Agreed - most programmers stink. I'm sitting here with a shedload of C++ in front of me (for a commercial program) and the line: catch (...) { }
appears with a depressing frequency.
I know the guy who did this and he's not dumb - just lazy. Even an assert in there would give you a fighting chance of finding out that something went wrong. Jeez.
Not sure about that. I use the web for shopping for stuff I couldn't get easily elsewhere (isn't that the beauty of it?). Most hideous recent example of "you want to go there to buy that" would be this specimen. I quite like the Pixies (although I'm going right off them after 5 minutes of that) but nobody needs a site like that - even though the prices are good (for the UK).
There's "evolving technology" and there's just plain fucking annoying. I don't need this shit when I just want to buy some new bindings.
Maybe I should send more abuse to more webmasters.
PS Glad someone else on here can put an apostrophe in the right place.
for most businesses, already in the Win2000 migration, XP is not a good choice
I'd be interested to know what proportion of MS OS license income comes from business Vs consumer. From a business perspective, why the hell should we re-train 10 support people and 1000 users on a n.0 MS OS? OK, so this one might be different and work properly before SP1, but we know the drivers aren't there and we aint gonna replace half our hardware just to get the Fischer Price look on the desktop.
What the advertising campaign is doing to causing (l)users to constantly question why we aren't upgrading. This causes a lot of pressure on senior (ie stupid) management and that is a PITA. We must resist.
Maybe, one of the reasons is that, as long as they keep adding more complicated stuff to web sites, it's never viable to produce a cheap (ie, all in hardware) web browsing device. (I did rant on this subject before.)
First it was frames (which fucked the Web TVs), then all this layers, DHTML crap. What next?
If the stuff is cheap and in hardware then upgrading your browser or downloading this week's plugin is not really an option.
I bet that lots of companies could produce a cheap, simple browsing device - providing all you wanted to browse was/. and Need To Know.
Does anyone else want to join a Keep Websites Simple/Kill All Graphic Designers movement?
Actually, you're both wrong. In it's first incarnation, local.NET components do sit on top of COM - IIRC remoting is done via SOAP, not DCOM. Dunno what this does to security - at least DCOM has a security model, even if it's only there to drive you mad and blind (try it across NT domains with connection points). This is a bit like the way OLE used to sit on top of DDE (when nobody used it) until they sorted that out.
As to when the "real thing" will be available, well reading this makes me wonder.
Since I wrote the original, I did a bit of digging and found this (scroll down a bit for a really flash diagram). By affecting the measured polarisation on one end, they are, instantly, affecting the measured polarisation on the other.
Yes, you are correct that the reason it can't be intercepted is that because it would break the message. And, of course, it's totally impractical. Interesting though - and, as a lot of Quantam Physics things are - totally counter-intuitive.
For years, I've been reading about the idea of data transmission using quantum entangled pairs of particles (possibly photons). The idea (Bell's Hypothesis) being that measurement of a property (eg spin) of a quantum particle will affect the property of another particle (which it has previously interracted with) instantly. That's instantly - not at the speed in light. This has been tested in the lab and proved to be true.
This effect could be used for communication and would imply two things:
1. As stated above, the communication would be instant, regardless of distance.
2. It is impossible to intercept the message with affecting it as any measurement will affect the result.
If it could be made to work, then you would have instant, uninterceptable communications. The problem being how you separate entangled pairs and get them to each end of the line. It's only been tested with distances of about 10 feet so far.
I'm talking about a shrink-wrapped app (dumb user install) with roughly 1m lines of code using threads and DCOM onto 9x and NT/2K. Single code base with some platform detection code. Never ever, ever, ever again. Fuck DCOM - you'd be better off using sockets. The documentation is all lies.
Calculation server code is now going onto Linux. Better.
You'll notice though that pretty much nobody actually runs it as flat 32 way SMP. The Unisys 32 CPU boxes are normally configured as a 4 x 8 way CMP machine (basically a cluster in a box). In fact, the 16/32 processor Compaq machines are rebadged Unisys ES7000s.
