By this reasoning, if he had handled the subprime crisis better, an appropriate severance pay would've been much smaller?
Well, yes, sort of -- as counter-intuitive as it may sound. The key is looking at it from the "Ok, I'm in deep shit, how do I get out?" angle. The more you stand to lose, the more money you can throw at it while still cutting your losses. It's all about the meaning you imply by "appropriate": if you have a bad infection in your hand leading to gangrene, medics might find it appropriate to sever your arm, whereas if the infection is small, an appropriate response is simply to administer antibiotics. Of course, we're talking business here. Just because something is worth a certain amount of money, that doesn't mean you shouldn't pay less if you can get away with it.
OTOH, good thing they didn't lose a whole $trillion under his watch- then $700M would've been reasonable severance (being only a week's loss, of course)...
It's not so much how much they lost as it is how much they stand to lose in the near future. The subprime crisis is not nearly over, and if his handling of it was bad in the near past, your first guess is that it'll remain bad in that near future (cf. derivative), and you cut your losses.
I used this particular example because the OP brought it up, and actual numbers were readily available (which is much more interesting -- and hygienic -- than pulling some out of my arse) but it seems it caused way too much confusion, though, as everybody's taking it as rewarding somebody for a cock up. Let's use something less actively bad as an example. Imagine an overly cautious CEO, who shies away from businesses at any sign of risk. At the end of the year, you might notice that the bottom line isn't as fat as it could've been, because in hindsight your company refused a number of profitable business ventures out of them being risky. The CEO, while not making any actively wrong decisions, failed to make some good ones, and, therefore, cost the company those businesses. You can pretty much expect his behaviour in the future to remain consistent with his past performance, and can reasonably say that he's effectively costing the company x amount of money, not because he's causing losses but because he's failing to achieve earnings. If you can get him to go away right now and replace him with someone more daring (but ultimately more "profitable") for less than the amount it'd cost to keep him aboard until he moved from his own accord, you're basically fattening the bottom line.
The big problem I see in all this is that US executives have a huge upside (Goldman Sachs CEO got a $68 million dollar bonus this year), but with no downside (Merrill Lynch fired its failed CEO with a $160 million golden parachute)
I used to think the same, but was explained recently how the "injustice" of the gold parachute thing is actually a misconception, and that it actually makes perfect business sense.
Let's use Merrill Lynch, which you brought up, as example. According to Wikipedia, they reported a net income of $7.49 Billion in 2006 (all further numbers are derived from simple arithmetic or taken from this same article). That's about $625M a month. Or $146M every week. Keep that number in mind.
Now, imagine you're in the board for a big high-stakes company (banking, insurance, that sort of stuff). If you have an incompetent CEO (hey, if Merrill Lynch had one, almost anyone is liable to get one at some point). You want to get rid of him, but, since being CEO for this sort of company is an intrinsically high-paying job, he obviously resists getting the boot. Assuming you don't have anything strong enough to outright fire the CEO, how much money can he make your company lose between now and you actually getting him sacked? That's the monetary value of getting him to leave of his own free will right now.
Stanley O'Neal got about $161M in stock options and retirement benefits as "severance pay". Based on my earlier math, that's just over 1 week of net income, which is, simply put, peanuts, seeing as he reportedly lost Merrill Lynch some $2.24 Billion (over 3 months' worth of net income) in how he was handling the sub prime crisis. How much would the company stand to lose by keeping him on-board any longer?
"Lighter to render" is actually a perfectly reasonable blanket statement to make in this case (though I'm sure someone can come up with some corner cases). PostScript is a full-fledged imperative style Turing-strong programming language, whereas PDF is a simplified version of PS, which removes all conditionals and loops, thus being just a nice list of statements.
