The company plans to unveil its first car in London later this month, a small two-seater that weighs roughly 700 pounds.
A car that will never sell anywhere in the US due to total inability to pass crash safety test. I'm actually surprised that it can be sold anywhere in the first world, to be honest.
Technetium, element 43, has no stable isotopes. Do you want to forbid people from referring to it as an element? That would be kind of silly. Chemists can do reactions with technetium, form compounds with it, etc.
Or if you want to arbitrarily pick some minimum half-life, what is that half-life going to be?
I think you just gave me a perfectly non-arbitrary standard -- if it lives long enough to do chemistry, it's an element. Certainly many of the very heavy atoms with half-life in the nanoseconds don't qualify. This makes a lot of sense to me now too -- an element is defined by its chemical properties and so without chemistry, no element.
No, the nuclei of elements below (and including) lead are infinitely stable* ** -- you would have to input energy to break those nuclei apart. Be grateful for this -- this allows the sun to generate energy from 4H->He -- that reaction is downhill in energy, albeit with a huge activation barrier.
If all elements decayed into hydrogen, nuclear fusion would be impossible and the universe would be a colder, darker place than it already is.
* That is, there exists at least one infinitely-stable isotope for that element, for those that want to quibble.
** According to all proven physics (e.g. the Standard Model -- you don't get a name like that without being pretty well hashed out). String theory suggests the decay of the proton, which might change this. On the other hand, string theory suggests all kinds of wacky and sometime contradictory (there are lots of implementations) things that need to be sorted out by experiment before we start quoting them like fact.
If you want to quibble about semantics, here's one for you: an atom with a half-life
That's a good quibble, but so long has passed from when the term was coined (back when the atom was thought to be atomic in the indivisible sense) that it's impossible to go back and rework the language. By contrast, my quibble here involves naming an "element" right now, so it's a current dispute.
Reading Democritus is fun though! Better than the infinite-divisibility theories of the time -- bones are made of smaller bones which are themselves made of smaller bones ad infinitum!
A nucleus with a half-life measured in milliseconds or smaller doesn't seem to qualify, at least in my sort of language-to-thought translator, as really as an "element". That word carries with it the connotation of actual material existence which seems incompatible with its inability to actually exist for any period of time on the human scale.
I freely admit this is a quibble, but this sort of thing bugs me. Yes, IAAP and this rant has no bearing whatsoever on the scientific merits of the research (not my field, so I'll pass on that) and is just about the naming.
Especially since, from what I hear, areas around many airports have been essentially turned into wetlands.
(1) Flight 1549 was 5 miles from the airport at the time of the bird strike, meaning that they have to patrol a huge area (especially hard in the NYC metro area) to get rid of all the nesting sites there.
(2) The Canada (blame Canada!) geese that were ingested into the engine were just passing through the area on their migration route. So any sort of habitat destruction on the ground would have zero effect on them anyway. Good luck changing their migration routes too.
So, at least in this instance, there was basically nothing you could do about it except have trained pilots well-versed in emergency procedures. In fact, as a general matter, I think it's silly to invest in technology/training/whatever that solves an individual problem when you can invest in other measures that will accrue benefits across a wide variety of (perhaps unexpected) problems.
The other thing to note is that there are some CPUs that by this metric are clearly just not very worth it where their are cheaper ones that perform better.
Depends on how much you value the time it takes to complete $PROCESSOR_INTENSIVE_TASK. If I hired software engineers for $100,000 a year, that's $1 a minute and so it might very well be "worth it" to spend $1000 on the latest N-core processor (compilation being nearly infinitely parallel) if that save 5 minutes per compile over 200 compiles on the year.
The only thing the plot proves is that there are diminishing returns, not that any particular price/performance point is not "worth it".
The LGPL expressly allows closed-source and even non-free-as-in-beer software to link to an LGPL library, either statically or dynamically, without violation of its terms. That's what makes it lesser than the GPL.
I agree with sentiment in the last link though, this is none of the FSF's business -- the FFMPEG people are the only ones that can claim to be aggrieved here. Until that happens, this is much ado about nothing since there can be no violation of license terms to which the holder of the copyright does not object.
