is that a commercial entity is ponying up half the cost for something that could/should be handled by the government. From TFA:
"This project is a mobility improvement for the area as a whole," said Lou Gellos, a spokesman for Microsoft. An existing bridge a few blocks away is congested and a nightmare for pedestrians and bicycle riders, he said.
So, we have the relatively common phenomenon that commercial development has outgrown the infrastructure. Big deal. Usually the government handles this as part of its own work, without direct commercial assistance. In this case, MSFT is offering money to help solve the problem. They deserve kudos, not punishment, since they could alternatively be lobbying/strongarming the relevant government entities to foot the bill at 100%.
Even if you hold the (inane) view that MSFT should foot the bill at 100%, they don't have the authority to just build a bridge over any highway they want. So you need some kind of legislation anyway.
You make a good point, but it's worth getting even more brain-twisting.
Assuming Einstein is correct about the speed of light c being an absolute speed limit, it literally doesn't make sense to talk about events "occurring" X years ago in a location greater than X light-years away. Since there is literally nothing we can do to observe later events from that location any faster than they are already arriving - i.e. no faster than light space travel - then those events may as well not have happened yet for our location. There's no way to "get behind" the wave of events to get an advance preview of what's coming next for our original location. There is no way for objects or information to travel faster than c, and the implications are profound.
Suppose we are X light years away from a star, and we see the light/events from it just as it goes supernova. Suppose we want to know what's happening with the blast wave coming in our direction, to see if there's something we can do to protect ourselves. (In Larry Niven's Ringworld the ring's ancient inhabitants apparently created the ringworld as an edge-on shield for exactly this purpose.)
We might start by thinking we can shoot a rocket with people/sensors on it really fast toward the star, get some information, and then come back so we can plan our defenses. The problem is that we can only travel up to the speed of light; so even though the people on the rocket are seeing the subsequent events "before" the people on earth, they can't send the information back any sooner than it was already going to arrive there.
Put another way - suppose there's a star 100 light-years away and a rocket 50 light-years away between us and the star. As soon as it detects the star going nova, the rocket zooms back home to warn us. Unfortunately, even with near-infinite acceleration it could only get back to us just at the same time that we see the nova for ourselves (since both the rocket and the original signal are traveling at/near c). As far as we're concerned, the rocket "saw" the nova at the same time we did. This is borne out by the experience of the rocket's crewmembers, who experienced little/no aging or time lapse between the moment they saw the nova and the moment they arrived back at earth (due to the near-infinite acceleration).
BTW I put "before" in quotes above because, by the same principles, there is no external observer who could time-correlate the observations at the rocket and at earth to establish ordering. This is the brain-bending consequence of unifying space and time...
It's useful to understand the way large software companies think about this kind of problem space (I work at one, 5000 employees, Microsoft partner). The forward-compatibility problem is highly significant throughout any long-lived software project. There is a standing requirement from customers that stuff does not break when you release a new version. If you think this is a problem, blame the customer, not the vendor. Blaming the vendor is like blaming a retailer for raising prices when sales tax goes up - you're looking at the symptom, not the problem.
Given this, we (large ISVs) pay very serious attention to versioning and compatibility. If designed well from the outset, a robust versioning system from can make life much easier for both users and developers. Basically you have to allow customers to install your new version (so they get security and stability fixes) and then opt-in to your new systems and features, either via licensing or configuration.
The compatibility story for the web is actually pretty fragile. If you accept that change can and will happen, i.e. that the technological needs of the HTML-based web will change, i.e. that it's not fair to ask the W3C to foresee everything, then you have to accept that versioning is needed. META is the main way to handle safe versioning, in practice. An opt-in META tag that says "I want this document rendered as ACID2 compliant, please disable your workarounds for past non-ACID2 transgressions" would be completely sensible. It would be much better than a solution that said "Render for IE8 plz kthx"
Consider an alternative path: IE8 or Firefox drops support for everything except ACID2 rendering. That browser's market share would drop precipitously overnight. Why? Because the customers wouldn't stand for it. The problem lies in the mirror, not in Redmond.
