M$ owns and controls and can enforce copyright on M$ Office, its source,
even if only to kill (abandon) it...
OoO is free, current source is in thousands of places, and, at the very least it secures legacy.doc files, they __CAN__ be read and re-formatted
Re:Itanic, was ... deja vue, DEC
on
Oracle Buys Sun
·
· Score: 1
I really despair at some of the comment comming from the USA, it is absolutely irrational.
Simplistically, when the desktop was seen as king WIntel was seen as the way, then people realised they still needed servers, and HP+Intel decided to start over, using an EPIC design.
The implementation was a goat-fuck, late, performance and delivery a disaster, personally I dont care for EPIC, but it did not work... in fact.
The old DEC Alpha would, in my view, have been a better base, but that is not how the industry went. So what?
86_64 is morphing to RISC with the 4.2 ISA extensions though the 86 screws compilers to some extent, but it is, for now the commercial winner.
Your very long comment is unfortunately valueless, and completely misses the point, __AND__ is yet further evidence of the lack of education in the USA. I do not have either the time, nor inclination to dissect your arguments in detail.
Pluri-potent cells, unlike differentiated cells, have the ability to morph into the needed tissue type, differentiated cells, alone cannot effectively repair injury or organ failure, stem cells, embryonic or not can. Autonomic (self generated, Pluri-potent cells) are to be preferred, since they cannot, per se, generate rejection. BUT genetic defects need new, working genes to provide the factory that the defective gene implies the lack of.
To do the research you need the diversity, so Bush's decision was doubly idiotic, in limiting the stem cell lines to a small number of already contaminated cytotypes. Having been a grad student at the beginning of the IT age, in the 60s, let me tell you that most VC funding comes when someone else has made money from what you need to do. Biotechnology is in that phase now.
This has been almost 10 years overdue, when Scott McNealy started behaving like Ken Olsen, c. 1978-80, SUN was doomed, and makes the "No Computer Company makes two full generations" law up there with Moore's and Bell's laws.
The get to have a death wish, and wont adapt to market conditions:
Use VMS/Bliss not Unix and C, Unix is Snake Oil
s/VMS/Solaris/g
It is soo sad, and in some ways the better the product they the worse the delusional thinking is. If HP/Intel had got the Itanium right this would have been over 10 years ago.
The other sad thing is how the aging Solaris sysadmins still insist that Lintel is less reliable than SPARC+Solaris. As one who worked extensively with the Solaris core kernel let me tell you the Linux code is far better than its Solaris counterpart. As some Intel hardware vendors, HP, Dell & IBM were forced to 86_64 when the Itanic sunk, one got similar high quality engineering for servers that SUN did, hot-swap, ECC ram carefully designed boards with diagnostic capability, good ground plane, equalised clock distribution, quality thermal design... as SUN.
It is so easy to get blind sided by prejudice, I remember the first DEC ethernet controller, for the Unibus, with AMD2900 bit-slice and loadable u-code, wonderfull engineering, but it drew 7A of 5V, and you often needed to install an additional backplane and PSU to cope with the heat and +5V drain. Madness!
Please read my original comment about paying attention in school, the point is simply that lots of different embryos give you lots of different genes, which is what you need to screen for cause.
Once you have found the right gene you can prevent the cell from differentiating and mass produce in vitro.
A case of a nail, 2 4 8... you have what you need, so long as you dont loose the cell line but micro-bioligists understand that.
This is precisely what I meant about truely dumb-ass comments from lack of education, even if you need a human cell at all, since often it is easier to gene splice into another organism that is easier and quicker to grow.
The detail depends on what you are trying to do. Eg type 1 diabetes is often caused by destruction, by disease of the necessary cells, you want to replace those cells, same nerve damage; type 2 diabetes have the working cells, but faulty biochemistry and may need a drug not new cells, or a vectored gene implant, in either case you wont use whole cells, and the stem cell may not be used at all once the _process_ is understood.
The real problem is that the US has fostered the notion that any opinion, no matter how stupid, and especially if religiously based must be given a balanced hearing. Dumb is dumb! Corrupt is corrupt!
Pandering to 'your base' on cable is dishonest, and no way to build a successful society. You need to re-learn that quick and run your creationalists, intelligent-d-twats, and most TV pastors out on the rail. Else, in free fall you are soon a third world country.