Before that, switched networking had taken over. No-one was using T connectors and terminators by then.. and switching hubs were cheaper than broadcast hubs for UTP and active ARP attacks hadn't been demonstrated. Still, its amazing that such a good implementation of ssh came along when it did.
So the irony is that people were fooled into a false sense of security by switched networks on the theory that you can trust the network only to deliver your packets to the intended addressee? How is that ironic? It's just foolishly confusing a performance issue with a security promise -- if that were ironic then the whole internet would be a pile of sarcastic innuendo.
For the most part you don't "reuse" libraries because of storage space but in order to reuse the code.
Then distribute the library with the application or statically linked into the binary.
But would you gladly sacrifice having a security hole and instead of correcting at one place and then have corrected all apps that use it having to upgrade each and every program that happens to use it and then not knowing if per chance there's still another copy hidden overthere?
Most apps are updated fairly often anyway. Let them reuse the Firefox updater, which I think is the perfect example of usability and transparency -- if there's an update, grab it and apply it on next launch. No problems whatsoever.
SDL, GTK+, etc, and the dlls have to be included with the binary downloads because Windows/Mac don't have an easy to use package manager.
Please, please, please stop with this stingy attitude towards distributing libraries. We have megabits of bandwidth and terabytes of disk space and yet there is an almost herculean effort made to economize a few GB by using dynamic linking where it's not strictly needed. The time and effort spent on these measures dwarfs the extra download times associated with statically linking as much as humanly possible.
As much as possible, every application should be entirely self-contained and reliant only on system calls in the underlying OS. I will gladly sacrifice any amount of diskspace not to deal with my package manager and insane requirements.
For example, maybe it's cheaper to reuse an int than to throw away the old one and create a new one. If you clustered similar objects together in memory you could use such a technique to try to avoid memory fragmentation.
Not *trying* to rag on your idea, but this just isn't true in x86. Not by an order of magnitude actually, since an int on the stack can be operated on for free while one on the heap has an extra indirection operator required before it can be operated on. In many contexts, it's cheaper to actually make a local copy of an int stored in some object elsewhere and update it when you are down.
Moreover, locality of reference means putting together data that will be accessed together, not data of the same basic type. If "Customer" is defined by ulong _id, float _balance, enum_t (int) type, all those should be contiguous in memory (the C++ style) instead of having an array of floats for the balances, an array of ulongs for the ids (the Fortran style) so that they can fit in a cache line.
Not too long ago people where word-processing, spread-sheeting, data-basing, developing software and even Windows, heck, even using AutoCAD on a Pentium II. Or a 486 if you go farther back a bit more.
Not too long ago, I remember having to wait 15-20 minutes to TeX up my research papers, only to find out that I missed a curly brace somewhere.
Not too long ago, my spreadsheet couldn't import data from a MySQL database halfway around the world through the internet.
Not too long ago, the database that I run on that other computer would need a refrigerator-sized mainframe.
Not too long ago, developing software meant that it was faster to manually read for syntax errors than to just compile and have the IDE flag the errors. On a project 1/20th the size, at least half of which was implementing things that are now in libraries. Actually, as I recall, I didn't have an IDE, just a dumb terminal. The debugger was crap to -- it pales in comparison to what I have today.
Never used CAD software, but I bet dollars to donuts that in the 12 years since the Pentium II, it's also come a damned long way. And that's the problem with these comparisons -- people may have been doing the same tasks but they were still doing much much less than we casually do today. In many ways, we the usefulness of the tasks themselves expands to fill the available power -- our programs and environments get better and better.
If 10 years ago you would have told me that I'd be running a miniature search engine on my computer, crawling and indexing my filesystems to save me the trouble of finding files, I'd say you were nuts. Today, I can't remember how I lived without Google Desktop: ctrl ctrl + filename and the results are there. To say that somehow this is comparable to my computer 10 years ago because they both perform the same basic function -- allowing access to saved files -- is disingenuous. They are the same in the way that a steak knife and a chainsaw are the same. That all goes for the modern web, AJAX and all, versus the web that I browsed back in the dark old days. Same for programming, same for just about everything I can think of.
Computers do more than they did. This is a GOOD THING. Stop convincing yourself that somehow what they do now is good enough for the future. I hope it's not, and I'm working to make sure that it's not by pursuing more ways that my computer can do more for me.
