent platform specific (eg. some #ifdef to handle 32/64 bit ints on x86/UtraSparc code are inevitable; and I'd expect to find some assembly in most of the modules anyway). Second, Linux has some logic separated in clear modules, even though they are part of a "monolithic" kernel (eg. swapfs can be used as kernel module; embedded Linux kernels may not provide paging or vm at all.)
It comes down to fundamental issues on OS design. The article presents the clean & layered design as a portability advantage versus the quick & dirty design as a speed advantage. It's just a basic view that's been teached for years.
But wrt. portability in this case it turns out to be not that much of an issue in practice. I would not be surprised either if in practice Solaris turns out to be competitive in latency too, even though the books teach differently...
That's a bit easy, isn't it? The support for my view is statistics, but you haven't (yet) presented a single point that relies on relevant empirical evidence. I just tried to point out that crime statistics should be taken with a grain of salt; that not to say there is no basis.
OK, back on topic. The problem is that in the division between "those who believe they won't get caught, and the rest", the majority of people beleive they don't get caught. This is not typical for criminals but for all people (consider that most criminals - especially internet related - don't even consider themselves a criminal until they get caught).
This is essentially the reason why higher punishments have limited effect. Even if a speeding ticket could bankrupt you for life, many people would break speeding laws. Primarily, because they think they won't get caught or cause an accident and second because most people don't get caught. A negative side effect of stiffer punishment would be that people will do *anything* not to get caught. It works the same for "real" criminals, hence the high number of deadly highspeed car chases in countries that still have a death penalty. That explains why your line is controversial:
Obviously, stiffer sentencing will deter those who don't believe 100% that they will get away with it
Conversely, if what you were saying is just a fact, all crime would have been solved eons ago. (I mean, our current societies are not the ones with highest punishments, historically.) As - apparently - crime still exists that would mean either that the only criminals left are the ones that think they don't get caught or that your phrase is just not true. In both cases stiffer punishment will not significantly influence crime.
As for you last paragraph: Yes, these people should be punished. Not to prevent others from doing it, but just because they did something immoral and incidentally broke a few laws (not really incidentally, its what the laws are designed for). So I hope that once convicted, they get a fair punishment. Fair does not mean it's just a joke and it does not mean chopping of heads or hands. By fair I just mean reasonable considering what has actually happened.
>PS. Whilst I don't agree with cutting people's hands off. It IS a very effect way of reducing crime, that has been proven to work.
No it is not. Do you mean "proven", as in a mathematical proof? Cutting off a hand (or both hands) may be effective in preventing a specific criminal from stealing but it has no significant effect on "crime" whatsoever. There's more than enough statistical proof about that. (Although statistics - crime statistics in particular - of course lie)
The (current) GPL actually has a phrase about future versions of the GPL. It comes down to this: you may choose to stick with the old version (GPL2) as long as you like, but you can't fall back to an older version (GPL1) than the one used for the code you have changed.
In other words: the problem would not be that companies like Google have to cease using their current linux codebase & modifications, but it does mean they loose support for projects that choose to upgrade to GPL3.
So the question is: will all projects upgrade to GPL3 (likely, eventually) and - if so - will companies like google have enough momentum to do all maintenance on GPL2 legacy *without* support from the FOSS movement?
Will they just cheat, and how long will it take for FOSS people to find out and take action? Or will they simply release all changes they made and try to sift proprietry stuff from FOSS and play "nice" on that only?
-- And remember: Nothing is illegal if one hundred businessmen decide to do it -- Andrew Young
Somehow you seem to think that there is an intrinsic danger in living sub-sea level area's. As though flooding is intrincically an unmeasurably higher danger than those in areas with eartquakes, forrest fires, or than being hit by a meteor. As though it's a danger that can't be circumvented by proper management, like building dams & canals and maintaining those.
Actually, living sub sea level has proven to be pretty safe in other countries, but it doesn't work if people would rather have their tax wasted on less life-threatening dangers, like a war in Iraq.
