In actuality, most FAA regs are to protect (a) people (and property) on the ground and (b) passengers. They don't really care much if a pilot kills himself (or herself -- although most of the female pilots I've known were a little less reckless than the males) as long as he doesn't hurt anyone else. (Unless, of course, it was a commercially built (vs homebuilt) aircraft at fault. And then they're still more concerned with the other folks who might get hurt by similar.)
You cannot patent an idea, only the implementation of an idea.
You're confusing patent and copyright law there. You cannot copyright an idea, only the expression of the idea. Patents are all about ideas, as in methods, processes and apparatus. Your points about non-obviousness and originality are correct, which weeds out most "pure idea" patents. (Although in the software and business process fields we can come up with too many counter-examples).
Classic examples of patenting an idea rather than the expression abound in the drug field, where a patent can be obtained for treating disease X with drug Y that was initially developed for treating disease Z. (This may actually make some sense, since the drugco has to go through expensive trials to get FDA approval for treating disease X with drug Y, even if drug Y has proven harmless in years of using it for disease Z. Sigh.)
Cutting down on the tumbling when it hits also eliminates the argument that some have made that the.223 violated the spirit of the Geneva Convention banning dum-dum rounds, even if technically legal.
(Note that most police ammunition violates that spirit -- the cops want hollowpoint so that the bullet tends to stop when it hits something, rather than travelling on through and endangering bystanders.)
Perhaps a more compelling reason for retaining/etc,/usr,/var, and so on is that they're not language-specific. Oh, sure, they're vaguely english-derived, but short names are going to be a heck of a lot easier for some non-anglophone to remember when editing his french, finnish, or turkish distro than "/programs" or "/settings" (or is that last "/configuration" or "/parameters" or "/registry"?).
Nobody's telling new users to "shut up and use the GUI", they're telling whiners who complain about/etc and/usr to shut up a write a GUI. New users can RTFM like everyone else -- and as they'd have to figure out what they're doing with the config files, whether they're in/etc or/settings -- or/paramètres or/asetukset or/ayarlar, for that matter.
Even though/usr,/var,/tmp,/etc are ridiculously cryptic, changing them would be horrible?
Yes, and unnecessary.
Anyone with the knowledge to be messing around with those directories (other than via some pointy-clicky GUI) will have no problem remembering their names. There's only a half-dozen of them, for pete's sake.
If you don't like the names, create pointy-clicky GUIs. They're wonderful for things like adjusting settings, and they don't care what the directory name is.
R[edhat]PM is pretty crap how it dumps all your apps into/usr or/usr/bin without any thought.
Allow me to quote the Linux Standard Base Specification 1.2, Chapter 18, File System Hierarchy (The Filesystem Hierarchy Standard, Version 2.2, Section 3.12.1) "/opt is reserved for the installation of add-on application software packages." [emphasis added]
Also "No other package files may exist outside the/opt,/var/opt, and/etc/opt hierarchies except for those package files that must reside in specific locations within the filesystem tree in order to function properly. For example, device lock files must be placed in/var/lock and devices must be located in/dev."
Minor nit-pick: if you're using an LCD screen, the image may never go to D/A converters. It might be a bit tricky to decode that interface back into text -- but probably easier than the tempest tricks that let you duplicate a CRT image by analyzing its RF emissions. (Which, of course, would be another way of taking a copy.)
I believe that Sun published a reference version of NFS source, but didn't open source their optimized internal version until much later.
I'd also add NIS (aka yellow pages) to the list of Sun's notable contributions. That made NFS a whole lot more useful. (You really don't want the situation where user 'fred' has different UIDs on two machines both NFS-mounting/home (or/usr/users, back when).)
Nope. At least not where I learned to drive (Ontario). On the driving test, turning left at a light, I stopped before entering the intersection. The examiner told me to pull out far enough to block the cross traffic so that I could proceed when the oncoming traffic stopped. (He didn't flunk me for not doing that in the first place.) The point being that with heavy oncoming traffic (and no left turn arrow on the signals), if you don't edge out into the intersection you might be sitting waiting to turn left for a long time (many cycles).
