The article was written by a journalist for Forbes. It also says nowhere that this is the only way to use the storage.
Well, the Forbes article might not, the Official Google Blog post on the new offering does say just that: That's why the Picasa team is pleased to tell you that in a few hours we'll be rolling out extra storage that you can purchase to use across several Google products (today, Picasa Web Albums and Gmail; soon, other applications like Google Docs & Spreadsheets).
Google's paid storage is, as far as anything I can find from Google, only planning to support Google Applications, and currently only supports Gmail and Picasa.
There is a "fittest" of, under the usual definition of fitness, the recently-deceased past generations. What there is not is a static external ideal that is the fittest toward which evolution is leading. I was not taking issue with the idea that there exists a definable "fittest" at any given point in time (though at any given point in time that "fittest" is only defined for the past), I was taking issue with the suggestion that there was a static fittest which over time evolution approaches, justifying the conclusion that decreasing diversity is consistently what should be expected.
It's just not absolute or measurable and changes when the environment (or other variables) in which the species lives changes.
Well, no, it is measurable, its just not static. But since the post that I was responding to required a particular type of "fittest" for its logic (that evolutions consistent approach to the "fittest" should generally lead to decreasing diversity) to hold, and that is precisely the kind of static "fittest" that does not exist that I was taking issue with...
"fittest" is a confusing word, we should probably use a different term since, like you, many people seem to confuse it with meaning "best" whereas it really is closer to "highest chance of survival in it's environment".
The usual definition is more like highest likelihood of having offspring survive to reproduce in its environment. But I didn't confuse anything, I was addressing the kind of "fittest" that was required for the argument being justified on the basis of the approach to the "fittest" to hold. Essentially, I was arguing that the poster to whom I was responding didn't understand what the "fittest" in the survival of the fittest meant and entailed, since the concept of "fittest" necessary for the argument to work was different from the kind of "fittest" actually relevant to evolution.
If you take the meaning to be the last, you'll see that "fittest" in no way implies strongest, smartest, prettiest or any other absolute measure.
The actual definition usually used has a quite clear absolute measure, though its not strongest, smartest, or prettiest (at least one of which isn't an absolute objective measure, either.)
. However, I feel that the tests provide much more value in that once a test for a piece of code works, you now have confidence in that piece of code. If you miss something with the test, its a simple matter of adding a new test for that case.
A test doesn't replace documentation. Documentation describes what the code does. A test verifies that, in the case covered by the test, the code does, in fact, do what it is intended to do. But you can't infer the information in good documentation from an (undocumented) test suite, because while the test cases will specify behavior in specific conditions, you can't (generally) unambiguously infer what behavior is expected in other conditions, particularly when the task being tested is nontrivial. A test is always a specific case of an operationalization of a specification, even if no one has ever reduce the specification to writing. But it would be good if they did. Its true that good documentation isn't always worth the effort, but its equally true, I think, that it is almost always more worth the effort than it would seem intuitively at the time the code is written, so if the impression at the time of first developing the code is the guide to documentation, code always ends up underdocumented which tends to cause problems down the road.
If goto were not useful or necessary then languages would have dropped it years ago
Uh, no. Languages, particularly ones in wide use where continued support of legacy code is important, don't, in the real world, necessarily drop features just because they aren't useful or necessary if they are in wide use in existing code, even if that code is poorly written. So you can't infer utility from the fact that it still exists.
Plenty of newer languages (and, for that matter, some older ones), of course, don't have gotos, and get along just fine without it. Gotos may or may not be useful in languages more abstract than assembly, but they certainly aren't necessary.
Layout is really a 2D constraint problem. Internally, you have constraints like "boxes can't overlap", which turns into constraints like "upper left corner of box B must be below lower left corner of box A", or "right edge of box A and left edge of box B must have same X coordinate". Browsers really ought to do layout that way.
Doesn't XSL:FO handle that? I remember back when there were XSL:FO browsers in the works (and early versions released even), but it looks like that, sadly, all fell apart, and XSL:FO mostly got relegate to generating output aimed at print, where it is used at all.
IIRC, Google has a policy of not using metadata to rank search results, because (a) it's easy to abuse, (b) it's easy to use incorrectly or inconsistently in ways that are unhelpful, and more importantly, (c) it's not information that's visible to the users of the page, and therefore, it tends to be irrelevant to whether the users will find the page useful.