-- There's a fine line between courage and foolishness - Too bad it's not a fence
I suppose not having any moving parts in your laptop has its merits: - No noise (vs. low noise) - More tolerant to physical stress - Longer batery life due to less power consumption (while reading, no spin-up/spin down latency)
Eventually, solid state will replace mechanical storage in anything portable, and demands for higher capacity will obviously remain.
-- The reason computer chips are so small is computers don't eat much
>There can't be anyone on/. who ENJOY Zonk's postings.
Hey, it's just your opinion; you shouldn't try to force that onto others.
>All that needs to be done is reign him in and school him in some basics and he could be OK.
I completely agree there. Zonk incidentally has posted one or two interesting stories, which is why I haven't blocked him from my homepage altogether... But teaching him some basics - and while they're at it give all "editors" an update - seems like a great idea! I mean: one or two dupes a day is just a poor sign that they don't eat their own stuff.
Now here is something Slashcode can help with. From the UnZonk-HOWTO:
From/. main page click "Preferences" (top left), then click the "Homepage" tab. You'll end up somewehere like here. At right hand side you see a list of authors. Uncheck the bottom one (called "Zonk"). Then scroll down and press "Save".
That's it: you've fired Zonk from your personal/. page!
It appears your angle is not against DRM, but rather against Microsoft DRM. Ie., DReaM Sun's open DRM initiative should be okay from your point of view.
As a sincere, open question I'd like to ask you: do you believe Open Source can (or cannot) coexist or even cooperate with DRMed media?
I responded that, perhaps it was coincidence but most of the programmers I knew with certs weren't any good and most developers who were good didn't bother with certs. The interviewer grinned and responded "oh, it's no coincidence..."
I guess that's the only right answer. I could even imagine popping a certificate related question just to test an applicant's view on the topic...
A friend of mine used to lecture MS certification courses - it paid his bills. In the meantime he was trying to setup a small software company. My guess was that in his position as a teacher he would have no problem recruiting a bunch of high-profile developers from his classes. But on the contrary he told me that in the hundreds of faces he'd seen, there were maybe two or three that he'd considered "bright". For the rest of the guys (and occasional gals) he estimated that it would cost him more to have them in his company than they could ever deliver.
Hence, he choose to be in the teaching business, which - the/. crowd agrees - is quite lucrative!
Certification makes you a tradable good. Hence: - Certification get you beyond HR/headhunter/unsavvy manager - Certification gets you a job that requires that certificate
And sure, you may pick-up some info that may be interesting, or even useful. The main question is, however, why would you want to learn about / get into a market where *THOUSANDS* of souls could replace you? I understand the benefits for a company, but isn't the fun in *any* job in the stuff that's on the fringe of what you learned; on the edge of what you can do?
Personally, I'm wary of jobs that require some certificate because more often than not they really expect you to do the kind of mind bogging coding that you learned in the mind bogging certification course. I'm much more interested in companies that take the time to go through my resume to see that I apt for the job. If they need labels to judge me by, then why would I want to work there?
But, for example, it doesn't really hurt Microsoft all that much if a bunch of "Paper MCSEs" don't get a job as a result of their certification. They have other huge venues and it's not like it's MS' job to get you a job. They could stop offering certs tommorow and be fine.
Certification has never been profitable to MS on its own. Rather, MSCEs are likely to find themselves a job as developer for the windows platform, which means one less soul working for the competition (ie, Apple, FOSS, or previously OS/2, Borland, Sun, Zortech,..). The only way to get qualified programmers to doing a mind bogging certification is if you can convince them there's a job waiting.
Fortunately, certification has always attracted the less talented (in hope of leveraging either their skills or their credentials) while leaving the higher "caste" to just prove themselves through their CV.
With bevarages the actual test is relatively simple: its simply a matter of taste.
Comparing software will take a few more minutes. Agreeing even over what should be tested is an open ended question. Having an independent study is one thing, but interpreting it and emphasizing what is "relavant" is another.
So sure, this is an oportunity for OSDL to get some exposure, but Microsoft has a headstart with billions of marketing dollars to spend on spinning the story the way they like it.
