Re:How to fix trade secrets
on
Apple vs Bloggers
·
· Score: 4, Insightful
The problem with (1) is that Apple is a *big* company. They have lots of people working for them, most of whom will know *something* about some secret project/gizmo/whatever.
Apple has a strong tradition of corporate secrecy, but what *exactly* can they do if some employee lets something slip after a few too many drinks, or for money, or for the "thrill" of revealing an exclusive ? I mean, it's not as though they're going to send 'steve@apple.com' an email the next morning confessing all, is it ?
From what I read earlier on digg, although he makes a big deal about the money Apple make/have in the bank, he's not being sued for cash or penalities, he's being sued to reveal his source, because it's the only way Apple can find and fire the guilty employee.
As for 'first amendment rights' being abused, I thought they were only rights that the government couldn't abuse, not that any limits on civil suits by corporations were imposed, but hell, I'm a bloody foreigner, so I'm probably wrong.
If that's the only reason you have a 100Mb line (to utilize 0.5MB of it), you are wasting roughly 95% of your money. However, I'm assuming there are numerous servers and dozens of users -- not exactly the typical home user's environment
Indeed, the files are large, and they're there for several thousand users in a closed group. The machine can easily saturate that bandwidth at intensely busy times (sometimes the photos are topical...)
With the possible exception of p2p (with very local peers), or being extremely close to a source of data to which you must have high speed access, the article is correct. If you are going over the backbone, you're bottle-necked. Other than bragging rights, what's the point of having 80% un-utilized bandwidth versus wasting 20%? It's like duct-taping a fire hose to the end of your garden hose and calling it faster.
But that's the point, you see, it's *not* bottlenecked by the backbone - which is why I reported my download speeds. There's no way I could get that speed on a 1.5Mbps network. So, basically, he's talking rubbish.
Now, (and this is more aimed at the rest of the comments than your good self) I realise (duh!) that I'm not your average 2.5 kids, suburban family-type with apposite networking requirements, but he didn't say anything about demographics. He just made a (wrong) blanket statement...
So, the COO of company A who provide a worse service than company B says that there's no service-level difference in practice. Well, he *would*, wouldn't he ? It's always worth remembering the wisdom of ages... "cui bono"
IMHO (and it's only a single datapoint) it's certainly worth it for me... I have servers located in the UK on a 100mbit link, and at least 80% of the time I can download at ~500 kBytes/sec (sometimes more) from there to San Jose (CA). Since I transfer large numbers of multi-megapixel images, it's important to me that I have a fast link.
So, basically, picture me blowing a loud raspberry at Mr. Stephenson, thumb on the end of my nose, and waggling my fingers at him. I'll take the Comcast service, thanks.
Oh, BTW, you can get HDTV down the same wire too:-)
Not sure what I think of this... On the one hand, if it's possible to save loadsamanny by automating non-critical jobs, then fair-enough, sounds cool. And the brits have something of a history in designing warships - presumably they'll not have forgotten too many of the important bits...
On the other hand, during a conflict, a carrier is a pretty juicy target, and one thing humans *are* good at in combat [apart from dying:( ] is being adaptible. It'd be a real shame if the plug fell out of the automated aircraft-landing computer because of a nearby explosion... Yes, I'm being facetious, but the point isn't. Machines can only perform within their limitations, and people frequently perform outside their normal potential when (a) their life depends on it, and (b) there's no other option...
So, as long as we don't go to war, it'll probably be excellent. If we do, I hope they've thought of the consequences...
The scary thing here is nothing to do with what 'BadAnalogyGuy' was posting, it's that you're too intellectually-challenged to recognise it as humour - it was in bad taste, malicious, and morally bankrupt, just like most good humour. I would point out that there is a difference between satire and genocide, and that a lot of the most lauded works of literature exhibit "the pen is mightier than the sword" characteristics.
Humour is a sophisticated weapon, no-one likes being the butt of a joke, and cracking jokes at these folks expense is one way of getting them to examine their beliefs in a social context. It may not be the most persuasive of options, but hell, we've *tried* reasoning with them... [sigh] the problem is that they insist on believing their 2000-year-old fairytale. Just because it's old, doesn't make it right.
