Cookie recipe? (yes, i know the latter may best be covered by a copyright... but hell, should a copyright even be granted?)
FYI you can't copyright a food recipe in the USA. A recipe is a mere compilation of facts. Facts can only be patended. You can copyright a recipe if in a certain creative form, such as in your own handwriting with some doodles on the side. But this doesn't prevent anyone from publishing the recipe themselves after typesetting it themselves and rewriting any clever introductory text you may have added.
It was the[sic] 9th Circuit Federal court who usually do the right thing and then get overturned on appeal by Scalia and the Supremes.
Furthermore, the number of decisions out of the 9th Circuit that get overturned is high enough that the rest of the country rarely pays any attention to what they have to say, except possibly to determine what the law isn't (or won't be).
Actually they have if not the lowest then one of the lowest percentages of overturned rulings. But they do hear most of the important cases just by virtue of their territory. And the most important cases get heard by the SC. When they get overturned most people know they will ahead of time, given the biases of our current crop of nine, but the 9th circuit is also careful in these cases to write the type of strong opinions that law students will read in their classes and that they know will eventually become the law of the land.
I thought that SSE and MMX both had significantly lower precision than standard IEEE floating point ops. If I'm wrong, please correct me, but if it is lower precision, it makes it useless for Real Work(tm).
It performs precise math by default. You can only use 32 or 64 bit floats, the "long double" 80 bit floats are not supported. But this often isn't a problem. You can also turn off denormals, and with interupts on bad math (divide-by-zero type stuff). Turning those off hasn't given me any performance boost, but I still consider these things features not bugs. There are some low precision operations available, but no compiler I know of uses them unless you ask for em. I do in some cases but then I know what I'm getting.
A math person may give you a better answer than me. I'm a graphics person, a field where SSE2 is a godsend compared to the stack based floating point units that came before.
I actually told my prof about my concerns, and his only reply was to tell me to choose between taking the class (which I need to do) and running my papers through the system. I wish I had the balls to do what this kid in McGill did, but I want to go to grad school, and thus need to suck up to profs. Such is the magic of academia I suppose.
I think your chances of getting into the grad school of your choice will be significantly better if you fight this type of bullshit. If the prof really decides to fight you on this you will find more than a few professors on your side when you take this to the administration. And many of them will be happy to write you a glowing essay when you need one. Professors want grad students who will fight for what they believe in. Tilting at windmills is practically a hallmark of the profession.
I never understood the whole "implied binary digit" thing. I have no idea what watch you are talking about. But if they count from 1 to 24 on for the hours, as opposed to 0 to 23 and they have a symbol for both 1 and 0, then the implied digit is simple. A number between 1 and 24 in binary always begins with a one, so you can leave it off. ''(1) '0'(2) '1'(3) '00'(4) '01'(5) etc etc This is idea is also used to save a bit in IEEE floating point, where the special strings of all zeros and all zeros except for the sign bit signify +0 and -0 respectively, and all other real numbers begin with an implied 1 bit (with the exponent placing it in the right place).
Unless someone has come up with a full-spectrum LED, I don't think this kind of lighting is going to see wide adoption outside of perpetually sun-drenched areas, and then only as a "Gee Whiz" item because of the high cost.
They have, sort of. The "White LED" is actually a blue LED with set of phosphors to approximate white light. Basically it's the same as those flourescent fixtures you installed. Flourescents typically emit UV light (slightly higher energy than the blue LED in white LEDs.) and then use phosphors to step it down to a series of different lower energy frequencies. This step down is what makes them a little warm, the energy lost converting ultra-violet to red is lost as heat.
Of course, "White LED's" aren't super efficient. Stick with the flourescents for now. Also, not all white LED's are created the same, just like flourescent bulbs. There is a potential though for LEDs as full spectrum lighting, they come in many natural frequencies and combining those with a nice diffusor and maybe some phosphors to fill in the blanks could give you a relatively cheap to run "full spectrum light." LED's have advantages over flourescents, like you can change the color mixture simply, you can dim them, they are small and mechanically resilient They do have problems that we wouldn't think of off-hand, I know an architect that wanted to use some in a building and was prevented from doing so because 12 volt lamps can't be UL listed, and none of her suppliers were willing to wire 5 or more of them together and apply for the somewhat expensive UL listing. The NYC building inspector could not be convinced that a battery operated 12 volt light was perfectly safe without a UL listing.
Compare it the current versions of icc, butt monkey. icc still beats gcc by 2:1, if not more. Your comparison is meaningless. Its like comparing a 2004 Chevy to a Ford Model T and saying, "See, Chevy really is better than Ford!"
Umm, yeah, well icc 8.0 is the newest release. And I checked against several versions of gcc 3.2.3, 3.3.2 and the latest 3.4 from CVS. The 3.3.2 seemed to have the best performance overall, with 3.4 a close contender. Except the 3.4 had worse performance on a couple benchmarks, not unexpected for a pre-beta. icc was left far far behind, if you look at my posting history you will see I heaped praise on them in the 6.0 period, said nothing after the 7.0 release, and now let people know that with certain code gcc outperforms icc. If you have a legacy app maybe icc will do better, I don't know, with my application it is significantly slower.
icc also won't use my default libstdc++ which was a bit of a pain (limits used gcc builtins which icc doesn't have). I used the Dinkumware standard library for the benchmarks, but if I could have used the same sytem libraries as gcc I might have had the best of both worlds...