Thinking ahead, I could see IE9 making the ACID2 render path the default. So in a weird way this could be a blessing in disguise since it would enable content authors to safely target ACID2.
I've found that setting up a fulltext search system on the code can be invaluable. Much of the code I deal with is dynamic, so it's hard or impossible to statically determine a call graph. (Think about objects that register for events at runtime.) In these cases it's nice to be able to search for the "glue" (e.g. interface or event names) that binds functions together.
I've used dtSearch in the past with success, even though the version I used didn't have syntax highlighting since it was aimed at human languages. I don't know if more recent versions have that.
that the entire thing will be open to the public. No government would permit countries like North Korea to easily acquire sophisticated ballistic missile technology. The "hard parts" will remain under tight control indefinitely.
Well said. And for the people wondering whether this is just nitpicking - no it isn't. When there is a serious possibility of rendering your hardware permanently useless (say, when flashing your Linksys router's firmware with the excellent DD-WRT replacement), it needs to be clearly communicated that what you're about to do is NOT the usual might-only-wipe-your-hard-drive change.
If the term "bricking" comes to mean "oh man, i had to reinstall the OS, bummer", then someday someone reading "Danger: This could brick your machine" will misunderstand and not fully appreciate the risk.
in a speech at Google last week Democratic presidential candidate Barack Obama called for data to be stored in 'universally accessible formats.
With the availability of the free (as in beer) Word document viewer, it's arguable that Word.doc files are in fact universally accessible already, for some reasonable definition of universal (cf. universal telephone access). You might argue that people still have to buy Windows, which could constitute an obstacle to universal access; but going one level further, they also have to buy a computer regardless of which OS runs on it, so even a free software solution isn't actually without cost.
You didn't RTFA. Ars never claimed your strawman argument that the OS shouldn't defend against viruses.
Examples cited by Ars about how Vista "employs design and tactics against viruses" better than XP:
"... the issue of buffer overflows, which has been addressed well in Vista by most accounts. This was a major weak spot with XP, and so far, Vista looks strong in this area, strong enough that Vista may never get its own 'SQL Slammer.'"
"IE7 in protected mode forces such scripts to run at a very restricted user privilege level, unlike XP which will allow those same scripts to run at the same privilege level as a user." XP does not provide the new protected mode environment for IE7.
The reason for all the food being made dead is precisely because of health issues. Most bacteria/mold/fungi/etc. that would otherwise be present are potentially significant public health risks. When you have a gigantic national food distribution system, a few E. Coli on your spinach can kill quite a few people. From the FDA administrator's perspective, it would be hard to defend setting the public health policy to a lower standard of "deadness" than we have today.
Although I agree there's something to be said for obtaining the freshest possible food, it only works well when you're connected to the original source in some way (e.g. farmers' market). In a supermarket, I want everything dead, dead, dead.
Where do the pollutants go, or what are they transformed into? If they're just transferred to groundwater, or transformed into some other potentially-toxic form, then this might help air quality but it wouldn't address the global pollution problem (as suggested in some of the posts here). The article refers to the pollutants being "broken down" more rapidly than normal, but it's not clear what that means.
Encryption is a technology used to implement DRM. Vanilla encryption becomes unmanageable when I have to distribute and manage keys for every document. This isn't just public-key cryptography, where I'm signing the document with my private key, and anyone can validate that the document is mine and hasn't been tampered with. This is 1) (the equivalent of) encrypting the document with each recipient's public key so that each of them (and nobody else) can read it, and 2) ensuring that they can't forward the decrypted document to someone I didn't authorize.
Microsoft is already highly active in this area. As far as I can see it's "under the radar" of the open source world, which is a shame, since businesses are finding this functionality to be increasingly valuable (think Sarbanes-Oxley, HIPAA, etc.). A couple of useful links for the Microsoft flavor of this technology:
TechNet documentation on the Rights Management Services included in Windows Server 2003 (ironic that the acronym is RMS, isn't it?).
