There's nothing wrong with cryptographic signing, nothing at all. In fact, it would be a pretty good thing for precisely the reasons you gave, and I would even extend it to banknotes.
But what I strongly object to is contactless transmission, including any kind of RFID.
Nowadays everybody and his dog can read out RFID chips. They don't have to decipher it, they don't have to forge it - it's bad enough they can read it. It's just none of their business! Back in 1890, the later US Supreme Court Justice Louis Brandeis said, privacy is "the right to be left alone." I couldn't agree more.
Your concept doesn't even require contactless transmission. Let the card have some smart card-like readout pins. At the airport, have the security guard put the ID card in some card reader, and pronto your hi-res photograph complete with signature validation is on the screen, the guard looks at it, looks at your face, and returns the card. Pretty painless, uh?
And in fact even more secure: the check-in lines are long at airports, so lots of people with their ID cards near the readout point. Makes RFID more difficult (though not impossible, just a bit more error-prone). No problem at all with contact-based readout.
Following the idea of RFID-signed banknotes, the drawbacks of contactless transmission become obvious: a criminal can check you for money by simply pointing an RFID reader in your direction. He knows in advance that you're worth a shot, figuratively speaking. And he'll find the money in your money belt, too. Stupid, stupid idea!
(And this example also demonstrates that the need for privacy has nothing to do with this "I have nothing to hide" bullshit simple-minded people often give you. The amount of money I'm carrying falls by no means in the "nothing to hide" category, nor is it embarrassing, or sinister, or something. It's just not your business. Period.)
Have it contact-based and all those severe security and privacy problems disappear.
Everybody who works for MS has to shoulder the blame/credit for everything MS does.
Riiiight. Just as some lowly developer is responsible for (or has any influence on) an upper level manager who is bullying vendors and/or competitors with next-to-illegal means.
Or how about some G.I. serving for example in Ramstein who hasn't fired a shot in anger for months. Should he shoulder the blame for a few evil freaks torturing Iraqi army regulars? Hell no.
Insighful my asshole. You cannot blame someone for something (s)he's no influence on.
Anyone else noticed that the "Abort" button changes place, from the right to the middle to the left, where it is most often and which is - the reference being Windows - the most unexpected place.
The point is, "Abort" is the safe choice. You maneuvered yourself into some arcane dialog and don't understand nothin' - you hit "Abort" a.k.a. get me the hell out of here. This emergency button (as in, "Oh shit I'm gonna wreck by document") should always be in the same place.
Now when you align the buttons to the lower right (which the Gnome dialogs do), then this place is of course just there, in the lower right corner. Unlike Gnome, Windows got this right.
My favorite is the property screenshot. Instant inconsistency. (Shakes head and leaves.)
yep, they just dont build things the way they used to
The Liberty ships were designed with one goal in mind: build ships faster than the German Uboat force could sink them. And they succeeded! The Liberty ships were assembled (from pre-manufractured components) by mostly unskilled labour on the shipyards of Henry J. Kaiser within only 80 hours! On these shipyards, 140 Liberty ships per month would be completed.
The Liberty ships were never built to last. Their quality was rather poor. Definately not up to todays standards in shipbuilding.
Anyone else notice that there is a little footer with teh "recycled" symbol and the phrase "printed on recycled paper" ? it's a PDF. what happens if i print it out on non recycled paper?
Everyone bitches and moans about systems like this that can prevent terrorist attacks
What does "high terrorist level" imply? That the Feds do a thorough background check on you? That you'll be shadowed? For how long? 120000 people being shadowed? Even if you accept that the 119900 non-terrorists are severely harassed, the chances to find one single real terrorist are miniscle.
Do you genuinely believe such a system could prevent a single act of terrorism?
Or is it far more likely that such a system, or rather, the subsequent harassment of innocent people, actually increases the chances for acts of terror (think people seriously pissed off by the state, like Tim McVeigh)?
If Microsoft is focused on making money, they are focused on their products. They cannot, at gunpoint, force consumers to purchase their products.
On a free market, they cannot. On a (nearly) monopolized "market", of course they can. The whole point of anti-trust legislation is to ensure a Free Market, the basis for capitalism (as opposed to corporatism).
Microsoft made computing mainstream and gives most consumers exactly what they want.
Do you think so? I have yet to meet a single Windows user wo wants virus mails, trojans, popups, to name a few.
Microsoft cannot give consumers what they want, not because they are Microsoft but because most consumers don't know what they want. A free market assumes more or less informed customers. (Making customers believe they were informed is the whole point in marketing.)
Isn't that kind enough?
But the original author said,
they [Microsoft] have done NOTHING that I can think of out of the kindness of their hearts.
He talks motivation here, and Microsoft's motivation is, as with any company, to make as much money as they can. Nothing special here, move on. But do not confuse this with kindness. Kindness is altruistic, which they are not.
And still customers HAVE THAT OPTION. No one is threatening them at gunpoint.
Well, if losing your job in these times is the "alternative" to buying Microsoft, guess what, I'm not counting that as "having an option". There is a very strong bias pressuring me into the Microsoft "option".