Back to the topic, as mentioned by other posters, you don't need IIS if you're just using the DC box a a big f*ck-off database server. In fact, you're better off not having IIS on there - especially if you're open to RDS (the s'kiddies favourite) as they can cause untold havoc using SQL server. Personally, if I had to use IIS for something public, I'd have a second firewall between the IIS box and the data box with no data at all on the IIS box.
Oh, and another point, if you are using SQL 7, then forget clusters/CMP. It just about works with SQL 2000, but has no load balancing (you can do it manually by partitioning the database) and the failover time is 1 - 5 minutes.
From what I can gather, DC and AS are very similar (although MS will deny this). Both have address extensions allowing over 4GB to be used, > 4 way SMP etc., but DC comes with all the drivers etc. checked out on the specific hardware so - hopefully - not as many BSODs. I wouldn't be surprised if someone comes up with the reg settings to turn AS into DC (Max_Processors=32, Max_Cluster_Boxes = 4) - but you'd have to frig with a DC box first - if you can get your hands on one.
That is why Microsoft has always sold its operating system cheaply and has done everything to make life easy for programmers.
Obviously not someone who is familiar with the joys of COM - especially pre-ATL. Also, not someone who ever spent weeks trying to get that new shiny feature of NT4, DCOM, working only to find out that it never worked at all (RPC layer broken) until SP3. Not someone who has ever tried to produce a system which runs perfectly on all Win32s. If he means "made life easy for VB programmers", then maybe - but I wouldn't dignify them with the name "programmer".
I could rant for hours about specific instances, but I wont.
How the lack of "tone of voice" in these sort of discussions can cause two people who agree with each other to argue like that. Do a "view source" on my HTML formatted comments and you'll see I'm not a content person either. BTW
Anyone still trying to make it look nicer to end users is living in 1998
Can I use that next time I have a set-to with our graphics designers?
PS I may only have had a/. login for a few weeks (karma going to be even lower after this OT post), but I bet I've been in the business longer than most on here.
If you don't get the joke, maybe in future, I'll tag them all up for you to make it easier.
<serious>Actually, <slightly_humorous>despite the hugh proportion of my working day which I spend browsing</slightly_humorous> I've yet to find a benefit to using IE6.0 over 5.01. This raises the question in my mind - what is the point of IE6.0? OK, so people aren't using the new features, but what are these features which are so damn crucial? What end-user benefit is there in what is, mostly, eye-candy?</serious>
*You* can do it (good - the world seems to be short of people who can configure a firewall properly), but how many pr0n crazed ADSL users (with "always on" connections) are going to have firewalls? Apart, that is, from the shit one in XP provided by MS - which probably can't be configured to prevent Redmond doing as they please.
Yes, us techies will always find a way round (legal or illegal) but it's the vast majority of users who just want to play games, listen to CDs (which they bought) on whatever device and jerk off who will suffer. Even if they don't know it yet.
Did I say anything about running flash type stuff on the server? When I say complicated, I'm not referring to graphics and buttons and stuff. Yes, a mechanism is needed which can do some things without having to go back and forward to the server all the time (like to say "you must enter an email address").
Oh, and yes, I do appreciate that, for IE, COM is used for plugins. Looking at IE, how else would you do it? "Standard plugins" though are a world away from any bit of executable code which someone might feel like running on your machine - and don't give me crap about "certificates" making it all OK.
"leaps ahead of any other browser". What does that actually mean? That people have keep buggering about with HTML (yes, I know there's more to it than HTML) so new versions of browsers are constantly needed in order to keep up. What effect does that have? Well, with an MS browser, upgrading generally means replacing half your OS with files (OLE*.DLL amongst others) that cause old apps to not work - even if they had any disk space left. And why is it changing? Generally to please a load of graphic designers (make the bastards work with a 14.4 modem) plus the people who market stuff like Dreamweaver and C*ntpage. Do you think it's rewarding work for web developers to have to code and test for all these different browsers?