I continue to be appalled by the bigoted and histrionic comments of religious Americans who do not seem to understand that not even all Christians agree with them let alone the rest of us, and then Squabble endlessly over exactly what Bush's disastrous decision in Health was, again never mind his contributions to Foreign Affairs or the Economy. He set health research back eight years while presiding over un-necessary wars and the de-regulation of the financial system which has resulted in the greatest depression in four generations and the rise of more crooks, fraudsters and scam agents to shake a stick at.
For the record I, and most outside the US, do not care whether the stem cells are embryonic or not, so long as the medics have the genetic material, and that means diversity of genes, to search for cure to debiliting illnesses especially as there is a surplus of fetuses for other reasons.
Restoring pluri-potence to the patients own cells may well be desirable but is not essential as various gene splicing and gene injection techniques generate neither cancer or rejection as some comments, clearly FUD from the US rabid right suggested.
What is clearly necessary is better scientific education and that is better done by paying attention in school rather than church.
You note that an Australian politician is so entranced by "Global Warming", insisting that the ice is melting when his own scientists contradict this view.
Truly Environmentalism is becoming a new religon, and needs to be treated similarly, separated from government.
Solaris is(NT) more stable, on Intel it is junky, BUT the point is this type of comment is so DEC, 15 years ago, but SUN was the descendent of DEC, and much the same arrogance is there.
The sad thing is they died some time ago, and the world is just noticing.
It is very easy to program round this restriction, but you do need kernel awareness which AFAIK is in.30 which will be with us as STABLE 30.1+ in 10-16 weeks. Distributions 6-12 months.
The trick is to keep a smallish pool of blocks in your back pocket, and a table, usually only normally read, of where the wearing blocks are, if you use two blocks/wearing block and keep, in the first block a count of how many writes happened to the second block, you can re-vector critical writes to get as big a MTTCW (mean time to cant write) as big as you need. For example, considering just the Superblock, with 5 M back pocket space you multipy the MTTCW by 5000.
The skill range is actually infinity, as there are people who just cant program.
Of those that can 100:1 is normal
The US is going to have to grow up, and get educated again, or other places really will eat their lunch. Fortunately Obama seems to realise this, but neither Mainstreet or the rest of the world will put up with the MBA/Wall street culture again.
If you listened to the G20 proceedings you will realise just how surely the game is up
No, No, The anti-virus and security upgrade treadmill is a farce, as long as Active-X and extension based execution are pervasive Win X has no security, if the Security Policy is modernised then there is massive application level incompatibility, the only way to run Win-x securely is to virtualize it under Linux.
Linux is as usable/more than anything made by M$, especially if pre-configured by your hardware vendor or by your Enterprise IT department, and everything works, out of the box. If you want to play own Systems Integrator, then you need to know what you are doing and your job will be infinitely harder if you use cheap, rip-off hardware that came out last week, and even more so if its USB and its USB Man:Dev code has been changed to 'protect the innocent', a frequent asian trick.
If you want painless Google is your friend, but even that is very conservative, eg I have been using Brother printers for years, and using their raster utility, and my own Perl Fax/Printer driver it takes 5 minutes to get a new model up.
In these days of virtualization you can run, say XP, in a linux virtual and keep all your data safe on the Linux side. You will be very unlucky/careless if your v Win is ever compromised, but if it is you re-image it from a CD in 5 minutes. You install more Win software re-cut the CD from the Linux side, no Ghost, no hassle. Finally YOU have control of what your Win system can do, eg no phone home. This is how I use my private Linux laptop in the enterprise and that way you can Outlook PIM functions, read corporate mail... , all without intruding on the enterprise host network management, you can even do C#/.Net development on either side, and cross test and you can ensure that your web pages render under IE/Firefox/Opera/Konqueror...
Critics, real and imaginary, and Priorities
on
Linux Needs Critics
·
· Score: 1
In many ways thus guy has it exactly backward, we see far too much re-inventing the wheel when we have already a perfectly working one, but this is only serious because of its prevalence, rather than the tendancy itself.