It's a bit simplistic (and at the same time, melodramatic) to claim death was an option. Hell, now that's a slogan: Give me Microsoft, or give me death!
the hardware vendors have more resources to throw at the software, and inevitably produce better software that works well with their product
Depends.
A friend of mine bought a Sony Cybershot (which is certainly no cheap digicam at over $1000) and the software that came with it is outrageously bad and just plain stupid. A special, severely crippled download software instead of using Explorer on the USB storage device; a viewer that behaves irregular at best (zooming into previews and the like - looks really great).
Anyway, nowadays more often than not, the software that comes with your favourite device is outperformed in both function and usability by independent third-party software (be it Open Source or Closed).
I highly doubt that Linux authors would think twice about breaking buggy apps to force the issue.
(Nods agreement.)
Trouble is, there's not much choice but install the service pack, thus not much choice but breaking existing applications. You are now completely at mercy of the application vendors, which may or may not provide fixes. Most likely, they will force you to buy an upgraded version... which in turn may or may not work with all your legacy stuff. (Note that Microsoft is itself one of these application vendors, thus the temptation to generate a stream of revenues via forced upgrades is definetely given.)
With Open Source applications, on the other hand, even with older apps, you have a decend chance that broken applications will be fixed, even if you (as a company) have to pay someone to do it.
Open Source helps you mitigate the risk of being left out in the cold by software vendors. Depending on your business, this risk far outweights the financial benefit of a (really or perceivedly) smaller TCO.
no access at all (much less write access) to most of your own C: drive.
Much like not being allowed read access to the entire/etc and/sbin on Unix, much less write access to/
I don't know what exactly you do for work, but unless you're a software developer, there should be no reason for you to tamper with the install. And I would argue that even application developers should do just fine without these privilegues. They can write, compile and test their programs without advanced privilegues.
(Application programmers will of course need to test the installation process as well, but that should be done on dedicated machines, preferrably virtual ones.)
As an administrator I would refuse to take any responsibility for any setup where users are not strictly confined to what is essential for their job. And I'm saying this as a developer. Of course I work under Unix, where I can install any programs I really need in my home directory, without any advanced privilegues.
Wilhelm Conrad Roentgen, 1845-1923, German physicist who in 1895 discovered what he called and most people know as X-Rays. (For some obscure reason, in German they're called Roentgen-Strahlung.) Got the Prize for it in 1901.
one should be doing what they are doing in science to make a difference.
If you as a mechanic make someone's car work again, you do make a difference.
As with every other proper job.
I was doing research in particle physics for almost six years. It was interesting all right, but it wasn't me who made the difference, it was a team... a team of 500. And if we had discovered the Higgs boson, 499 of us would not get any recognition to speak of: only the spokesman would get the Nobel prize. The same spokesman who probably didn't even comment on the paper because he's drowning in political activity as his job requires him to.
From my observations in the field of research, I doubt very much that many people "make a difference". Those who do have fun working in science (quite a remarkable percentage considering the pay and the career options) just plain like the work they do. The idea of "making a difference" to speak of is most common among first semester students.
My personal guess would be that anyone who thinks she, personally, can make a difference is basically deluding herself. But science is fun anyway! Learning stuff is fun! Discovering things is fun! It really is. As you surely know, scanning through your list of publications.
it has always bothered me that we insist on believing there are only 4 types of force [...] Why can't there be other forces that operate on too large a scale or too small a scale for us to observe?
"Grand Unification Theories" (GUT) propose that for very high energies, the coupling constants of the various forces would converge into one, effectively leaving us with one unified force.
Is the postulate of "dark force" effectively a theory about a fifth type of force?
Not at all. First off, you've been watching Star Wars again, haven't you? It's "dark matter", not "dark force". And dark matter is an attempt to maintain the four currently known forces.
To the best of my knowledge the following companies make hardware that does not run Windows (tm): [...] BMW
Bzzzt. Wrong answer. BMW's iDrive runs on Windows CE. Don't really know about the other car manufracturers.
And this is more relevant that you'ld think, since e.g. tuning kits for modern cars are essentially or at least include firmware upgrades for the motor management. Add DRM to that and only the car vendor can provide tuning kits. Voila - Instant lock-out of competitors, thanks to DRM.
And even if the masses act like lemmings, there will always be a market of those of us who just won't play that game.
Yeah, just like that SGI workstation I have at home. Not.
All these "lemmings", as you aptly call them, constitute the mass market, and only a mass market makes high-tech products cheap enought for the masses and, well, for me. Pervasive DRM computing forces everyone who is not filthy rich to by DRM computers, or no computer at all.
If you believe the numbers, running a drive in RAID mirror will double the effective MTBF, we have done that by choosing the Maxline series vs a standard consumer IDE hard drive.
(Shakes head and bangs it violently against concrete wall)
MTBF and RAID is about entirely different things. The R in RAID stands for REDUNDANCY. You can have a MTBF approaching infinity and you would still have no redundancy.