A professional-grade compiler like MIPSpro or ICC can generate code over twice as fast as GCC on the same processor, because it's smarter about processor-specifics.
gcc 3.3.2 beats the pants off icc 8.0 on my SSE2 code. Up to a 50:1 ratio on speed tests, 4:1 on average. With earlier revisions of gcc and icc the ratio was 2:1 with icc being faster. This code is written with explicit parallelism so all the fancy loop unrolling icc does doesn't help, and the register allocation algorithm in gcc seems to be the thing giving it the advantage, icc spills to memory on simple matrix vector multiplies. I think it just validates the old maxim about optimization, trust nothing, test everything.
Basically, with given energy inputs, pure hydrogen combining with oxygen give you the most "bang for your buck" since it only requires a small amount of activation energy. Hydrocarbons are more contained, less volitale (that's why we use them in cars instead of hydrogen gas for the 'combustion'), and the emitted energy from the reaction is less.
Sure hydrogen gives you the most bang per weight. But remember it's also the lightest substance around. Since we live in an atmosphere, and live with a significant amount of gravity the hydrogen doesn't just burn hot, it burns away from your body faster than any other matter in the universe would. If there is a continuous pointed at you leak, then like a blowtorch it's not so good to be near. If we are talking about a container that undergoes some kind of catastorpic failure while near my body, I would rather it be hydrogen than gasoline. I don't want a substance infiltrating my clothing and burning slowly as it evaporates.
(More "bang for your buck" is a little inaccurate since it is still pretty expensive compared to a mined fuel, which is why we use gas instead of the cleaner burning hydrogen.)
Yes, napalm has other things than gasoline. If you notice polystyrene is a longer more stable molecule, yet napalm is a pretty nasty substance. It's the sticking around longer that makes it's flame more dangerous. Same as with gasolene vs. hydrogen. There are two things people worry about with hydrogen, a non-flame rupture of the container since when it is used as a fuel it is under pressure and it will explode. Then there is the fire that can follow, and I think here it is important to realize that the fire will envelop you only for a second or so in an open environment, and will die out quickly in a closed one, unlike gasoline, or any of the heavier slower burning hydrocarbon fuels.
Hydrogen is more dangerous than gesoline. Much much more dangerous.
Huh? Hydrogen quickly moves up and away from you while gasoline sticks to you. You did see the pictures of those babies our boys burned alive with napalm in the 70's?
A lot of big companies still use COBOL, and COBOL is over fifty years old, which means we can keep using X even if it's not being developed anymore.
Yeah COBOL. COBOL reminds me of high school. I had to get special permission to not take COBOL before Pascal (which the AP exam was in.) I'm guessing there are a lot of people who weren't as stubborn as I and have some knowledge of it. Nothing wrong with that, the next year I learned C at the local community college, but Turbo Pascal 3 was still more fun, it compiled as fast as assembler and ran faster to boot. I'm sure COBOL has it's own advantages.
Still, I hope XFree86 doesn't go the way of COBOL. I think this announcement may just mean that it won't. The Linux development model is just so much more robust for an important project like X. I may even contribute to X now, in the bad old days I could even figure out to get a bug report into the system.
BTW I have only been involved in 2 accidents involving other people each were under 10 mph bumper dings and over 10 years ago I have been driving for 12 years.
Hehe, I don't consider myself a good driver and haven't been in any "accidents," involving people or not. Not in the last 12 years, not ever.
Not that I'm perfect, I've been known to go 30 mph over the speed limit on a highway, I've forgotten the turn signal a few times (not in the last decade) when driving in farm country, and once when I was 17 I ran a red light because I thought I could make the yellow and I couldn't. After that incident at 17 I never drove tired again, my life is worth more than that. I think maybe you should buy a stick shift without any fancy anti-lock breaks and learn to drive again, maybe turn off the radio so you can concentrate better.
If i recall the 'stacker' case correctly. MS and Stacker had a contract to put Stackers tech into DOS. Then when MS sold a bunch of software with their stuff in it and people stopped buying stacker software.
Well sort of, as I remember it they couldn't come to terms on the contract and so the sale was called off. But MS hired some of their lead coders with seven digit signing bonuses and MS claimed it didn't know they had taken the Stacker code with them, but the judge seemed to think they should have at least looked considering the timeframe in which these coders incorporated the feature. And it was trivial to find this copyright infringement even without looking at the source. In the end it was settled out of court with a 9 digit number, so the ugly details aren't all public. My guess is the settlement number is something like the sale number Stacker wanted, it was a large number because they knew selling it to MS was the end of their company it was essentially a company sale they were negotiating. Microsoft is known for making silly offers, less than the investor's initial investments, and backing it up with the threat of taking billion dollar losses killing the little company out of some sense of spite mixed with their need to set an example of ruthlessness to would be future negotiating partners.
So, you have proof that MS has been involved in "technology theft" then? Either you're wrong in that they're successfully covering up after themselves, or you're wrong in that they're doing it. Either way, you're talking crap.