Sometimes, re-engineering what already exists is very beneficial, especially when it is part of the core toolkit itself and this is particularly true in the kernel and in the core tools arena, binutils (gold) and the gcc C++ effort by Ian Taylor. The simple range of Linux has reached the point where Windows, with all M$ cash, cannot compete and this is becoming increasingly obvious in the server, embedded and netbook spaces, and we certainly have not seen the end of this trend given the downturn and an emerging appliance marketplace.
That said we do have some tech-culture problems which have not been adaquately addressed by the Linux Businesses and the Distributions, and these either are or will be serious hinderances to linux adoption.
The twin lack of (a) Exchange/Outlook compatibility, which often mandates the running of (a vitualised copy) of Windows on Linux desktops to effectively collaberate in the Enterprise and ActiveDirectory/SingleSignOn, and (b) C# and.Net. In both these areas we do _NOT_ want to be too late, and allow the Word-Macro, Excell,... type lock in.
Since we are now clearly nearing the tipping point of OS choice by merit not marketing it is important not to drop this ball, and timelyness is of the essence here so we do not have the luxury of having good solutions mature and emerge slowly here.
One other are where we must be careful is the "lets re-write it" because it isn't written in C++, Ruby... for example and this is one of the our most common failure modes.
Another common failure mode is to adopt over complex frameworks, and then force the solution to meet the framework needs, an example is the recent redesign of freshmeat.net, which has migrated a simple convenient solution to a form-over-function nightmare.
1/ Measuring "speed" of a human interaction limited UI editor is simply daft, and the idea that you should reboot ever few minutes indicates a Win mentality, re-boot often since the OS memory leaks. Next since the scripting is unlike real behavior.
2./ What matters is how good the layout algorithms are, and what functionallity you get. I use vanilla OOO all the time, on my laptop for all personal and business purposes, even though I can/do run M$ Word under Wine or virtualised in an XP environment and all are fast enough for all realistic practical purposes, at least on a dual core with 2GB.
Subjectively OOO is easier to use, and less likely to cause idiots to miss-use its functionality, but much more important, with improved and now almost bug free font management, the output looks good. For a publisher/word-processor that is good. If I want it to look beautiful, I use La(Tex), OOO is acceptable and if, never, I dont mind ugly then M$ Word.
Exactly like IE, Word is a hostage to M$ marketing, so they waste time with ribbons in the UI rather than making the layout right.
You can just tell what typeset stuff, if you have reasonable eyesight, and an understanding of typography, you will easibly recognise TeX, OOO, and M$-Word output and I leave the rest to your own judgement.
As has already been explained, Non-Sequential thinking is hard, you postulate double speed, BUT the producer thread, the app finished and handed of the buffer to the OS to send to the GPU, and you say it threads this. Well fine, so the threaded part can run on another core, but then hardware DMAs the data and waits for a GPU interrupt/done-queue ack so how does this speed things up on multicore. Not at all, someone has to set up the DMA and wait, not run, while it completes, so unless all cores are at 100% you have saved nothing, and created additional overhead spawning a new thread
The real problems with exploiting parallelism are (a) a solution is needed since Moor's law has run into a brick wall, excepting major process improvements that the semi-manufacturers dont see, and (b) all current algorithm design going right back to 1948, John von Neumann, has been essentially serial. Threads, and multi-cores are an essentially serial solution to a parallel problem.
Parallel computation, is very hard, see how many kernel (OS) developers we have v app. developers. This is because of problems related to timing and computational order. This produces problems with data sharing and correctness.
Then there is the problem space, in some problems, easily artificially constructed, then the next step depends on the completion of all earlier steps, and the solution cannot be parallelized eg Fibonnachi where you can show a trivial parallel decomposition that wins nothing. In other, and more interesting cases eg the Partial Differential Equations of Mathematical Physics, Routing, and Finite Element systems some to large amounts of parallelism may be possible, particularly will well thought out analysis that capitalizes on special features of the problem or the known solution, eg multimode solution and boundrey condition matching in Navier-Stokes or Elasticity.
The point is that this is at the algorithmic level, it is not about code optimization or other programming paradigms so the kool aid of we need a better tool chain or parallelizing compilers is hope, hand-waving and optimism and of course product support.