Maybe in your experience, but not in mine. I used to have to use the Mac a lot (in a fairly controlled environment, 5 people in an office), and it rarely made it through the day. It seemed to have the knack of crashing just after I'd spent a couple of hours perfecting my latest drivel, or diagram, or whatever... To give you an idea, I ended up using LaTeX rather than the GUI apps on the Mac, simply because I couldn't use the Mac for any reasonable time...
I'm no windows apologist (take a peek at my posting history!) but at the time, and in the same environment, the win-98 machine held up a lot better. I can't remember if we had an NT machine at the same time as the MacOS-9 one (we ditched the macs), but I agree it was a quantum step from Win98 to WinNT.
I'm aware of the differences in the OS types - hell, I've even helped write a small OS before. My point was that pre-emptive is better, though more complex, and therefore incurs an extra overhead, I wasn't saying it was the alpha and omega...
The original poster was claiming OS9 was a better OS just because it was faster. My point is that you get what you pay for, and OSX gives a lot more back.
As an exmaple, try writing a program that spawns two copies of itself. The only reason most preemptive OSs won't stay on their knees is they have work arounds to prevent this particular problem (but didn't always have them)
Well, to be fair, *any* OS will choke on that one (through RAM, process limits, CPU time, whatever resource is limited first), unless it has workarounds to stop it - it's not as though co-operative succeeds where pre-emptive fails in that regard...
For all the talk about the speed of OS X, Apple has never addressed the most obvious issue: on a machine that can run either OS 9 or OS X, OS 9 is very much faster.
OS9 did a lot less than OSX, which is why it was faster. OSX is a *lot* more reliable. Examples:
OS9 didn't have pre-emptive multitasking, so one bug somewhere in one program could bring the system to its knees. I saw that happen far too many times...
OS9 didn't have memory protection, so if a pointer went outside the current app's address space, it could quite happily scribble random data all over another application.
... I could go on...
Before OSX, the mac had the reputation of the machine that crashed all the time. By comparison, Windows was actually pretty reliable (this was before all the spyware/malware/crap that affects it recently, remember). Linux was best, of course...
They took an OS written from the ground up in the early 80s to be graphical, and replaced it with an OS written the 70s to be textual, with the GUI glued on top of it
Now you're just displaying your ignorance
The mac UI isn't the same as most unix ones - it's not X.
Even if it were X, for "glued on top", you really need to use "seamlessly integrated". The 'everything is a file' mantra of unix design actually works really well for X.
The core of the OS is a micro-kernel message-passing system (mach), which was developed between 1985 and 1994
... etc....
And then even worse, the people who wrote Carbon, the MacOS backward-compatibility layer, had no idea how to write it to be fast - simple calls like HLock which used to be two instructions on the original 128K Mac are now thousands of cycles under OS X
newsflash:when you need to do more work because you're in a far-more-capable and complex environment, it can take more machine-instructions to perform the task. This is just griping - the world has moved on from buggy, insecure, crappy-old OS9. Move with it.
They didn't throw any babies away, they did what they needed to do (ditch the abortion that was OS9) and move onto a new platform which provided the security, flexibility, and reliability that any modern OS provides. A brave decision, under the circumstances, and one well-conceived and executed.
It's very simple. Every time someone comes up with "most apps are useless at multi-processing", it's always a windows app. Most Apple apps are already multi-threaded for the reasons I state.
It seems you can't point out a technical achievement (on either side of the fence) without some 'fanboy' accusation being levelled. [sigh]
If you look at the way most OSX apps are designed, it's easy to multi-thread them. Cocoa pretty much imposes a model/view/controller pattern, and when your model manipulation is separate from your UI, it's pretty simple to spawn a background thread to calculate long tasks, or adopt a divide & conquer approach.
The other nice thing they have is the Accelerate.framework - if you link against that, you automatically get the fastest possible approach to a lot of compute-intensive problems (irrespective of architecture), and they put effort into making them multi-CPU friendly.
Then there's xcode which automatically parallelises builds to the order of the number of CPUs you have. If you have more than one mac on your network, it'll use distcc to (seamlessly) distribute the compilation. I notice my new Mac Mini is significantly faster than my G5 at producing PPC code. Gcc is a cross-compiler, after all...