-- The path of least resistence is steady deterioration: without constant effort/energy keeping information free, it becomes not free, as thermodynamics clearly st ates
The article points to Magdalena Urbanska and Thomas Urbanski's original research paper which "reveal" its valididy through a "mathematical proof". (8 pages of formulas, so it must be true) Of course, anyone with a post 1990's knowledge of the thing called "internet" would know that mentioning the existence of a research paper has been replaced by the thing called a "hyper link".
Googling, I found little more than this link to ARF, an unknown organization boldly calling itself "The Research Authority" (with capitals, mind you).
No paper (at least not online) and no references to the institute or organization that Magdalena Urbanska and Thomas Urbanski might be affiliated with. Their daily lifes seem to be spent as "researchers" - whatever that may mean.
It looks like a hoax. A sad hoax. Because why would anyone want to hoax a story this sad?
Tough call - DRM is coming (Or is already here), one way or another, and is better to work on creating something done right, or to object to it on moral grounds?
Helping on the creation of an open DRM platform is more likely to make it unbreakable than keeping it proprietary. Also, on open source, free source, copyright, IP, fair use, etc. issues, supporting DRM is a clear sign that above all these you value the ad-hoc naturel of EULA as a legit tool for regulation of information use. IOW, you don't have any objections on content providers to decided how much you should pay, how and when you should use it, and who is allowed to do so. Therefore, I think DReaM is not the way to go if you're having second thoughts on DRM.
As long as no law forces you to adopt DRM and meanwhile breaking DRM protection is consisdered (mostly) illegal, my advice is: don't buy DRMed stuff. And that leaves fewer and fewer options...
POWER, V9 SPARC, Alpha and PA-RISC are all (more or less) RISC processors, which have very simple instruction decoding and relatively small pipelines (compared to contemporary x86 implementations). Therefore, emulating them in a VLIW processors ADDS a lot of the translation overhead, while little is gained from removing minor decoding and branch prediction units.
Roughly: if you can build a (multi) core that outperforms (in raw MIPS) any of the current POWER, V9 SPARC, Alpha or PA-RISC implementations by a factor of five to ten at the same number of Watts, you may indeed stand a good chance to be able to emulate them at a speed that would persuade customers to switch. Since RISC processors, however, are already (relatively) lean-and-mean, you'll never succee.
And indeed, as both Crusoe and Itanium efforts have show, emulation on VLIW isn't all that performant in absolute numbers.
You seem to know your stuff pretty well (or at least, a lot better than I do), while also your opponent seems to have a good point in that it's theoretically possible. Summarizing, you need rotating joints and stronger beams than currently avaialble, you may be able to relieve some of the strain from the beams by moving some equipment to the rotational center, but on the rotors, nevertheless. (Theoretically, you could do away with the stable spaceship altogether by simply distributing it over the two rotor centres). I'd like to believe this is all very simple, cheap and light, but that probably isn't true. The lack of seriuous funding for these kind of tech is mentioned in half of the posts on this topic. So my question is: How much would it cost to do some serious research? In actual dollars. Could a privately owned company like Virgin Galactic ever come up with that kind of money, or will space tourism end a few footsteps from where it's currently at?
In another scenario the entire crew running (at evenly distributed locations) will increase or decrease the rotational speed, hence increasing/decreasing "gravitation".
Also, the influence of personel movement on rotation speed and center point would depend on height relative to the centre. Climbing up and down to the core (and as/decending at the other side) should be noticeable, but even standing/lying down has minor effects that might accumulate.
In all business cases, supporting two competing, parallel systems is going to cost you more, regardless of what they are.
I think the confusing is with the word "you". If referring to a single company/entity as with competing teams within IBM, (relative) cost is higher then going for a single solution (though quality may be greater too, depending to project size and budget).
The sentence can also be read as referring to the end user, who in terms of price will probably benefit from competing parallel systems. Or at least, so say the rules of free market.