However, you should only complete the turn when it's clear, i.e. you don't want to be blocking the oncoming traffic lanes.
I suppose it comes down to the exact wording of the traffic regs in each state/province, and where in the intersection you stop your car.
Last night (circa 9PM MDT) I could only reach sites via Level3 or ATT (via L3), according to traceroute, for about an hour or so. Nothing else got anywhere at all, so it might have been local ISP problem.
Apparently 'the testing companies make money by certifying products, not catching problems' thus implying that the seal of approval might not mean a whole lot."
Sounds familiar. Oh yeah -- the US Patent Office makes its money off of "user fees", and it collects a bunch more when it issues a patent then when it denies one.
Hmm...
(The real answer may be in just charging for the application (either for certification or for patent, depending on whom we're talking about), and the cert is issued (or not) without extra fee. I say "may" be the answer because that approach might encourage agencies to trivially deny applications to boost re-application fee income.)
Iridium failed because of the nightmare maze of regulation and red tape they had to fight through.
Consider, an Iridium phone is usable from potentially any place on the planet (minor technical exceptions aside). That means they had to deal with every radio frequency and telco regulating agency on the planet, practically. And everybody wants their cut, er, license fees and tarriffs. (Not to mention the sheer nightmare of wading through all those application processes, and the lawyers' fees involved.)
It was a great idea, but government bureaucracy killed it just as much (or more) as any internal politics.
a dream of mine is to move up north to Muskoka, Ontario, Canada and live there permanently
Hey, there's a reason that Muskoka is cottage country: it's only habitable in summer. (And that's debatable, depending on how tasty you are to mosquitoes.) You want to live there in winter, too?
(Just kidding. I spent half my life in various places in southern Ontario. Yeah, there's some pretty country up there. Parts of the year, anyway;-)
If you really wanted to, you could probably do it with just one hard drive if you partitioned it enough. (I dunno, is there a limit to the number of logical partitions Windows will recognize?)
This wouldn't be the first time Intel has screwed the pooch. Do a google on "iapx432" for something that turned out possibly even worse than Itanium might. (It was a nice chip design, on paper -- and it eventually met all its design goals, of which performance was not one, alas.)
The reality, as you point out, is that the people in China is mainly poor, and don't have the ability to own a computer.
True, but misleading. Factor in China's huge population, and even though most Chinese may fall into the above category, you're still left with a middle class population of a couple of hundred million. There are as many potential customers for these sorts of goods in China as there are, say, in the US or Europe.
(And that's why American companies are so eager to get into Chinese markets.)
MIPS makes sense if they're also looking at the consumer electronics market (which they should be -- the volume demands there is pretty high). A lot of the consumer electronics (eg DVD players) already made in China use chips with MIPS cores. The architecture seems suited to video apps for some reason (hello SGI).
Yeah, Blender is cool. It does help if you have some kind of manual to get started, though. ("The Blender Book" -- Linux Journal Press/No Starch Press -- worked for me) Blender (nor, I imagine, any similar program) doesn't lend itself well to learning by just playing around with it.
There is this fantastically common misconception that centralising your various digital identities will somehow decrease security. Not true!
Absolutely true. The annals of computer crime are full of cases where crackers have accessed systems B, C, D and E by harvesting passwords from system A and users re-used the same password on those other systems. Now true, if those other systems had some other gaping hole that would let them be compromised without a password, then in some theoretical absolute sense the security isn't any less because of the shared password (since there was no real security to start with), but such holes are bugs and fixable by the sysadmin, whereas shared passwords are not.