Dont get me wrong, I see this as a very good thing, seas of CPU (core = 1 CPU), will fully solve lots of problems, and improve robustness, but it will not help with many problems, 4-16 cores will be generally handy, but 1024.. 1048576 will need new algorithms, for the first time in 60 years of computing and 500 of Mathematics.
A last thought, when we get to ~1073741824 cores we may start to make progress with AI and need to worry about the Singularity.
It wasn't C, if it was BCPL when David Barron developed the Transputer, at Southampton I would be surprised, but the world turns and now we have the AMD Hyperchannel and Infini-x
I sometimes despair at what I read here, sin(x), cos(x) are both very smooth, ie infinitely differentiable periodic functions, so why would it surprise you that interpolation off a table, spacing determined to give desired accuracy would not be quicker than the Taylor series, and since the function is periodic the table size is bounded.
Try that with any _ROUGH_ non periodic function and see where you get
There is a balance between complexity, weight, and wide availability in tools, BASH is a shell, not a programming language, although you and write __simple__ programs in it In a Linux or modern Unix environment you also have Perl, which from the look of your sig will be right up your street, Python, PHP or Java as interpreted programming languages. But without extra cost you also have C C++ C# FORTRAN ADA LISP ML HASKAL...
My take is that Bash is almost too powerful since it seduces people into writing long and complex scripts when they should be writing a program, and all that entails eg it is hard to debug bash, but easy in Perl.
This is history in that HP, IBM and SUN resisted installing Perl by default, so only with the comming of Linux coud you assume Perl, now of course Perl is standard in OpenSolaris and the NIH syndrome is over.
The whole sh - awk - sed - test - eval stuff is nonsense, Perl, or for the more traditional programmer, Python is much cleaner.
Windows is uniformly slower and more un-reliable than Linux. That is when it has competant enterprise support.
It is inherrently insecure, due to Active X.
It is buggy, and you cant trust the M$ fixes, so support is a huge enterprise problem.
The lock-in applications are mostly un-auditable Excel/Word macros and crap proprietary VB code, which should have been banned under Sarbanes-Oxley already. This must STOP, All busnisses are responsible for their conduct even if it is the consequence of crock closed source code. When CEO' fear to go to jail the closed EULAs will stop and proper financial oversight will insist that, at least in major financial applications, all code and applications are to be auditable. The same rule must apply to any computer aided voting machinery.
Nothing runs fast on Itanium, it is the biggest architectural design mistake in a century of CPU design, even Babage did better, a triumph of marketing { what MBA's call strategic planning over reality}, however, the AMD 86_64 contains a perfectly good embedded RISC engine to which Intel has no essential rights.
If its native IS [of the AMD 86_64] were to be exposed would be both easily re-targetable, say USD 50 M for a commercial (quick, 3 month (eg Code Sourcery)) GCC back end. would re-target the silicon, much as Intel did for 86_64. In this model the large register file and all RISC gains would be manifest.
The normal (div 50) cross emulation downside performance can be cut since the data path and ALU logic is already 86 native. So the data-path and hi=bit management on both the 64/32 bit boundry is ok.
Since the downturn cannot but hurt Windows, already stressed in embedded and netbook marketplaces, this means that, if Intel prevails, the Wintel market place fragments. This will require unprecedented agility from Microsoft.
The real news is that this reinforces the failure of US patent policy and has created another closed US markets.
Not only did it fly quite high, it flew VERY Fast which made shooting it down a lot harder.
It is very hard to shoot down a Mach 5 aircraft with a Mach 4 missile or Mach 1.5 cannon, and also, remember with the zoom manouvre, you only get one try and survive only if you are a very good pilot since your re-entry is likely flamed out due to compressor stall and in an un-gentlemanly stance eg inverted flat spin.
So this is not as safe and easy as it sounds, the pilots need to know how to recover the airframe from any/kind of spin/stall, without engine power and then do an in flight engine restart, which may require a 90 deg. dive with a fixed geometry intake.
BTW you cant eject either till you get down to c. 32000 ft as the partial gas pressure will mean that you cant even breath pure oxygen as you wait 10+ mins to get into thicker air.
Slashdot Mirror
2 inch hemp rope, and a yard-arm and some guts from the ever stupid pols in the swamp by the Potomac.
M$ owns and controls and can enforce copyright on M$ Office, its source,
...