Woah there partner, you haven't been paying attention. Search your old Windows 3.1 executable for "Stacker"... or google. You might also want to look into some of the other settlements, like the one in France last year. A lot of them involved some very nasty unethical stuff, much of it under the category of theft. You could also buy some drinks for someone you know that worked for a company targeted for ruin by Microsoft, a few hours later you'll not want to partner with it ever again.
It relies on memory latency so it really doesn't take up resources from your system. It appears the only disadvantage is distributed mailing and progress of technology.
Memory latency has been the biggest bottleneck on PC's for the last 5-10 years. So yes it eats up gobs of resources. I haven't seen the actual algorithm but it may be possible that this is completely defeated by a special purpose machine with gobs of memory bandwidth, such as a Playstation 2. That being said my posting history will show I'm a big fan of hash cash. What it will do is force mass e-mailers to either spend money on computers, lowering their profit, or break the law by taking over millions of zombie machines. Lowering their profit will make this a less attractive business to get into, and breaking the law will make them liable to run ins with local police in their country of residence. Both should reduce the number of spams. They may constuct Amway type schemes where they get people to install programs willingly much like Gator and the like work now. But unless they pay these victims, lowering profit, they will likely be prosecuted for fraud. Prosecutors may ignore Gator which just creates problems for ordinary people, but they won't be so happy with hundreds of thousands of unwanted e-mail in their inbox everyday, nor the bill for 3rd party filtering services they shouldn't need.
Well, yes but what happens when the ISP's network is flooded with worm traffic? They really don't have much choice.
Wouldn't it be much better to just disable the ports where virus floods are coming from and have an auto-dialer call up the customer and tell them their computer is infected, giving them a phone number to call once the system is fixed? Then they would be aware of their problem and probably take some more measures in the future to prevent it...
Why is the judge letting SCO get away with this coy "they infringed, but we're not going to show you how yet" stuff? Why hasn't he said, "Put the infringing source code in a brief and hand it over tomorrow, or I'm tossing this"?
The judge has done this. Except they have thirty (30) days to do it. The clock is ticking, it's just ticking very very slowly since this is the US legal system we're dealing with. If their evidence is anything like what they have shown to date even a non-technically literate judge might just issue a summary judgement against them with prejudice. After which they would have a hard time bringing up any more copyright infringement libel against anyone before a US court in the future.
I'm starting to doubt this will even see a trial, and not because I see a settlement as happening. The Boise team must be kicking themselves over this. The damage to their reputation is going to cost them a lot more than the few million they can collect in lawyer fees on this turkey. Someone screewed up, someone thought IBM would just buy the company to save themselves some cash in the short term. Instead they are now going to have to show up in court and enjoy some judge yelling at them for wasting everyone's time.
To answer the question: In the U.S., per FAA regulations, landowners have control up to 500 feet. Above that is public domain airspace. It has been this way since very early on in aviation, since it is critical to have public airspace and federal jurisdiction thereof in order to have a viable air transportation system.
Huh? is that 500ft above your building? Say your building is 800ft tall...
Not that this matters much, in most states you can pretty much assume that anything you say or do outside your 12 ft thick lead walls can be observed legally these days. At least the police can't use extrardinary measures, like thermal imaging, in court; but once that becomes a ordinary, look out.
Not to mention Iceland. Its legislature, the Althing, has been meeting since CE 930 -- nearly 1100 years.
When growing up in Iceland I used to think of America's strengths and problems stemming from it being one of the youngest countries around. It was a total shock when I moved here and heard it's citizen's speak of America as the world's oldest democracy. If you put a bunch of caveats on the phrase it can be interpreted as having some truth, but this foolish. America is America because it is a young country. One without history or language or religion, and so it functions a beautiful microcosm of the world. Our government is a stitched together hodge-podge of democracy and totalitarianism that almost never displays any signs of conscious thought being applied to a problem; this can be incredibly frustrating and even scary at times, but also gives us that lawlessness countries like Iceland lack, this can be really liberating when the laws are insane or just innane or rather silly.
You can check the web page for patent och registreringsverket which handles trademarks.
It looks like it's a set of EU trademarks not a Swedish one, it was probably granted on the basis of the US trademark and assuming Lindows wins in the US by invalidating the "Windows" trademark, it will be vacated on an EU wide basis. It seems to be registered to three entities each having a different part of the "class". Notably there isn't even a company name or address registered for any of the "Windows" trademarks.
And how did Microsoft get a trademark when X Windows was already being sold in Sweden by a number of UNIX vendors? I don't know. Maybe X Windows wasn't trademarked in Sweden/Europe. Or the trademark holders failed to protect their trademark.
But you don't need to register a trademark to have one, you just have to use it. Registering just lets you sue someone else for using it assuming they didn't own it before you did, and puts others on notice that you already took that name. For instance, a long time ago I started a company called Bungee Software making DOS shareware, but I never registered the trademark so I don't have much standing to sue the game maker with that name. They probably didn't even know I existed, and I didn't know they existed early enough to get them to change their name. It's a PITA because, practically, I have to come up with a new name and a new logo to avoid confusion, even though I could legally continue to use the name since I was first to use it for software.
Lindows point is, that "Windows" is a generic term anyway No, in Sweden (Swedish) it is not.