What he was referring to (and challenging me in the process) was document structure, not metadata. There is a substantial difference. Its much harder to abuse, its fairly easy to use consistently in a way that is helpful, and its information that while it can be obscured through clever styling, will generally be visible to page viewers.
Isn't this basically what that whole "survival of the fittest" thing does? End less suitable genetic traits and combine the surviving ones in an ever repeating cycle, ever closer to the "fittest" genetic blend?
No, there's not some Platonic ideal of "fittest" that is being progressively approached.
Depending on the timescale, environment, and the universe of analysis, evolution may tend toward a more or less diversity over time (e.g., overall, diversity has increased from the first life, while diversity in a narrow population over a limited time may decrease.)
Years from now, are we still going to see IE 10, Firefox 5, and Safari 3.1 support deprecated tags?
A "deprecated" tag is one that implementations should support, but developers should not use for new pages, since it is planned to be removed as "obsolete" in a future version of the standard.
An "obsolete" tag is one that it is no longer expected that implementations will support.
Supporting deprecated tags is what browsers should do.
An article, as division of a document, is semantically no different than any other division of a document. So there is no semantic difference between:
<article>...</article> and <div class="article">...</div>
Inasmuch as there is a difference between an "article" and another division of the document, the distinction is purely a matter of the internal heirarchical structure of a specific document. There is no consistent interpretation that can or should be applied to an "article" beyond that of a division of a document.
Similar concerns have also been raised about the award of a smaller machine to Oak Ridge national lab, which is a Department of Energy laboratory, not a site one would expect to house an NSF machine.
Really? Seems to me that the NSF and the DOE Office of Science tend to work together a lot, and sharing facilities, cross-detailing personnel, etc., are pretty common between those two organizations.
Look it up. Ballot readers are compromised as easily as the original machines.
Yes, but you have the ballots. All-electronic systems with no separate ballots don't allow random-precinct hand-count confirmation, or even full-election hand recounts if there is cause.
Systems that generate a ballot, which the voter than confirms and turns in and is then counted by a separate machine do allow that, so even if the ballot reader is just as easily compromised as a voting machine in an all-electronic system would be, the election is more secure.
This issue is not how secure is the machine, it is how secure is an election using the machines; using vote counting machines like, e.g., optical scan machines, that count voter-produced or voter-verified ballots which are both human and machine readable and which are available for auditing produces a more secure election than using systems where the "ballots" exist only as electronic data within the system.
Whenever a story on the voting machines comes up many people present your argument. I find it fundamentally flawed however as counting by hand is extremely inefficient. Not only is it a slow, labor intensive task but it is also open to human error and other technical issues (hanging chads, etc).
Um, hanging chads are a problem with with ballots designed for machine counting that are resolvable only by hand counting, not a problem with ballots designed to be counted by hand.
There is no real point of denying it, computer voting is coming.
What kind of machine (if any) you produce a ballot on is a logically orthogonal question to how the ballots are counted. Computer voting can be done with hand counted ballots. Hand-filled ballots can be counted by machines. Whether ballots should be counted by machine, and what kind, and what kind of relation that machine should have (if any) to the production of the ballots are all open questions that cannot just be waved away out of the discussion.
Instead of saying "Oh this new system doesn't work in it's current incarnation, we should go back to the other method" we should be asking "The new method we are trying to implement is flawed, how should we change it?"
And the answer to that may well be "we should change it back into the old way". Or it may not. But its certainly a valid response, and you've provided no counterargument to it except that you apparently don't like the option even being considered and wish people would eliminate it without thought.
Invoking the Digital Millennium Copyright Act of 1998, Federal Custom's Agents have raided over 30 homes and businesses looking to confiscate so-called 'mod chips', or other devices that allow the playback of pirated video games. This raises an important question: Are legitimate backup copies of a piece of software you own illegal under the DMCA?
1. Modchips are not "backup copies", legitimate or otherwise. They are, arguably, circumvention devices under the DMCA. Whether some of the games that some people play with them may be "legitimate backup copies" is a different issue.
2. I suspect the number of people using modchips to play "legitimate backups" is small compared to the number using modchips to play pirated copies.
3. If a particular copy is illegal (under the DMCA or otherwise), it is not a "legitimate" backup.
4. When a coordinated raid like this is done, its usually (there are exceptions) done based on evidence of some kind of coordinated enterprise. Even if people use modchips to play backups (technically illegal or not) of games they own legitimate copies of, I doubt very much that that's the focus of these raids.