.doc files, they __CAN__ be read and re-formatted
even if only to kill (abandon) it
OoO is free, current source is in thousands of places, and, at the very least
it secures legacy
I really despair at some of the comment comming from the USA, it is absolutely irrational.
... in fact.
Simplistically, when the desktop was seen as king WIntel was seen as the way, then people realised they still needed servers, and HP+Intel decided to start over, using an EPIC design.
The implementation was a goat-fuck, late, performance and delivery a disaster, personally I dont care for EPIC, but it did not work
The old DEC Alpha would, in my view, have been a better base, but that is not how the industry went. So what?
86_64 is morphing to RISC with the 4.2 ISA extensions though the 86 screws compilers to some extent, but it is, for now the commercial winner.
Itanic is dead.
The next frontier is parallelism.
Your very long comment is unfortunately valueless, and completely misses the point, __AND__ is yet further evidence of the lack of education in the USA. I do not have either the time, nor inclination to dissect your arguments in detail.
Pluri-potent cells, unlike differentiated cells, have the ability to morph into the needed tissue type, differentiated cells, alone cannot effectively repair injury or organ failure, stem cells, embryonic or not can. Autonomic (self generated, Pluri-potent cells) are to be preferred, since they cannot, per se, generate rejection. BUT genetic defects need new, working genes to provide the factory that the defective gene implies the lack of.
To do the research you need the diversity, so Bush's decision was doubly idiotic, in limiting the stem cell lines to a small number of already contaminated cytotypes. Having been a grad student at the beginning of the IT age, in the 60s, let me tell you that most VC funding comes when someone else has made money from what you need to do. Biotechnology is in that phase now.
Put shortly, you are clueless.
This has been almost 10 years overdue, when Scott McNealy started behaving like Ken Olsen, c. 1978-80, SUN was doomed, and makes the "No Computer Company makes two full generations" law up there with Moore's and Bell's laws.
... as SUN.
The get to have a death wish, and wont adapt to market conditions:
Use VMS/Bliss not Unix and C, Unix is Snake Oil
s/VMS/Solaris/g
It is soo sad, and in some ways the better the product they the worse the delusional thinking is. If HP/Intel had got the Itanium right this would have been over 10 years ago.
The other sad thing is how the aging Solaris sysadmins still insist that Lintel is less reliable than SPARC+Solaris. As one who worked extensively with the Solaris core kernel let me tell you the Linux code is far better than its Solaris counterpart. As some Intel hardware vendors, HP, Dell & IBM were forced to 86_64 when the Itanic sunk, one got similar high quality engineering for servers that SUN did, hot-swap, ECC ram carefully designed boards with diagnostic capability, good ground plane, equalised clock distribution, quality thermal design
It is so easy to get blind sided by prejudice, I remember the first DEC ethernet controller, for the Unibus, with AMD2900 bit-slice and loadable u-code, wonderfull engineering, but it drew 7A of 5V, and you often needed to install an additional backplane and PSU to cope with the heat and +5V drain. Madness!
NO you dont,
... you have what you need, so long as you dont loose the cell line but micro-bioligists understand that.
Please read my original comment about paying attention in school, the point is simply that lots of different embryos give you lots of different genes, which is what you need to screen for cause.
Once you have found the right gene you can prevent the cell from differentiating and mass produce in vitro.
A case of a nail, 2 4 8
This is precisely what I meant about truely dumb-ass comments from lack of education, even if you need a human cell at all, since often it is easier to gene splice into another organism that is easier and quicker to grow.
The detail depends on what you are trying to do. Eg type 1 diabetes is often caused by destruction, by disease of the necessary cells, you want to replace those cells, same nerve damage; type 2 diabetes have the working cells, but faulty biochemistry and may need a drug not new cells, or a vectored gene implant, in either case you wont use whole cells, and the stem cell may not be used at all once the _process_ is understood.
The real problem is that the US has fostered the notion that any opinion, no matter how stupid, and especially if religiously based must be given a balanced hearing. Dumb is dumb! Corrupt is corrupt!
Pandering to 'your base' on cable is dishonest, and no way to build a successful society. You need to re-learn that quick and run your creationalists, intelligent-d-twats, and most TV pastors out on the rail. Else, in free fall you are soon a third world country.