In principle, having a monopoly is most efficient - at least from the monopolies point of view. State monopolies are at the base of communism, corporate monopolies are what made Ford and AT&T great. Still, monopolies just don't seem to last and in the mean time they have a fierce stiffling effect on innovation.
Hence, in macro economic terms, competing parallel systems are commonly valued as a good thing.
Hmm. sorry. It was a joke hinting at the loyalty of Scottish police officers. They'd never rotate to non-police jobs and would *never* work on English soil.
Hmm. It may not be funny afterall. sorry
-- If you explain so clearly that nobody can misunderstand, somebody will.
Slashdot Mirror
Open source RISCOS (or what's left of it)
A couple of years ago - when Acorn was still Acorn (well, actually Olivetti) - a representative hinted to open sourcing the OS in case they'd go bust. At that time the downward spiral was pretty evident, and he didn't seem to think the monetary value of the OS' IP was worth much anyway due to their small market share.
That's not the way things went, unfortunately. Far from it. In my opinion the IP has been the hostage victim of a number of quarreling dinosaurs - RISC OS Ltd. and Pace being the most prominent parties. For obvious budget reasons nothing worth mentioning has evolved on the RISCOS front for the last decade, while a lot of disapointed developers and users have turned
to open source alternatives.
Free RISCOS now! Open *all* sources - and let's hope it's not too late...
It has a fantastic UI and was in its day a supurb platform [..] but it no longer suits my needs and the OS architecture is frankly antiquated despite various moves in the past to drag it up into the modern world (notably some attempts to introduce pre-emptive multitasking and the Hydra multiprocessor boards).
...If only they had done so - I guess it could have been a viable complement to Linux which IMHO lacked a decent GUI until at least 2000. The rest of the OS -though a bit entangled with the hardware - had great design quality as well. (snif)
Couldn't agree more. My father is a RISCOS devotee as well. Recently bought an Iyonic and actually *uses* it as well on a dayly basis. Personally I think even Microsoft Windows has moved beyond any competitive advantage RISCOS has or might have had. Although IMHO archimedes and especially the ARM processor had great design value; I just can't imagine using ARM-based hardware for running RISCOS instead of Linux.
To me, RISCOS died when I found out RISCOS3 yet again depended on cooperative multitasking. (I guess, that's about the same time the Hydra thing turned out to be vapourware and BeOS became hype). With intel processors quickly catching up on ARM it became evident that only radical changes could keep Archimedes on rails. But the uninspiring A5000 lacked soft and hardware features to compete in any way with Apple or even Microsoft (even in early nineties).
I remember speaking to an Acorn representative a few years later, probing him as to whether their business was still viable. He told me they were still planning pre-emptive scheduling and multiprocessors. (Not a word about 26-bit issues.) I asked about opensourcing RISCOS in case they'd go bust. He frowned but admitted the'd played with the idea.
As noted in other reply, the reason for Solaris' small set of supported platforms is not primarily technical.
However, your point does illustrate that "the lack of abstraction" in Linux isn't that much of an issue in portablity. Moreover, I think the paragraph exaggerates the whole issue. First, I guess all mentioned solaris modules *are* to some extent platform specific (eg. some #ifdef to handle 32/64 bit ints on x86/UtraSparc code are inevitable; and I'd expect to find some assembly in most of the modules anyway). Second, Linux has some logic separated in clear modules, even though they are part of a "monolithic" kernel (eg. swapfs can be used as kernel module; embedded Linux kernels may not provide paging or vm at all.)
It comes down to fundamental issues on OS design. The article presents the clean & layered design as a portability advantage versus the quick & dirty design as a speed advantage. It's just a basic view that's been teached for years.
But wrt. portability in this case it turns out to be not that much of an issue in practice. I would not be surprised either if in practice Solaris turns out to be competitive in latency too, even though the books teach differently...
I.e. what you have just said has no basis.
That's a bit easy, isn't it? The support for my view is statistics, but you haven't (yet) presented a single point that relies on relevant empirical evidence. I just tried to point out that crime statistics should be taken with a grain of salt; that not to say there is no basis.