And, all the "base" libraries (Core Image, Core Video, Core Graphics etc.) are designed to be either (a) currently multi-cpu aware, or (b) upgradeable to being multi-cpu aware when development cycles become available.
You get a hell of a lot "for free" just by using the stuff they give away. This all came about because they had slower CPUs (G4's and G5's) but they had dual-proc systems. It made sense for them to write code that handled multi-cpu stuff well. I fully expect the competition to do the same now that dual-CPU is becoming mainstream in the intel world, as well as in the Apple one...
As France had recently sold Super Entendard aircraft and Exocet missiles to Argentina, when war broke out there was still a French team in Argentina helping to fit out the Exocets and aircraft for Argintinean use. The French team continued to assist the Argentines throughout the war, in spite of the NATO embargo and official French government policy. [2]... and Argentina did (after all) use Exocets to sink British ships. I don't necessarily blame them - we were at war with them, and killing them just as happily! But it caused a *lot* of anti-French feeling in the UK at the time.
Not that it matters either way. Can you tell me how to rig up logic gates in order to implement that function? I can, it was required in my CS program to take an introductory digital circuit design class that went all the way up to building whole CPUs
Well, to be fair, that sounds hard, but it's only ~200 lines of verilog for a simple 16-bit CPU with an ALU and no pipelining/RAM prefetch. Designing CPU's is a hobby of mine... (So, yes, I can tell you how to do it [grin]). Although I'm more interested in compression systems at the moment - I'm currently trying to build an HDTV H.264 digitiser/encoder in hardware to connect via firewire to my Mac Mini...
He had a typo in his example (which is why it'd have to be square), but I believe his point was to do with locality of reference - the second example would miss the cache almost every time (with most image-storage algorithms, anyway. Images are generally stored left-to-right,top-to-bottom rather than vice-versa).
From your DOJ article: "The higher volume of crime in the United States is due, at least in part,to the greater population size of the United States. A more meaningful comparison is between the crime rates of the two countries."
The UK is has a far higher density of population than the USA. There are ~65 million people in the UK, and ~320 million in the USA. 4x the people, 50x the area. A "more meaningful" comparison would take that into account.
An additional reason for the UK to have higher violent crime is that the victim often survives in the UK. Even if attacked with a knife. You are far more likely to end up dead in the USA, so the figures are artifically low when comparing the two.
The USA has just sustained a huge drop [warning: PDF] in murder rates/year (in 2002, the latest figures I could find, it was ~5 murders per 10,000 people per year). During the time period you linked to in the "higher violent crime rate", it was ~8 murders per 10,000 people per year. Or, put another way ~260,000 murders. In the UK in 2003/2004 there were 853. In the period you linked to, it was ~700.
853 x4 is nowhere near 160,000 (both recent figures). The UK is no panacea, but to paint it as more-violent than the USA is just plain wrong.
For the purpose of this essay, let's forget about whether Apple failed to live up to its own PR. In fact, let's ignore the PR strategy altogether and focus on one of the product announcements: the new Mac Minis.
... and in any event, what hype is it that we're talking about here ? All Apple did was issue invitations (with no details on what was about to be disclosed) to a bunch of people in the press. In no way, shape, or form is this "hype" (verb: to promote or publicize a product or idea intensively, often exaggerating its importance or benefits) - it was an invite to a product release (with no details of the product) for crying out loud!
As soon as the invitations hit the 'net, all sorts of rumours (note: these are *not* Apple-created) surface. Some people publicly projected their own desires onto the event, irrespective of how likely it is, and then these self-same people get all disappointed when their fantasy doesn't come true. These people need to (a) get out more, (b) have more sex, and (c) move on from the mental state of a five-year-old ("Me want", "Me want", "Me want"). [aside: note that (c), as applied to (b), is more likely after (a). Just a hint to get you started...]
The fault here lies solely, completely, and utterly with those who raise Apple on too high a pedestal. There's only so much cool stuff any one company can make (although I thought the new mini *was* pretty cool, personally).
See my reply above (vcl v2) and look on the linux for PS2 website for VCL.
VCL takes sequential code and splits it up into parallel code based on the constraints of the vector-units (each VU is dual-issue, with some restrictions). It'll re-order code, insert wait states, etc. Certainly it's a good start at auto-parallelisation of the code. It's supposed to do as well as a skilled engineer...