I continue to be appalled by the bigoted and histrionic comments of religious Americans who do not seem to understand that not even all Christians agree with them let alone the rest of us, and then Squabble endlessly over exactly what Bush's disastrous decision in Health was, again never mind his contributions to Foreign Affairs or the Economy. He set health research back eight years while presiding over un-necessary wars and the de-regulation of the financial system which has resulted in the greatest depression in four generations and the rise of more crooks, fraudsters and scam agents to shake a stick at.
For the record I, and most outside the US, do not care whether the stem cells are embryonic or not, so long as the medics have the genetic material, and that means diversity of genes, to search for cure to debiliting illnesses especially as there is a surplus of fetuses for other reasons.
Restoring pluri-potence to the patients own cells may well be desirable but is not essential as various gene splicing and gene injection techniques generate neither cancer or rejection as some comments, clearly FUD from the US rabid right suggested.
What is clearly necessary is better scientific education and that is better done by paying attention in school rather than church.
Thank goodness Obama at least sounds rational.
You note that an Australian politician is so entranced by "Global Warming", insisting that the ice is melting when his own scientists contradict this view.
Truly Environmentalism is becoming a new religon, and needs to be treated similarly, separated from government.
Solaris is(NT) more stable, on Intel it is junky, BUT the point is this type of comment is so DEC, 15 years ago, but SUN was the descendent of DEC, and much the same arrogance is there.
The sad thing is they died some time ago, and the world is just noticing.
It is very easy to program round this restriction, but you do need kernel awareness which AFAIK is in .30 which will be with us as STABLE 30.1+ in 10-16 weeks. Distributions 6-12 months.
The trick is to keep a smallish pool of blocks in your back pocket, and a table, usually only normally read, of where the wearing blocks are, if you use two blocks/wearing block and keep, in the first block a count of how many writes happened to the second block, you can re-vector critical writes to get as big a MTTCW (mean time to cant write) as big as you need. For example, considering just the Superblock, with 5 M back pocket space you multipy the MTTCW by 5000.
The skill range is actually infinity, as there are people who just cant program.
Of those that can 100:1 is normal
The US is going to have to grow up, and get educated again, or other places really will eat their lunch. Fortunately Obama seems to realise this, but neither Mainstreet or the rest of the world will put up with the MBA/Wall street culture again.
If you listened to the G20 proceedings you will realise just how surely the game is up
No, No,
The anti-virus and security upgrade treadmill is a farce, as long as Active-X and extension based execution are pervasive Win X has no security, if the Security Policy is modernised then there is massive application level incompatibility, the only way to run Win-x securely is to virtualize it under Linux.
Linux is as usable/more than anything made by M$, especially if pre-configured by your hardware vendor or by your Enterprise IT department, and everything works, out of the box. If you want to play own Systems Integrator, then you need to know what you are doing and your job will be infinitely harder if you use cheap, rip-off hardware that came out last week, and even more so if its USB and its USB Man:Dev code has been changed to 'protect the innocent', a frequent asian trick.
... , all without intruding on the enterprise host network management, you can even do C#/.Net development on either side, and cross test and you can ensure that your web pages render under IE/Firefox/Opera/Konqueror ...
If you want painless Google is your friend, but even that is very conservative, eg I have been using Brother printers for years, and using their raster utility, and my own Perl Fax/Printer driver it takes 5 minutes to get a new model up.
In these days of virtualization you can run, say XP, in a linux virtual and keep all your data safe on the Linux side. You will be very unlucky/careless if your v Win is ever compromised, but if it is you re-image it from a CD in 5 minutes. You install more Win software re-cut the CD from the Linux side, no Ghost, no hassle. Finally YOU have control of what your Win system can do, eg no phone home. This is how I use my private Linux laptop in the enterprise and that way you can Outlook PIM functions, read corporate mail
In many ways thus guy has it exactly backward, we see far too much re-inventing the wheel when we have already a perfectly working one, but this is only serious because of its prevalence, rather than the tendancy itself.
.Net. In both these areas we do _NOT_ want to be too late, and allow the Word-Macro, Excell, ... type lock in.
... for example and this is one of the our most common failure modes.
