On POSIX systems, creating PostScript is not generally considered part of the driver.
On Systems using CUPS, the application is responsible for creating PostScript. (Cups also has some support for converting other common formats, such as plain text or raster images, but PostScript (or equivalently PDF) is the platform's preferred format.)
The application can Query CUPS about the printers capabilities, and produce optimized PostScript, but producing generic PostScript is also completely acceptable. One of the filter passes in CUPS should read the PPD file, and make some small tweaks to the file to help optimize it for the printer, such as cutting down the resolution of embedded raster graphics where they would significantly exceed the printer's DPI, utilizing of well known PostScript extensions if the printer supports them, and by embedding any fonts needed that the printer does not already have.
At this point if the printer actually supports PostScript, CUPS simply sends it. If not, it goes on to use one of many PostScript to other format converters. These are are the equivalent of printer drivers. ---
Now you might call CUPS a printer driver itself. That is fairly misleading though. If a Linux application knew the printer spoke PostScript, it could simply send the generic PostScript directly to the printer. Not being optimized for the printer means it would take a bit longer to send and longer for the printer to process, but it would work[1]. That is clearly driver-less, since the app simply sent its normal print format directly to the printer.
Footnote: [1] Unless the PostScript required fonts not present on the printer, which were not embedded in the PostScript. ---
Now one could argue that the PostScript interpreter in a printer is effectively a printer driver. But who cares. The printer is a black box. All I care about is that if I regularly print to 20 different printers that I don't need to have 20 print drivers on my computer, or that I don't need a full CUPS install on my phone, but can get away with a simple PostScript print queue, and perhaps a quick optimizing pass using a PDD file obtained by asking the printer directly.
I'm emphatically not a FPS enthusiast. The only FPSs I've ever really played are the half-life series, the Portal series (hardly a FPS at all anyway), and the metroid prime series, which are primarily a 3d platformer and only secondarily a FPS.
So why have I played and enjoyed the Half life series (including the original HL, HL:OF, HL:BS, HL2, HL3:EP1, HL3:EP2)? (I refuse to call the episodes HL2, since even Valve admits that was a mistake.)
I've enjoyed the story and atmosphere. Sure I've heard claim that the original HL hardly had a story. That is not true even in the slightest. It had a very rich story. Consider the excellent introduction sequence. Quite a bit of the story is present there. If you stop and listen to the scientists and other characters you get quite a bit more. Then the sequence of locations and events were carefully chosen to add more to the story. As long as you include the implied portions of the story it had substantially more than the average FPS of its day.
With HL2 the plot was even more explicitly story driven, with a bit less being only implied, but if you carefully check out some of the textures you'll see even more attention to detail. HL3 has so far been rather similar, which is why I await the release of HL3:EP3.
As for what is taking valve so long, consider that the companies attention is now diviided between licensing the source engine, managing the steam platform, and quite a few distinct game series. Its little surprise that HL3:EP3 is on the back burner.
With a stolen car, the owner cannot utilize nearly any their rights.
Without a copy of the copyrighted work, the copyright holder cannot utilize some of their rights. For example without a copy, the copyright holder can still go after infringes, but would experience great difficulty in licensing the work to others, since in addition to a license, most licensees require an initial copy of the work, and telling them to buy it from one of the other licensees is unlikely to go over very well, and may not even be possible. Not to mention that if you are suing for infringement you basically need to register the work, which requires a copy. For works of US origins it is officially a requirement, although it is plausible that a court would waive the requirement if registration were not possible, but you would face a greatly increased burden of proof, since you would need to prove that you created the work, purchased the rights, or commissioned the creation of the work under the works-for-hire doctrine in order to establish holding copyright on the work, plus the normal burden of showing infringement.
Both cases (stolen car and lack of a copy) can be resolved by re-obtaining the item in question, either the stolen car, or a copy of the work.
They are by no means an exact match, but there is a distinct similarity between them.
This could happen if a person were to fraudulently convince the state agency that administers copyrights that the owner of the copyright has assigned it to him.
sounds a lot more like registering a fraudulent title to the car.