I wholeheartedly agree. I saw the trailers, played the flash version (which, by the way, translated the source engine-powered gameplay beautifully into a "measly" 2D flash game), was all hyped up to try the original, and when I got my hands on the orange box (figured I'd give HL2 a try too, hadn't played it yet), it didn't disappoint. What surprised me was that the game is actually two games in one: a puzzler if you have your sound off, or a psychological thriller/survival horror type of thing if the sound is on (which, incidentally, is necessary to get some cues here and there).
Curiously, even though I tend to get motion sickness from FPSs (and, indeed, HL2 does give me a bit of that), all the huge jumping and camera twisting and turning that defines much of portal's visuals seems to be A-OK with my brain.
It's a stretch to suggest that the folks directly responsible for innovation would, in the normal course of their work, have more than a parenthetical involvement with any of this.
May I counterpoint that with:
And while Microsoft came in second to IBM in The Patent Board's 2006 survey, its upcoming 2007 report has Microsoft besting IBM (and even its 2006 report had Microsoft #1 in terms of the "scientific strength" of its patent portfolio) Which suggests, quite to the contrary of your statement, that their patent portfolio is backed up by some serious R&D (I call that more than "parenthetical involvement").
Accepting the reported superior scientific quality behind the vast portfolio of patents Microsoft has, the really interesting question is, as mentioned by other posters, where is this R&D going to?
Eh, not really. The iPod was a minor refinement of what was already out there, (...)
That was the GP's point all along. While you see "minor refinement of what was already out there" and state "not innovative", what I see (and I guess the GP agrees) is that the streamlined, well-designed interface they made with parts already available was a breath of fresh air.
Enlightenment (release E16) does that. A wee bit laggy on my old P2-generation Celeron, 300 MHz laptop, with 320 megs RAM, so it should run fine on modern systems.
Welcome to the wonderful world of lasers: They're high energy, highly focused, parallel, beams of light. Which is why they make such nice pinpoints. Since the pinpoint at 500 feet is as small as the pinpoint at 10 feet, the number of photons going out at the source is about the same as the number hitting the pinpoint, so 500 feet or 20 feet or 1 foot is exactly the same bloody thing.
Note that helicopters tend to tilt forwards when moving forwards, and also that a lot of choppers have a very large portion of their front made out of "glass" (probably not glass, but something to that effect). A bit of refraction on the helmet lens and the glass front would be enough to turn it towards the eyes of the pilot, and any bit of interference along the way would probably spread the beam a bit, making it bigger than the pinpoint it usually is. Honestly, the extended disorientation/pain/discomfort the pilot claims don't seem that far fetched, especially when we're talking about night-time surveillance, and a laser (aka an "inordinately large amount of light") was (supposedly) shone into your eyes.
I wonder what they had in mind exactly when they put that maximum on that category?
Perhaps doing it on purpose, with the intent to harm? (versus doing it on purpose as an irresponsible prank, or doing it out neglect, like would probably be the case for this couple)
Sorry to burst your bubble, but there are several issues with the idea you posted.
PDF is a great format for publication, but crap for information exchange. You need some sort of "work format" to do the heavy lifting for you before you can commit the document to its published.pdf form. Regarding that "work format", I'll assume that by "XHTML" you actually meant "XML", since the latter is a general purpose mark-up language, and the former a domain-specific application of the latter, and this whole discussion is overkill if you really meant XHTML, since that is simply not enough for a functional office application format.
XML might have many virtues, but it has one major flaw: It's not a standard. It's a meta-standard. You need a DTD to turn XML into a usable standard to work on -- like XHTML. Guess what, exactly, ODF and OOXML are? Yip, they're at their core just DTDs for specific applications of XML. Funny you should mention SVG for vector graphics: It's just yet another DTD for XML. Effectively, your statement that XML is the solution is in direct contradiction with later saying that OOXML and ODF are unnecessary.
The idea of "powering a spreadsheet with JavaScript" kind of implies that you're going to embed the actual calculation logic in the spreadsheet, rather than just having a formula language. Nice and light -- or perhaps not. As far as I can see, the only other way to read that statement (which is more or less equivalent in performance) is that you're suggesting writing the calculator core (the one single part of the application you'd really really want to write in highly optimized C) in JavaScript, which is really not that good an idea either.