On the PS2, there are two vector units (vu0 and vu1), which are basically where all the grunt work is done - the mips chip is there for housekeeping and non-time-critical code. Each VU has 2 code-paths (the instruction word is 64-bit, and there are two 32-bit instructions in each word). There are limitations on what you can do in each of the two words simultaneously. Sony have a GUI tool (in their professional kit) which allows the programmer to write essentially sequential code, and have it take full advantage of the vector units. According to Sony, it performs as well as a skilled programmer.
For the linux kit, they only released vcl (a commandline version). It's a bit like a compiler-stage. It takes sequential assembly language for a single VU and re-orders code, inserts wait-states etc. Finally producing another assembly output which is optimised for the dual-issue nature of a VU.
It strikes me that optimising for constraints over 2 code paths in a single unit isn't too far a stretch from optimising for constraints over 8 code paths in 8 units. The differences are mainly to do with locality of reference. On a VU it was up to the programmer to DMA data into scratch-space RAM, and set flags as semaphores on operation. There's no real reason why a computer program can't do that - a basic approach would be to do it on a function-by-function approach, or use #pragma constraints in the code. There's no need to have the all-singing, all-dancing version of the optimiser as version 1...
So, I've got a Garmin Streetpilot 2610. Last year I drove from Mountain View (where I live) to Las Vegas, without any maps, purely by putting (I think 3) waypoints - ("yes, I do want to go through Death Valley and Yosemite" etc.), a start and an end destination.
I've done the trip before, but I wouldn't say I was familiar with it, and I tried to see different things this time anyway - we went somewhat out of our way to Monument Valley on the way back, 'cos that's fairly spectacular...
The thing performs perfectly - I use it a lot. The confidence it brings when you know it will work out the best route and take you that way is just great. No panic if (when) you miss a junction on a freeway because you were in the wrong lane. No problem if you miss a turn in a city you don't know. I can't praise the thing enough.
My parents came over to visit - never been to the US before. Because I had to work that week, they went on a trip down the coast, with the GPS guiding them all the way. 10 minutes instruction was all it took to get them started, and they were experts after a few days, putting their own waypoints in etc. You have to understand that my parents really *are* VCR-challenged...
I'd never try and put in directions when the car was moving though - just pull over, type the details, and drive on afterwards. News-flash: driving without looking at the road (no matter what gizmo is involved) is not a good idea...
Fitts' Law is of course valid, but it has limitations as well. On the original Macs (with what ? A 9" screen ?), you can see the point - it's an easy move from wherever you are on the screen to the top left corner - a definite win.
However, I'm currently sitting in front of 2 23" monitors (and I have 2 30" monitors at work!). That gives me a screen resolution of 3840x1200, and moving the mouse from the right-most screen to the top-left of the left-most screen is by no means a trivial operation. It would be far more convenient for me to move the pointer 3" up to the top of the window than 36" (I've just measured from where I'm typing) to the menubar.
So it's just taken me 3 move-the-mouse, lift-and-reposition-the-mouse-back-where-it-was cycles to get to the menu, and one movement to position the mouse over a link in the quick-links bar in Safari. The latter was far faster than the former, and easier too. It just gets worse with 2 30" monitors, of course...
I've lost count of the number of times I've moved the mouse *almost* to the menubar and clicked too soon, so the Finder menu appears. Damn! Now I have to move the mouse all the way back again to the application, click there, and try again. This never happens with embedded menus.
Fitts' law was useful, but certainly when you're using two displays, it's had its day.
Two solutions spring to mind - either place a menubar optionally on *both* screens, or allow apps (even if just newer cocoa ones) to bind their menubar to their windows (in which case the Finder could be substituted on the menu at the top of the screen). I used to think this would never happen, but now that we have two-button mice....
And your proof that miracles don't happen is... what exactly? That looks like an a priori assumption to me. It is impossible to do formal criticism of the New Testament documents and other corroborating literature without coming to the conclusion that Jesus, whatever people thought of him, was a miracle worker of some sort. The onus is on the sceptic to come up with an alternative explanation that holds water for more than 30 seconds
Er, no it's not. Just look at the number of gullible fools that treat a 2000-year-old fable as absolute truth today. Assuming gullibility is roughly equivalent today (ie: the null nypothesis - nothing has changed) it ought to be relatively easy to convince people of falsehoods. Today we regard magicians tricks as sleight-of-hand, but historically they had far more credibility.