Mirroring does NOT just double MTBF. It folds two probability functions. With RAID1 not only have both disks to die for data loss, but both disks have to die at the same time! (Or in fact, during the recovery window.) With a MTBF of 1.2 mio hours and a recovery window of maybe 5 hours, this really makes the difference.
Using non-RAID IDE disks, especially on a server, no matter how small the budget, is just playing russian roulette with your data. With at least 5 chambers loaded. It's wantonly negligent. It's unprofessional. Don't do it.
(As a side node, the MTBF is an utterly useless bit of information. It is determined by e.g. running 10,000 disks for 10 hours, with one disk failing. That is one dead disk in 100,000 hours of operation, so MTBF is 100,000. It's a bit like saying that if one woman can make a child in 9 months, 9 women can make a child in 1 month. Reality just doesn't work like that.)
There is soooo much less polution from nuclear reactors given the probability of worst case scenarios versus the diesel they are currently using. Why are we still burning fossel fuels!
[...] Such a backwards society we live in, when technology is available and safe, and we delay in implementation.
(A bit late, so I just hope you get notified by slashcode...)
Point is, even if operating a nuclear reactor was perfectly safe (which it isn't, because both men and machines will fail eventually, as history shows), even then, nuclear power is still not at all "clean" because you have radioactive nuclear waste afterwards:-( And no matter what you do, there simply is no way to store the nuclear waste for 10000 years in a safe manner.
So many people are not too pleased using a technology which even in a best case scenario is simply unmanageable. But in the end (ie. if we don't cut down our energy needs and the use of fossil fuels) we might end up HAVING TO use nuclear energy, no matter what are the consequences, let alone the sheer cost. (As you might know, nuclear power is highly subsidized, and at least in Germany the entire D&R for nuclear power plants was done with tax money.)
So don't be too quick to condemn a society that actually consideres the (pretty evident) drawbacks of a technology before it is utilized.
First, the signal/slot mechanism really bugs me. I am annoyed with the need to use non-ANSI C++ techniques (e.g. public slots, moc) to achieve results that could easily be done with legal C++ code.
But it can't. C++ is notorious for its inspection features, or rather, lack thereof. C++ offers virtually no meta class information. That's the sole point of moc's existance.
(And not only Qt desparately needs meta classes. The ROOT framework also uses a meta class generator, here creferred to as a "dictionary generator", but doing the same thing moc does: build a meta class from a class.)
Specifically the reliance on macros to achieve basic GUI functionality violates a key principle in Meyers' "Effective C++"
You can't just use a mixin here - you had to inherit from a generated (by your much-hated moc!) class. That would not only pollute the name space but also make your program less readable.
Maybe it's best to think of moc as an aspect weaver that weaves in meta object information. This style of programming (AOP) is current research in Computer Science and going far beyond mere OOP techniques.
[...] not have a proper separation of data from view. I am thinking specifically of QTable and QListView.
I agree fully-hearted (esp. on your gripes about QTable). Then again, the article reads:
Model/view classes for list box, tree view, icon view and table
The fact that 'weird places' means that there are a half-dozen places for binaries to go (/bin,/sbin,/usr/bin,/usr/sbin/, etc...)... in fact, I find the whole/usr heirarchy annoying. Why was that necessary? Weren't the six other folders for binaries enough?
The structure of the Unix filesystem is aimed at professional computing centres, not home use.
The reason for the separation of */bin and */sbin is simply the distiction between user commands and system administrator commands.
The distiction between (/bin,/sbin) and (/usr/bin,/usr/sbin) is that the entire/usr tree is meant to be mountable via NFS while leaving the rudimentary system on a local disk. (Or at least make/usr a separate partition.) So the commands to set up networking and to mount filesystems need to reside on the root partition (i.e. in/bin/lib and/sbin), while the other stuff goes to/usr.
If you put everything in/bin and/sbin, you will need a huge and unshareable root partition.
If you put everything in/usr/bin and/usr/sbin, you could theoretically make/usr a separate partition but wouldn't be able to mount it:-)
Now, both/bin and/usr/bin contain vendor software, ie. whatever your distro maker considered essential. You don't really want to mess up this with your downloaded, self-compiled (or at least self-installed, in any case not vendor-supplied) software. Consequently, third party software goes to/usr/local/...
Sometimes the size of third-party software justifies an entire directory tree of its own right. These massive packages are usually installed under/opt/<packagename>.
You see: to every complex problem, there is a solution that is simple, elegant, and wrong:-)
That said, it is very delicate to decide which program goes where. Take for example GNOME and KDE. Basically all distros include them, in this sense they're not third party software - so some distro makers put them in/usr/bin. OTOH the sheer size of the packages easily justifies an extra directory under the/opt hierarchy, which is what other distro makers do.
It's really not that easy.
You should ask yourself this question, though: why do you bother? Why do you even care? Although I'm about to celebrate my tenth year of Unix, I still have to which(1) many executables because I don't bother to remember actually where that particular binary resides. The PATH handles this just fine, and the package managers take care of the package integrity.
/root is not under/home
Same reason here:/home is not guaranteed to be on the same partition or even machine, and you still want to log in as root when the network (and thus,/home) is down.