CSS might be quite powerful for the web, but for book formatting I'll stick to TeX, thank you very much. That's just an example of a particular application where CSS is underwhelming compared to the alternatives. I'm not much of a fan of writing a gazillion different standards for slightly different application uses, but using CSS as the baseline layout description language for your whole office document format is a hardcore case of shoehorning.
Finally, and just to nitpick, PNG is really underwhelming for photography and similar image types, where JPEG is far better.
HTML does not handle all the use cases of office documents smoothly and is a pretty terrible format for exchanging documents since in many cases you'd be exchanging entire directories of files instead of a single file since all the resources in HTML are stored by reference.
I wholeheartedly agree that HTML isn't really the optimal format for document exchange, and your first point (that it doesn't handle all the necessary use cases) is quite valid. However, the latter point is not necessarily a big problem. Everybody and their dog has an implementation of zip by now, so something as trivial as a zip file with an index.html file and a resources/ directory with all the needed external stuff would be a pretty open and portable way to address the issue. Also, refer to the Apple implementation of.app,.wdgt and probably a few more "rich folder formats" (just made that name up), which work quite well for application storage, for another potential approach.
I don't think that anybody's trying to argue about what's right and what's wrong in a logical sense, but merely what makes sense to have in an encyclopaedia, and what doesn't. So don't get your panties in a twist (though I bet you were just begging for an opportunity either way...)
I'd say that rather than full proofs, they should have sketches of proofs. Like, say, explaining that Gödel's incompleteness theorem was originally proved by constructing a self-referencing cycle in number theory, effectively a mathematical statement stating something about itself (which blows up pretty bad).
More often than not, the full extent of a formal proof is unnecessary outside of publications on the subject, and the information I personally would like to find is enough of a sketch of a proof (or more than one, if it helps illustrate an important relationship with another concept) such that would allow me to understand the general gist of it (the whole purpose of an encyclopaedia), without burdening me with the full proof, which, for most non-trivial results, is probably a tangled mess.
if the data is encrypted then only those that know how to decrypt the data can read it, everyone else has no idea what that data is
Not completely true. Once you dive into the source, you can verify whether the information that's being packaged is indeed the information they say they're collecting. Their EULA (apparently) says they're collecting the information, so you know they have it. But what of anybody who intercepts it? Granted, it's not particularly useful information, but it's good standard procedure to encrypt this sort of thing anyway, especially when the client has the benefit of the transparency of OSS.
At whichever point of decay you are, another half life and you double the amount of "B" you stick into your body to get the necessary amount of "A". Exponentials are a bitch, I know.
All jokes aside, both Emacs and Vim are available with fully (dis)functional graphic interfaces. I barely ever use vim proper, but rather some form or another of gvim (nowadays, mostly the windows or the OSX ports).
WRT power savings... the fact that you have the power adapter for the disk connected *at all* is already consuming power (I'm yet to see an external HDD with a power switch upstream from the adapter proper). I don't think that the power necessary to keep the USB connection alive is going to be that much higher than what it's passively eating up. Either way around, if you were Seagate, and you thought complete powerdown was a good idea, great. Document the behaviour and let everybody know what you're doing and why, hopefully more people will pick up on it (and work to cooperate nicely with it, from the OS point of view). Otherwise, I call shenanigans.
This is purely anecdotal, but my dogs (both of them), respond to dogs on TV (in a sense, the test they portrayed here), sometimes ramming against the TV paws first and barking at it, other times just staring intently. They also clearly identify at least some cartoon dogs (e.g, Scooby Doo), reacting in much the same way as with real dogs.
They also tend to react to other non-human animals on TV, with reactions varying dramatically with the species (they seem to have a fascination with bulls), and definitely can tell "humans" and "animals" apart on TV consistently. So this test doesn't come across as much of a surprise to me (though formal testing is always better than empirical or anecdotal evidence).
