Re:makes windows marginally bearable
on
Cygwin 1.7 Released
·
· Score: 2, Interesting
No, SFU/SUA is not the same as Cygwin... it is an implementation of a POSIX(?) userland that sits directly on top of the NT kernel.
Also unlike Cygwin, it does not rely on the Win32 subsystem.
It comes with numerous GNU, System V, and BSD utilities. One of its design goals is to be "source compatible", so that in general only a recompile of Unix apps is required.
And I, like many others, would like to know when the release of all the actual tools for Windows 7 and Server 2008 R2 version of SUA will show up... it has now been over 2 months since general release of Windows 7. Sigh.
I just saw your post as I was finishing researching mine... and I certainly agree with you that the summary is wrong.
The Microsoft KB article is quite explicit that the workaround is what disables the sleep states, leading to higher power usage - the hotfix itself does not exhibit this problem.
... the mouse was invented by Doug Engelbart at Stanford Research Institute (later just SRI). And yes, it is true that Xerox PARC was a very early adopter of the technology.
While a citation for something this widely known and used shouldn't be necessary, here are 2:
I certainly won't argue against paying for software... I am a developer myself, and one needs to live on something.:)
As a matter of fact, I use VMware Server myself for most of my virtualization needs, as so far, it has offered the right mix of features and stability.
But I brought up the fact that you were actually talking about Workstation and that it is not free, because the overall discussion is about the free (for personal use etc) VirtualBox product.
"... obligation to the shareholders... to maximise the value of the company."
I keep seeing this meme repeated - without any citation to any civil or criminal code or statute.
Nonetheless, even if we accept it more or less at face value, there clearly is a difference between a company valuation based solely on next quarter's (or next week's) profits, and one based on planning and actions that may not bear fruit for months or (gasp) years.
At least some stockholders understand this, and at the least, a given company's leaders have an "obligation" to understand what their shareholders expect/demand as to which type of corporate valuation model they should consider.
Sadly, if their shareholders insist on throwing away the company's future for possible extremely short-sighted^H^H^H^H^H^H^Hterm gains, then they may have to sadly do just that.
I really don't know what you are talking about... I *chose* Vista x64 SP1, and have never looked back.
To address your specific points:
I am happily running VMware Server 2.0 with a 64-bit VMX.
No comment on Google desktop, I don't use it.
I have used Rocketdock on my Vista x64 SP1 installation since the first day - I wouldn't want to be without it! Yes, I know Punk Labs (creators of Rocketdock) says that it is "not supported"... nonetheless, I can say WFM.
When you remember your "other programs", I will comment on those if I have any direct knowledge of their status WRT x64.
"... All such orders will be reported to our entire customer base."
Ummm... dream on about this part (at least), as "Patriot Act"-backed demands (with or without a warrant) can forbid the disclosure of said demand.
And while an especially conscientious service provider might insist on dotting i's and crossing t's, it is doubtful any of their personnel (or bosses) will be willing to be jailed as a "terrorist".:(
"VMWare Server, the free edition only emulates a single processor environment for your virtualized host."
Actually, this is not correct. Server does have "experimental" support for supporting a virtual dual-core on a real dual-core... but only on a real CPU with VT (or maybe AMD's Pacifica, but I am not sure about this).
However, the point is still correct - it is costly to virtualize (and possibly even more so with virtualizing multi-core).
Sigh... didn't you even read the SA article you reference?
The part (the first sentence, actually) with the
'John Romero's long anticipated, infamous "Daikatana"...'?
The first two are widely regarded as being groundbreaking and classic, while the third may indeed not show up on the same lists - but then Daikatana was not a Carmack project.
I know this is Slashdot, and not reading the FA is somewhat expected, but your own "evidence"?:)
Not anyone outside the US... and not anyone *inside* - at least until they are required by law to "patriotically" only buy US-made networking gear.
It would have been nice if they had learned *something* from the years of the crypto export restrictions - stuff without the restrictions / backdoors / etc will be made somewhere, and will be purchased and used...
All this crap does is kill the viability in the global marketplace of products from US networking gear manufacturers. Sigh.
Performance evals with pre-release software can be problematic in general... and in the case of the current VMware "Server" pre-releases, you just need to look at their support forums to see that a) a number of users complain about disk performance, and b) that the product's DEBUG settings are forcibly in the ON state, which (it is claimed) is causing skewed performance results.
And then there is the notorious laptop disk [non]performance...
I saw the article and immediately said to myself "didn't this used to be called astroturfing?"
Then I noticed this comment and decided to add to it for those unfamiliar with the term...
http://en.wikipedia.org/wiki/Astroturfing
Hmmm... actually, I have built 10-12 systems since the mid-90's, and every one of them has used a PC Power & Cooling Silencer PSU!
New trend? I have never built a system for home-office use (where there is minimal if any "masking" noise) without carefully selecting the components (including PCP&C case fans) with an eye - OK, ear - to the noise output and temperature characteristics of the whole assembly.
Of course, this has kept me away from the monster CPU, CPU fans, and PSUs so popular these days.:)
I think my 1.4 GHz Pentium III-S is sweet! As long as you don't really need the high memory bandwidth usually found in say, a 3 GHz Pentium 4 (no, I'm not encoding movie-length material with MPEG-4), it kicks reasonable butt - and it's quiet, including the twin glass-platter IBM 60GXPs.
Who knows about the future? Longer term, I hope we move more towards asynchronously clocked systems, which solve a lot of power/heat/emmisions problems... closer than that, IBM's new PPC 970 (Apple "G5") represents a much more elegant approach than that of "let's just crank on a few more GHz!" If we get more of this kind of engineering, we won't need monster power supplies and all of those fans!:)
Now if we can just start seeing some high-performance/low-power GPUs...