The SH/BASH scripting language. (!!!!)
Though they all are more or less inconsistent compared to a properly designed language liek eg. Python, the Bourne shell family is a very powerful tool (don't get me started on (t)csh...).
Configuration files based on archaic paradigms like the SH/BASH scripting language.
The shell languages are more or less historically (hysterically) grown and offer quite some quirks, but the paradigm, procedural programming, is sound.
This is, of course, no excuse for ad-hoc or "defining-by-writing-a-parser" configuration languages - these are a royal PITA indeed!
Natural cosmic ray (probably created by supernovae or hypernovae) are far more energetic than any puny little collision we can muster.
First off, the origin of 10^20 eV cosmics is not at all understood. The Auger experiment for example is investigating this question.
Second, those very high energetic cosmic particles crash into earth (or whatever object in their path), which is basically at rest (compared to the speed of the cosmics). In particle physics, this is called "fixed target mode". Since both energy and momentum are conserved in the crash, the particles produced in the collision are not at rest but must carry the momentum of the cosmics (think billard). Thus, only a small part of the energy of the cosmics is avalable for forming new objects, namely sqrt(E), which is only 10 GeV, well within range of terrestral accelerators since over 10 years. The rest of the cosmics' energy just propels the new objects.
The Large Hadron Collider at CERN will crash protons at 7 TeV energy against other protons of the same energy/speed but opposite direction. This is called "collider mode", and the entire energy of 2x7=14 TeV is available for new objects.
(Well, not really, since protons are themselves compound objects, made of 3 quarks and lots of "gluons" which glue the quarks together. So really its only a quark-quark or gluon-gluon collision with less than a sixth of 14 TeV but still more than the 10 GeV above.)
There is of course the possibility of a cosmic particle colliding with another cosmic particle, but given the rate of 5 of those cosmics per 1000 km^2 per year, and the very low cross section of these high energetic particles, this isn't going to happen very often:-)
Imagine if the first signal we decode is: "don't build a particle accelerator larger than 5 kilometers in diameter or you will destroy your whole world."
That would be pretty darn bad - because it would be too late:-)
Fortunately though, the message probably reads, "don't build a particle accelerator with more than XXX TeV center of mass energy", and even this would be accurate only for lepton (i.e. electrons or muons; tau or neutrino accelerators being not very likely) accelerators. The Tevatron, for instance, uses protons (and antiprotons) as ammunition. The proton itself consists of quarks and gluons, and in a proton-antiproton collision, only one of them actually interacts - the rest is just fragments that clutter the forward regions of the detector.
I forsee distributed nets of the future attempting to produce results, in order to keep people interested and donating their computer cycles.
Ain't gonna happen. Unlike seti@home, the data per CPU minute throughput in high energy physics (aka particle physics) is much higher.
As an order of magnitude, simulating a single collision of the upcoming LHC's proton-proton beam (Large Hadron Collider is CERN's upcoming accelerator, supposed to start 2007/2008), takes about 1 CPU minute and generates ca 1 MB of data... that's 16 kB/s to send from your PC to CERN. Easy going for ethernet, but no fun with Modem, ISDN, or even DSL.
And consider that this is a high-CPU job. Most jobs in particle physics require way more bandwidth. Especially searches for (new or known) particles requires you to loop over huge data sets (some Terabytes for sure).
The Grid computing ansatz in high energy physics is intended for relatively few local computing centers (about one per country; Germany's center is here at Karlsruhe), interconnected with multi-Mbit links. (Our institute has a nice GigE link to the German Grid center:-)
NFS under Linux isn't there yet. It's unreliable under heavy load (eg. when saturating Gigabit ethernet). Linux NFS also has a very funny idea about RFCs, ie. it ignores them when it doesn't like what they say (NFS fragments packets in some cases when the RFC explicitly says not to).
NFS over TCP would also be nice. Linux has NFS over UDP only (in production code). Connection-based NFS gives a warm fussy feeling wrt. actually *knowing* about lost packets.
Multiple network cards are a pain, largely due to how Linux handles ARP. Routing table and ARP cache can contradict each other in certain situations (involving multiple network cards connected to the same switch, even on different subnets, caused by how Linux handles ARP requests).
The fact that the Linux packet filter doesn't track TCP sequence numbers is disconcerting. (This "feature" is used to allow TCP window scaling over NATted connections.)
In general, it seems that Linux networking tends to be rather sloppy about rules that concern security (TCP seq num) and interoperability (NFS frags) whenever it fits them, and don't care much about how other OSes cope with it. This mindset is almost Windowsian.
Re:Too bad the author is no good
on
F'd Companies
·
· Score: 1
The Calender.COM example stuck with me. "It could have been done with a script". Not true.
You got the CalendarCentral thingy and the OnlineChoice thingy mixed up. OnlineChoice was the one he claimed it could have been reduced to a script.
Slashdot Mirror
what's wrong with cryptographic signing?
There's nothing wrong with cryptographic signing, nothing at all. In fact, it would be a pretty good thing for precisely the reasons you gave, and I would even extend it to banknotes.
But what I strongly object to is contactless transmission, including any kind of RFID.