So, while I agree with the idea of reforming the system, and then enforcing it, I think it has to be done in that order if you are truly worried about "fairness."
Accepted. We'll increase the amount of legal immigrants allowed in by a factor of 10 and remove all the quotas on Monday. On Tuesday we will deport all the illegals.
Illegal immigration has nothing to do with mischievous Mexicans that don't want to play by the rules. If the current immigration laws were fair, do you really think so many of them would be risking their lives to come into the US?
I don't think that a foreigner has the right to judge whether our laws are fair and decide, based on that judgment, whether they should be followed. There are many laws that I find unfair, but I follow them because I respect my fellow citizens' right to pass laws that I don't approve of. If you don't understand that very basic concept (that you are not the final authority on the law) then you don't really understand the American system of government.
Like I've said 3 times now, I support massive legal immigration, but only if we get serious about deporting all the illegal immigrants. That's my compromise position.
Maybe if they would lighten up the rules a bit there would be fewer people seeking illegal methods of entry.
Maybe you should have read my post in which I offered massive increases in legal immigration as a policy alternative, but only if combined with massive deportation of illegal immigrants. That's my compromise offer.
I don't care if they pay taxes, I care about helping people that follow the rules and not the people that think it's their god-given right to enter a country against the law. Ultimately, that's what it comes down to -- illegal immigrants are breaking the law for their own benefit while legal immigrants (who I support fully and think we should admit much more of) get to wait in line for years. By giving jobs to illegals, you are rewarding criminality and you are rewarding the kind of people that don't respect the rule of law -- the basic cornerstone of the American Republic.
How about the contractors who accept forged SSNs and then proceed to duly withhold and file all payroll related taxes?
Don't care. They knowingly accepted a forged SSN and should be punished for violating a fairly simple and straightforward law. Yes, it's nice that they paid taxes, but my main concern with illegal immigration is not taxes -- it's basic fairness.
It is unfair to the legal immigrants who did things according to the rules to allow those that skipped the lines to have the same benefits.
It's not a matter of xenophobia. For most people anyway. Illegal immigration is a very real social and economic problem.
I'd like to second this. I'm not xenophobic -- I support allowing a large number of legal immigrants into the country each year under fairly generous terms. I oppose all forms of ethnic quotas and other restrictive immigration policies. I support giving legal immigrants nearly full access to the benefits of citizenship as soon as they arrive and additional services (if they want) to help them in adjusting to a different country. Hopefully, this is enough to convince people that I'm not, by any stretch of the imagination, anti-immigrant.
On the other hand, I am a firm believer in the need to enforce the law with regards to illegal immigrants -- deport them and bar them from reentry. These positions aren't contradictory and, in fact, I see them as complementary -- by increasing legal immigration and throwing out all the illegal immigrants, we will be rewarding those honest people that follow the rules instead of those that decide that they have the right to break the law to get what they want. Those are the kind of people that we ought to be allowing to immigrate. The incentives in our current system are perversely the opposite of this -- it punishes those that want to follow the rules with onerous waits and arbitrary terms while rewarding those that skip in line with amnesty and "safe haven". It's ludicrous, and I blame both the GOP for stymieing legal immigration and the Dems for stymieing systematic attempts to identify and deport illegals and punish unscrupulous employers (only the really negligent, of course -- not every contractor that accepts a forged SSN deserves to get canned, but the ones that intentionally look the other way certainly do).
Such a partisan football is made out of what I thought was just common sense -- it's depressing really. I can't understand it -- I just can't. It's some sort of collective insanity we've entered in this country.
If my understanding of the technology is correct, the seek time on most hard drives already limits drive access speed to typically be slower than 3Gb/sec. Would this rely on a transition to Solid State Drives for any noticeable difference in performance?
The seek time has nothing to do with the throughput. The seek time refers to the latency between when a read command is issued and when it begins to be fulfilled. The throughput refers to the data transferred per unit time during fulfillment.
Here's a nice car analogy for those of us in New England -- consider the Mass Pike versus I-93. The Mass Pike has a very long seek time from the onramp because of the toll lanes (and the mouth breathers that won't get a transponder even though they are now free and clog the automatic lanes) but once you get on the highway, you can go 80 MPH until your exit. On I-93, by contrast, you can get right on, but you will be going 30 MPH for the duration. Of course, if you drive down to CT and get on I-84, you have a low-latency AND high throughput highway but if you drive too far down to, say, the Bronx, it becomes high-latency and low throughput.