They don't have the political connections or the ENORMOUS resources that AT&T/Cingular has.
It is broken up, as I understand, both by portions of the spectrum and geographic regions. Assuming there aren't limits on how much of the spectrum in any one region any one interest can buy (I don't know that much about the particular auction), the arguably, if the FCC isn't going to impose freedom and Google really values it, its best move may be to find one region that lots of other carriers would like access to and put as much as it can afford into buying up spectrum in that market, and then provide access to other carriers on condition of them adopting Google's openness principles on the rest of their holdings in the 700MHz range.
This vote would seem to mean that Google won't bid in the spectrum auction.
Why would it seem to mean that? They announced that if their principles were adopted, they would bid up to a specific amount on a specific share of that spectrum. They did not say, that I recall, that if there demands were not (or incompletely) addressed, they would definitely bid nothing.
This is clearly a very unix oriented post as there are relatively few command-line windows apps and few window GUI apps that accept command-lines.
I dunno what you consider "few" but lots of Windows GUI apps I've encountered accept command-line arguments. Most users don't recognize it because they just click the pretty icon, of course, but that doesn't mean they aren't there.
I've said it before but I love to preach this idea: send all punitive damages to the federal government's general fund.
Doesn't make a whole lot of sense for (and probably unconstitutional, too) where the action is commenced under state law, and anyway, many class action suits aren't seeking punitive damages (and most settle, anyhow.)
Suits where punitive damages are awarded are generally more serious, and more deliberate wrongs, and are arguably the cases that it is most in the public interest to have brought; if the plaintiff didn't get the punitive damage award (and if they could not, therefore, be part of the basis for contingency fees), there'd be less incentive for plaintiffs to bring and less incentive for attorney's to take such cases, but while at the same time not touching the vast majority of controversial lawsuits, which primarily seek compensatory damages of an amount that observers (often ill-informed, but sometimes not) see as disporportionate to the wrong done.
What confuses me as soon as it says "$100 more" is that you are at $299 and for another $150 you can wander into BestBuy and splash $450 on a decent laptop that comes with Vista.
Perhaps, but that's for a time-and-a-half the cost of "high end" of the ASUS inexpensive laptops.
As for "rushing to save the rest of the world", the Russians did far more to defeat Hitler, at huge cost to themselves.
This is kind of ironic coming from someone who is dimissive of the US involvement on the basis that only Pearl Harbor got the US to joing at all, considering the position of Nazi Germany and Stalin's USSR prior to the point at which their totalitarian imperialistic ambitions could no longer coexist.
Perhaps they could - but I'm not convinced the "average Joe" would go for it.
The average Joe manifestly wouldn't if everything were exactly the way it is today.
I mean seriously - pay double the cost of a "normal" car, and then being able to use it for all your special needs?
I'm not sure what this is supposed to mean. I suspect that you left something out of this sentence, but I'm not sure what since I can think of a few different ways that it could have been meant.
I agree the current costs of "normal" cars don't include "real" costs (i.e. pollution).
This is a fairly big factor in the current market, since it is, in effect, a subsidy for fuel consumption which encourages people to buy wasteful vehicles (and/or use vehicles in a wasteful manner) compared to the situation if the external costs were internalized.
I'm certainly not arguing that we shouldn't be moving to battery technology - just that there are some significant things currently stopping people from going there. Hopefully in the future we will see these obsticles removed, and the problems solved.
Sooner or later we will; oil supplies alone should guarantee that. Its really more a matter of how controlled the transition is, and how much damage is done first.
Slashdot Mirror
Well, the Forbes article might not, the Official Google Blog post on the new offering does say just that: That's why the Picasa team is pleased to tell you that in a few hours we'll be rolling out extra storage that you can purchase to use across several Google products (today, Picasa Web Albums and Gmail; soon, other applications like Google Docs & Spreadsheets).
Google's paid storage is, as far as anything I can find from Google, only planning to support Google Applications, and currently only supports Gmail and Picasa.
There is a "fittest" of, under the usual definition of fitness, the recently-deceased past generations. What there is not is a static external ideal that is the fittest toward which evolution is leading. I was not taking issue with the idea that there exists a definable "fittest" at any given point in time (though at any given point in time that "fittest" is only defined for the past), I was taking issue with the suggestion that there was a static fittest which over time evolution approaches, justifying the conclusion that decreasing diversity is consistently what should be expected.