Sometimes, re-engineering what already exists is very beneficial, especially when it is part of the core toolkit itself and this is particularly true in the kernel and in the core tools arena, binutils (gold) and the gcc C++ effort by Ian Taylor. The simple range of Linux has reached the point where Windows, with all M$ cash, cannot compete and this is becoming increasingly obvious in the server, embedded and netbook spaces, and we certainly have not seen the end of this trend given the downturn and an emerging appliance marketplace.
That said we do have some tech-culture problems which have not been adaquately addressed by the Linux Businesses and the Distributions, and these either are or will be serious hinderances to linux adoption.
The twin lack of (a) Exchange/Outlook compatibility, which often mandates the running of (a vitualised copy) of Windows on Linux desktops to effectively collaberate in the Enterprise and ActiveDirectory/SingleSignOn, and (b) C# and
Since we are now clearly nearing the tipping point of OS choice by merit not marketing it is important not to drop this ball, and timelyness is of the essence here so we do not have the luxury of having good solutions mature and emerge slowly here.
One other are where we must be careful is the "lets re-write it" because it isn't written in C++, Ruby
Another common failure mode is to adopt over complex frameworks, and then force the solution to meet the framework needs, an example is the recent redesign of freshmeat.net, which has migrated a simple convenient solution to a form-over-function nightmare.
1/
Measuring "speed" of a human interaction limited UI editor is simply daft, and the idea that you should reboot ever few minutes indicates a Win mentality, re-boot often since the OS memory leaks. Next since the scripting is unlike real behavior.
2./
What matters is how good the layout algorithms are, and what functionallity you get. I use vanilla OOO all the time, on my laptop for all personal and business purposes, even though I can/do run M$ Word under Wine or virtualised in an XP environment and all are fast enough for all realistic practical purposes, at least on a dual core with 2GB.
Subjectively OOO is easier to use, and less likely to cause idiots to miss-use its functionality, but much more important, with improved and now almost bug free font management, the output looks good. For a publisher/word-processor that is good. If I want it to look beautiful, I use La(Tex), OOO is acceptable and if, never, I dont mind ugly then M$ Word.
Exactly like IE, Word is a hostage to M$ marketing, so they waste time with ribbons in the UI rather than making the layout right.
You can just tell what typeset stuff, if you have reasonable eyesight, and an understanding of typography, you will easibly recognise TeX, OOO, and M$-Word output and I leave the rest to your own judgement.
As has already been explained, Non-Sequential thinking is hard, you postulate double speed, BUT the producer thread, the app finished and handed of the buffer to the OS to send to the GPU, and you say it threads this. Well fine, so the threaded part can run on another core, but then hardware DMAs the data and waits for a GPU interrupt/done-queue ack so how does this speed things up on multicore. Not at all, someone has to set up the DMA and wait, not run, while it completes, so unless all cores are at 100% you have saved nothing, and created additional overhead spawning a new thread
Duh, Marketing Departments
The real problems with exploiting parallelism are (a) a solution is needed since Moor's law has run into a brick wall, excepting major process improvements that the semi-manufacturers dont see, and (b) all current algorithm design going right back to 1948, John von Neumann, has been essentially serial. Threads, and multi-cores are an essentially serial solution to a parallel problem.
.. 1048576 will need new algorithms, for the first time in 60 years of computing and 500 of Mathematics.
Parallel computation, is very hard, see how many kernel (OS) developers we have v app. developers. This is because of problems related to timing and computational order. This produces problems with data sharing and correctness.
Then there is the problem space, in some problems, easily artificially constructed, then the next step depends on the completion of all earlier steps, and the solution cannot be parallelized eg Fibonnachi where you can show a trivial parallel decomposition that wins nothing. In other, and more interesting cases eg the Partial Differential Equations of Mathematical Physics, Routing, and Finite Element systems some to large amounts of parallelism may be possible, particularly will well thought out analysis that capitalizes on special features of the problem or the known solution, eg multimode solution and boundrey condition matching in Navier-Stokes or Elasticity.
The point is that this is at the algorithmic level, it is not about code optimization or other programming paradigms so the kool aid of we need a better tool chain or parallelizing compilers is hope, hand-waving and optimism and of course product support.