Pure theft of the car would be far more like destroying all copies of the work the copyright owner has. The copyright owner is still lawfully regarded as the owner in that case, just like with the car theft case, but down must track down a copy in order to fully utilize their ownership rights.
The big difference is that with a car, there is only one that could be tracked down, while with a copyrighted work, there are potentially many, and the owner may be able to use any of them.
Considering that in all likelihood when you remade the processor, you used a smaller process size than the original[1], and that when you use a smaller process size you generally scale the elements to fit[2], the maximum stable clock rate of your remake should be higher substantially greater than the clock rate of the original If the turbo button was a simple clock rate switch, it could speed things up.
The only particularly tricky parts are sound and video, since if you don't compensate for the increased clock rate, these output would be garbled. (I'm assuming that this hypothetical USB device includes the audio video hardware, connected to a a modern video/audio capture chip for display/reproduction on the computer.
[1] Since it can be hard to find a fab for the original process size. [2] While you could use the smaller size to make the chip at the original scale, that is wasteful.
In the mathematical community, the word modulo is often used informally. Generally, to say "A is the same as B modulo C" means, more-or-less, "A and B are the same except for differences accounted for or explained by C".
With that in mind, the summary is saying this article and and the previous one cover the same topic except for differences accounted for by "kernel debugging abilities".
In this case that does mean pretty much the same thing as the word minus would have, and the word minus would probably be have a better choice.
The only way to obtain the private key other than breaking the key directly, is to be given access by the key's owner, or given access by somebody the owner gave access to, (recursively). Giving access in this sense means the person in question failed to protect the certificate sufficiently to prevent access. There is no further chain.
It is trivially the case that modified code on the client side could lie to the user. It is also the case that compromised client code could permit a man in the middle to obtain the session key, and then pretend to be the server.
However from a security model view that is equivalent to the recipient conspiring with the MITM to allow the MITM access.
In the US with the exception of tests/exams, it is almost unheard of to not have your name directly on an essay. A few Universities or programs may do that, but the vast majority do not.
On standardized tests/exams the essays never have the authors name.
On professor generated exams, the grader usually can see the name, but professors setting up exams such that graders do not know the student's name, while rare, is not unheard of. If such an exam contains an essay, then the essay would be graded without knowing the student's identity, (unless the identity can be inferred from the essay's content of course).
I think the artist's idea is that in the future we will have fully separate lanes on major roadways with cars driving automatically at really high velocities. With fully separate lanes an automatic car need only worry about not hitting the car ahead of them, since the car would be unable to leave the lane. So no worrying about the car in the next lane cutting you off, etc.
Obviously in practice the there would be the on and off ramps, and periodic opportunities to change lanes, but even so, the cars would basically be on rails, and it is far, far simpler to create a safe automatic driving system for vehicles on rails than free-driving vehicles.
Nevertheless I'm not at all sure if that is what the author meant, since that video was just terrible.
You are correct. A cert does not prove identity per se. Nothing can. Fundamentally the concept of identity is only a psychological/sociological construct.
There is no physical notion of identity. Since we insist on pretending there is such a concept (it is rather convenient, after all), we must accept the fact hat there is always some level of doubt.
Now what do SSL certs have to do with identity? Well lets ask some questions.
Does an SSL cert for Yahoo.com prove that the the other end is controlled by Yahoo! Inc? No it definitely does not.What it does indicate is that the machine is controlled by somebody who the owner of the cert in question gave access to the private key. [1] It gives a presumption that the person controlling the machine is authorized to speak on behalf of the key owner, but that is not absolute. But who is the key owner? Without considering who the CA is we don't have any real idea.
Sometimes this does not matter. In the case of an SSL Cert for http://somerandomsite.example/, we may not be making any assumptions about who is behind the domain. All we might want to know is when we come back later if we are still talking to machines run by the same entity. If the key is still the same, then we can be fairly confident that we are indeed talking to machines run by the same entity.
Or consider the case of being a reporter receiving a signed email from some anonymous tipster. We don't know who is at the other end, but over time the person can establish reliability. We don't necessarily even care who it is in the "real world". What we do care about is knowing if this latest Tip came from the same entity as the previous messages. If they were all valid and all used the same certificate we can be reasonably confident in that.