So if they ever start selling Lindows in Sweden they can call it the Swedish name for windows... Sheesh, I don't like Lindows for a number of reasons but it isn't right to sue someone without their knowledge, how is this legal for Microsoft to do? And how did Microsoft get a trademark when X Windows was already being sold in Sweden by a number of UNIX vendors? Did X Windows somehow get diluted when the X consurtium allowed Microsoft to market "Microsoft Windows"? Would they have better off suing and losing against Microsoft way back when so that Microsoft couldn't do this to them now (Lindows = Linux + X Windows)?
Also when did Microsoft get a trademark in Sweden? They couldn't get one for years in the USA...it wasn't until they got a "pro-business" government that they got it here, since there were pre-existing competing products with the same name and it's been a generic term for Windowing systems since the 1950's amoung English speakers. I believe the inventor Ivan Sutherland is still alive even, shouldn't his company (Evans Sutherland, former graphics giant, now a military industrial complex supplier) get first bids on the trademark a la how the US courts gave Linus the Linux trademark after someone else registered it?
I don't know about project managers being more advanced than coders, but I am sure architects are more advanced than coders. SO, if the project manager is an architect, yes he is more advanced than the coder.
When I read the headline I thought you were referring to building architects. As a programmer dating an architect I see a lot of similarity in what we do. When she creates the overall look of the building that's like when I architect my program on sheets of paper. When she fills in the detail, and she does this to the point of telling the builder how to lay out the ceiling tiles that's like actual coding.
Now she tells me that the builder knows how to lay out the ceiling tiles correctly, but if she doesn't spec it he will do the easiest thing which is to just start in one corner and lay it out so that all the cut pieces are along two of the walls which looks like shit (you end up with little slivers along the edges instead of small cuts off the sides of tiles all around the edges of the ceiling.) I think this phenomina of everything going to the lowest bidder in programming will end up the same way, instead of our designs going to a slightly more expensive team consisting of beginner programmers with good coding skills and limited architecting skill, the jobs will go to lowest common denominator 'coders' with no inclination to learn programming. This will mean we will need to spend inordinate amounts of time doing detail work in our specs which now are done at the 'coding' level by qualified programmers.
(*I've put coding in quotes since when I think of a coder I think of someone performing black magic with a mix of assembler and intimate knowledge of RAM bank loading times and 4 way-cache associativity when dealing with 7 simultanious arrays that need to be permuted, matrix multiplied and tested against a semaphore on an average time of 2/9 cycles per element with a worst case of 3.21 bus cycles. Which may exist in an outsourcing outfit, but certainly a coder by my definition isn't available on the outsourcing market, and I would never consider my coder a 'low-skill' worker. More like a "shit he's more stock options than the fucking CEO" worker.)
My hope is that some of these outsourcing shops will actually hire programmers and basically replicate the model we've had here for the last couple decades, except "for hire" instead of as product based startups. That might just produce the kind of stable upper middle class jobs that many yearn for, in other countries sure. But after the DMCA and the creation of software patents, the legal climate in America is not condusive to in country programming anymore anyway. This might be a superior model too, we all know the types of programming done when creating a new product requires a radically different skill-set and attitude than maintaining a product once created. This type of outsourcing will let us specialize without the product creation programmers needing to bounce around from company to company. Investors will know to go to ABC shop to create the a product, XYZ shop to maintain it, and ACME to run the consumer 'tech-support/manual reading' services.
As a final warning though... building architects don't make much money at all, a first year NYC school teacher can make more money than a Harvard trained architect with the same level of schooling. If you don't love programming, now might be the time to look into fields that may pay a middle class wage a few years hence.
I don't want my library to be 'quaint'. I want them to have a good selection and low cost to taxpayers (me). I see this only as a benefit.
Just replace taxpayers with something else, like patrons maybe? Every library I've worked at got little and sometimes no government funding. When they one did get money it was to scan in some 400-500 year old books and destroy the originals.
Yep, I decided if I ever become wealthy enough I will start a library whose charter specifies it may not accept any government funding or special tax treatment and may not remain open to the public if the local government wishes to censor it's contents. I envision something where you pay monthly membership dues, like a gym, with higher dues for more borrowing privledges. And it would not be in a country like the USA that bans software lending.
How good is the engineering solution when it can be brought low by bean counters?
The easier it can be brought low by bean counters the better it is. Engineers are trained to create solutions that are absolutely the lowest cost solution that meets the specification. So with the perfectly engineered bridge if you spend a dime less it will fall into the sea. In real life you build in wide tolerances because it's cheaper than a perfect spec. Cutting that dime may just increase the odds of failure from 0.001% to 5%, so you will probably get away with it a few times. It's probably still unwise, but we can't expect business school graduates to reach the level of mathematical competence that your average English major does.
It seems more of a pride issue. The more the rest of the world says they should stop the more the old guard in japan, the same ones who see nothing wrong with japans war crimes, insist on keeping whaling alive.
Sounds like Iceland. Whale meat isn't particularly tasty, but it's traditional. Sort of like canabalism was in certain populations in the last century. Hopefully someday we will find new traditions. At least with pre-European American settlers it's not just tradition but also their great poverty that sends them out on the kill, maybe low tax cigarette sales will help with that problem.