The problem with parallel is that you can't crank up the clock speed because you have to make sure that the signal on each line is combined with the ones from the other lines that were sent at the same time. This limits how fast you can send the send the bits (if the time being bits is comparable to the skew time, the receiver will not be able to reliably reassemble the data) and how long the interconnect can be (skew being linearly amplified by length). It's not for nothing that PCI has been replaced with PCI-E, PATA with SATA, SCSI with SAS. USB and IEE1394 would be impossible with parallel. Serial communications are more reliable and more scalable (one big exception -- wireless RF, but that's not what we are discussing here).
Multiprocessing, incidentally, has nothing to do with it -- the software interface to a storage device hides all the implementation details (PATA/SATA, for instance) anyway. The hard part in multi-threading IO-intensive apps has quite a bit more to do with latency issues and atomicity guarantees (the complete lack thereof) rather than the inability of the storage device to do 2 things at once (which, for a physical disk, is impossible anyway, meaning that it would have to back-convert into a serial process anyway).
Again (I'm starting to sound like a broken record), this happened because users demanded it:
User: Why do I lose my formatting when I copy/paste from one word document to the other? MS: Hmm, the clipboard, I guess we'll have to include some formatting. MS2: But what if a user pastes into notepad/IE/???? MS: Hmm, better make it secret-invisible formatting so that non-HTML aware apps can understand it.
There is just no way around the fact that users want two contradictory things in a shared resource -- they want all their apps to be able to copy/paste at the same time they want formatting preserved (as much as possible) when doing so. Those two things just can't be reconciled without a nasty hack because they don't make any sense together.
The clipboard is a shared-resource between apps -- making data both portable and rich (i.e. containing structure) in a shared-resource is one of those problems that everyone knows is hard!
This is a feature. If there is no mouse input to a window, Windows decides that it is a background process and assigns it a low priority so that it does not disrupt the performance of interactive tasks. Of course, that relegates it to the same category as all the other background process where it competes (when you wiggle the mouse, it's interactive and therefore preempts all the background junk). Easy workaround: (hahaha, I know, retarded) go to process manager and manually set the priority of the installer "above normal".
This happened, of course, because users were complaining that Word was slow because AV/backup/defrag was running in the background causing the latency to spike. It's not a perfect solution but it's damned hard to think of how to better implement it (aside from making it easier to manually set the priority by moving out of process manager into the window interface, which is clutter IMO).
Sometimes users don't seem to get that no set of system rules can work perfectly under all conditions, and it's better to have rules that screw everyone a little (by making installs slow) than screwing some cases (running Word while a backup is going) a lot.
Indeed. The most obvious example of the inability to consider threshold doses is the ubiquitous warning "This chemical is known to the State of CA to cause cancer" without noting "By the way, the amount in this product is 10 times lower than the threshold does and 1000000 lower than the LD50.
Among the many exceptions to copyright (regularly used by the BBC), is for a bona fide news organization to report the news. There are many other exceptions, such as home recording for personal use, fair use, backups, satire, parody, teaching & education. The fine print varies by jurisdiction.
One of the misunderstandings of 'fair use' is that it creates an affirmative right to do the thing in question. It does not. 'Fair use' is an affirmative defense to copyright infringement, it means that if someone sues you for violating their copyright, you can counter by saying "what I did is fair use" -- it functions as an exception to the criminal/civil statute.
What it doesn't do, is create an affirmative obligation on the part of the providers to allow you to make fair use of their works. That is to say, if act X (excerpting for a class) is fair use, they may not sue you for X but they need no help you do X and they may even have technological measures (CSS) that frustrate X.
Yes, it sucks, and yes, this does mean that content providers can use the DMCA + technological measures to frustrate your "right" to fair use. That's the way the law is written, I'd like to have it changed, but for until it gets changed, that's the way it is.
Slashdot Mirror
The company plans to unveil its first car in London later this month, a small two-seater that weighs roughly 700 pounds.