Now so far we have been ignoring the CA. All the above holds even with self-signed certificates. But the above is pretty weak. the above gives us absolutely no confidence that the certificate belongs to the company or person named in it.
However, if you check who the CA is for the certificate you may be able to establish some confidence that the certificate is owned by by the entity named in it. This requires trusting the CA to verify the identity of the entity requesting a certificate. I do trust that a Verisign extended Validation certificate does indeed belong to the named entity. Given that trust I have high confidence that the pages I recently received from a server using the certificate with serial number "2E:33:87:4F:6F:E2:D4:1E:D3:FF:FF:35:F6:A4:C9:18" did indeed come from PayPal Inc.
I also trust that a non-EV Verisign cert was indeed issued to somebody who once controlled web server or DNS servers of the indicated domain.
I don't however place any trust in some of the CA's found in my browser.'s trusted list. I agree that the current system is unworkable since it requires trusting the CA's, and yet even I cannot be bothered to remove the CA's from the browser that I don't trust. The average Joe definitely won't.
Turning it around to make it opt in would not work either. The average Joe does not even know who Verisign is, much less whether to trust them, and there is no way we can educate the masses enough to allow for this.
So I fully agree we need a change. I also agree that the browser accepting a certificate says nothing about the identity of the entity on the other end. But I disagree that certificates cannot establish identity with reasonably high certainty.
I'm kind of partial to a web of trust based solution in which the browser trusts the Browser writer's key by default (you probably trust them). To ensure better results the browser would also come with a set of certificates of large companies that people may trust, (Google, Microsoft, AT&T, Yahoo, Comcast, IBM, SONY, etc)[2], all of which the browser maker have verified belongs to the company in question. The browser would give the opportunity to chose which if any of those certificates to also m
It would be very interesting to see this redesigned today. Take the basic design, like the use of composite plastics for crash saftey, replace the old automatically controlled manual transmission with a automatic DSG, and add some front seatbelts!
Replace the engine with a modern design of similar weight and power, but which meets modern emissions standards. The improvements of a modern engine and an automatic DSG might be enough to bring the vehicle up to 40 mpg or so. The car should trivially get a 5 star rating on all standard crash tests. The maker may even be able to lobby for having higher speed testing available to earn a 6 star rating.
But you don't need to do that. You do need to actively protect your trademarks. However you only need to take actual action if likely confusion would result.
Since Notch's game is very unlikely to cause confusion, a cease and desist is not only not warranted, but might even open up Bethesda to potential liability for one of several torts. (I've never heard of anybody successfully suing for an overzealous C&D letter, but that does not make it impossible.)
What may been prudent is to send a letter reminding Notch that they own a trademark on "The Elder Scrolls", and that care should be taken in ensuring that the logo, gameplay, etc are sufficiently distinct to avoid any possible consumer confusion. The letter properly written would make it clear that they have no objection to the name as long as the rest of the game is sufficiently distinct.
That is all they need to do. Unfortunately most lawyers are not trained in writing such non-threatening letters, and they are very afraid of trying, since the wrong wording could be construed as a grant of rights to use the Trademark, which could result in being fired by their client, and possible a malpractice suit being filed against them.
On the other hand if it is changing on a weekly basis, them many times there will be no new features to familiarize yourself with, or there will be only one or two small features. One person could spend an hour or two testing in those cases. (If nobody can afford that, then the company should have hired another person for IT prior to rolling out the chrome-book, especially since the rollout process could easily monopolize the time of a member of IT for several weeks.
Educating support staff on an invisible bug fix consists of doing exactly nothing. For small features, it may also consist of extremely little.
Communicating changes to end users should always be even less involved than educating the support staff. Basically an extremely high level overview like "You can now plug your Android phones/tablets into the chrome-book and see the files on it" should be enough. Those afraid to experiment can ask support for more details.
Well if you wanted to write a Doom 3 like game, this would be an excellent place to start. You would need to develop graphics and levels, and make some changes for any unique features, but you would need to do that anyway even if starting from scratch, and at least here you already have a working engine. The communities that develop around these source drops often focus on making the code portable, and making minor changes here and there so that it can work well on modern hardware.