Nowadays everybody and his dog can read out RFID chips. They don't have to decipher it, they don't have to forge it - it's bad enough they can read it. It's just none of their business! Back in 1890, the later US Supreme Court Justice Louis Brandeis said, privacy is "the right to be left alone." I couldn't agree more.
Your concept doesn't even require contactless transmission. Let the card have some smart card-like readout pins. At the airport, have the security guard put the ID card in some card reader, and pronto your hi-res photograph complete with signature validation is on the screen, the guard looks at it, looks at your face, and returns the card. Pretty painless, uh?
And in fact even more secure: the check-in lines are long at airports, so lots of people with their ID cards near the readout point. Makes RFID more difficult (though not impossible, just a bit more error-prone). No problem at all with contact-based readout.
Following the idea of RFID-signed banknotes, the drawbacks of contactless transmission become obvious: a criminal can check you for money by simply pointing an RFID reader in your direction. He knows in advance that you're worth a shot, figuratively speaking. And he'll find the money in your money belt, too. Stupid, stupid idea!
(And this example also demonstrates that the need for privacy has nothing to do with this "I have nothing to hide" bullshit simple-minded people often give you. The amount of money I'm carrying falls by no means in the "nothing to hide" category, nor is it embarrassing, or sinister, or something. It's just not your business. Period.)
Have it contact-based and all those severe security and privacy problems disappear.
Everybody who works for MS has to shoulder the blame/credit for everything MS does.
Riiiight. Just as some lowly developer is responsible for (or has any influence on) an upper level manager who is bullying vendors and/or competitors with next-to-illegal means.
Or how about some G.I. serving for example in Ramstein who hasn't fired a shot in anger for months. Should he shoulder the blame for a few evil freaks torturing Iraqi army regulars? Hell no.
Insighful my asshole. You cannot blame someone for something (s)he's no influence on.
The screen shots do look cool indeed, but...
Anyone else noticed that the "Abort" button changes place, from the right to the middle to the left, where it is most often and which is - the reference being Windows - the most unexpected place.
The point is, "Abort" is the safe choice. You maneuvered yourself into some arcane dialog and don't understand nothin' - you hit "Abort" a.k.a. get me the hell out of here. This emergency button (as in, "Oh shit I'm gonna wreck by document") should always be in the same place.
Now when you align the buttons to the lower right (which the Gnome dialogs do), then this place is of course just there, in the lower right corner. Unlike Gnome, Windows got this right.
My favorite is the property screenshot. Instant inconsistency. (Shakes head and leaves.)
yep, they just dont build things the way they used to
The Liberty ships were designed with one goal in mind: build ships faster than the German Uboat force could sink them. And they succeeded! The Liberty ships were assembled (from pre-manufractured components) by mostly unskilled labour on the shipyards of Henry J. Kaiser within only 80 hours! On these shipyards, 140 Liberty ships per month would be completed.
The Liberty ships were never built to last. Their quality was rather poor. Definately not up to todays standards in shipbuilding.
Anyone else notice that there is a little footer with teh "recycled" symbol and the phrase "printed on recycled paper" ? it's a PDF. what happens if i print it out on non recycled paper?
Then the symbol disappears.
Dooh!
When you in a long narrow tunnel and you start to see light at the end of it, It's usually of the approaching train.
Or a fire-breathing dragon.
Or, worse yet, a burning truck.
(Emphasis mine.)
Everyone bitches and moans about systems like this that can prevent terrorist attacks
What does "high terrorist level" imply? That the Feds do a thorough background check on you? That you'll be shadowed? For how long? 120000 people being shadowed? Even if you accept that the 119900 non-terrorists are severely harassed, the chances to find one single real terrorist are miniscle.
Do you genuinely believe such a system could prevent a single act of terrorism?
Or is it far more likely that such a system, or rather, the subsequent harassment of innocent people, actually increases the chances for acts of terror (think people seriously pissed off by the state, like Tim McVeigh)?
If Microsoft is focused on making money, they are focused on their products. They cannot, at gunpoint, force consumers to purchase their products.
On a free market, they cannot. On a (nearly) monopolized "market", of course they can. The whole point of anti-trust legislation is to ensure a Free Market, the basis for capitalism (as opposed to corporatism).
Microsoft made computing mainstream and gives most consumers exactly what they want.
Do you think so? I have yet to meet a single Windows user wo wants virus mails, trojans, popups, to name a few.
Microsoft cannot give consumers what they want, not because they are Microsoft but because most consumers don't know what they want. A free market assumes more or less informed customers. (Making customers believe they were informed is the whole point in marketing.)
Isn't that kind enough?
But the original author said,
they [Microsoft] have done NOTHING that I can think of out of the kindness of their hearts.
He talks motivation here, and Microsoft's motivation is, as with any company, to make as much money as they can. Nothing special here, move on. But do not confuse this with kindness. Kindness is altruistic, which they are not.
And still customers HAVE THAT OPTION. No one is threatening them at gunpoint.
Well, if losing your job in these times is the "alternative" to buying Microsoft, guess what, I'm not counting that as "having an option". There is a very strong bias pressuring me into the Microsoft "option".