Add in the fact that we have a local native preaching about a new kingdom (which would resound well in the ears of an oppressed populace) and his credibility wouldn't even have to be that high - people are always more willing to believe things they want to believe...
And then there's the fact that most of the stories were made up ^W^W scribed much later - well after the death of this man, even if he ever existed.
Organised religion (at least in the West) is simply a power structure where one (wo)man attempts to gain power over peers by promising them fantastic gains/terrible suffering after death. It uses a 2000-year-old ("well it *must* be true") book written by corrupt church "leaders" (have you ever *read* about early church leaders??) to cow the common people into doing as they are told. It's nothing more than that.
You are born You live You die That's it
As for your comment about sceptics having to come up with proof of non-miracles - nothing could be further from the truth... Extraordinary claims require extraordinary evidence, and none has ever been provided.
Slashdot Mirror
The problem with (1) is that Apple is a *big* company. They have lots of people working for them, most of whom will know *something* about some secret project/gizmo/whatever.
Apple has a strong tradition of corporate secrecy, but what *exactly* can they do if some employee lets something slip after a few too many drinks, or for money, or for the "thrill" of revealing an exclusive ? I mean, it's not as though they're going to send 'steve@apple.com' an email the next morning confessing all, is it ?
From what I read earlier on digg, although he makes a big deal about the money Apple make/have in the bank, he's not being sued for cash or penalities, he's being sued to reveal his source, because it's the only way Apple can find and fire the guilty employee.
As for 'first amendment rights' being abused, I thought they were only rights that the government couldn't abuse, not that any limits on civil suits by corporations were imposed, but hell, I'm a bloody foreigner, so I'm probably wrong.
Simon
But that's the point, you see, it's *not* bottlenecked by the backbone - which is why I reported my download speeds. There's no way I could get that speed on a 1.5Mbps network. So, basically, he's talking rubbish.
Now, (and this is more aimed at the rest of the comments than your good self) I realise (duh!) that I'm not your average 2.5 kids, suburban family-type with apposite networking requirements, but he didn't say anything about demographics. He just made a (wrong) blanket statement...
Simon.
HELL, YEAH, but only briefly before I scraped my eyes out of my skull with my fingernails.
Simon
So, the COO of company A who provide a worse service than company B says that there's no service-level difference in practice. Well, he *would*, wouldn't he ? It's always worth remembering the wisdom of ages... "cui bono"
IMHO (and it's only a single datapoint) it's certainly worth it for me... I have servers located in the UK on a 100mbit link, and at least 80% of the time I can download at ~500 kBytes/sec (sometimes more) from there to San Jose (CA). Since I transfer large numbers of multi-megapixel images, it's important to me that I have a fast link.
So, basically, picture me blowing a loud raspberry at Mr. Stephenson, thumb on the end of my nose, and waggling my fingers at him. I'll take the Comcast service, thanks.
Oh, BTW, you can get HDTV down the same wire too
Simon
Not sure what I think of this... On the one hand, if it's possible to save loadsamanny by automating non-critical jobs, then fair-enough, sounds cool. And the brits have something of a history in designing warships - presumably they'll not have forgotten too many of the important bits ...
:( ] is being adaptible. It'd be a real shame if the plug fell out of the automated aircraft-landing computer because of a nearby explosion ... Yes, I'm being facetious, but the point isn't. Machines can only perform within their limitations, and people frequently perform outside their normal potential when (a) their life depends on it, and (b) there's no other option...
On the other hand, during a conflict, a carrier is a pretty juicy target, and one thing humans *are* good at in combat [apart from dying
So, as long as we don't go to war, it'll probably be excellent. If we do, I hope they've thought of the consequences...
Simon
The scary thing here is nothing to do with what 'BadAnalogyGuy' was posting, it's that you're too intellectually-challenged to recognise it as humour - it was in bad taste, malicious, and morally bankrupt, just like most good humour. I would point out that there is a difference between satire and genocide, and that a lot of the most lauded works of literature exhibit "the pen is mightier than the sword" characteristics.