Single sign-on, whether Passport or Liberty Alliance, seems like a disaster waiting to happen, although if properly designed and correctly implemented (bloody big "if"), it'd be safer than multiple sign-ons all using the same password (because the latter gives multiple points of attack). But it's also painting a huge target and sign on itself that says "crack me!". And it's still less-safe than multiple sign-on with different passwords. (Think about it -- if you're a big-time crook (or terrorist, etc), do you go for the high-stakes bank job, or just stick up a string of 7-11s? It all comes down to effort vs payoff.)
This has got to be the most irritating pending issue for linux to make it on the desktop. Even if CUPS can see the printer, that doesn't mean it's going to actually print anything.
And this differs from Windows how?
I just upgraded my wife's machine. She runs NT (some Windows-only apps). It promptly decided it could no longer see the laser printer on the network. At least, the print software decided that. I can ping the printer. I can log into its admin software. But but the Windows software refuses to acknowledge it's existence. I finally just gave up on that crap and connected it via a parallel cable (the printer's in her office). The Linux boxes and Macs in the house have no problem with it over the net.
Care to point out the ancient nuclear weapon? Or the ancient use of velcro? Radar? Full plate armor before the fall or Rome?
Well, the first one might be a bit tricky, but the other three all have analogs in biology: burrs, bat/dolphin sonar, and any critter with an exoskeleton.
I agree with your conclusion, you just happened to pick bad examples.
You're quite right, and in fact this predates Linux and NT -- Unix was always good at process creation, whereas VMS process startup was very heavy on overhead.
It's not surprising that Linux (modelled on Unix) and NT (originally modelled on VMS) show similar characteristics. It's the reason that many Unix applications tend to be written as a bunch of cooperating processes, whereas NT apps are monolithic monsters with lots of threads.
Unfortunately, thanks to a generation of CS students having learned bad habits on Windows, we're starting to see a lot of Linux apps written as monolithic monsters. (Of course there are few old Unix apps out there like that too, perhaps some old mainframe mentality leaking through.) There are advantages to cooperating/communicating processes vs the monolithic multithreaded approach: it's easier to test the components separately, it's easier to reuse the components to make different systems, and a bug in one place won't necessarily clobber the whole thing.
Slashdot Mirror
... and see how well she does ...
Just think of it as evolution in action.
In actuality, most FAA regs are to protect (a) people (and property) on the ground and (b) passengers. They don't really care much if a pilot kills himself (or herself -- although most of the female pilots I've known were a little less reckless than the males) as long as he doesn't hurt anyone else. (Unless, of course, it was a commercially built (vs homebuilt) aircraft at fault. And then they're still more concerned with the other folks who might get hurt by similar.)
You cannot patent an idea, only the implementation of an idea.
You're confusing patent and copyright law there. You cannot copyright an idea, only the expression of the idea. Patents are all about ideas, as in methods, processes and apparatus. Your points about non-obviousness and originality are correct, which weeds out most "pure idea" patents. (Although in the software and business process fields we can come up with too many counter-examples).
Classic examples of patenting an idea rather than the expression abound in the drug field, where a patent can be obtained for treating disease X with drug Y that was initially developed for treating disease Z. (This may actually make some sense, since the drugco has to go through expensive trials to get FDA approval for treating disease X with drug Y, even if drug Y has proven harmless in years of using it for disease Z. Sigh.)
Cutting down on the tumbling when it hits also eliminates the argument that some have made that the .223 violated the spirit of the Geneva Convention banning dum-dum rounds, even if technically legal.
(Note that most police ammunition violates that spirit -- the cops want hollowpoint so that the bullet tends to stop when it hits something, rather than travelling on through and endangering bystanders.)
Nobody's telling new users to "shut up and use the GUI", they're telling whiners who complain about /etc and /usr to shut up a write a GUI. New users can RTFM like everyone else -- and as they'd have to figure out what they're doing with the config files, whether they're in /etc or /settings -- or /paramètres or /asetukset or /ayarlar, for that matter.
Even though /usr, /var, /tmp, /etc are ridiculously cryptic, changing them would be horrible?
Yes, and unnecessary.