Examples of how Office integration with RMS can meet the requirements I outlined in my post.
Since this is Slashdot: No, I don't work for Microsoft, own their stock, or have friends/family/etc. who do either. I have no particular interest in seeing their solution dominate. I simply find the technology valuable; if that valuation causes FUD in some people, there's not much I can do about it.
DRM is far broader than just copy-protecting DVDs.
As a user of content-creation software I would like the freedom to control distribution of my work. This is not necessarily related to money -- for example, I may want to ensure that early copies of an article on a controversial subject don't get circulated too broadly before I'm ready (since my first drafts rarely hit the mark, and I want to be as persuasive as possible). I want to mark specifically who can read my work. (This is DRM, albeit different from the "anonymous recipient" model used by DVD authors.) If the programmer isn't free to provide me this feature, my freedom is reduced.
Another example, this time from the business world: If I'm working on my company's financial reports, I am required by law to restrict who may view them. There are many approaches in use today to meet that requirement, usually consisting of things like separate account domains, secure fileservers with special permissions, etc. These are cumbersome and expensive; I would prefer to be able to use my authoring software to mark a document as viewable only by a particular set of people, and stick it out on a file share somewhere secure in the knowledge that even if unauthorized users get a copy of the document, it's useless to them. This too is DRM, and again, if the programmer isn't free to provide the feature, I can't have the feature I want.
Think about this one: I'd actually prefer to have an open-source DRM system, as then I'd have increased confidence that there aren't any bugs or loopholes that allow unauthorized access.
Actually the point of natural selection is that it isn't amazing at all -- it's inevitable.
Natural selection suggests that there very likely have been thousands or millions of such fungal species, each of them happening to produce some chemical byproducts as part of its life processes. The vast majority of the time, those byproducts don't have a positive effect on the survival of the species because they're neutral or negative in their environment. But it just so happens that in this species, in this environment (the bug's body and the availability of trees), this particular chemical accidentally happened to have a positive effect on the species' survival, and therefore the species happens to be visible to us observers today.
It's only amazing if it's thought of from an intelligent design perspective. Nature is constantly running billions of little experiments, some of which occasionally contribute meaningfully to the next round of experiments.
(As an OT aside, this sort of thing is why the free market works. Stay with me here.:) Capitalism in the free market leads to entrepreneurs constantly running millions of little experiments (businesses), some of which occasionally contribute meaningfully to the next round of experiments (larger businesses/prosperity). In this mental model, Soviet-style central planning is equivalent to Intelligent Design; the problem is that the Central Committee is not infinitely intelligent (unlike the purported Intelligent Designer) and couldn't possibly do as well as the comparatively out-of-control but ultimately more productive free market.)
This new theory says that the force of gravity should be stronger near the galactic core, where the stars are packed most densely. So the core is even more massive than we thought, meaning that the rotation curve of the galaxy should be even more skewed - far from flat. So either New Scientist seriously misrepresnted his theory, or it doesn't even deserve a cursory thought. MOND at least seems plausible.
I wasn't aware of the flat rotation curve, but it's an interesting point. Assuming for the moment that it's correct, the theory matches the data quite well. The theory says that despite the greater mass concentration at the center of the galaxy, the force it exerts on the outer rim is weaker than Newtonian mechanics would suggest. The theory doesn't suggest the center is more massive, just that the gravitational effects of that mass decrease with distance. So there is a greater gravitational contribution from the mass between the outer rim and the core (compared to the Newtonian model). This would explain the flattening of the curve you describe.
The submitter should have RTFD (directive). He managed to cram several incorrect ideas into a single paragraph.
Incorrect: The content of the communications themselves are to be kept.
Correct: Only metadata (time, source, destination, duration, geo-location) are to be retained.
Incorrect: This is a fundamentally new kind of data that is being retained.