Humour is a sophisticated weapon, no-one likes being the butt of a joke, and cracking jokes at these folks expense is one way of getting them to examine their beliefs in a social context. It may not be the most persuasive of options, but hell, we've *tried* reasoning with them... [sigh] the problem is that they insist on believing their 2000-year-old fairytale. Just because it's old, doesn't make it right.
Simon.
That, sir, is an *excellent* riposte.
Excuse me while I start to 'fish' my tuna sandwich out of the gaps in my keyboard...
Simon.
Maybe in your experience, but not in mine. I used to have to use the Mac a lot (in a fairly controlled environment, 5 people in an office), and it rarely made it through the day. It seemed to have the knack of crashing just after I'd spent a couple of hours perfecting my latest drivel, or diagram, or whatever... To give you an idea, I ended up using LaTeX rather than the GUI apps on the Mac, simply because I couldn't use the Mac for any reasonable time...
I'm no windows apologist (take a peek at my posting history!) but at the time, and in the same environment, the win-98 machine held up a lot better. I can't remember if we had an NT machine at the same time as the MacOS-9 one (we ditched the macs), but I agree it was a quantum step from Win98 to WinNT.
Simon
I'm aware of the differences in the OS types - hell, I've even helped write a small OS before. My point was that pre-emptive is better, though more complex, and therefore incurs an extra overhead, I wasn't saying it was the alpha and omega...
The original poster was claiming OS9 was a better OS just because it was faster. My point is that you get what you pay for, and OSX gives a lot more back.
Simon
As an exmaple, try writing a program that spawns two copies of itself. The only reason most preemptive OSs won't stay on their knees is they have work arounds to prevent this particular problem (but didn't always have them)
Well, to be fair, *any* OS will choke on that one (through RAM, process limits, CPU time, whatever resource is limited first), unless it has workarounds to stop it - it's not as though co-operative succeeds where pre-emptive fails in that regard...
Simon
Before OSX, the mac had the reputation of the machine that crashed all the time. By comparison, Windows was actually pretty reliable (this was before all the spyware/malware/crap that affects it recently, remember). Linux was best, of course...
Now you're just displaying your ignorance
newsflash:when you need to do more work because you're in a far-more-capable and complex environment, it can take more machine-instructions to perform the task. This is just griping - the world has moved on from buggy, insecure, crappy-old OS9. Move with it.
They didn't throw any babies away, they did what they needed to do (ditch the abortion that was OS9) and move onto a new platform which provided the security, flexibility, and reliability that any modern OS provides. A brave decision, under the circumstances, and one well-conceived and executed.
Simon
It's very simple. Every time someone comes up with "most apps are useless at multi-processing", it's always a windows app. Most Apple apps are already multi-threaded for the reasons I state.
It seems you can't point out a technical achievement (on either side of the fence) without some 'fanboy' accusation being levelled. [sigh]
Simon.
If you look at the way most OSX apps are designed, it's easy to multi-thread them. Cocoa pretty much imposes a model/view/controller pattern, and when your model manipulation is separate from your UI, it's pretty simple to spawn a background thread to calculate long tasks, or adopt a divide & conquer approach.
The other nice thing they have is the Accelerate.framework - if you link against that, you automatically get the fastest possible approach to a lot of compute-intensive problems (irrespective of architecture), and they put effort into making them multi-CPU friendly.
Then there's xcode which automatically parallelises builds to the order of the number of CPUs you have. If you have more than one mac on your network, it'll use distcc to (seamlessly) distribute the compilation. I notice my new Mac Mini is significantly faster than my G5 at producing PPC code. Gcc is a cross-compiler, after all...
And, all the "base" libraries (Core Image, Core Video, Core Graphics etc.) are designed to be either (a) currently multi-cpu aware, or (b) upgradeable to being multi-cpu aware when development cycles become available.
You get a hell of a lot "for free" just by using the stuff they give away. This all came about because they had slower CPUs (G4's and G5's) but they had dual-proc systems. It made sense for them to write code that handled multi-cpu stuff well. I fully expect the competition to do the same now that dual-CPU is becoming mainstream in the intel world, as well as in the Apple one...
Simon
From the article you link to:
... and Argentina did (after all) use Exocets to sink British ships. I don't necessarily blame them - we were at war with them, and killing them just as happily! But it caused a *lot* of anti-French feeling in the UK at the time.