Slashdot Mirror
Cookie recipe? (yes, i know the latter may best be covered by a copyright... but hell, should a copyright even be granted?)
FYI you can't copyright a food recipe in the USA. A recipe is a mere compilation of facts. Facts can only be patended. You can copyright a recipe if in a certain creative form, such as in your own handwriting with some doodles on the side. But this doesn't prevent anyone from publishing the recipe themselves after typesetting it themselves and rewriting any clever introductory text you may have added.
It was the[sic] 9th Circuit Federal court who usually do the right thing and then get overturned on appeal by Scalia and the Supremes.
Furthermore, the number of decisions out of the 9th Circuit that get overturned is high enough that the rest of the country rarely pays any attention to what they have to say, except possibly to determine what the law isn't (or won't be).
Actually they have if not the lowest then one of the lowest percentages of overturned rulings. But they do hear most of the important cases just by virtue of their territory. And the most important cases get heard by the SC. When they get overturned most people know they will ahead of time, given the biases of our current crop of nine, but the 9th circuit is also careful in these cases to write the type of strong opinions that law students will read in their classes and that they know will eventually become the law of the land.
I thought that SSE and MMX both had significantly lower precision than standard IEEE floating point ops. If I'm wrong, please correct me, but if it is lower precision, it makes it useless for Real Work(tm).
It performs precise math by default. You can only use 32 or 64 bit floats, the "long double" 80 bit floats are not supported. But this often isn't a problem. You can also turn off denormals, and with interupts on bad math (divide-by-zero type stuff). Turning those off hasn't given me any performance boost, but I still consider these things features not bugs. There are some low precision operations available, but no compiler I know of uses them unless you ask for em. I do in some cases but then I know what I'm getting.
A math person may give you a better answer than me. I'm a graphics person, a field where SSE2 is a godsend compared to the stack based floating point units that came before.
I actually told my prof about my concerns, and his only reply was to tell me to choose between taking the class (which I need to do) and running my papers through the system. I wish I had the balls to do what this kid in McGill did, but I want to go to grad school, and thus need to suck up to profs. Such is the magic of academia I suppose.
I think your chances of getting into the grad school of your choice will be significantly better if you fight this type of bullshit. If the prof really decides to fight you on this you will find more than a few professors on your side when you take this to the administration. And many of them will be happy to write you a glowing essay when you need one. Professors want grad students who will fight for what they believe in. Tilting at windmills is practically a hallmark of the profession.
I never understood the whole "implied binary digit" thing.
I have no idea what watch you are talking about. But if they count from 1 to 24 on for the hours, as opposed to 0 to 23 and they have a symbol for both 1 and 0, then the implied digit is simple. A number between 1 and 24 in binary always begins with a one, so you can leave it off. ''(1) '0'(2) '1'(3) '00'(4) '01'(5) etc etc
This is idea is also used to save a bit in IEEE floating point, where the special strings of all zeros and all zeros except for the sign bit signify +0 and -0 respectively, and all other real numbers begin with an implied 1 bit (with the exponent placing it in the right place).
Unless someone has come up with a full-spectrum LED, I don't think this kind of lighting is going to see wide adoption outside of perpetually sun-drenched areas, and then only as a "Gee Whiz" item because of the high cost.
They have, sort of. The "White LED" is actually a blue LED with set of phosphors to approximate white light. Basically it's the same as those flourescent fixtures you installed. Flourescents typically emit UV light (slightly higher energy than the blue LED in white LEDs.) and then use phosphors to step it down to a series of different lower energy frequencies. This step down is what makes them a little warm, the energy lost converting ultra-violet to red is lost as heat.
Of course, "White LED's" aren't super efficient. Stick with the flourescents for now. Also, not all white LED's are created the same, just like flourescent bulbs. There is a potential though for LEDs as full spectrum lighting, they come in many natural frequencies and combining those with a nice diffusor and maybe some phosphors to fill in the blanks could give you a relatively cheap to run "full spectrum light." LED's have advantages over flourescents, like you can change the color mixture simply, you can dim them, they are small and mechanically resilient They do have problems that we wouldn't think of off-hand, I know an architect that wanted to use some in a building and was prevented from doing so because 12 volt lamps can't be UL listed, and none of her suppliers were willing to wire 5 or more of them together and apply for the somewhat expensive UL listing. The NYC building inspector could not be convinced that a battery operated 12 volt light was perfectly safe without a UL listing.
Compare it the current versions of icc, butt monkey. icc still beats gcc by 2:1, if not more. Your comparison is meaningless. Its like comparing a 2004 Chevy to a Ford Model T and saying, "See, Chevy really is better than Ford!"
Umm, yeah, well icc 8.0 is the newest release. And I checked against several versions of gcc 3.2.3, 3.3.2 and the latest 3.4 from CVS. The 3.3.2 seemed to have the best performance overall, with 3.4 a close contender. Except the 3.4 had worse performance on a couple benchmarks, not unexpected for a pre-beta. icc was left far far behind, if you look at my posting history you will see I heaped praise on them in the 6.0 period, said nothing after the 7.0 release, and now let people know that with certain code gcc outperforms icc. If you have a legacy app maybe icc will do better, I don't know, with my application it is significantly slower.
icc also won't use my default libstdc++ which was a bit of a pain (limits used gcc builtins which icc doesn't have). I used the Dinkumware standard library for the benchmarks, but if I could have used the same sytem libraries as gcc I might have had the best of both worlds...