What do you actually need in a browser aside from fast HTML rendering (with CSS), a consistent Javascript model (so you can do stuff without having to go back to the server) and an architecture which supports common plugins (Flash, SVG). OK, you can make an argument for Java Applets if something more complicated needs to be done on the machine, but downloading and running some other muppets native executable code (ActiveX) and running it with my priveledges - no way. I can do enough damage with my own code. If it's complicated, why isn't it running on the server? What ever happened to thin clients?
Stop the madness.
PS I appreciate the irony that I'm posting using IE 6, but I'm at work and I'm testing whether it offers anything over our standard IE 5.01. It doesn't -/. looks the same to me.
How can you possibly run a capitalist economy when people are going to start using less of things? There can only be economic growth if people spend more money on stuff. To make more stuff requires more energy. You can reduce this cost by either reducing the cost of energy or the amount you use to produce stuff, but if you reduce the cost of stuff (by reducing the energy cost), then people just buy more stuff - because they can. Either way, you don't reduce the amount of energy used. Unless you reverse economic growth, total energy use can only increase.
PS I know nothing about economics and the above is probably stupid and wrong.
The RIAA believes that this kind of technological "self-help" against online pirates, if done carefully, is legal under current federal law.
How carefully? I see no loopholes in the link. In the UK it's certainly not legal (Computer Misuse Act). If you decide to get revenge on some script kiddie who's messing around with your site, you can't. You just have to call PC Plod - which is going to be great: Plod: "Evening all. What did you say the address of this person was?" BOFH: "203.41.14.3" Plod: "I'm sorry, but what part of Essex is that in?"
... I always felt that a closed-cycle gas turbine (Rover experimented with one in the 50s) combined with CVT (Continuously Variable transmission - eg DAF variomatic, Uno Selecta) would be a good way forward.
Gas turbines are effecient (insert something clever to do with thermodynamics here)and can run on anything from coal-dust to hydrogen. The problem is - IIRC - that they only really work well within a narrow range of speed so coupling them to either conventional (stick-shift) or auto transmissions never really worked. Coupling to CVT should allow the engine to always spin at an efficient speed. Piston engined cars with CVT get good gas mileage - but people don't like the fact that the engine note stays the same as they accelerate.
It would run good on hydrogen (should be very little H2O2 in the exhaust burning like that), but I still don't have a solution to producing and storing H2.
My company exchanges a shedload of confidential data with customers - some of whom use PGPG. I tried the eval of PGPmail last week and couldn't get it going with Notes (no Outlook - no virus). Even waving the prospect of 12,000 seats at them they wouldn't respond. Should've guessed something was up.
We'll just have to stick to our normal encryption method - making our documents too boring for anyone to remain concious while they read them.
OK, I must admit that I haven't got round to benchmarking CLR versus native yet - but then of course it's only in beta. For some reason, I load up the IDE and then lose the will to live.
I can see them producing a decent CLR for Itanium, but for a decent high-end *nix box? They'd need to charge so much for the CLR to make up for the fact that you didn't need 50 Win?? boxes.
Still sceptical.
Re:This is good news...
on
J#
·
· Score: 3, Insightful
Is it a good idea? I can see that it would be - if it weren't for the small point that it only runs on Windows. If you're only going single platform then surely you'd be better off going native code and just using the same compiler back-end for all languages (as Borland do for Delphi and C++ Builder - which both run like sh*t off a shiny shovel).
Sure they say they can build a CLR for other platforms, but I'll believe it when I see it.
Of course, if they didn't have this CLR, then they couldn't make the claim that they were going multi-platform so it looks like they just found a way to slow down your Windows code for no good reason.
... recumbent. People have experimented with recumbent supine and prone in the past in an effort to get just that extra aerodynamic advantage. Since friction and rolling resistance are pretty much negligable on a racing bike, air resistance is the major force. I remember an article many years ago (before the web existed) in Scientific American on the subject where someone calculated that if you could cycle in a spacesuit in a vacuum (lots of ifs) a decent rider should be able to do over 200mph.