OK, back on topic. The problem is that in the division between "those who believe they won't get caught, and the rest", the majority of people beleive they don't get caught. This is not typical for criminals but for all people (consider that most criminals - especially internet related - don't even consider themselves a criminal until they get caught).
This is essentially the reason why higher punishments have limited effect. Even if a speeding ticket could bankrupt you for life, many people would break speeding laws. Primarily, because they think they won't get caught or cause an accident and second because most people don't get caught. A negative side effect of stiffer punishment would be that people will do *anything* not to get caught. It works the same for "real" criminals, hence the high number of deadly highspeed car chases in countries that still have a death penalty. That explains why your line is controversial:
Obviously, stiffer sentencing will deter those who don't believe 100% that they will get away with it
Conversely, if what you were saying is just a fact, all crime would have been solved eons ago. (I mean, our current societies are not the ones with highest punishments, historically.) As - apparently - crime still exists that would mean either that the only criminals left are the ones that think they don't get caught or that your phrase is just not true. In both cases stiffer punishment will not significantly influence crime.
As for you last paragraph: Yes, these people should be punished. Not to prevent others from doing it, but just because they did something immoral and incidentally broke a few laws (not really incidentally, its what the laws are designed for). So I hope that once convicted, they get a fair punishment. Fair does not mean it's just a joke and it does not mean chopping of heads or hands. By fair I just mean reasonable considering what has actually happened.
>PS. Whilst I don't agree with cutting people's hands off. It IS a very effect way of reducing crime, that has been proven to work.
No it is not. Do you mean "proven", as in a mathematical proof? Cutting off a hand (or both hands) may be effective in preventing a specific criminal from stealing but it has no significant effect on "crime" whatsoever. There's more than enough statistical proof about that. (Although statistics - crime statistics in particular - of course lie)
The (current) GPL actually has a phrase about future versions of the GPL. It comes down to this: you may choose to stick with the old version (GPL2) as long as you like, but you can't fall back to an older version (GPL1) than the one used for the code you have changed.
In other words: the problem would not be that companies like Google have to cease using their current linux codebase & modifications, but it does mean they loose support for projects that choose to upgrade to GPL3.
So the question is: will all projects upgrade to GPL3 (likely, eventually) and - if so - will companies like google have enough momentum to do all maintenance on GPL2 legacy *without* support from the FOSS movement?
Will they just cheat, and how long will it take for FOSS people to find out and take action? Or will they simply release all changes they made and try to sift proprietry stuff from FOSS and play "nice" on that only?
--
And remember: Nothing is illegal if one hundred businessmen decide to do it -- Andrew Young
> anyone and their family pet can create and/or render PDF. e.g. GhostScript.
That would be xpdf, gpdf, pdftex, pdflatex or pdfcreator.
GhostScript renders PostScript (ps, not pdf), which is an "open format" as well.
Somehow you seem to think that there is an intrinsic danger in living sub-sea level area's. As though flooding is intrincically an unmeasurably higher danger than those in areas with eartquakes, forrest fires, or than being hit by a meteor. As though it's a danger that can't be circumvented by proper management, like building dams & canals and maintaining those.
Actually, living sub sea level has proven to be pretty safe in other countries, but it doesn't work if people would rather have their tax wasted on less life-threatening dangers, like a war in Iraq.
--
There's a fine line between courage and foolishness - Too bad it's not a fence
I suppose not having any moving parts in your laptop has its merits:
- No noise (vs. low noise)
- More tolerant to physical stress
- Longer batery life due to less power consumption (while reading, no spin-up/spin down latency)
Eventually, solid state will replace mechanical storage in anything portable, and demands for higher capacity will obviously remain.
--
The reason computer chips are so small is computers don't eat much
>There can't be anyone on /. who ENJOY Zonk's postings.
Hey, it's just your opinion; you shouldn't try to force that onto others.
>All that needs to be done is reign him in and school him in some basics and he could be OK.