A professional-grade compiler like MIPSpro or ICC can generate code over twice as fast as GCC on the same processor, because it's smarter about processor-specifics.
gcc 3.3.2 beats the pants off icc 8.0 on my SSE2 code. Up to a 50:1 ratio on speed tests, 4:1 on average. With earlier revisions of gcc and icc the ratio was 2:1 with icc being faster. This code is written with explicit parallelism so all the fancy loop unrolling icc does doesn't help, and the register allocation algorithm in gcc seems to be the thing giving it the advantage, icc spills to memory on simple matrix vector multiplies. I think it just validates the old maxim about optimization, trust nothing, test everything.
Basically, with given energy inputs, pure hydrogen combining with oxygen give you the most "bang for your buck" since it only requires a small amount of activation energy. Hydrocarbons are more contained, less volitale (that's why we use them in cars instead of hydrogen gas for the 'combustion'), and the emitted energy from the reaction is less.
Sure hydrogen gives you the most bang per weight. But remember it's also the lightest substance around. Since we live in an atmosphere, and live with a significant amount of gravity the hydrogen doesn't just burn hot, it burns away from your body faster than any other matter in the universe would. If there is a continuous pointed at you leak, then like a blowtorch it's not so good to be near. If we are talking about a container that undergoes some kind of catastorpic failure while near my body, I would rather it be hydrogen than gasoline. I don't want a substance infiltrating my clothing and burning slowly as it evaporates.
(More "bang for your buck" is a little inaccurate since it is still pretty expensive compared to a mined fuel, which is why we use gas instead of the cleaner burning hydrogen.)
Yes, napalm has other things than gasoline. If you notice polystyrene is a longer more stable molecule, yet napalm is a pretty nasty substance. It's the sticking around longer that makes it's flame more dangerous. Same as with gasolene vs. hydrogen. There are two things people worry about with hydrogen, a non-flame rupture of the container since when it is used as a fuel it is under pressure and it will explode. Then there is the fire that can follow, and I think here it is important to realize that the fire will envelop you only for a second or so in an open environment, and will die out quickly in a closed one, unlike gasoline, or any of the heavier slower burning hydrocarbon fuels.
Hydrogen is more dangerous than gesoline. Much much more dangerous.
Huh? Hydrogen quickly moves up and away from you while gasoline sticks to you. You did see the pictures of those babies our boys burned alive with napalm in the 70's?
A lot of big companies still use COBOL, and COBOL is over fifty years old, which means we can keep using X even if it's not being developed anymore.
Yeah COBOL. COBOL reminds me of high school. I had to get special permission to not take COBOL before Pascal (which the AP exam was in.) I'm guessing there are a lot of people who weren't as stubborn as I and have some knowledge of it. Nothing wrong with that, the next year I learned C at the local community college, but Turbo Pascal 3 was still more fun, it compiled as fast as assembler and ran faster to boot. I'm sure COBOL has it's own advantages.
Still, I hope XFree86 doesn't go the way of COBOL. I think this announcement may just mean that it won't. The Linux development model is just so much more robust for an important project like X. I may even contribute to X now, in the bad old days I could even figure out to get a bug report into the system.
BTW I have only been involved in 2 accidents involving other people each were under 10 mph bumper dings and over 10 years ago I have been driving for 12 years.
Hehe, I don't consider myself a good driver and haven't been in any "accidents," involving people or not. Not in the last 12 years, not ever.
Not that I'm perfect, I've been known to go 30 mph over the speed limit on a highway, I've forgotten the turn signal a few times (not in the last decade) when driving in farm country, and once when I was 17 I ran a red light because I thought I could make the yellow and I couldn't. After that incident at 17 I never drove tired again, my life is worth more than that. I think maybe you should buy a stick shift without any fancy anti-lock breaks and learn to drive again, maybe turn off the radio so you can concentrate better.
If i recall the 'stacker' case correctly. MS and Stacker had a contract to put Stackers tech into DOS. Then when MS sold a bunch of software with their stuff in it and people stopped buying stacker software.
Well sort of, as I remember it they couldn't come to terms on the contract and so the sale was called off. But MS hired some of their lead coders with seven digit signing bonuses and MS claimed it didn't know they had taken the Stacker code with them, but the judge seemed to think they should have at least looked considering the timeframe in which these coders incorporated the feature. And it was trivial to find this copyright infringement even without looking at the source. In the end it was settled out of court with a 9 digit number, so the ugly details aren't all public. My guess is the settlement number is something like the sale number Stacker wanted, it was a large number because they knew selling it to MS was the end of their company it was essentially a company sale they were negotiating. Microsoft is known for making silly offers, less than the investor's initial investments, and backing it up with the threat of taking billion dollar losses killing the little company out of some sense of spite mixed with their need to set an example of ruthlessness to would be future negotiating partners.
So, you have proof that MS has been involved in "technology theft" then?
Either you're wrong in that they're successfully covering up after themselves, or you're wrong in that they're doing it. Either way, you're talking crap.