It's a bit simplistic (and at the same time, melodramatic) to claim death was an option. Hell, now that's a slogan: Give me Microsoft, or give me death!
the hardware vendors have more resources to throw at the software, and inevitably produce better software that works well with their product
Depends.
A friend of mine bought a Sony Cybershot (which is certainly no cheap digicam at over $1000) and the software that came with it is outrageously bad and just plain stupid. A special, severely crippled download software instead of using Explorer on the USB storage device; a viewer that behaves irregular at best (zooming into previews and the like - looks really great).
Anyway, nowadays more often than not, the software that comes with your favourite device is outperformed in both function and usability by independent third-party software (be it Open Source or Closed).
I highly doubt that Linux authors would think twice about breaking buggy apps to force the issue.
(Nods agreement.)
Trouble is, there's not much choice but install the service pack, thus not much choice but breaking existing applications. You are now completely at mercy of the application vendors, which may or may not provide fixes. Most likely, they will force you to buy an upgraded version... which in turn may or may not work with all your legacy stuff. (Note that Microsoft is itself one of these application vendors, thus the temptation to generate a stream of revenues via forced upgrades is definetely given.)
With Open Source applications, on the other hand, even with older apps, you have a decend chance that broken applications will be fixed, even if you (as a company) have to pay someone to do it.
Open Source helps you mitigate the risk of being left out in the cold by software vendors. Depending on your business, this risk far outweights the financial benefit of a (really or perceivedly) smaller TCO.
no access at all (much less write access) to most of your own C: drive.
/etc and /sbin on Unix, much less write access to /
Much like not being allowed read access to the entire
I don't know what exactly you do for work, but unless you're a software developer, there should be no reason for you to tamper with the install. And I would argue that even application developers should do just fine without these privilegues. They can write, compile and test their programs without advanced privilegues.
(Application programmers will of course need to test the installation process as well, but that should be done on dedicated machines, preferrably virtual ones.)
As an administrator I would refuse to take any responsibility for any setup where users are not strictly confined to what is essential for their job. And I'm saying this as a developer. Of course I work under Unix, where I can install any programs I really need in my home directory, without any advanced privilegues.
text says "microroengen per hour"
It's micro Roentgen per hour.
Wilhelm Conrad Roentgen, 1845-1923, German physicist who in 1895 discovered what he called and most people know as X-Rays. (For some obscure reason, in German they're called Roentgen-Strahlung.) Got the Prize for it in 1901.
one should be doing what they are doing in science to make a difference.
If you as a mechanic make someone's car work again, you do make a difference.
As with every other proper job.
I was doing research in particle physics for almost six years. It was interesting all right, but it wasn't me who made the difference, it was a team... a team of 500. And if we had discovered the Higgs boson, 499 of us would not get any recognition to speak of: only the spokesman would get the Nobel prize. The same spokesman who probably didn't even comment on the paper because he's drowning in political activity as his job requires him to.
From my observations in the field of research, I doubt very much that many people "make a difference". Those who do have fun working in science (quite a remarkable percentage considering the pay and the career options) just plain like the work they do. The idea of "making a difference" to speak of is most common among first semester students.
My personal guess would be that anyone who thinks she, personally, can make a difference is basically deluding herself. But science is fun anyway! Learning stuff is fun! Discovering things is fun! It really is. As you surely know, scanning through your list of publications.
it has always bothered me that we insist on believing there are only 4 types of force [...] Why can't there be other forces that operate on too large a scale or too small a scale for us to observe?
"Grand Unification Theories" (GUT) propose that for very high energies, the coupling constants of the various forces would converge into one, effectively leaving us with one unified force.
Is the postulate of "dark force" effectively a theory about a fifth type of force?
Not at all. First off, you've been watching Star Wars again, haven't you? It's "dark matter", not "dark force". And dark matter is an attempt to maintain the four currently known forces.
Because unlike the English Channel, the Straits of Gibraltar is 900+ meters deep.
To the best of my knowledge the following companies make hardware that does not run Windows (tm): [...] BMW
Bzzzt. Wrong answer. BMW's iDrive runs on Windows CE. Don't really know about the other car manufracturers.
And this is more relevant that you'ld think, since e.g. tuning kits for modern cars are essentially or at least include firmware upgrades for the motor management. Add DRM to that and only the car vendor can provide tuning kits. Voila - Instant lock-out of competitors, thanks to DRM.
And even if the masses act like lemmings, there will always be a market of those of us who just won't play that game.
Yeah, just like that SGI workstation I have at home. Not.
All these "lemmings", as you aptly call them, constitute the mass market, and only a mass market makes high-tech products cheap enought for the masses and, well, for me. Pervasive DRM computing forces everyone who is not filthy rich to by DRM computers, or no computer at all.
Isn't trusted computing a joy!
If you believe the numbers, running a drive in RAID mirror will double the effective MTBF, we have done that by choosing the Maxline series vs a standard consumer IDE hard drive.