I completely agree there. Zonk incidentally has posted one or two interesting stories, which is why I haven't blocked him from my homepage altogether... But teaching him some basics - and while they're at it give all "editors" an update - seems like a great idea! I mean: one or two dupes a day is just a poor sign that they don't eat their own stuff.
Now here is something Slashcode can help with. From the UnZonk-HOWTO:
/. main page click "Preferences" (top left), then click the "Homepage" tab. You'll end up somewehere like here. At right hand side you see a list of authors. Uncheck the bottom one (called "Zonk"). Then scroll down and press "Save".
/. page!
From
That's it: you've fired Zonk from your personal
Glad to have been of assistance.
So are you saying the FEMA should be aquited?
It appears your angle is not against DRM, but rather against Microsoft DRM. Ie., DReaM Sun's open DRM initiative should be okay from your point of view.
As a sincere, open question I'd like to ask you: do you believe Open Source can (or cannot) coexist or even cooperate with DRMed media?
I responded that, perhaps it was coincidence but most of the programmers I knew with certs weren't any good and most developers who were good didn't bother with certs. The interviewer grinned and responded "oh, it's no coincidence..."
/. crowd agrees - is quite lucrative!
I guess that's the only right answer. I could even imagine popping a certificate related question just to test an applicant's view on the topic...
A friend of mine used to lecture MS certification courses - it paid his bills. In the meantime he was trying to setup a small software company. My guess was that in his position as a teacher he would have no problem recruiting a bunch of high-profile developers from his classes. But on the contrary he told me that in the hundreds of faces he'd seen, there were maybe two or three that he'd considered "bright". For the rest of the guys (and occasional gals) he estimated that it would cost him more to have them in his company than they could ever deliver.
Hence, he choose to be in the teaching business, which - the
Certification makes you a tradable good. Hence:
- Certification get you beyond HR/headhunter/unsavvy manager
- Certification gets you a job that requires that certificate
And sure, you may pick-up some info that may be interesting, or even useful. The main question is, however, why would you want to learn about / get into a market where *THOUSANDS* of souls could replace you? I understand the benefits for a company, but isn't the fun in *any* job in the stuff that's on the fringe of what you learned; on the edge of what you can do?
Personally, I'm wary of jobs that require some certificate because more often than not they really expect you to do the kind of mind bogging coding that you learned in the mind bogging certification course. I'm much more interested in companies that take the time to go through my resume to see that I apt for the job. If they need labels to judge me by, then why would I want to work there?
But, for example, it doesn't really hurt Microsoft all that much if a bunch of "Paper MCSEs" don't get a job as a result of their certification. They have other huge venues and it's not like it's MS' job to get you a job. They could stop offering certs tommorow and be fine.
Certification has never been profitable to MS on its own. Rather, MSCEs are likely to find themselves a job as developer for the windows platform, which means one less soul working for the competition (ie, Apple, FOSS, or previously OS/2, Borland, Sun, Zortech,..). The only way to get qualified programmers to doing a mind bogging certification is if you can convince them there's a job waiting.
Fortunately, certification has always attracted the less talented (in hope of leveraging either their skills or their credentials) while leaving the higher "caste" to just prove themselves through their CV.
With bevarages the actual test is relatively simple: its simply a matter of taste.
Comparing software will take a few more minutes. Agreeing even over what should be tested is an open ended question. Having an independent study is one thing, but interpreting it and emphasizing what is "relavant" is another.
So sure, this is an oportunity for OSDL to get some exposure, but Microsoft has a headstart with billions of marketing dollars to spend on spinning the story the way they like it.
--
The path of least resistence is steady deterioration: without constant effort/energy keeping information free, it becomes not free, as thermodynamics clearly st ates
The article points to Magdalena Urbanska and Thomas Urbanski's original research paper which "reveal" its valididy through a "mathematical proof". (8 pages of formulas, so it must be true) Of course, anyone with a post 1990's knowledge of the thing called "internet" would know that mentioning the existence of a research paper has been replaced by the thing called a "hyper link".
Googling, I found little more than this link to ARF, an unknown organization boldly calling itself "The Research Authority" (with capitals, mind you).