You can buy recumbent HPVs for road use, but down there the trucks aint gonna see you. keeping out the way should give you an extra speed boost even if nothing else does. Some models have things like flags on long poles to give you a chance of being seen.
I haven't used Beowolf at all so I'd appreciate any advice you guys could give. Years ago I worked with Transputers - which have similar problems: throwing the data round the transputer network is the bottleneck.
I work on actuarial calculations - which tend to be just a shedload of FP calcs done over and over again. Monte Carlo simulations are popular at the moment - which means really just running the calcs a few thousand times with some random elements changing, then getting a few percentiles out at the end. The result being, you end up doing 5 billion plus FP ops to get 4 numbers out the end. We'd like to produce this as a web service ('cos we have the client data - they just need a simple SOAP request saying what numbers they want) with a 5 seconds response. Yeah, I know.
Assuming I can design so that bandwidth isn't too much of a problem, does anyone think this will work? Is there latency inherent in the clustering software itself? Would I be better off looking at some sort of hardware vector processor, or writing my own stuff to send the data round?
I'll probably get modded down for ignorance and having the nerve to ask an actual question, but if anyone's got any insights, I'd appreciate it.
That's interesting. If the pitch variations did invalidate the claim, then that would mean that, in theory, I could take my old 'cello out and play any copyrighted composition on the basis that it would be out of tune. And shit.
Seriously though, they have missed a trick. I seem to recall that Penderecki developed what he called an "optical notation" for his "Threnody to the victims of Hiroshima" in 1960 as it contains quarter tones (sounds bloody awful IMO). I can't find any details so I don't know if it would allow an accurate representation. There are plenty of modern works which use "modified" (fucked) instruments to produce sounds not in the euqitempered scale (Arvo Part - Tabula Rasa, Sonic Youth - Death to our friends) which are still under copyright.
Some of my own compositions use effects like mixing in the sound of people leaving the building. Quickly.
I'm going to get modded down now aren't I for being cultural. And boring.
Slashdot Mirror
Agreed - most programmers stink. I'm sitting here with a shedload of C++ in front of me (for a commercial program) and the line:
catch (...) { }
appears with a depressing frequency.
I know the guy who did this and he's not dumb - just lazy. Even an assert in there would give you a fighting chance of finding out that something went wrong. Jeez.
Nobody's forcing you to.
Not sure about that. I use the web for shopping for stuff I couldn't get easily elsewhere (isn't that the beauty of it?). Most hideous recent example of "you want to go there to buy that" would be this specimen. I quite like the Pixies (although I'm going right off them after 5 minutes of that) but nobody needs a site like that - even though the prices are good (for the UK).
There's "evolving technology" and there's just plain fucking annoying. I don't need this shit when I just want to buy some new bindings.
Maybe I should send more abuse to more webmasters.
PS Glad someone else on here can put an apostrophe in the right place.
for most businesses, already in the Win2000 migration, XP is not a good choice
I'd be interested to know what proportion of MS OS license income comes from business Vs consumer. From a business perspective, why the hell should we re-train 10 support people and 1000 users on a n.0 MS OS? OK, so this one might be different and work properly before SP1, but we know the drivers aren't there and we aint gonna replace half our hardware just to get the Fischer Price look on the desktop.
What the advertising campaign is doing to causing (l)users to constantly question why we aren't upgrading. This causes a lot of pressure on senior (ie stupid) management and that is a PITA. We must resist.
Maybe, one of the reasons is that, as long as they keep adding more complicated stuff to web sites, it's never viable to produce a cheap (ie, all in hardware) web browsing device. (I did rant on this subject before.)
/. and Need To Know.
First it was frames (which fucked the Web TVs), then all this layers, DHTML crap. What next?
If the stuff is cheap and in hardware then upgrading your browser or downloading this week's plugin is not really an option.
I bet that lots of companies could produce a cheap, simple browsing device - providing all you wanted to browse was
Does anyone else want to join a Keep Websites Simple/Kill All Graphic Designers movement?