Slashdot Mirror
Ultimately, in this day and age, I fear the two are taken as synonyms. Much like the original and present meanings of the word 'perfect'.
I know the one you mean. It's the Anatidae commutator subgroup!
I think it's actually "they produce fertile offspring" -- horses and donkeys producing viable, non-fertile mules.
The unstated but crucial point being that, obviously, nobody was ever fired for being careful.
Well, yes, sort of -- as counter-intuitive as it may sound. The key is looking at it from the "Ok, I'm in deep shit, how do I get out?" angle. The more you stand to lose, the more money you can throw at it while still cutting your losses. It's all about the meaning you imply by "appropriate": if you have a bad infection in your hand leading to gangrene, medics might find it appropriate to sever your arm, whereas if the infection is small, an appropriate response is simply to administer antibiotics. Of course, we're talking business here. Just because something is worth a certain amount of money, that doesn't mean you shouldn't pay less if you can get away with it.
OTOH, good thing they didn't lose a whole $trillion under his watch- then $700M would've been reasonable severance (being only a week's loss, of course)...It's not so much how much they lost as it is how much they stand to lose in the near future. The subprime crisis is not nearly over, and if his handling of it was bad in the near past, your first guess is that it'll remain bad in that near future (cf. derivative), and you cut your losses.
I used this particular example because the OP brought it up, and actual numbers were readily available (which is much more interesting -- and hygienic -- than pulling some out of my arse) but it seems it caused way too much confusion, though, as everybody's taking it as rewarding somebody for a cock up. Let's use something less actively bad as an example. Imagine an overly cautious CEO, who shies away from businesses at any sign of risk. At the end of the year, you might notice that the bottom line isn't as fat as it could've been, because in hindsight your company refused a number of profitable business ventures out of them being risky. The CEO, while not making any actively wrong decisions, failed to make some good ones, and, therefore, cost the company those businesses. You can pretty much expect his behaviour in the future to remain consistent with his past performance, and can reasonably say that he's effectively costing the company x amount of money, not because he's causing losses but because he's failing to achieve earnings. If you can get him to go away right now and replace him with someone more daring (but ultimately more "profitable") for less than the amount it'd cost to keep him aboard until he moved from his own accord, you're basically fattening the bottom line.
I used to think the same, but was explained recently how the "injustice" of the gold parachute thing is actually a misconception, and that it actually makes perfect business sense.
Let's use Merrill Lynch, which you brought up, as example. According to Wikipedia, they reported a net income of $7.49 Billion in 2006 (all further numbers are derived from simple arithmetic or taken from this same article). That's about $625M a month. Or $146M every week. Keep that number in mind.
Now, imagine you're in the board for a big high-stakes company (banking, insurance, that sort of stuff). If you have an incompetent CEO (hey, if Merrill Lynch had one, almost anyone is liable to get one at some point). You want to get rid of him, but, since being CEO for this sort of company is an intrinsically high-paying job, he obviously resists getting the boot. Assuming you don't have anything strong enough to outright fire the CEO, how much money can he make your company lose between now and you actually getting him sacked? That's the monetary value of getting him to leave of his own free will right now.
Stanley O'Neal got about $161M in stock options and retirement benefits as "severance pay". Based on my earlier math, that's just over 1 week of net income, which is, simply put, peanuts, seeing as he reportedly lost Merrill Lynch some $2.24 Billion (over 3 months' worth of net income) in how he was handling the sub prime crisis. How much would the company stand to lose by keeping him on-board any longer?
"Lighter to render" is actually a perfectly reasonable blanket statement to make in this case (though I'm sure someone can come up with some corner cases). PostScript is a full-fledged imperative style Turing-strong programming language, whereas PDF is a simplified version of PS, which removes all conditionals and loops, thus being just a nice list of statements.
the cake is a lie!