Dont get me wrong, I see this as a very good thing, seas of CPU (core = 1 CPU), will fully solve lots of problems, and improve robustness, but it will not help with many problems, 4-16 cores will be generally handy, but 1024
A last thought, when we get to ~1073741824 cores we may start to make progress with AI and need to worry about the Singularity.
It wasn't C, if it was BCPL when David Barron developed the Transputer, at Southampton I would be surprised, but the world turns and now we have the AMD Hyperchannel and Infini-x
I sometimes despair at what I read here, sin(x), cos(x) are both very smooth, ie infinitely differentiable periodic functions, so why would it surprise you that interpolation off a table, spacing determined to give desired accuracy would not be quicker than the Taylor series, and since the function is periodic the table size is bounded.
Try that with any _ROUGH_ non periodic function and see where you get
eg 1/(1-e^x)
There is a balance between complexity, weight, and wide availability in tools, BASH is a shell, not a programming language, although you and write __simple__ programs in it In a Linux or modern Unix environment you also have Perl, which from the look of your sig will be right up your street, Python, PHP or Java as interpreted programming languages. But without extra cost you also have C C++ C# FORTRAN ADA LISP ML HASKAL ...
My take is that Bash is almost too powerful since it seduces people into writing long and complex scripts when they should be writing a program, and all that entails eg it is hard to debug bash, but easy in Perl.
This is history in that HP, IBM and SUN resisted installing Perl by default, so only with the comming of Linux coud you assume Perl, now of course Perl is standard in OpenSolaris and the NIH syndrome is over.
The whole sh - awk - sed - test - eval stuff is nonsense, Perl, or for the more traditional programmer, Python is much cleaner.
Absolutely RIGHT, HEAR HEAR.
Windows is uniformly slower and more un-reliable than Linux. That is when it has competant enterprise support.
It is inherrently insecure, due to Active X.
It is buggy, and you cant trust the M$ fixes, so support is a huge enterprise problem.
The lock-in applications are mostly un-auditable Excel/Word macros and crap proprietary VB code, which should have been banned under Sarbanes-Oxley already. This must STOP, All busnisses are responsible for their conduct even if it is the consequence of crock closed source code. When CEO' fear to go to jail the closed EULAs will stop and proper financial oversight will insist that, at least in major financial applications, all code and applications are to be auditable. The same rule must apply to any computer aided voting machinery.
I notice the trend to re-engineer sites to be more complex and less useful
/. are awful
The parent is right, the UI pages on
and the new Freshmeat3 is worse.
Nothing runs fast on Itanium, it is the biggest architectural design mistake in a century of CPU design, even Babage did better, a triumph of marketing { what MBA's call strategic planning over reality}, however, the AMD 86_64 contains a perfectly good embedded RISC engine to which Intel has no essential rights.
If its native IS [of the AMD 86_64] were to be exposed would be both easily re-targetable, say USD 50 M for a commercial (quick, 3 month (eg Code Sourcery)) GCC back end. would re-target the silicon, much as Intel did for 86_64. In this model the large register file and all RISC gains would be manifest.
The normal (div 50) cross emulation downside performance can be cut since the data path and ALU logic is already 86 native. So the data-path and hi=bit management on both the 64/32 bit boundry is ok.
Since the downturn cannot but hurt Windows, already stressed in embedded and netbook marketplaces, this means that, if Intel prevails, the Wintel market place fragments. This will require unprecedented agility from Microsoft.
The real news is that this reinforces the failure of US patent policy and has created another closed US markets.
Not only did it fly quite high, it flew VERY Fast which made shooting it down a lot harder.
It is very hard to shoot down a Mach 5 aircraft with a Mach 4 missile or Mach 1.5 cannon, and also,
remember with the zoom manouvre, you only get one try and survive only if you are a very good pilot since your re-entry is likely flamed out due to compressor stall and in an un-gentlemanly stance eg inverted flat spin.
So this is not as safe and easy as it sounds, the pilots need to know how to recover the airframe from any/kind of spin/stall, without engine power and then do an in flight engine restart, which may require a 90 deg. dive with a fixed geometry intake.
BTW you cant eject either till you get down to c. 32000 ft as the partial gas pressure will mean that you cant even breath pure oxygen as you wait 10+ mins to get into thicker air.