Actually, you're both wrong. In it's first incarnation, local .NET components do sit on top of COM - IIRC remoting is done via SOAP, not DCOM. Dunno what this does to security - at least DCOM has a security model, even if it's only there to drive you mad and blind (try it across NT domains with connection points). This is a bit like the way OLE used to sit on top of DDE (when nobody used it) until they sorted that out.
As to when the "real thing" will be available, well reading this makes me wonder.
Since I wrote the original, I did a bit of digging and found this (scroll down a bit for a really flash diagram). By affecting the measured polarisation on one end, they are, instantly, affecting the measured polarisation on the other.
Yes, you are correct that the reason it can't be intercepted is that because it would break the message. And, of course, it's totally impractical. Interesting though - and, as a lot of Quantam Physics things are - totally counter-intuitive.
For years, I've been reading about the idea of data transmission using quantum entangled pairs of particles (possibly photons). The idea (Bell's Hypothesis) being that measurement of a property (eg spin) of a quantum particle will affect the property of another particle (which it has previously interracted with) instantly. That's instantly - not at the speed in light. This has been tested in the lab and proved to be true.
This effect could be used for communication and would imply two things:
1. As stated above, the communication would be instant, regardless of distance.
2. It is impossible to intercept the message with affecting it as any measurement will affect the result.
If it could be made to work, then you would have instant, uninterceptable communications. The problem being how you separate entangled pairs and get them to each end of the line. It's only been tested with distances of about 10 feet so far.
I'm talking about a shrink-wrapped app (dumb user install) with roughly 1m lines of code using threads and DCOM onto 9x and NT/2K. Single code base with some platform detection code. Never ever, ever, ever again. Fuck DCOM - you'd be better off using sockets. The documentation is all lies.
Calculation server code is now going onto Linux. Better.
You'll notice though that pretty much nobody actually runs it as flat 32 way SMP. The Unisys 32 CPU boxes are normally configured as a 4 x 8 way CMP machine (basically a cluster in a box). In fact, the 16/32 processor Compaq machines are rebadged Unisys ES7000s.
Back to the topic, as mentioned by other posters, you don't need IIS if you're just using the DC box a a big f*ck-off database server. In fact, you're better off not having IIS on there - especially if you're open to RDS (the s'kiddies favourite) as they can cause untold havoc using SQL server. Personally, if I had to use IIS for something public, I'd have a second firewall between the IIS box and the data box with no data at all on the IIS box.
Oh, and another point, if you are using SQL 7, then forget clusters/CMP. It just about works with SQL 2000, but has no load balancing (you can do it manually by partitioning the database) and the failover time is 1 - 5 minutes.
From what I can gather, DC and AS are very similar (although MS will deny this). Both have address extensions allowing over 4GB to be used, > 4 way SMP etc., but DC comes with all the drivers etc. checked out on the specific hardware so - hopefully - not as many BSODs. I wouldn't be surprised if someone comes up with the reg settings to turn AS into DC (Max_Processors=32, Max_Cluster_Boxes = 4) - but you'd have to frig with a DC box first - if you can get your hands on one.
That is why Microsoft has always sold its operating system cheaply and has done everything to make life easy for programmers.
Obviously not someone who is familiar with the joys of COM - especially pre-ATL. Also, not someone who ever spent weeks trying to get that new shiny feature of NT4, DCOM, working only to find out that it never worked at all (RPC layer broken) until SP3. Not someone who has ever tried to produce a system which runs perfectly on all Win32s. If he means "made life easy for VB programmers", then maybe - but I wouldn't dignify them with the name "programmer".
I could rant for hours about specific instances, but I wont.
How the lack of "tone of voice" in these sort of discussions can cause two people who agree with each other to argue like that. Do a "view source" on my HTML formatted comments and you'll see I'm not a content person either. BTW
/. login for a few weeks (karma going to be even lower after this OT post), but I bet I've been in the business longer than most on here.
Anyone still trying to make it look nicer to end users is living in 1998
Can I use that next time I have a set-to with our graphics designers?
PS I may only have had a
If you don't get the joke, maybe in future, I'll tag them all up for you to make it easier.