Well, no, it is measurable, its just not static. But since the post that I was responding to required a particular type of "fittest" for its logic (that evolutions consistent approach to the "fittest" should generally lead to decreasing diversity) to hold, and that is precisely the kind of static "fittest" that does not exist that I was taking issue with...
The usual definition is more like highest likelihood of having offspring survive to reproduce in its environment. But I didn't confuse anything, I was addressing the kind of "fittest" that was required for the argument being justified on the basis of the approach to the "fittest" to hold. Essentially, I was arguing that the poster to whom I was responding didn't understand what the "fittest" in the survival of the fittest meant and entailed, since the concept of "fittest" necessary for the argument to work was different from the kind of "fittest" actually relevant to evolution.
The actual definition usually used has a quite clear absolute measure, though its not strongest, smartest, or prettiest (at least one of which isn't an absolute objective measure, either.)
A test doesn't replace documentation. Documentation describes what the code does. A test verifies that, in the case covered by the test, the code does, in fact, do what it is intended to do. But you can't infer the information in good documentation from an (undocumented) test suite, because while the test cases will specify behavior in specific conditions, you can't (generally) unambiguously infer what behavior is expected in other conditions, particularly when the task being tested is nontrivial. A test is always a specific case of an operationalization of a specification, even if no one has ever reduce the specification to writing. But it would be good if they did. Its true that good documentation isn't always worth the effort, but its equally true, I think, that it is almost always more worth the effort than it would seem intuitively at the time the code is written, so if the impression at the time of first developing the code is the guide to documentation, code always ends up underdocumented which tends to cause problems down the road.
Actually, in this case, I think the previous poster wanted 0/0 = 1, not 0.
And, IIRC, division by zero isn't always undefined, its just always not a real number. I think 0/0 is still undefined, though I could be mistaken.
Uh, no. Languages, particularly ones in wide use where continued support of legacy code is important, don't, in the real world, necessarily drop features just because they aren't useful or necessary if they are in wide use in existing code, even if that code is poorly written. So you can't infer utility from the fact that it still exists.
Plenty of newer languages (and, for that matter, some older ones), of course, don't have gotos, and get along just fine without it. Gotos may or may not be useful in languages more abstract than assembly, but they certainly aren't necessary.
Doesn't XSL:FO handle that? I remember back when there were XSL:FO browsers in the works (and early versions released even), but it looks like that, sadly, all fell apart, and XSL:FO mostly got relegate to generating output aimed at print, where it is used at all.
You still need as well as , and, of course, you still need hyperlinks and images and other things which don't define structure.
What he was referring to (and challenging me in the process) was document structure, not metadata. There is a substantial difference. Its much harder to abuse, its fairly easy to use consistently in a way that is helpful, and its information that while it can be obscured through clever styling, will generally be visible to page viewers.
No, there's not some Platonic ideal of "fittest" that is being progressively approached.
Depending on the timescale, environment, and the universe of analysis, evolution may tend toward a more or less diversity over time (e.g., overall, diversity has increased from the first life, while diversity in a narrow population over a limited time may decrease.)
A "deprecated" tag is one that implementations should support, but developers should not use for new pages, since it is planned to be removed as "obsolete" in a future version of the standard.
An "obsolete" tag is one that it is no longer expected that implementations will support.
Supporting deprecated tags is what browsers should do.
An article, as division of a document, is semantically no different than any other division of a document. So there is no semantic difference between:
<article>...</article> and <div class="article">...</div>
Inasmuch as there is a difference between an "article" and another division of the document, the distinction is purely a matter of the internal heirarchical structure of a specific document. There is no consistent interpretation that can or should be applied to an "article" beyond that of a division of a document.
Really? Seems to me that the NSF and the DOE Office of Science tend to work together a lot, and sharing facilities, cross-detailing personnel, etc., are pretty common between those two organizations.
Yes, but you have the ballots. All-electronic systems with no separate ballots don't allow random-precinct hand-count confirmation, or even full-election hand recounts if there is cause.
Systems that generate a ballot, which the voter than confirms and turns in and is then counted by a separate machine do allow that, so even if the ballot reader is just as easily compromised as a voting machine in an all-electronic system would be, the election is more secure.