Correct: Phone companies have retained exactly that kind of data for decades -- for billing purposes. Law enforcement has successfully used that data for decades without excessively infringing citizens' privacy rights (see next bullet for why). The primary justification for the directive (as stated in the linked document) is to react to the fact that companies no longer need to retain the metadata for billing purposes; the directive acknowledges that fact but makes them keep it anyway for law enforcement purposes. Without it there would actually be a reduction in legitimate law enforcement capability.
Incorrect: There are no controls on accessing the data; law enforcement can access it at will.
Correct: Individual countries' existing legal frameworks dictate when law enforcement may access it. If a country requires a court order to get the metadata for phone conversations, they'll still need one for this new data.
Disclaimer: I'm no pro-government shill -- I have problems with the USA PATRIOT act and I believe government must be denied any authority it is not explicitly granted. But this legislation is just reasonable adaptation to technological reality.
Unless these people are being illegally denied representation in government, I honestly don't give a flying flip about why they feel "driven" to violence. Democracies have mechanisms for peacefully airing grievances; if you can't convince others of your position, you can't get what you want. End of story.
This is about cost. The manufacturers' request is perfectly reasonable, given their costs, and the kernel developers' concerns are perfectly valid given their costs.
Initial costs associated with a manufacturer-supported driver:
Initial development against a specific kernel version. This is the item of least concern to the manufacturers.
Initial testing against some specific subset of officially supported distributions. Tweaks or rewrites to handle additional kernel versions. This is where the costs start to go up.
Corresponding documentation, release notes, etc.
Make driver available on company website (ongoing cost).
Put driver on the CD in the box (include documentation on which releases it was tested against)
Ongoing costs associated with a manufacturer-supported driver:
Development and test resources to check compatibility with subsequent kernel versions.
Test resources to check compatibility with subsequent specific officially supported distributions.
Update documentation for any changes.
Update website with updated versions. Handle support calls from confused users.
Slipstream CD in the box with updated versions.
These costs exist even if a version of the driver is merged into the mainline kernel. The only problem solved by such source-level merging is compatibility with the latest kernel version. It is not acceptable to the manufacturers' customers to be required to update to the latest kernel/distribution to be able to use the device.
Here's the key point: If there is no binary interface between the driver and the kernel, all of the above costs skyrocket. You have M kernel versions against N distributions, with the total increasing over the life of the product. If there is a binary interface guarantee from the kernel development team to change only very slowly and only extremely rarely breaking compatibility -- like the guarantee Windows provides -- then the incremental costs are containable. It is reasonable to expect that 95% of their testing on 2.6.5 is valid on 2.6.14.
The perfectly reasonable response from kernel developers is that with closed-source drivers they get stuck debugging problems that are't kernel-related (I don't hold ideology to be economically significant so I'll ignore it here, without insult to people's strong opinions on the subject). Their proposed solution is to require the driver's source before they'll help with the debugging.
From the manufacturers' point of view that's a very draconian requirement. They are justifiably concerned about intellectual property (availability of the source makes it much easier for competitors to reverse-engineer the hardware/firmware). Surely there must be a middle ground. Is there some way to have a relationship between the device manufacturers and the kernel developers that minimizes everyone's costs?
I think there is. Note that all of the above costs and issues are just as valid in the Windows world as in the Linux world. Microsoft doesn't want to deal with bad drivers crashing their systems, costing them both development/debugging time and reduced perceived stability (--> lower sales). Their solution is the Windows Hardware Quality Lab (WHQL).
The WHQL is a separate entity from Microsoft. Device manufacturers are required to submit their driver source (effectively under NDA) along with their device. The WHQL staff runs the driver through a battery of tests, probably mostly automated. If the device and driver meet stability standards set by Microsoft, the driver is signed by WHQL. Windows checks for this signature at installation time and warns the administrator if it is not present. Microsoft can reasonably refuse to support non-WHQL-signed drivers when crashes occur, for exactly the same reasons that Linux kernel developers refuse to support drivers without the source. This system has been the single most important factor in Windows' significan
As a software engineer who has been on the recruiting side of the picture for seven years at a major ISV, I have to say that an online-only degree is a yellow flag (not a red one) on your resume. There are several reasons for this.