<serious>Actually, <slightly_humorous>despite the hugh proportion of my working day which I spend browsing</slightly_humorous> I've yet to find a benefit to using IE6.0 over 5.01. This raises the question in my mind - what is the point of IE6.0? OK, so people aren't using the new features, but what are these features which are so damn crucial? What end-user benefit is there in what is, mostly, eye-candy?</serious>
*You* can do it (good - the world seems to be short of people who can configure a firewall properly), but how many pr0n crazed ADSL users (with "always on" connections) are going to have firewalls? Apart, that is, from the shit one in XP provided by MS - which probably can't be configured to prevent Redmond doing as they please.
Yes, us techies will always find a way round (legal or illegal) but it's the vast majority of users who just want to play games, listen to CDs (which they bought) on whatever device and jerk off who will suffer. Even if they don't know it yet.
Did I say anything about running flash type stuff on the server? When I say complicated, I'm not referring to graphics and buttons and stuff. Yes, a mechanism is needed which can do some things without having to go back and forward to the server all the time (like to say "you must enter an email address").
Oh, and yes, I do appreciate that, for IE, COM is used for plugins. Looking at IE, how else would you do it? "Standard plugins" though are a world away from any bit of executable code which someone might feel like running on your machine - and don't give me crap about "certificates" making it all OK.
Done arguing now.
"leaps ahead of any other browser". What does that actually mean? That people have keep buggering about with HTML (yes, I know there's more to it than HTML) so new versions of browsers are constantly needed in order to keep up. What effect does that have? Well, with an MS browser, upgrading generally means replacing half your OS with files (OLE*.DLL amongst others) that cause old apps to not work - even if they had any disk space left. And why is it changing? Generally to please a load of graphic designers (make the bastards work with a 14.4 modem) plus the people who market stuff like Dreamweaver and C*ntpage. Do you think it's rewarding work for web developers to have to code and test for all these different browsers?
/. looks the same to me.
What do you actually need in a browser aside from fast HTML rendering (with CSS), a consistent Javascript model (so you can do stuff without having to go back to the server) and an architecture which supports common plugins (Flash, SVG). OK, you can make an argument for Java Applets if something more complicated needs to be done on the machine, but downloading and running some other muppets native executable code (ActiveX) and running it with my priveledges - no way. I can do enough damage with my own code. If it's complicated, why isn't it running on the server? What ever happened to thin clients?
Stop the madness.
PS I appreciate the irony that I'm posting using IE 6, but I'm at work and I'm testing whether it offers anything over our standard IE 5.01. It doesn't -
How can you possibly run a capitalist economy when people are going to start using less of things? There can only be economic growth if people spend more money on stuff. To make more stuff requires more energy. You can reduce this cost by either reducing the cost of energy or the amount you use to produce stuff, but if you reduce the cost of stuff (by reducing the energy cost), then people just buy more stuff - because they can. Either way, you don't reduce the amount of energy used. Unless you reverse economic growth, total energy use can only increase.
PS I know nothing about economics and the above is probably stupid and wrong.
According to the article:
The RIAA believes that this kind of technological "self-help" against online pirates, if done carefully, is legal under current federal law.
How carefully? I see no loopholes in the link. In the UK it's certainly not legal (Computer Misuse Act). If you decide to get revenge on some script kiddie who's messing around with your site, you can't. You just have to call PC Plod - which is going to be great:
Plod: "Evening all. What did you say the address of this person was?"
BOFH: "203.41.14.3"
Plod: "I'm sorry, but what part of Essex is that in?"
In case anyone's not familiar with CVT.
It does look a bit fragile (rubber band drive anyone?) but I'm assured that they work well up to several hundred HP.
... I always felt that a closed-cycle gas turbine (Rover experimented with one in the 50s) combined with CVT (Continuously Variable transmission - eg DAF variomatic, Uno Selecta) would be a good way forward.