Anyone with the knowledge to be messing around with those directories (other than via some pointy-clicky GUI) will have no problem remembering their names. There's only a half-dozen of them, for pete's sake.
If you don't like the names, create pointy-clicky GUIs. They're wonderful for things like adjusting settings, and they don't care what the directory name is.
R[edhat]PM is pretty crap how it dumps all your apps into /usr or /usr/bin without any thought.
/opt, /var/opt, and /etc/opt hierarchies except for those package files that must reside in specific locations within the filesystem tree in order to function properly. For example, device lock files must be placed in /var/lock and devices must be located in /dev."
Allow me to quote the Linux Standard Base Specification 1.2, Chapter 18, File System Hierarchy (The Filesystem Hierarchy Standard, Version 2.2, Section 3.12.1) "/opt is reserved for the installation of add-on application software packages." [emphasis added]
Also "No other package files may exist outside the
Minor nit-pick: if you're using an LCD screen, the image may never go to D/A converters. It might be a bit tricky to decode that interface back into text -- but probably easier than the tempest tricks that let you duplicate a CRT image by analyzing its RF emissions. (Which, of course, would be another way of taking a copy.)
I believe that Sun published a reference version of NFS source, but didn't open source their optimized internal version until much later.
/home (or /usr/users, back when).)
I'd also add NIS (aka yellow pages) to the list of Sun's notable contributions. That made NFS a whole lot more useful. (You really don't want the situation where user 'fred' has different UIDs on two machines both NFS-mounting
Nope. At least not where I learned to drive (Ontario). On the driving test, turning left at a light, I stopped before entering the intersection. The examiner told me to pull out far enough to block the cross traffic so that I could proceed when the oncoming traffic stopped. (He didn't flunk me for not doing that in the first place.) The point being that with heavy oncoming traffic (and no left turn arrow on the signals), if you don't edge out into the intersection you might be sitting waiting to turn left for a long time (many cycles).
However, you should only complete the turn when it's clear, i.e. you don't want to be blocking the oncoming traffic lanes.
I suppose it comes down to the exact wording of the traffic regs in each state/province, and where in the intersection you stop your car.
Last night (circa 9PM MDT) I could only reach sites via Level3 or ATT (via L3), according to traceroute, for about an hour or so. Nothing else got anywhere at all, so it might have been local ISP problem.
Apparently 'the testing companies make money by certifying products, not catching problems' thus implying that the seal of approval might not mean a whole lot."
Sounds familiar. Oh yeah -- the US Patent Office makes its money off of "user fees", and it collects a bunch more when it issues a patent then when it denies one.
Hmm...
(The real answer may be in just charging for the application (either for certification or for patent, depending on whom we're talking about), and the cert is issued (or not) without extra fee. I say "may" be the answer because that approach might encourage agencies to trivially deny applications to boost re-application fee income.)
Iridium failed because of the nightmare maze of regulation and red tape they had to fight through.
Consider, an Iridium phone is usable from potentially any place on the planet (minor technical exceptions aside). That means they had to deal with every radio frequency and telco regulating agency on the planet, practically. And everybody wants their cut, er, license fees and tarriffs. (Not to mention the sheer nightmare of wading through all those application processes, and the lawyers' fees involved.)
It was a great idea, but government bureaucracy killed it just as much (or more) as any internal politics.
a dream of mine is to move up north to Muskoka, Ontario, Canada and live there permanently
;-)
Hey, there's a reason that Muskoka is cottage country: it's only habitable in summer. (And that's debatable, depending on how tasty you are to mosquitoes.) You want to live there in winter, too?
(Just kidding. I spent half my life in various places in southern Ontario. Yeah, there's some pretty country up there. Parts of the year, anyway
If you really wanted to, you could probably do it with just one hard drive if you partitioned it enough. (I dunno, is there a limit to the number of logical partitions Windows will recognize?)
Like owning anything from micro, soft does?