Actually, the mouse didn't come from either Xerox [PARC] or Apple... that credt is usually given to Doug Engelbart, while he was at SRI.
At a finer grain, it was Engelbart's idea and sketches that were the beginnings of what came to be known as the "mouse", but an engineer named Bill English that actually followed through and made it happen (from the history here).
Slashdot Mirror
No, SFU/SUA is not the same as Cygwin... it is an implementation of a POSIX(?) userland that sits directly on top of the NT kernel. Also unlike Cygwin, it does not rely on the Win32 subsystem. It comes with numerous GNU, System V, and BSD utilities. One of its design goals is to be "source compatible", so that in general only a recompile of Unix apps is required. And I, like many others, would like to know when the release of all the actual tools for Windows 7 and Server 2008 R2 version of SUA will show up... it has now been over 2 months since general release of Windows 7. Sigh.
Nuke it from orbit... it's the only way to be sure. :)
I just saw your post as I was finishing researching mine... and I certainly agree with you that the summary is wrong.
The Microsoft KB article is quite explicit that the workaround is what disables the sleep states, leading to higher power usage - the hotfix itself does not exhibit this problem.
... the mouse was invented by Doug Engelbart at Stanford Research Institute (later just SRI). And yes, it is true that Xerox PARC was a very early adopter of the technology.
While a citation for something this widely known and used shouldn't be necessary, here are 2:
http://en.wikipedia.org/wiki/Douglas_Engelbart
http://en.wikipedia.org/wiki/Computer_mouse
I certainly won't argue against paying for software... I am a developer myself, and one needs to live on something. :)
As a matter of fact, I use VMware Server myself for most of my virtualization needs, as so far, it has offered the right mix of features and stability.
But I brought up the fact that you were actually talking about Workstation and that it is not free, because the overall discussion is about the free (for personal use etc) VirtualBox product.
Ummm... for those playing along at home, you are talking about the Workstation product - which as you note, costs money.
The Server product, which is free, does not support the more interesting graphics APIs.
"... obligation to the shareholders ... to maximise the value of the company."
I keep seeing this meme repeated - without any citation to any civil or criminal code or statute.
Nonetheless, even if we accept it more or less at face value, there clearly is a difference between a company valuation based solely on next quarter's (or next week's) profits, and one based on planning and actions that may not bear fruit for months or (gasp) years.
At least some stockholders understand this, and at the least, a given company's leaders have an "obligation" to understand what their shareholders expect/demand as to which type of corporate valuation model they should consider.
Sadly, if their shareholders insist on throwing away the company's future for possible extremely short-sighted^H^H^H^H^H^H^Hterm gains, then they may have to sadly do just that.
To address your specific points:
"... All such orders will be reported to our entire customer base."
:(
Ummm... dream on about this part (at least), as "Patriot Act"-backed demands (with or without a warrant) can forbid the disclosure of said demand.
And while an especially conscientious service provider might insist on dotting i's and crossing t's, it is doubtful any of their personnel (or bosses) will be willing to be jailed as a "terrorist".
"VMWare Server, the free edition only emulates a single processor environment for your virtualized host."
Actually, this is not correct. Server does have "experimental" support for supporting a virtual dual-core on a real dual-core... but only on a real CPU with VT (or maybe AMD's Pacifica, but I am not sure about this).
However, the point is still correct - it is costly to virtualize (and possibly even more so with virtualizing multi-core).
I know this is Slashdot, and not reading the FA is somewhat expected, but your own "evidence"? :)
Who is going to want to buy this stuff?
Not anyone outside the US... and not anyone *inside* - at least until they are required by law to "patriotically" only buy US-made networking gear.
It would have been nice if they had learned *something* from the years of the crypto export restrictions - stuff without the restrictions / backdoors / etc will be made somewhere, and will be purchased and used...
All this crap does is kill the viability in the global marketplace of products from US networking gear manufacturers. Sigh.
Performance evals with pre-release software can be problematic in general... and in the case of the current VMware "Server" pre-releases, you just need to look at their support forums to see that a) a number of users complain about disk performance, and b) that the product's DEBUG settings are forcibly in the ON state, which (it is claimed) is causing skewed performance results.
And then there is the notorious laptop disk [non]performance...
I saw the article and immediately said to myself "didn't this used to be called astroturfing?" Then I noticed this comment and decided to add to it for those unfamiliar with the term... http://en.wikipedia.org/wiki/Astroturfing
Actually, they can't... hence the interest in multi-core and the Pentium M! :)
New trend? I have never built a system for home-office use (where there is minimal if any "masking" noise) without carefully selecting the components (including PCP&C case fans) with an eye - OK, ear - to the noise output and temperature characteristics of the whole assembly.
Of course, this has kept me away from the monster CPU, CPU fans, and PSUs so popular these days. :)
I think my 1.4 GHz Pentium III-S is sweet! As long as you don't really need the high memory bandwidth usually found in say, a 3 GHz Pentium 4 (no, I'm not encoding movie-length material with MPEG-4), it kicks reasonable butt - and it's quiet, including the twin glass-platter IBM 60GXPs.
Who knows about the future? Longer term, I hope we move more towards asynchronously clocked systems, which solve a lot of power/heat/emmisions problems... closer than that, IBM's new PPC 970 (Apple "G5") represents a much more elegant approach than that of "let's just crank on a few more GHz!" If we get more of this kind of engineering, we won't need monster power supplies and all of those fans! :)
Now if we can just start seeing some high-performance/low-power GPUs...
At a finer grain, it was Engelbart's idea and sketches that were the beginnings of what came to be known as the "mouse", but an engineer named Bill English that actually followed through and made it happen (from the history here).