I wholeheartedly agree. I saw the trailers, played the flash version (which, by the way, translated the source engine-powered gameplay beautifully into a "measly" 2D flash game), was all hyped up to try the original, and when I got my hands on the orange box (figured I'd give HL2 a try too, hadn't played it yet), it didn't disappoint. What surprised me was that the game is actually two games in one: a puzzler if you have your sound off, or a psychological thriller/survival horror type of thing if the sound is on (which, incidentally, is necessary to get some cues here and there).
Curiously, even though I tend to get motion sickness from FPSs (and, indeed, HL2 does give me a bit of that), all the huge jumping and camera twisting and turning that defines much of portal's visuals seems to be A-OK with my brain.
May I counterpoint that with:
And while Microsoft came in second to IBM in The Patent Board's 2006 survey, its upcoming 2007 report has Microsoft besting IBM (and even its 2006 report had Microsoft #1 in terms of the "scientific strength" of its patent portfolio) Which suggests, quite to the contrary of your statement, that their patent portfolio is backed up by some serious R&D (I call that more than "parenthetical involvement").Accepting the reported superior scientific quality behind the vast portfolio of patents Microsoft has, the really interesting question is, as mentioned by other posters, where is this R&D going to?
That was the GP's point all along. While you see "minor refinement of what was already out there" and state "not innovative", what I see (and I guess the GP agrees) is that the streamlined, well-designed interface they made with parts already available was a breath of fresh air.
Enlightenment (release E16) does that. A wee bit laggy on my old P2-generation Celeron, 300 MHz laptop, with 320 megs RAM, so it should run fine on modern systems.
Welcome to the wonderful world of lasers: They're high energy, highly focused, parallel, beams of light. Which is why they make such nice pinpoints. Since the pinpoint at 500 feet is as small as the pinpoint at 10 feet, the number of photons going out at the source is about the same as the number hitting the pinpoint, so 500 feet or 20 feet or 1 foot is exactly the same bloody thing.
1 word: refraction
Note that helicopters tend to tilt forwards when moving forwards, and also that a lot of choppers have a very large portion of their front made out of "glass" (probably not glass, but something to that effect). A bit of refraction on the helmet lens and the glass front would be enough to turn it towards the eyes of the pilot, and any bit of interference along the way would probably spread the beam a bit, making it bigger than the pinpoint it usually is. Honestly, the extended disorientation/pain/discomfort the pilot claims don't seem that far fetched, especially when we're talking about night-time surveillance, and a laser (aka an "inordinately large amount of light") was (supposedly) shone into your eyes.
Perhaps doing it on purpose, with the intent to harm? (versus doing it on purpose as an irresponsible prank, or doing it out neglect, like would probably be the case for this couple)
Sorry to burst your bubble, but there are several issues with the idea you posted.
PDF is a great format for publication, but crap for information exchange. You need some sort of "work format" to do the heavy lifting for you before you can commit the document to its published .pdf form. Regarding that "work format", I'll assume that by "XHTML" you actually meant "XML", since the latter is a general purpose mark-up language, and the former a domain-specific application of the latter, and this whole discussion is overkill if you really meant XHTML, since that is simply not enough for a functional office application format.
XML might have many virtues, but it has one major flaw: It's not a standard. It's a meta-standard. You need a DTD to turn XML into a usable standard to work on -- like XHTML. Guess what, exactly, ODF and OOXML are? Yip, they're at their core just DTDs for specific applications of XML. Funny you should mention SVG for vector graphics: It's just yet another DTD for XML. Effectively, your statement that XML is the solution is in direct contradiction with later saying that OOXML and ODF are unnecessary.
The idea of "powering a spreadsheet with JavaScript" kind of implies that you're going to embed the actual calculation logic in the spreadsheet, rather than just having a formula language. Nice and light -- or perhaps not. As far as I can see, the only other way to read that statement (which is more or less equivalent in performance) is that you're suggesting writing the calculator core (the one single part of the application you'd really really want to write in highly optimized C) in JavaScript, which is really not that good an idea either.