As France had recently sold Super Entendard aircraft and Exocet missiles to Argentina, when war broke out there was still a French team in Argentina helping to fit out the Exocets and aircraft for Argintinean use. The French team continued to assist the Argentines throughout the war, in spite of the NATO embargo and official French government policy. [2]
Simon
Well, to be fair, that sounds hard, but it's only ~200 lines of verilog for a simple 16-bit CPU with an ALU and no pipelining/RAM prefetch. Designing CPU's is a hobby of mine... (So, yes, I can tell you how to do it [grin]). Although I'm more interested in compression systems at the moment - I'm currently trying to build an HDTV H.264 digitiser/encoder in hardware to connect via firewire to my Mac Mini...
He had a typo in his example (which is why it'd have to be square), but I believe his point was to do with locality of reference - the second example would miss the cache almost every time (with most image-storage algorithms, anyway. Images are generally stored left-to-right,top-to-bottom rather than vice-versa).
Simon
Ok, sorry to respond to my own post, but the figures I was quoting for the USA were per 100,000 population, not per 10,000. Mea culpa.
The conclusion is still the same though:
853x320/65 (to normalise population) =~4200
~5.5 per 100,000 = 17600.
Or a factor of ~4.
Simon
From your DOJ article: "The higher volume of crime in the United States is due, at least in part,to the greater population size of the United States. A more meaningful comparison is between the crime rates of the two countries."
The UK is has a far higher density of population than the USA. There are ~65 million people in the UK, and ~320 million in the USA. 4x the people, 50x the area. A "more meaningful" comparison would take that into account.
An additional reason for the UK to have higher violent crime is that the victim often survives in the UK. Even if attacked with a knife. You are far more likely to end up dead in the USA, so the figures are artifically low when comparing the two.
The USA has just sustained a huge drop [warning: PDF] in murder rates/year (in 2002, the latest figures I could find, it was ~5 murders per 10,000 people per year). During the time period you linked to in the "higher violent crime rate", it was ~8 murders per 10,000 people per year. Or, put another way ~260,000 murders. In the UK in 2003/2004 there were 853. In the period you linked to, it was ~700.
853 x4 is nowhere near 160,000 (both recent figures). The UK is no panacea, but to paint it as more-violent than the USA is just plain wrong.
Simon
As soon as the invitations hit the 'net, all sorts of rumours (note: these are *not* Apple-created) surface. Some people publicly projected their own desires onto the event, irrespective of how likely it is, and then these self-same people get all disappointed when their fantasy doesn't come true. These people need to (a) get out more, (b) have more sex, and (c) move on from the mental state of a five-year-old ("Me want", "Me want", "Me want"). [aside: note that (c), as applied to (b), is more likely after (a). Just a hint to get you started...]
The fault here lies solely, completely, and utterly with those who raise Apple on too high a pedestal. There's only so much cool stuff any one company can make (although I thought the new mini *was* pretty cool, personally).
Simon.
Enough said.
Simon
Um, the cell has 8 SPE's (all identical) and 1 PPC core. This is the same (although on a larger scale) as the PS2 (2 VU's, and 1 MIPS chip)
Simon.
See my reply above (vcl v2) and look on the linux for PS2 website for VCL.
VCL takes sequential code and splits it up into parallel code based on the constraints of the vector-units (each VU is dual-issue, with some restrictions). It'll re-order code, insert wait states, etc. Certainly it's a good start at auto-parallelisation of the code. It's supposed to do as well as a skilled engineer...
Simon
On the PS2, there are two vector units (vu0 and vu1), which are basically where all the grunt work is done - the mips chip is there for housekeeping and non-time-critical code. Each VU has 2 code-paths (the instruction word is 64-bit, and there are two 32-bit instructions in each word). There are limitations on what you can do in each of the two words simultaneously. Sony have a GUI tool (in their professional kit) which allows the programmer to write essentially sequential code, and have it take full advantage of the vector units. According to Sony, it performs as well as a skilled programmer.
For the linux kit, they only released vcl (a commandline version). It's a bit like a compiler-stage. It takes sequential assembly language for a single VU and re-orders code, inserts wait-states etc. Finally producing another assembly output which is optimised for the dual-issue nature of a VU.