(Shakes head and bangs it violently against concrete wall)
MTBF and RAID is about entirely different things. The R in RAID stands for REDUNDANCY. You can have a MTBF approaching infinity and you would still have no redundancy.
Mirroring does NOT just double MTBF. It folds two probability functions. With RAID1 not only have both disks to die for data loss, but both disks have to die at the same time! (Or in fact, during the recovery window.) With a MTBF of 1.2 mio hours and a recovery window of maybe 5 hours, this really makes the difference.
Using non-RAID IDE disks, especially on a server, no matter how small the budget, is just playing russian roulette with your data. With at least 5 chambers loaded. It's wantonly negligent. It's unprofessional. Don't do it.
(As a side node, the MTBF is an utterly useless bit of information. It is determined by e.g. running 10,000 disks for 10 hours, with one disk failing. That is one dead disk in 100,000 hours of operation, so MTBF is 100,000. It's a bit like saying that if one woman can make a child in 9 months, 9 women can make a child in 1 month. Reality just doesn't work like that.)
There is soooo much less polution from nuclear reactors given the probability of worst case scenarios versus the diesel they are currently using. Why are we still burning fossel fuels!
:-( And no matter what you do, there simply is no way to store the nuclear waste for 10000 years in a safe manner.
[...]
Such a backwards society we live in, when technology is available and safe, and we delay in implementation.
(A bit late, so I just hope you get notified by slashcode...)
Point is, even if operating a nuclear reactor was perfectly safe (which it isn't, because both men and machines will fail eventually, as history shows), even then, nuclear power is still not at all "clean" because you have radioactive nuclear waste afterwards
So many people are not too pleased using a technology which even in a best case scenario is simply unmanageable. But in the end (ie. if we don't cut down our energy needs and the use of fossil fuels) we might end up HAVING TO use nuclear energy, no matter what are the consequences, let alone the sheer cost. (As you might know, nuclear power is highly subsidized, and at least in Germany the entire D&R for nuclear power plants was done with tax money.)
So don't be too quick to condemn a society that actually consideres the (pretty evident) drawbacks of a technology before it is utilized.
Cheers, psc
First, the signal/slot mechanism really bugs me. I am annoyed with the need to use non-ANSI C++ techniques (e.g. public slots, moc) to achieve results that could easily be done with legal C++ code.
But it can't. C++ is notorious for its inspection features, or rather, lack thereof. C++ offers virtually no meta class information. That's the sole point of moc's existance.
(And not only Qt desparately needs meta classes. The ROOT framework also uses a meta class generator, here creferred to as a "dictionary generator", but doing the same thing moc does: build a meta class from a class.)
Specifically the reliance on macros to achieve basic GUI functionality violates a key principle in Meyers' "Effective C++"
You can't just use a mixin here - you had to inherit from a generated (by your much-hated moc!) class. That would not only pollute the name space but also make your program less readable.
Maybe it's best to think of moc as an aspect weaver that weaves in meta object information. This style of programming (AOP) is current research in Computer Science and going far beyond mere OOP techniques.
[...] not have a proper separation of data from view. I am thinking specifically of QTable and QListView.
I agree fully-hearted (esp. on your gripes about QTable). Then again, the article reads:
Model/view classes for list box, tree view, icon view and table
Sure sounds good to me!
The fact that 'weird places' means that there are a half-dozen places for binaries to go (/bin, /sbin, /usr/bin, /usr/sbin/, etc...) ... in fact, I find the whole /usr heirarchy annoying. Why was that necessary? Weren't the six other folders for binaries enough?
/sbin) and (/usr/bin, /usr/sbin) is that the entire /usr tree is meant to be mountable via NFS while leaving the rudimentary system on a local disk. (Or at least make /usr a separate partition.) So the commands to set up networking and to mount filesystems need to reside on the root partition (i.e. in /bin /lib and /sbin), while the other stuff goes to /usr.
/bin and /sbin, you will need a huge and unshareable root partition.
/usr/bin and /usr/sbin, you could theoretically make /usr a separate partition but wouldn't be able to mount it :-)
/bin and /usr/bin contain vendor software, ie. whatever your distro maker considered essential. You don't really want to mess up this with your downloaded, self-compiled (or at least self-installed, in any case not vendor-supplied) software. Consequently, third party software goes to /usr/local/...
/opt/<packagename>.
:-)
/usr/bin. OTOH the sheer size of the packages easily justifies an extra directory under the /opt hierarchy, which is what other distro makers do.
/root is not under /home
/home is not guaranteed to be on the same partition or even machine, and you still want to log in as root when the network (and thus, /home) is down.
The structure of the Unix filesystem is aimed at professional computing centres, not home use.
The reason for the separation of */bin and */sbin is simply the distiction between user commands and system administrator commands.
The distiction between (/bin,
If you put everything in
If you put everything in
Now, both
Sometimes the size of third-party software justifies an entire directory tree of its own right. These massive packages are usually installed under
You see: to every complex problem, there is a solution that is simple, elegant, and wrong
That said, it is very delicate to decide which program goes where. Take for example GNOME and KDE. Basically all distros include them, in this sense they're not third party software - so some distro makers put them in
It's really not that easy.