One of the key factors we look for is participation in interesting team projects. There are many subtle lessons learned in these projects that simply aren't available without face-to-face interaction. Team socialization, leadership, etc. If our engineers all worked remotely, and therefore we needed people who are skilled at teamwork in an online-only environment, we might see online degrees as a bonus. But that's not how we're set up.
As others have pointed out, the school's reputation is key. Even if your online school is actually excellent, we'd never know it without interviewing you. Your school is still "guilty by association" with diploma mills. We have limited bandwidth to go through large numbers of resumes, so an online-only place we've never heard of might be the thing that ticks it over from "interview" to "shredder". An online-only degree from a top-five engineering school, however, would be something else entirely.
As I said, this is a yellow flag, not a red one. We recognize that it's possible to be a great software engineer without even having a CS degree, much less a degree from a good school (we have such people on our staff right now). But if we had to choose who to interview between two otherwise equally-qualified candidates, we'll give the interview to the candidate with the "traditional" degree.
One final caveat -- your degree matters much less than your experience. If you have five years of experience in a field we're interested in, we won't even ask about your college projects or your grades.
Slashdot Mirror
"This project is a mobility improvement for the area as a whole," said Lou Gellos, a spokesman for Microsoft. An existing bridge a few blocks away is congested and a nightmare for pedestrians and bicycle riders, he said.
So, we have the relatively common phenomenon that commercial development has outgrown the infrastructure. Big deal. Usually the government handles this as part of its own work, without direct commercial assistance. In this case, MSFT is offering money to help solve the problem. They deserve kudos, not punishment, since they could alternatively be lobbying/strongarming the relevant government entities to foot the bill at 100%.
Even if you hold the (inane) view that MSFT should foot the bill at 100%, they don't have the authority to just build a bridge over any highway they want. So you need some kind of legislation anyway.
You make a good point, but it's worth getting even more brain-twisting.
Assuming Einstein is correct about the speed of light c being an absolute speed limit, it literally doesn't make sense to talk about events "occurring" X years ago in a location greater than X light-years away. Since there is literally nothing we can do to observe later events from that location any faster than they are already arriving - i.e. no faster than light space travel - then those events may as well not have happened yet for our location. There's no way to "get behind" the wave of events to get an advance preview of what's coming next for our original location. There is no way for objects or information to travel faster than c, and the implications are profound.
Suppose we are X light years away from a star, and we see the light/events from it just as it goes supernova. Suppose we want to know what's happening with the blast wave coming in our direction, to see if there's something we can do to protect ourselves. (In Larry Niven's Ringworld the ring's ancient inhabitants apparently created the ringworld as an edge-on shield for exactly this purpose.)
We might start by thinking we can shoot a rocket with people/sensors on it really fast toward the star, get some information, and then come back so we can plan our defenses. The problem is that we can only travel up to the speed of light; so even though the people on the rocket are seeing the subsequent events "before" the people on earth, they can't send the information back any sooner than it was already going to arrive there.
Put another way - suppose there's a star 100 light-years away and a rocket 50 light-years away between us and the star. As soon as it detects the star going nova, the rocket zooms back home to warn us. Unfortunately, even with near-infinite acceleration it could only get back to us just at the same time that we see the nova for ourselves (since both the rocket and the original signal are traveling at/near c). As far as we're concerned, the rocket "saw" the nova at the same time we did. This is borne out by the experience of the rocket's crewmembers, who experienced little/no aging or time lapse between the moment they saw the nova and the moment they arrived back at earth (due to the near-infinite acceleration).
BTW I put "before" in quotes above because, by the same principles, there is no external observer who could time-correlate the observations at the rocket and at earth to establish ordering. This is the brain-bending consequence of unifying space and time...
- B
Agreed on all points.