It strikes me that optimising for constraints over 2 code paths in a single unit isn't too far a stretch from optimising for constraints over 8 code paths in 8 units. The differences are mainly to do with locality of reference. On a VU it was up to the programmer to DMA data into scratch-space RAM, and set flags as semaphores on operation. There's no real reason why a computer program can't do that - a basic approach would be to do it on a function-by-function approach, or use #pragma constraints in the code. There's no need to have the all-singing, all-dancing version of the optimiser as version 1...
Simon.
So, I've got a Garmin Streetpilot 2610. Last year I drove from Mountain View (where I live) to Las Vegas, without any maps, purely by putting (I think 3) waypoints - ("yes, I do want to go through Death Valley and Yosemite" etc.), a start and an end destination.
...
I've done the trip before, but I wouldn't say I was familiar with it, and I tried to see different things this time anyway - we went somewhat out of our way to Monument Valley on the way back, 'cos that's fairly spectacular
The thing performs perfectly - I use it a lot. The confidence it brings when you know it will work out the best route and take you that way is just great. No panic if (when) you miss a junction on a freeway because you were in the wrong lane. No problem if you miss a turn in a city you don't know. I can't praise the thing enough.
My parents came over to visit - never been to the US before. Because I had to work that week, they went on a trip down the coast, with the GPS guiding them all the way. 10 minutes instruction was all it took to get them started, and they were experts after a few days, putting their own waypoints in etc. You have to understand that my parents really *are* VCR-challenged...
I'd never try and put in directions when the car was moving though - just pull over, type the details, and drive on afterwards. News-flash: driving without looking at the road (no matter what gizmo is involved) is not a good idea...
Simon
Fitts' Law is of course valid, but it has limitations as well. On the original Macs (with what ? A 9" screen ?), you can see the point - it's an easy move from wherever you are on the screen to the top left corner - a definite win.
However, I'm currently sitting in front of 2 23" monitors (and I have 2 30" monitors at work!). That gives me a screen resolution of 3840x1200, and moving the mouse from the right-most screen to the top-left of the left-most screen is by no means a trivial operation. It would be far more convenient for me to move the pointer 3" up to the top of the window than 36" (I've just measured from where I'm typing) to the menubar.
So it's just taken me 3 move-the-mouse, lift-and-reposition-the-mouse-back-where-it-was cycles to get to the menu, and one movement to position the mouse over a link in the quick-links bar in Safari. The latter was far faster than the former, and easier too. It just gets worse with 2 30" monitors, of course...
I've lost count of the number of times I've moved the mouse *almost* to the menubar and clicked too soon, so the Finder menu appears. Damn! Now I have to move the mouse all the way back again to the application, click there, and try again. This never happens with embedded menus.
Fitts' law was useful, but certainly when you're using two displays, it's had its day.
Two solutions spring to mind - either place a menubar optionally on *both* screens, or allow apps (even if just newer cocoa ones) to bind their menubar to their windows (in which case the Finder could be substituted on the menu at the top of the screen). I used to think this would never happen, but now that we have two-button mice....
Simon
Er, no it's not. Just look at the number of gullible fools that treat a 2000-year-old fable as absolute truth today. Assuming gullibility is roughly equivalent today (ie: the null nypothesis - nothing has changed) it ought to be relatively easy to convince people of falsehoods. Today we regard magicians tricks as sleight-of-hand, but historically they had far more credibility.
Add in the fact that we have a local native preaching about a new kingdom (which would resound well in the ears of an oppressed populace) and his credibility wouldn't even have to be that high - people are always more willing to believe things they want to believe...
And then there's the fact that most of the stories were made up ^W^W scribed much later - well after the death of this man, even if he ever existed.
Organised religion (at least in the West) is simply a power structure where one (wo)man attempts to gain power over peers by promising them fantastic gains/terrible suffering after death. It uses a 2000-year-old ("well it *must* be true") book written by corrupt church "leaders" (have you ever *read* about early church leaders??) to cow the common people into doing as they are told. It's nothing more than that.
As for your comment about sceptics having to come up with proof of non-miracles - nothing could be further from the truth... Extraordinary claims require extraordinary evidence, and none has ever been provided.