You should ask yourself this question, though: why do you bother? Why do you even care? Although I'm about to celebrate my tenth year of Unix, I still have to which(1) many executables because I don't bother to remember actually where that particular binary resides. The PATH handles this just fine, and the package managers take care of the package integrity.
Same reason here:
The SH/BASH scripting language. (!!!!)
Though they all are more or less inconsistent compared to a properly designed language liek eg. Python, the Bourne shell family is a very powerful tool (don't get me started on (t)csh...).
Configuration files based on archaic paradigms like the SH/BASH scripting language.
The shell languages are more or less historically (hysterically) grown and offer quite some quirks, but the paradigm, procedural programming, is sound.
This is, of course, no excuse for ad-hoc or "defining-by-writing-a-parser" configuration languages - these are a royal PITA indeed!
Natural cosmic ray (probably created by supernovae or hypernovae) are far more energetic than any puny little collision we can muster.
:-)
First off, the origin of 10^20 eV cosmics is not at all understood. The Auger experiment for example is investigating this question.
Second, those very high energetic cosmic particles crash into earth (or whatever object in their path), which is basically at rest (compared to the speed of the cosmics). In particle physics, this is called "fixed target mode". Since both energy and momentum are conserved in the crash, the particles produced in the collision are not at rest but must carry the momentum of the cosmics (think billard). Thus, only a small part of the energy of the cosmics is avalable for forming new objects, namely sqrt(E), which is only 10 GeV, well within range of terrestral accelerators since over 10 years. The rest of the cosmics' energy just propels the new objects.
The Large Hadron Collider at CERN will crash protons at 7 TeV energy against other protons of the same energy/speed but opposite direction. This is called "collider mode", and the entire energy of 2x7=14 TeV is available for new objects.
(Well, not really, since protons are themselves compound objects, made of 3 quarks and lots of "gluons" which glue the quarks together. So really its only a quark-quark or gluon-gluon collision with less than a sixth of 14 TeV but still more than the 10 GeV above.)
There is of course the possibility of a cosmic particle colliding with another cosmic particle, but given the rate of 5 of those cosmics per 1000 km^2 per year, and the very low cross section of these high energetic particles, this isn't going to happen very often
Imagine if the first signal we decode is: "don't build a particle accelerator larger than 5 kilometers in diameter or you will destroy your whole world."
:-)
That would be pretty darn bad - because it would be too late
Fortunately though, the message probably reads, "don't build a particle accelerator with more than XXX TeV center of mass energy", and even this would be accurate only for lepton (i.e. electrons or muons; tau or neutrino accelerators being not very likely) accelerators. The Tevatron, for instance, uses protons (and antiprotons) as ammunition. The proton itself consists of quarks and gluons, and in a proton-antiproton collision, only one of them actually interacts - the rest is just fragments that clutter the forward regions of the detector.
I forsee distributed nets of the future attempting to produce results, in order to keep people interested and donating their computer cycles.
:-)
Ain't gonna happen. Unlike seti@home, the data per CPU minute throughput in high energy physics (aka particle physics) is much higher.
As an order of magnitude, simulating a single collision of the upcoming LHC's proton-proton beam (Large Hadron Collider is CERN's upcoming accelerator, supposed to start 2007/2008), takes about 1 CPU minute and generates ca 1 MB of data... that's 16 kB/s to send from your PC to CERN. Easy going for ethernet, but no fun with Modem, ISDN, or even DSL.
And consider that this is a high-CPU job. Most jobs in particle physics require way more bandwidth. Especially searches for (new or known) particles requires you to loop over huge data sets (some Terabytes for sure).
The Grid computing ansatz in high energy physics is intended for relatively few local computing centers (about one per country; Germany's center is here at Karlsruhe), interconnected with multi-Mbit links. (Our institute has a nice GigE link to the German Grid center
NFS under Linux isn't there yet. It's unreliable under heavy load (eg. when saturating Gigabit ethernet). Linux NFS also has a very funny idea about RFCs, ie. it ignores them when it doesn't like what they say (NFS fragments packets in some cases when the RFC explicitly says not to).
NFS over TCP would also be nice. Linux has NFS over UDP only (in production code). Connection-based NFS gives a warm fussy feeling wrt. actually *knowing* about lost packets.
Multiple network cards are a pain, largely due to how Linux handles ARP. Routing table and ARP cache can contradict each other in certain situations (involving multiple network cards connected to the same switch, even on different subnets, caused by how Linux handles ARP requests).
The fact that the Linux packet filter doesn't track TCP sequence numbers is disconcerting. (This "feature" is used to allow TCP window scaling over NATted connections.)
In general, it seems that Linux networking tends to be rather sloppy about rules that concern security (TCP seq num) and interoperability (NFS frags) whenever it fits them, and don't care much about how other OSes cope with it. This mindset is almost Windowsian.
The Calender .COM example stuck with me. "It could have been done with a script". Not true.
You got the CalendarCentral thingy and the OnlineChoice thingy mixed up. OnlineChoice was the one he claimed it could have been reduced to a script.