This issue is not how secure is the machine, it is how secure is an election using the machines; using vote counting machines like, e.g., optical scan machines, that count voter-produced or voter-verified ballots which are both human and machine readable and which are available for auditing produces a more secure election than using systems where the "ballots" exist only as electronic data within the system.
Um, hanging chads are a problem with with ballots designed for machine counting that are resolvable only by hand counting, not a problem with ballots designed to be counted by hand.
What kind of machine (if any) you produce a ballot on is a logically orthogonal question to how the ballots are counted. Computer voting can be done with hand counted ballots. Hand-filled ballots can be counted by machines. Whether ballots should be counted by machine, and what kind, and what kind of relation that machine should have (if any) to the production of the ballots are all open questions that cannot just be waved away out of the discussion.
And the answer to that may well be "we should change it back into the old way". Or it may not. But its certainly a valid response, and you've provided no counterargument to it except that you apparently don't like the option even being considered and wish people would eliminate it without thought.
And a good thing, too, as that is really not the business of The Fed.
I think you mean to say "The feds don't" rather than "The fed doesn't".
1. Modchips are not "backup copies", legitimate or otherwise. They are, arguably, circumvention devices under the DMCA. Whether some of the games that some people play with them may be "legitimate backup copies" is a different issue.
2. I suspect the number of people using modchips to play "legitimate backups" is small compared to the number using modchips to play pirated copies.
3. If a particular copy is illegal (under the DMCA or otherwise), it is not a "legitimate" backup.
4. When a coordinated raid like this is done, its usually (there are exceptions) done based on evidence of some kind of coordinated enterprise. Even if people use modchips to play backups (technically illegal or not) of games they own legitimate copies of, I doubt very much that that's the focus of these raids.
That's a very long-winded way of saying "never".
It is broken up, as I understand, both by portions of the spectrum and geographic regions. Assuming there aren't limits on how much of the spectrum in any one region any one interest can buy (I don't know that much about the particular auction), the arguably, if the FCC isn't going to impose freedom and Google really values it, its best move may be to find one region that lots of other carriers would like access to and put as much as it can afford into buying up spectrum in that market, and then provide access to other carriers on condition of them adopting Google's openness principles on the rest of their holdings in the 700MHz range.
Why would it seem to mean that? They announced that if their principles were adopted, they would bid up to a specific amount on a specific share of that spectrum. They did not say, that I recall, that if there demands were not (or incompletely) addressed, they would definitely bid nothing.
Why, the Windows Genuine Advantage, of course.
I dunno what you consider "few" but lots of Windows GUI apps I've encountered accept command-line arguments. Most users don't recognize it because they just click the pretty icon, of course, but that doesn't mean they aren't there.
Doesn't make a whole lot of sense for (and probably unconstitutional, too) where the action is commenced under state law, and anyway, many class action suits aren't seeking punitive damages (and most settle, anyhow.)
Suits where punitive damages are awarded are generally more serious, and more deliberate wrongs, and are arguably the cases that it is most in the public interest to have brought; if the plaintiff didn't get the punitive damage award (and if they could not, therefore, be part of the basis for contingency fees), there'd be less incentive for plaintiffs to bring and less incentive for attorney's to take such cases, but while at the same time not touching the vast majority of controversial lawsuits, which primarily seek compensatory damages of an amount that observers (often ill-informed, but sometimes not) see as disporportionate to the wrong done.
Perhaps, but that's for a time-and-a-half the cost of "high end" of the ASUS inexpensive laptops.
This is kind of ironic coming from someone who is dimissive of the US involvement on the basis that only Pearl Harbor got the US to joing at all, considering the position of Nazi Germany and Stalin's USSR prior to the point at which their totalitarian imperialistic ambitions could no longer coexist.
The average Joe manifestly wouldn't if everything were exactly the way it is today.
I'm not sure what this is supposed to mean. I suspect that you left something out of this sentence, but I'm not sure what since I can think of a few different ways that it could have been meant.
This is a fairly big factor in the current market, since it is, in effect, a subsidy for fuel consumption which encourages people to buy wasteful vehicles (and/or use vehicles in a wasteful manner) compared to the situation if the external costs were internalized.
Sooner or later we will; oil supplies alone should guarantee that. Its really more a matter of how controlled the transition is, and how much damage is done first.