It's useful to understand the way large software companies think about this kind of problem space (I work at one, 5000 employees, Microsoft partner). The forward-compatibility problem is highly significant throughout any long-lived software project. There is a standing requirement from customers that stuff does not break when you release a new version. If you think this is a problem, blame the customer, not the vendor. Blaming the vendor is like blaming a retailer for raising prices when sales tax goes up - you're looking at the symptom, not the problem.
Given this, we (large ISVs) pay very serious attention to versioning and compatibility. If designed well from the outset, a robust versioning system from can make life much easier for both users and developers. Basically you have to allow customers to install your new version (so they get security and stability fixes) and then opt-in to your new systems and features, either via licensing or configuration.
The compatibility story for the web is actually pretty fragile. If you accept that change can and will happen, i.e. that the technological needs of the HTML-based web will change, i.e. that it's not fair to ask the W3C to foresee everything, then you have to accept that versioning is needed. META is the main way to handle safe versioning, in practice. An opt-in META tag that says "I want this document rendered as ACID2 compliant, please disable your workarounds for past non-ACID2 transgressions" would be completely sensible. It would be much better than a solution that said "Render for IE8 plz kthx"
Consider an alternative path: IE8 or Firefox drops support for everything except ACID2 rendering. That browser's market share would drop precipitously overnight. Why? Because the customers wouldn't stand for it. The problem lies in the mirror, not in Redmond.
Thinking ahead, I could see IE9 making the ACID2 render path the default. So in a weird way this could be a blessing in disguise since it would enable content authors to safely target ACID2.
-B
I've found that setting up a fulltext search system on the code can be invaluable. Much of the code I deal with is dynamic, so it's hard or impossible to statically determine a call graph. (Think about objects that register for events at runtime.) In these cases it's nice to be able to search for the "glue" (e.g. interface or event names) that binds functions together.
I've used dtSearch in the past with success, even though the version I used didn't have syntax highlighting since it was aimed at human languages. I don't know if more recent versions have that.
- B
- B
If the term "bricking" comes to mean "oh man, i had to reinstall the OS, bummer", then someday someone reading "Danger: This could brick your machine" will misunderstand and not fully appreciate the risk.
With the availability of the free (as in beer) Word document viewer, it's arguable that Word .doc files are in fact universally accessible already, for some reasonable definition of universal (cf. universal telephone access). You might argue that people still have to buy Windows, which could constitute an obstacle to universal access; but going one level further, they also have to buy a computer regardless of which OS runs on it, so even a free software solution isn't actually without cost.
You didn't RTFA. Ars never claimed your strawman argument that the OS shouldn't defend against viruses.
Examples cited by Ars about how Vista "employs design and tactics against viruses" better than XP:
Ars Technica is right.
The reason for all the food being made dead is precisely because of health issues. Most bacteria/mold/fungi/etc. that would otherwise be present are potentially significant public health risks. When you have a gigantic national food distribution system, a few E. Coli on your spinach can kill quite a few people. From the FDA administrator's perspective, it would be hard to defend setting the public health policy to a lower standard of "deadness" than we have today.
Although I agree there's something to be said for obtaining the freshest possible food, it only works well when you're connected to the original source in some way (e.g. farmers' market). In a supermarket, I want everything dead, dead, dead.
Nit: It's far, far worse than O(n^2), it's O(n!).
Where do the pollutants go, or what are they transformed into? If they're just transferred to groundwater, or transformed into some other potentially-toxic form, then this might help air quality but it wouldn't address the global pollution problem (as suggested in some of the posts here). The article refers to the pollutants being "broken down" more rapidly than normal, but it's not clear what that means.
/b
Correct.
Encryption is a technology used to implement DRM. Vanilla encryption becomes unmanageable when I have to distribute and manage keys for every document. This isn't just public-key cryptography, where I'm signing the document with my private key, and anyone can validate that the document is mine and hasn't been tampered with. This is 1) (the equivalent of) encrypting the document with each recipient's public key so that each of them (and nobody else) can read it, and 2) ensuring that they can't forward the decrypted document to someone I didn't authorize.