This wouldn't be the first time Intel has screwed the pooch. Do a google on "iapx432" for something that turned out possibly even worse than Itanium might. (It was a nice chip design, on paper -- and it eventually met all its design goals, of which performance was not one, alas.)
The reality, as you point out, is that the people in China is mainly poor, and don't have the ability to own a computer.
True, but misleading. Factor in China's huge population, and even though most Chinese may fall into the above category, you're still left with a middle class population of a couple of hundred million. There are as many potential customers for these sorts of goods in China as there are, say, in the US or Europe.
(And that's why American companies are so eager to get into Chinese markets.)
MIPS makes sense if they're also looking at the consumer electronics market (which they should be -- the volume demands there is pretty high). A lot of the consumer electronics (eg DVD players) already made in China use chips with MIPS cores. The architecture seems suited to video apps for some reason (hello SGI).
Yeah, Blender is cool. It does help if you have some kind of manual to get started, though. ("The Blender Book" -- Linux Journal Press/No Starch Press -- worked for me) Blender (nor, I imagine, any similar program) doesn't lend itself well to learning by just playing around with it.
A great little multiplatform Postscript and PDF (Acrobat) viewer.
Here's the Ghostscript, home page, and the GSview-specific page.
You forgot your tags.
(Or you forgot that slash would eat them if unescaped.)
There is this fantastically common misconception that centralising your various digital identities will somehow decrease security. Not true!
Absolutely true. The annals of computer crime are full of cases where crackers have accessed systems B, C, D and E by harvesting passwords from system A and users re-used the same password on those other systems. Now true, if those other systems had some other gaping hole that would let them be compromised without a password, then in some theoretical absolute sense the security isn't any less because of the shared password (since there was no real security to start with), but such holes are bugs and fixable by the sysadmin, whereas shared passwords are not.
Single sign-on, whether Passport or Liberty Alliance, seems like a disaster waiting to happen, although if properly designed and correctly implemented (bloody big "if"), it'd be safer than multiple sign-ons all using the same password (because the latter gives multiple points of attack). But it's also painting a huge target and sign on itself that says "crack me!". And it's still less-safe than multiple sign-on with different passwords. (Think about it -- if you're a big-time crook (or terrorist, etc), do you go for the high-stakes bank job, or just stick up a string of 7-11s? It all comes down to effort vs payoff.)
This has got to be the most irritating pending issue for linux to make it on the desktop. Even if CUPS can see the printer, that doesn't mean it's going to actually print anything.
And this differs from Windows how?
I just upgraded my wife's machine. She runs NT (some Windows-only apps). It promptly decided it could no longer see the laser printer on the network. At least, the print software decided that. I can ping the printer. I can log into its admin software. But but the Windows software refuses to acknowledge it's existence. I finally just gave up on that crap and connected it via a parallel cable (the printer's in her office). The Linux boxes and Macs in the house have no problem with it over the net.
Care to point out the ancient nuclear weapon? Or the ancient use of velcro? Radar? Full plate armor before the fall or Rome?
Well, the first one might be a bit tricky, but the other three all have analogs in biology: burrs, bat/dolphin sonar, and any critter with an exoskeleton.
I agree with your conclusion, you just happened to pick bad examples.
You're quite right, and in fact this predates Linux and NT -- Unix was always good at process creation, whereas VMS process startup was very heavy on overhead.
It's not surprising that Linux (modelled on Unix) and NT (originally modelled on VMS) show similar characteristics. It's the reason that many Unix applications tend to be written as a bunch of cooperating processes, whereas NT apps are monolithic monsters with lots of threads.
Unfortunately, thanks to a generation of CS students having learned bad habits on Windows, we're starting to see a lot of Linux apps written as monolithic monsters. (Of course there are few old Unix apps out there like that too, perhaps some old mainframe mentality leaking through.) There are advantages to cooperating/communicating processes vs the monolithic multithreaded approach: it's easier to test the components separately, it's easier to reuse the components to make different systems, and a bug in one place won't necessarily clobber the whole thing.