A car that will never sell anywhere in the US due to total inability to pass crash safety test. I'm actually surprised that it can be sold anywhere in the first world, to be honest.
Technetium, element 43, has no stable isotopes. Do you want to forbid people from referring to it as an element? That would be kind of silly. Chemists can do reactions with technetium, form compounds with it, etc.
Or if you want to arbitrarily pick some minimum half-life, what is that half-life going to be?
I think you just gave me a perfectly non-arbitrary standard -- if it lives long enough to do chemistry, it's an element. Certainly many of the very heavy atoms with half-life in the nanoseconds don't qualify. This makes a lot of sense to me now too -- an element is defined by its chemical properties and so without chemistry, no element.
Thanks for helping me out there!
No, the nuclei of elements below (and including) lead are infinitely stable* ** -- you would have to input energy to break those nuclei apart. Be grateful for this -- this allows the sun to generate energy from 4H->He -- that reaction is downhill in energy, albeit with a huge activation barrier.
If all elements decayed into hydrogen, nuclear fusion would be impossible and the universe would be a colder, darker place than it already is.
* That is, there exists at least one infinitely-stable isotope for that element, for those that want to quibble.
** According to all proven physics (e.g. the Standard Model -- you don't get a name like that without being pretty well hashed out). String theory suggests the decay of the proton, which might change this. On the other hand, string theory suggests all kinds of wacky and sometime contradictory (there are lots of implementations) things that need to be sorted out by experiment before we start quoting them like fact.
Further Reading:
http://en.wikipedia.org/wiki/Stable_isotope
http://en.wikipedia.org/wiki/List_of_elements_by_nuclear_stability
http://en.wikipedia.org/wiki/Proton_decay
If you want to quibble about semantics, here's one for you: an atom with a half-life
That's a good quibble, but so long has passed from when the term was coined (back when the atom was thought to be atomic in the indivisible sense) that it's impossible to go back and rework the language. By contrast, my quibble here involves naming an "element" right now, so it's a current dispute.
Reading Democritus is fun though! Better than the infinite-divisibility theories of the time -- bones are made of smaller bones which are themselves made of smaller bones ad infinitum!
A nucleus with a half-life measured in milliseconds or smaller doesn't seem to qualify, at least in my sort of language-to-thought translator, as really as an "element". That word carries with it the connotation of actual material existence which seems incompatible with its inability to actually exist for any period of time on the human scale.
I freely admit this is a quibble, but this sort of thing bugs me. Yes, IAAP and this rant has no bearing whatsoever on the scientific merits of the research (not my field, so I'll pass on that) and is just about the naming.
Especially since, from what I hear, areas around many airports have been essentially turned into wetlands.
(1) Flight 1549 was 5 miles from the airport at the time of the bird strike, meaning that they have to patrol a huge area (especially hard in the NYC metro area) to get rid of all the nesting sites there.
(2) The Canada (blame Canada!) geese that were ingested into the engine were just passing through the area on their migration route. So any sort of habitat destruction on the ground would have zero effect on them anyway. Good luck changing their migration routes too.
So, at least in this instance, there was basically nothing you could do about it except have trained pilots well-versed in emergency procedures. In fact, as a general matter, I think it's silly to invest in technology/training/whatever that solves an individual problem when you can invest in other measures that will accrue benefits across a wide variety of (perhaps unexpected) problems.
The other thing to note is that there are some CPUs that by this metric are clearly just not very worth it where their are cheaper ones that perform better.
Depends on how much you value the time it takes to complete $PROCESSOR_INTENSIVE_TASK. If I hired software engineers for $100,000 a year, that's $1 a minute and so it might very well be "worth it" to spend $1000 on the latest N-core processor (compilation being nearly infinitely parallel) if that save 5 minutes per compile over 200 compiles on the year.
The only thing the plot proves is that there are diminishing returns, not that any particular price/performance point is not "worth it".
It isn't when it's inside of Chrome.
The LGPL expressly allows closed-source and even non-free-as-in-beer software to link to an LGPL library, either statically or dynamically, without violation of its terms. That's what makes it lesser than the GPL.
I agree with sentiment in the last link though, this is none of the FSF's business -- the FFMPEG people are the only ones that can claim to be aggrieved here. Until that happens, this is much ado about nothing since there can be no violation of license terms to which the holder of the copyright does not object.