Microsoft is already highly active in this area. As far as I can see it's "under the radar" of the open source world, which is a shame, since businesses are finding this functionality to be increasingly valuable (think Sarbanes-Oxley, HIPAA, etc.). A couple of useful links for the Microsoft flavor of this technology:
Since this is Slashdot: No, I don't work for Microsoft, own their stock, or have friends/family/etc. who do either. I have no particular interest in seeing their solution dominate. I simply find the technology valuable; if that valuation causes FUD in some people, there's not much I can do about it.
/B
DRM is far broader than just copy-protecting DVDs.
As a user of content-creation software I would like the freedom to control distribution of my work. This is not necessarily related to money -- for example, I may want to ensure that early copies of an article on a controversial subject don't get circulated too broadly before I'm ready (since my first drafts rarely hit the mark, and I want to be as persuasive as possible). I want to mark specifically who can read my work. (This is DRM, albeit different from the "anonymous recipient" model used by DVD authors.) If the programmer isn't free to provide me this feature, my freedom is reduced.
Another example, this time from the business world: If I'm working on my company's financial reports, I am required by law to restrict who may view them. There are many approaches in use today to meet that requirement, usually consisting of things like separate account domains, secure fileservers with special permissions, etc. These are cumbersome and expensive; I would prefer to be able to use my authoring software to mark a document as viewable only by a particular set of people, and stick it out on a file share somewhere secure in the knowledge that even if unauthorized users get a copy of the document, it's useless to them. This too is DRM, and again, if the programmer isn't free to provide the feature, I can't have the feature I want.
Think about this one: I'd actually prefer to have an open-source DRM system, as then I'd have increased confidence that there aren't any bugs or loopholes that allow unauthorized access.
/B
Here's my fave along those lines. :)
Minor correction: It was Dubya's father, George H.W. Bush (our 41st president), who once ran the CIA, not Dubya himself.
-- bandannarama
Ahem. A $10-a-month dial-up connection will NOT get me all the porn I want.
You're male, aren't you? I suspect women have been very familiar with the "usual" hand position all along.
Natural selection suggests that there very likely have been thousands or millions of such fungal species, each of them happening to produce some chemical byproducts as part of its life processes. The vast majority of the time, those byproducts don't have a positive effect on the survival of the species because they're neutral or negative in their environment. But it just so happens that in this species, in this environment (the bug's body and the availability of trees), this particular chemical accidentally happened to have a positive effect on the species' survival, and therefore the species happens to be visible to us observers today.
It's only amazing if it's thought of from an intelligent design perspective. Nature is constantly running billions of little experiments, some of which occasionally contribute meaningfully to the next round of experiments.
(As an OT aside, this sort of thing is why the free market works. Stay with me here. :) Capitalism in the free market leads to entrepreneurs constantly running millions of little experiments (businesses), some of which occasionally contribute meaningfully to the next round of experiments (larger businesses/prosperity). In this mental model, Soviet-style central planning is equivalent to Intelligent Design; the problem is that the Central Committee is not infinitely intelligent (unlike the purported Intelligent Designer) and couldn't possibly do as well as the comparatively out-of-control but ultimately more productive free market.)
This new theory says that the force of gravity should be stronger near the galactic core, where the stars are packed most densely. So the core is even more massive than we thought, meaning that the rotation curve of the galaxy should be even more skewed - far from flat. So either New Scientist seriously misrepresnted his theory, or it doesn't even deserve a cursory thought. MOND at least seems plausible.
I wasn't aware of the flat rotation curve, but it's an interesting point. Assuming for the moment that it's correct, the theory matches the data quite well. The theory says that despite the greater mass concentration at the center of the galaxy, the force it exerts on the outer rim is weaker than Newtonian mechanics would suggest. The theory doesn't suggest the center is more massive, just that the gravitational effects of that mass decrease with distance. So there is a greater gravitational contribution from the mass between the outer rim and the core (compared to the Newtonian model). This would explain the flattening of the curve you describe.
-
Incorrect: The content of the communications themselves are to be kept.