CSS might be quite powerful for the web, but for book formatting I'll stick to TeX, thank you very much. That's just an example of a particular application where CSS is underwhelming compared to the alternatives. I'm not much of a fan of writing a gazillion different standards for slightly different application uses, but using CSS as the baseline layout description language for your whole office document format is a hardcore case of shoehorning.
Finally, and just to nitpick, PNG is really underwhelming for photography and similar image types, where JPEG is far better.
I wholeheartedly agree that HTML isn't really the optimal format for document exchange, and your first point (that it doesn't handle all the necessary use cases) is quite valid. However, the latter point is not necessarily a big problem. Everybody and their dog has an implementation of zip by now, so something as trivial as a zip file with an index.html file and a resources/ directory with all the needed external stuff would be a pretty open and portable way to address the issue. Also, refer to the Apple implementation of .app, .wdgt and probably a few more "rich folder formats" (just made that name up), which work quite well for application storage, for another potential approach.
Rupert Murdoch is Australian? I had no idea. I guess that he proves that for Australia, the saying rings true: Better out than in.
just replace "???" with "wrap it in flashy graphics", and you've got a full fledged path to profit.
I don't think that anybody's trying to argue about what's right and what's wrong in a logical sense, but merely what makes sense to have in an encyclopaedia, and what doesn't. So don't get your panties in a twist (though I bet you were just begging for an opportunity either way...)
I'd say that rather than full proofs, they should have sketches of proofs. Like, say, explaining that Gödel's incompleteness theorem was originally proved by constructing a self-referencing cycle in number theory, effectively a mathematical statement stating something about itself (which blows up pretty bad).
More often than not, the full extent of a formal proof is unnecessary outside of publications on the subject, and the information I personally would like to find is enough of a sketch of a proof (or more than one, if it helps illustrate an important relationship with another concept) such that would allow me to understand the general gist of it (the whole purpose of an encyclopaedia), without burdening me with the full proof, which, for most non-trivial results, is probably a tangled mess.
Not completely true. Once you dive into the source, you can verify whether the information that's being packaged is indeed the information they say they're collecting. Their EULA (apparently) says they're collecting the information, so you know they have it. But what of anybody who intercepts it? Granted, it's not particularly useful information, but it's good standard procedure to encrypt this sort of thing anyway, especially when the client has the benefit of the transparency of OSS.
At whichever point of decay you are, another half life and you double the amount of "B" you stick into your body to get the necessary amount of "A". Exponentials are a bitch, I know.
All jokes aside, both Emacs and Vim are available with fully (dis)functional graphic interfaces. I barely ever use vim proper, but rather some form or another of gvim (nowadays, mostly the windows or the OSX ports).
I fully agree with you on the file system front.
WRT power savings... the fact that you have the power adapter for the disk connected *at all* is already consuming power (I'm yet to see an external HDD with a power switch upstream from the adapter proper). I don't think that the power necessary to keep the USB connection alive is going to be that much higher than what it's passively eating up. Either way around, if you were Seagate, and you thought complete powerdown was a good idea, great. Document the behaviour and let everybody know what you're doing and why, hopefully more people will pick up on it (and work to cooperate nicely with it, from the OS point of view). Otherwise, I call shenanigans.
This is purely anecdotal, but my dogs (both of them), respond to dogs on TV (in a sense, the test they portrayed here), sometimes ramming against the TV paws first and barking at it, other times just staring intently. They also clearly identify at least some cartoon dogs (e.g, Scooby Doo), reacting in much the same way as with real dogs.
They also tend to react to other non-human animals on TV, with reactions varying dramatically with the species (they seem to have a fascination with bulls), and definitely can tell "humans" and "animals" apart on TV consistently. So this test doesn't come across as much of a surprise to me (though formal testing is always better than empirical or anecdotal evidence).