Woah there partner, you haven't been paying attention. Search your old Windows 3.1 executable for "Stacker"... or google. You might also want to look into some of the other settlements, like the one in France last year. A lot of them involved some very nasty unethical stuff, much of it under the category of theft. You could also buy some drinks for someone you know that worked for a company targeted for ruin by Microsoft, a few hours later you'll not want to partner with it ever again.
It relies on memory latency so it really doesn't take up resources from your system. It appears the only disadvantage is distributed mailing and progress of technology.
Memory latency has been the biggest bottleneck on PC's for the last 5-10 years. So yes it eats up gobs of resources. I haven't seen the actual algorithm but it may be possible that this is completely defeated by a special purpose machine with gobs of memory bandwidth, such as a Playstation 2. That being said my posting history will show I'm a big fan of hash cash. What it will do is force mass e-mailers to either spend money on computers, lowering their profit, or break the law by taking over millions of zombie machines. Lowering their profit will make this a less attractive business to get into, and breaking the law will make them liable to run ins with local police in their country of residence. Both should reduce the number of spams. They may constuct Amway type schemes where they get people to install programs willingly much like Gator and the like work now. But unless they pay these victims, lowering profit, they will likely be prosecuted for fraud. Prosecutors may ignore Gator which just creates problems for ordinary people, but they won't be so happy with hundreds of thousands of unwanted e-mail in their inbox everyday, nor the bill for 3rd party filtering services they shouldn't need.
Well, yes but what happens when the ISP's network is flooded with worm traffic? They really don't have much choice.
Wouldn't it be much better to just disable the ports where virus floods are coming from and have an auto-dialer call up the customer and tell them their computer is infected, giving them a phone number to call once the system is fixed? Then they would be aware of their problem and probably take some more measures in the future to prevent it...
Why is the judge letting SCO get away with this coy "they infringed, but we're not going to show you how yet" stuff? Why hasn't he said, "Put the infringing source code in a brief and hand it over tomorrow, or I'm tossing this"?
The judge has done this. Except they have thirty (30) days to do it. The clock is ticking, it's just ticking very very slowly since this is the US legal system we're dealing with. If their evidence is anything like what they have shown to date even a non-technically literate judge might just issue a summary judgement against them with prejudice. After which they would have a hard time bringing up any more copyright infringement libel against anyone before a US court in the future.
I'm starting to doubt this will even see a trial, and not because I see a settlement as happening. The Boise team must be kicking themselves over this. The damage to their reputation is going to cost them a lot more than the few million they can collect in lawyer fees on this turkey. Someone screewed up, someone thought IBM would just buy the company to save themselves some cash in the short term. Instead they are now going to have to show up in court and enjoy some judge yelling at them for wasting everyone's time.
To answer the question: In the U.S., per FAA regulations, landowners have control up to 500 feet. Above that is public domain airspace. It has been this way since very early on in aviation, since it is critical to have public airspace and federal jurisdiction thereof in order to have a viable air transportation system.
Huh? is that 500ft above your building? Say your building is 800ft tall...
Not that this matters much, in most states you can pretty much assume that anything you say or do outside your 12 ft thick lead walls can be observed legally these days. At least the police can't use extrardinary measures, like thermal imaging, in court; but once that becomes a ordinary, look out.
Not to mention Iceland. Its legislature, the Althing, has been meeting since CE 930 -- nearly 1100 years.
When growing up in Iceland I used to think of America's strengths and problems stemming from it being one of the youngest countries around. It was a total shock when I moved here and heard it's citizen's speak of America as the world's oldest democracy. If you put a bunch of caveats on the phrase it can be interpreted as having some truth, but this foolish. America is America because it is a young country. One without history or language or religion, and so it functions a beautiful microcosm of the world. Our government is a stitched together hodge-podge of democracy and totalitarianism that almost never displays any signs of conscious thought being applied to a problem; this can be incredibly frustrating and even scary at times, but also gives us that lawlessness countries like Iceland lack, this can be really liberating when the laws are insane or just innane or rather silly.
You can check the web page for patent och registreringsverket which handles trademarks.
It looks like it's a set of EU trademarks not a Swedish one, it was probably granted on the basis of the US trademark and assuming Lindows wins in the US by invalidating the "Windows" trademark, it will be vacated on an EU wide basis. It seems to be registered to three entities each having a different part of the "class". Notably there isn't even a company name or address registered for any of the "Windows" trademarks.
And how did Microsoft get a trademark when X Windows was already being sold in Sweden by a number of UNIX vendors?
I don't know. Maybe X Windows wasn't trademarked in Sweden/Europe. Or the trademark holders failed to protect their trademark.
But you don't need to register a trademark to have one, you just have to use it. Registering just lets you sue someone else for using it assuming they didn't own it before you did, and puts others on notice that you already took that name. For instance, a long time ago I started a company called Bungee Software making DOS shareware, but I never registered the trademark so I don't have much standing to sue the game maker with that name. They probably didn't even know I existed, and I didn't know they existed early enough to get them to change their name. It's a PITA because, practically, I have to come up with a new name and a new logo to avoid confusion, even though I could legally continue to use the name since I was first to use it for software.
Lindows point is, that "Windows" is a generic term anyway
No, in Sweden (Swedish) it is not.