-
Incorrect: This is a fundamentally new kind of data that is being retained.
-
Incorrect: There are no controls on accessing the data; law enforcement can access it at will.
Disclaimer: I'm no pro-government shill -- I have problems with the USA PATRIOT act and I believe government must be denied any authority it is not explicitly granted. But this legislation is just reasonable adaptation to technological reality.Correct: Only metadata (time, source, destination, duration, geo-location) are to be retained.
Correct: Phone companies have retained exactly that kind of data for decades -- for billing purposes. Law enforcement has successfully used that data for decades without excessively infringing citizens' privacy rights (see next bullet for why). The primary justification for the directive (as stated in the linked document) is to react to the fact that companies no longer need to retain the metadata for billing purposes; the directive acknowledges that fact but makes them keep it anyway for law enforcement purposes. Without it there would actually be a reduction in legitimate law enforcement capability.
Correct: Individual countries' existing legal frameworks dictate when law enforcement may access it. If a country requires a court order to get the metadata for phone conversations, they'll still need one for this new data.
-- Bandannarama
I had the same thought. Fortunately this has already been covered by the IPv6 Working Group, apparently as far back as 1999.
Unless these people are being illegally denied representation in government, I honestly don't give a flying flip about why they feel "driven" to violence. Democracies have mechanisms for peacefully airing grievances; if you can't convince others of your position, you can't get what you want. End of story.
Initial costs associated with a manufacturer-supported driver:
Ongoing costs associated with a manufacturer-supported driver:
These costs exist even if a version of the driver is merged into the mainline kernel. The only problem solved by such source-level merging is compatibility with the latest kernel version. It is not acceptable to the manufacturers' customers to be required to update to the latest kernel/distribution to be able to use the device.
Here's the key point: If there is no binary interface between the driver and the kernel, all of the above costs skyrocket. You have M kernel versions against N distributions, with the total increasing over the life of the product. If there is a binary interface guarantee from the kernel development team to change only very slowly and only extremely rarely breaking compatibility -- like the guarantee Windows provides -- then the incremental costs are containable. It is reasonable to expect that 95% of their testing on 2.6.5 is valid on 2.6.14.
The perfectly reasonable response from kernel developers is that with closed-source drivers they get stuck debugging problems that are't kernel-related (I don't hold ideology to be economically significant so I'll ignore it here, without insult to people's strong opinions on the subject). Their proposed solution is to require the driver's source before they'll help with the debugging.
From the manufacturers' point of view that's a very draconian requirement. They are justifiably concerned about intellectual property (availability of the source makes it much easier for competitors to reverse-engineer the hardware/firmware). Surely there must be a middle ground. Is there some way to have a relationship between the device manufacturers and the kernel developers that minimizes everyone's costs?
I think there is. Note that all of the above costs and issues are just as valid in the Windows world as in the Linux world. Microsoft doesn't want to deal with bad drivers crashing their systems, costing them both development/debugging time and reduced perceived stability (--> lower sales). Their solution is the Windows Hardware Quality Lab (WHQL).
The WHQL is a separate entity from Microsoft. Device manufacturers are required to submit their driver source (effectively under NDA) along with their device. The WHQL staff runs the driver through a battery of tests, probably mostly automated. If the device and driver meet stability standards set by Microsoft, the driver is signed by WHQL. Windows checks for this signature at installation time and warns the administrator if it is not present. Microsoft can reasonably refuse to support non-WHQL-signed drivers when crashes occur, for exactly the same reasons that Linux kernel developers refuse to support drivers without the source. This system has been the single most important factor in Windows' significan
As I said, this is a yellow flag, not a red one. We recognize that it's possible to be a great software engineer without even having a CS degree, much less a degree from a good school (we have such people on our staff right now). But if we had to choose who to interview between two otherwise equally-qualified candidates, we'll give the interview to the candidate with the "traditional" degree.
One final caveat -- your degree matters much less than your experience. If you have five years of experience in a field we're interested in, we won't even ask about your college projects or your grades.
-- Bandannarama