Before that, switched networking had taken over. No-one was using T connectors and terminators by then.. and switching hubs were cheaper than broadcast hubs for UTP and active ARP attacks hadn't been demonstrated. Still, its amazing that such a good implementation of ssh came along when it did.
So the irony is that people were fooled into a false sense of security by switched networks on the theory that you can trust the network only to deliver your packets to the intended addressee? How is that ironic? It's just foolishly confusing a performance issue with a security promise -- if that were ironic then the whole internet would be a pile of sarcastic innuendo.
Oh, shit.
For the most part you don't "reuse" libraries because of storage space but in order to reuse the code.
Then distribute the library with the application or statically linked into the binary.
But would you gladly sacrifice having a security hole and instead of correcting at one place and then have corrected all apps that use it having to upgrade each and every program that happens to use it and then not knowing if per chance there's still another copy hidden overthere?
Most apps are updated fairly often anyway. Let them reuse the Firefox updater, which I think is the perfect example of usability and transparency -- if there's an update, grab it and apply it on next launch. No problems whatsoever.
SDL, GTK+, etc, and the dlls have to be included with the binary downloads because Windows/Mac don't have an easy to use package manager.
Please, please, please stop with this stingy attitude towards distributing libraries. We have megabits of bandwidth and terabytes of disk space and yet there is an almost herculean effort made to economize a few GB by using dynamic linking where it's not strictly needed. The time and effort spent on these measures dwarfs the extra download times associated with statically linking as much as humanly possible.
As much as possible, every application should be entirely self-contained and reliant only on system calls in the underlying OS. I will gladly sacrifice any amount of diskspace not to deal with my package manager and insane requirements.
For example, maybe it's cheaper to reuse an int than to throw away the old one and create a new one. If you clustered similar objects together in memory you could use such a technique to try to avoid memory fragmentation.
Not *trying* to rag on your idea, but this just isn't true in x86. Not by an order of magnitude actually, since an int on the stack can be operated on for free while one on the heap has an extra indirection operator required before it can be operated on. In many contexts, it's cheaper to actually make a local copy of an int stored in some object elsewhere and update it when you are down.
Moreover, locality of reference means putting together data that will be accessed together, not data of the same basic type. If "Customer" is defined by ulong _id, float _balance, enum_t (int) type, all those should be contiguous in memory (the C++ style) instead of having an array of floats for the balances, an array of ulongs for the ids (the Fortran style) so that they can fit in a cache line.
Not too long ago people where word-processing, spread-sheeting, data-basing, developing software and even Windows, heck, even using AutoCAD on a Pentium II. Or a 486 if you go farther back a bit more.
Not too long ago, I remember having to wait 15-20 minutes to TeX up my research papers, only to find out that I missed a curly brace somewhere.
Not too long ago, my spreadsheet couldn't import data from a MySQL database halfway around the world through the internet.
Not too long ago, the database that I run on that other computer would need a refrigerator-sized mainframe.
Not too long ago, developing software meant that it was faster to manually read for syntax errors than to just compile and have the IDE flag the errors. On a project 1/20th the size, at least half of which was implementing things that are now in libraries. Actually, as I recall, I didn't have an IDE, just a dumb terminal. The debugger was crap to -- it pales in comparison to what I have today.
Never used CAD software, but I bet dollars to donuts that in the 12 years since the Pentium II, it's also come a damned long way. And that's the problem with these comparisons -- people may have been doing the same tasks but they were still doing much much less than we casually do today. In many ways, we the usefulness of the tasks themselves expands to fill the available power -- our programs and environments get better and better.
If 10 years ago you would have told me that I'd be running a miniature search engine on my computer, crawling and indexing my filesystems to save me the trouble of finding files, I'd say you were nuts. Today, I can't remember how I lived without Google Desktop: ctrl ctrl + filename and the results are there. To say that somehow this is comparable to my computer 10 years ago because they both perform the same basic function -- allowing access to saved files -- is disingenuous. They are the same in the way that a steak knife and a chainsaw are the same. That all goes for the modern web, AJAX and all, versus the web that I browsed back in the dark old days. Same for programming, same for just about everything I can think of.