So if they ever start selling Lindows in Sweden they can call it the Swedish name for windows... Sheesh, I don't like Lindows for a number of reasons but it isn't right to sue someone without their knowledge, how is this legal for Microsoft to do? And how did Microsoft get a trademark when X Windows was already being sold in Sweden by a number of UNIX vendors? Did X Windows somehow get diluted when the X consurtium allowed Microsoft to market "Microsoft Windows"? Would they have better off suing and losing against Microsoft way back when so that Microsoft couldn't do this to them now (Lindows = Linux + X Windows)?
Also when did Microsoft get a trademark in Sweden? They couldn't get one for years in the USA...it wasn't until they got a "pro-business" government that they got it here, since there were pre-existing competing products with the same name and it's been a generic term for Windowing systems since the 1950's amoung English speakers. I believe the inventor Ivan Sutherland is still alive even, shouldn't his company (Evans Sutherland, former graphics giant, now a military industrial complex supplier) get first bids on the trademark a la how the US courts gave Linus the Linux trademark after someone else registered it?
I don't know about project managers being more advanced than coders, but I am sure architects are more advanced than coders. SO, if the project manager is an architect, yes he is more advanced than the coder.
When I read the headline I thought you were referring to building architects. As a programmer dating an architect I see a lot of similarity in what we do. When she creates the overall look of the building that's like when I architect my program on sheets of paper. When she fills in the detail, and she does this to the point of telling the builder how to lay out the ceiling tiles that's like actual coding.
Now she tells me that the builder knows how to lay out the ceiling tiles correctly, but if she doesn't spec it he will do the easiest thing which is to just start in one corner and lay it out so that all the cut pieces are along two of the walls which looks like shit (you end up with little slivers along the edges instead of small cuts off the sides of tiles all around the edges of the ceiling.) I think this phenomina of everything going to the lowest bidder in programming will end up the same way, instead of our designs going to a slightly more expensive team consisting of beginner programmers with good coding skills and limited architecting skill, the jobs will go to lowest common denominator 'coders' with no inclination to learn programming. This will mean we will need to spend inordinate amounts of time doing detail work in our specs which now are done at the 'coding' level by qualified programmers.
(*I've put coding in quotes since when I think of a coder I think of someone performing black magic with a mix of assembler and intimate knowledge of RAM bank loading times and 4 way-cache associativity when dealing with 7 simultanious arrays that need to be permuted, matrix multiplied and tested against a semaphore on an average time of 2/9 cycles per element with a worst case of 3.21 bus cycles. Which may exist in an outsourcing outfit, but certainly a coder by my definition isn't available on the outsourcing market, and I would never consider my coder a 'low-skill' worker. More like a "shit he's more stock options than the fucking CEO" worker.)
My hope is that some of these outsourcing shops will actually hire programmers and basically replicate the model we've had here for the last couple decades, except "for hire" instead of as product based startups. That might just produce the kind of stable upper middle class jobs that many yearn for, in other countries sure. But after the DMCA and the creation of software patents, the legal climate in America is not condusive to in country programming anymore anyway. This might be a superior model too, we all know the types of programming done when creating a new product requires a radically different skill-set and attitude than maintaining a product once created. This type of outsourcing will let us specialize without the product creation programmers needing to bounce around from company to company. Investors will know to go to ABC shop to create the a product, XYZ shop to maintain it, and ACME to run the consumer 'tech-support/manual reading' services.
As a final warning though... building architects don't make much money at all, a first year NYC school teacher can make more money than a Harvard trained architect with the same level of schooling. If you don't love programming, now might be the time to look into fields that may pay a middle class wage a few years hence.
I don't want my library to be 'quaint'. I want them to have a good selection and low cost to taxpayers (me). I see this only as a benefit.
Just replace taxpayers with something else, like patrons maybe? Every library I've worked at got little and sometimes no government funding. When they one did get money it was to scan in some 400-500 year old books and destroy the originals.
Yep, I decided if I ever become wealthy enough I will start a library whose charter specifies it may not accept any government funding or special tax treatment and may not remain open to the public if the local government wishes to censor it's contents. I envision something where you pay monthly membership dues, like a gym, with higher dues for more borrowing privledges. And it would not be in a country like the USA that bans software lending.
How good is the engineering solution when it can be brought low by bean counters?
The easier it can be brought low by bean counters the better it is. Engineers are trained to create solutions that are absolutely the lowest cost solution that meets the specification. So with the perfectly engineered bridge if you spend a dime less it will fall into the sea. In real life you build in wide tolerances because it's cheaper than a perfect spec. Cutting that dime may just increase the odds of failure from 0.001% to 5%, so you will probably get away with it a few times. It's probably still unwise, but we can't expect business school graduates to reach the level of mathematical competence that your average English major does.
It seems more of a pride issue. The more the rest of the world says they should stop the more the old guard in japan, the same ones who see nothing wrong with japans war crimes, insist on keeping whaling alive.
Sounds like Iceland. Whale meat isn't particularly tasty, but it's traditional. Sort of like canabalism was in certain populations in the last century. Hopefully someday we will find new traditions. At least with pre-European American settlers it's not just tradition but also their great poverty that sends them out on the kill, maybe low tax cigarette sales will help with that problem.