Gas turbines are effecient (insert something clever to do with thermodynamics here)and can run on anything from coal-dust to hydrogen. The problem is - IIRC - that they only really work well within a narrow range of speed so coupling them to either conventional (stick-shift) or auto transmissions never really worked. Coupling to CVT should allow the engine to always spin at an efficient speed. Piston engined cars with CVT get good gas mileage - but people don't like the fact that the engine note stays the same as they accelerate.
It would run good on hydrogen (should be very little H2O2 in the exhaust burning like that), but I still don't have a solution to producing and storing H2.
My company exchanges a shedload of confidential data with customers - some of whom use PGPG. I tried the eval of PGPmail last week and couldn't get it going with Notes (no Outlook - no virus). Even waving the prospect of 12,000 seats at them they wouldn't respond. Should've guessed something was up.
We'll just have to stick to our normal encryption method - making our documents too boring for anyone to remain concious while they read them.
OK, I must admit that I haven't got round to benchmarking CLR versus native yet - but then of course it's only in beta. For some reason, I load up the IDE and then lose the will to live.
I can see them producing a decent CLR for Itanium, but for a decent high-end *nix box? They'd need to charge so much for the CLR to make up for the fact that you didn't need 50 Win?? boxes.
Still sceptical.
Is it a good idea? I can see that it would be - if it weren't for the small point that it only runs on Windows. If you're only going single platform then surely you'd be better off going native code and just using the same compiler back-end for all languages (as Borland do for Delphi and C++ Builder - which both run like sh*t off a shiny shovel).
Sure they say they can build a CLR for other platforms, but I'll believe it when I see it.
Of course, if they didn't have this CLR, then they couldn't make the claim that they were going multi-platform so it looks like they just found a way to slow down your Windows code for no good reason.
... recumbent. People have experimented with recumbent supine and prone in the past in an effort to get just that extra aerodynamic advantage. Since friction and rolling resistance are pretty much negligable on a racing bike, air resistance is the major force. I remember an article many years ago (before the web existed) in Scientific American on the subject where someone calculated that if you could cycle in a spacesuit in a vacuum (lots of ifs) a decent rider should be able to do over 200mph.
You can buy recumbent HPVs for road use, but down there the trucks aint gonna see you. keeping out the way should give you an extra speed boost even if nothing else does. Some models have things like flags on long poles to give you a chance of being seen.
I haven't used Beowolf at all so I'd appreciate any advice you guys could give. Years ago I worked with Transputers - which have similar problems: throwing the data round the transputer network is the bottleneck.
I work on actuarial calculations - which tend to be just a shedload of FP calcs done over and over again. Monte Carlo simulations are popular at the moment - which means really just running the calcs a few thousand times with some random elements changing, then getting a few percentiles out at the end. The result being, you end up doing 5 billion plus FP ops to get 4 numbers out the end. We'd like to produce this as a web service ('cos we have the client data - they just need a simple SOAP request saying what numbers they want) with a 5 seconds response. Yeah, I know.
Assuming I can design so that bandwidth isn't too much of a problem, does anyone think this will work? Is there latency inherent in the clustering software itself? Would I be better off looking at some sort of hardware vector processor, or writing my own stuff to send the data round?
I'll probably get modded down for ignorance and having the nerve to ask an actual question, but if anyone's got any insights, I'd appreciate it.
That's interesting. If the pitch variations did invalidate the claim, then that would mean that, in theory, I could take my old 'cello out and play any copyrighted composition on the basis that it would be out of tune. And shit.
Seriously though, they have missed a trick. I seem to recall that Penderecki developed what he called an "optical notation" for his "Threnody to the victims of Hiroshima" in 1960 as it contains quarter tones (sounds bloody awful IMO). I can't find any details so I don't know if it would allow an accurate representation. There are plenty of modern works which use "modified" (fucked) instruments to produce sounds not in the euqitempered scale (Arvo Part - Tabula Rasa, Sonic Youth - Death to our friends) which are still under copyright.
Some of my own compositions use effects like mixing in the sound of people leaving the building. Quickly.
I'm going to get modded down now aren't I for being cultural. And boring.