Computers do more than they did. This is a GOOD THING. Stop convincing yourself that somehow what they do now is good enough for the future. I hope it's not, and I'm working to make sure that it's not by pursuing more ways that my computer can do more for me.
So, while I agree with the idea of reforming the system, and then enforcing it, I think it has to be done in that order if you are truly worried about "fairness."
Accepted. We'll increase the amount of legal immigrants allowed in by a factor of 10 and remove all the quotas on Monday. On Tuesday we will deport all the illegals.
Illegal immigration has nothing to do with mischievous Mexicans that don't want to play by the rules. If the current immigration laws were fair, do you really think so many of them would be risking their lives to come into the US?
I don't think that a foreigner has the right to judge whether our laws are fair and decide, based on that judgment, whether they should be followed. There are many laws that I find unfair, but I follow them because I respect my fellow citizens' right to pass laws that I don't approve of. If you don't understand that very basic concept (that you are not the final authority on the law) then you don't really understand the American system of government.
Like I've said 3 times now, I support massive legal immigration, but only if we get serious about deporting all the illegal immigrants. That's my compromise position.
Maybe if they would lighten up the rules a bit there would be fewer people seeking illegal methods of entry.
Maybe you should have read my post in which I offered massive increases in legal immigration as a policy alternative, but only if combined with massive deportation of illegal immigrants. That's my compromise offer.
I don't care if they pay taxes, I care about helping people that follow the rules and not the people that think it's their god-given right to enter a country against the law. Ultimately, that's what it comes down to -- illegal immigrants are breaking the law for their own benefit while legal immigrants (who I support fully and think we should admit much more of) get to wait in line for years. By giving jobs to illegals, you are rewarding criminality and you are rewarding the kind of people that don't respect the rule of law -- the basic cornerstone of the American Republic.
How about the contractors who accept forged SSNs and then proceed to duly withhold and file all payroll related taxes?
Don't care. They knowingly accepted a forged SSN and should be punished for violating a fairly simple and straightforward law. Yes, it's nice that they paid taxes, but my main concern with illegal immigration is not taxes -- it's basic fairness.
It is unfair to the legal immigrants who did things according to the rules to allow those that skipped the lines to have the same benefits.
It's not a matter of xenophobia. For most people anyway. Illegal immigration is a very real social and economic problem.
I'd like to second this. I'm not xenophobic -- I support allowing a large number of legal immigrants into the country each year under fairly generous terms. I oppose all forms of ethnic quotas and other restrictive immigration policies. I support giving legal immigrants nearly full access to the benefits of citizenship as soon as they arrive and additional services (if they want) to help them in adjusting to a different country. Hopefully, this is enough to convince people that I'm not, by any stretch of the imagination, anti-immigrant.
On the other hand, I am a firm believer in the need to enforce the law with regards to illegal immigrants -- deport them and bar them from reentry. These positions aren't contradictory and, in fact, I see them as complementary -- by increasing legal immigration and throwing out all the illegal immigrants, we will be rewarding those honest people that follow the rules instead of those that decide that they have the right to break the law to get what they want. Those are the kind of people that we ought to be allowing to immigrate. The incentives in our current system are perversely the opposite of this -- it punishes those that want to follow the rules with onerous waits and arbitrary terms while rewarding those that skip in line with amnesty and "safe haven". It's ludicrous, and I blame both the GOP for stymieing legal immigration and the Dems for stymieing systematic attempts to identify and deport illegals and punish unscrupulous employers (only the really negligent, of course -- not every contractor that accepts a forged SSN deserves to get canned, but the ones that intentionally look the other way certainly do).
Such a partisan football is made out of what I thought was just common sense -- it's depressing really. I can't understand it -- I just can't. It's some sort of collective insanity we've entered in this country.
If my understanding of the technology is correct, the seek time on most hard drives already limits drive access speed to typically be slower than 3Gb/sec. Would this rely on a transition to Solid State Drives for any noticeable difference in performance?
The seek time has nothing to do with the throughput. The seek time refers to the latency between when a read command is issued and when it begins to be fulfilled. The throughput refers to the data transferred per unit time during fulfillment.