No paper (at least not online) and no references to the institute or organization that Magdalena Urbanska and Thomas Urbanski might be affiliated with. Their daily lifes seem to be spent as "researchers" - whatever that may mean.
It looks like a hoax. A sad hoax. Because why would anyone want to hoax a story this sad?
Tough call - DRM is coming (Or is already here), one way or another, and is better to work on creating something done right, or to object to it on moral grounds?
Helping on the creation of an open DRM platform is more likely to make it unbreakable than keeping it proprietary. Also, on open source, free source, copyright, IP, fair use, etc. issues, supporting DRM is a clear sign that above all these you value the ad-hoc naturel of EULA as a legit tool for regulation of information use. IOW, you don't have any objections on content providers to decided how much you should pay, how and when you should use it, and who is allowed to do so. Therefore, I think DReaM is not the way to go if you're having second thoughts on DRM.
As long as no law forces you to adopt DRM and meanwhile breaking DRM protection is consisdered (mostly) illegal, my advice is: don't buy DRMed stuff. And that leaves fewer and fewer options...
POWER, V9 SPARC, Alpha and PA-RISC are all (more or less) RISC processors, which have very simple instruction decoding and relatively small pipelines (compared to contemporary x86 implementations). Therefore, emulating them in a VLIW processors ADDS a lot of the translation overhead, while little is gained from removing minor decoding and branch prediction units.
Roughly: if you can build a (multi) core that outperforms (in raw MIPS) any of the current POWER, V9 SPARC, Alpha or PA-RISC implementations by a factor of five to ten at the same number of Watts, you may indeed stand a good chance to be able to emulate them at a speed that would persuade customers to switch. Since RISC processors, however, are already (relatively) lean-and-mean, you'll never succee.
And indeed, as both Crusoe and Itanium efforts have show, emulation on VLIW isn't all that performant in absolute numbers.
You seem to know your stuff pretty well (or at least, a lot better than I do), while also your opponent seems to have a good point in that it's theoretically possible.
Summarizing, you need rotating joints and stronger beams than currently avaialble, you may be able to relieve some of the strain from the beams by moving some equipment to the rotational center, but on the rotors, nevertheless. (Theoretically, you could do away with the stable spaceship altogether by simply distributing it over the two rotor centres).
I'd like to believe this is all very simple, cheap and light, but that probably isn't true. The lack of seriuous funding for these kind of tech is mentioned in half of the posts on this topic. So my question is: How much would it cost to do some serious research? In actual dollars. Could a privately owned company like Virgin Galactic ever come up with that kind of money, or will space tourism end a few footsteps from where it's currently at?
In another scenario the entire crew running (at evenly distributed locations) will increase or decrease the rotational speed, hence increasing/decreasing "gravitation".
Also, the influence of personel movement on rotation speed and center point would depend on height relative to the centre. Climbing up and down to the core (and as/decending at the other side) should be noticeable, but even standing/lying down has minor effects that might accumulate.
--
There's no future in time travel
It's just the same news posted by a different /. editor!
In all business cases, supporting two competing, parallel systems is going to cost you more, regardless of what they are.
I think the confusing is with the word "you". If referring to a single company/entity as with competing teams within IBM, (relative) cost is higher then going for a single solution (though quality may be greater too, depending to project size and budget).
The sentence can also be read as referring to the end user, who in terms of price will probably benefit from competing parallel systems. Or at least, so say the rules of free market.
In principle, having a monopoly is most efficient - at least from the monopolies point of view. State monopolies are at the base of communism, corporate monopolies are what made Ford and AT&T great. Still, monopolies just don't seem to last and in the mean time they have a fierce stiffling effect on innovation.
Hence, in macro economic terms, competing parallel systems are commonly valued as a good thing.
Hmm. sorry.
It was a joke hinting at the loyalty of Scottish police officers. They'd never rotate to non-police jobs and would *never* work on English soil.
Hmm. It may not be funny afterall. sorry
--
If you explain so clearly that nobody can misunderstand, somebody
will.