Here's a nice car analogy for those of us in New England -- consider the Mass Pike versus I-93. The Mass Pike has a very long seek time from the onramp because of the toll lanes (and the mouth breathers that won't get a transponder even though they are now free and clog the automatic lanes) but once you get on the highway, you can go 80 MPH until your exit. On I-93, by contrast, you can get right on, but you will be going 30 MPH for the duration. Of course, if you drive down to CT and get on I-84, you have a low-latency AND high throughput highway but if you drive too far down to, say, the Bronx, it becomes high-latency and low throughput.
The problem with parallel is that you can't crank up the clock speed because you have to make sure that the signal on each line is combined with the ones from the other lines that were sent at the same time. This limits how fast you can send the send the bits (if the time being bits is comparable to the skew time, the receiver will not be able to reliably reassemble the data) and how long the interconnect can be (skew being linearly amplified by length). It's not for nothing that PCI has been replaced with PCI-E, PATA with SATA, SCSI with SAS. USB and IEE1394 would be impossible with parallel. Serial communications are more reliable and more scalable (one big exception -- wireless RF, but that's not what we are discussing here).
Multiprocessing, incidentally, has nothing to do with it -- the software interface to a storage device hides all the implementation details (PATA/SATA, for instance) anyway. The hard part in multi-threading IO-intensive apps has quite a bit more to do with latency issues and atomicity guarantees (the complete lack thereof) rather than the inability of the storage device to do 2 things at once (which, for a physical disk, is impossible anyway, meaning that it would have to back-convert into a serial process anyway).
Again (I'm starting to sound like a broken record), this happened because users demanded it:
User: Why do I lose my formatting when I copy/paste from one word document to the other?
MS: Hmm, the clipboard, I guess we'll have to include some formatting.
MS2: But what if a user pastes into notepad/IE/????
MS: Hmm, better make it secret-invisible formatting so that non-HTML aware apps can understand it.
There is just no way around the fact that users want two contradictory things in a shared resource -- they want all their apps to be able to copy/paste at the same time they want formatting preserved (as much as possible) when doing so. Those two things just can't be reconciled without a nasty hack because they don't make any sense together.
The clipboard is a shared-resource between apps -- making data both portable and rich (i.e. containing structure) in a shared-resource is one of those problems that everyone knows is hard!
This is a feature. If there is no mouse input to a window, Windows decides that it is a background process and assigns it a low priority so that it does not disrupt the performance of interactive tasks. Of course, that relegates it to the same category as all the other background process where it competes (when you wiggle the mouse, it's interactive and therefore preempts all the background junk). Easy workaround: (hahaha, I know, retarded) go to process manager and manually set the priority of the installer "above normal".
This happened, of course, because users were complaining that Word was slow because AV/backup/defrag was running in the background causing the latency to spike. It's not a perfect solution but it's damned hard to think of how to better implement it (aside from making it easier to manually set the priority by moving out of process manager into the window interface, which is clutter IMO).
Sometimes users don't seem to get that no set of system rules can work perfectly under all conditions, and it's better to have rules that screw everyone a little (by making installs slow) than screwing some cases (running Word while a backup is going) a lot.
Indeed. The most obvious example of the inability to consider threshold doses is the ubiquitous warning "This chemical is known to the State of CA to cause cancer" without noting "By the way, the amount in this product is 10 times lower than the threshold does and 1000000 lower than the LD50.
Among the many exceptions to copyright (regularly used by the BBC), is for a bona fide news organization to report the news. There are many other exceptions, such as home recording for personal use, fair use, backups, satire, parody, teaching & education. The fine print varies by jurisdiction.
One of the misunderstandings of 'fair use' is that it creates an affirmative right to do the thing in question. It does not. 'Fair use' is an affirmative defense to copyright infringement, it means that if someone sues you for violating their copyright, you can counter by saying "what I did is fair use" -- it functions as an exception to the criminal/civil statute.
What it doesn't do, is create an affirmative obligation on the part of the providers to allow you to make fair use of their works. That is to say, if act X (excerpting for a class) is fair use, they may not sue you for X but they need no help you do X and they may even have technological measures (CSS) that frustrate X.
Yes, it sucks, and yes, this does mean that content providers can use the DMCA + technological measures to frustrate your "right" to fair use. That's the way the law is written, I'd like to have it changed, but for until it gets changed, that's the way it is.