So you get to start with a portablized version of a proven game engine. The existing engines are designed with performance on relatively slow machines (by today's standards) in mind. That sounds like a reasonably good fit for something like a modern cell phone. Rewriting the graphics code to take advantage of a mobile's platform's limited 3d acceleration is often not nearly as much work as it sounds.
A properly written balanced budget amendment would only require that the original budget be balanced. Any sane version would permit Congress to authorize emergency deficit spending, and may even permit deficit spending in an executive declared state-of-emergency if Congress is not in session and an emergency session cannot be called quickly enough.
Whether the amendments being proposed are sane or not, I have absolutely no idea. But sane ones are possible.
Sure, but only up to a point. If for example, you limit trades to once per minute, but break ties by who submitted a buy order/sell order first, then there would still be emphasis on getting your offers in as quickly as possible, (i.e. within the first few microseconds). Some, but not all of of the existing HFT practices would change. the HFT would become every-block traders, and would still still try to be first to the transaction.
Breaking ties by last to submit wins would produce something similar. Randomly breaking ties would be best, but people who accuse the system of being based, even if it were not.
Sure, but a such an adjustment is not binding on the actual investor. It cannot be, since no contract at the adjusted price was ever made by the trader. The trader is free to refuse to settle at the adjusted price, and no sane court would compel payment.
It will degrade the signal of downstream recipients. So does absolutely every radio receiver, with no exceptions. However, please consider that the only downstream recipient may well be the earth or space, considering that the path between a transmitter and receiver often does not pass particularly close to another receiver. How much one of these would impact the downstream signal quality anyway depends on just how much power this is extracting, and just how weak the signal would have been at the downstream receiver without this being present.
Also keep in mind that radio waves can be rather fickle. Placing these devices in certain locations may actually increase the received signal strength downstream, perhaps by absorbing an interference source, or by attenuating a secondary path of the signal which would have interfered with the primary signal.
You don't have access to integrated profiles, but the last I knew there was nothing preventing you from using your apps email address to create a standard Google Account (just like a Yahoo mail user would do), and create a profile using the Google Account, rather than the Google Apps Account.
The site is no longer invite only. I know this for an absolute fact, since I have a Google+ account but never received an invite.
As far as I can tell, they are turning on and off the ability to register several times a day, or perhaps it has something to so with which server you get connected to. Regardless, I did need to try several times to avoid the "temporarily exceeded our capacity" message.
The security regulations apply to repeating scheduled flights. If you own an airplane you are quite free to offer charter flights with reduced security. (Such as only a metal detector, or even no security at all.)
On the other hand, the average Linksys box actually uses a switch that supports multiple VLANs. The Wan port is actually on the same switch as the 4 LAN ports, but just on a separate VLAN. 6th port (to the router proper) is the only one configured by default to be in multiple VLANs and to see tagged frames.
As you might be able to guess, it is relatively straightforward to merge all ports onto a single VLAN, or to sacrifice one of the LAN ports to create a second WAN port. Obviously the default firmware for the routers don't supply this option, but custom firmware can do this.
A Microsoft hosted Sharepoint Server (including Office Web Apps)
A Microsoft Hosted Lync (Imstant messaging) server
That is literally all there is to it. They have tweaked the Web-app UIs to be branded Office 365 and put the whole thing behind a single sign-in portal, but you could set up the exact same service locally if you were a large business.
The more expensive plans also include subscription based access to the Office 2010 "Professional Plus" (previously known as "Enterprise").
Microsoft knows that many small businesses will end up buying full versions of the desktop apps. Since the "Cloud" is merely Exchange and SharePoint, there is no real problem using the desktop apps in conjunction with the web-based apps.
True, but the artifacts skips and drops only happen if you are right near the point of losing signal entirely, or you experienced exceptionally strong interference. However those are quite noticeable unlike quite a bit of analog degradation.
Jitter on the other hand can be far more insidious. If the cable was properly constructed and tested it should not be an issue. If it is very short (6 feet or less) it is not likely to be much of an issue either. But for long runs it can definitely be a problem.
Slashdot Mirror
On POSIX systems, creating PostScript is not generally considered part of the driver.
On Systems using CUPS, the application is responsible for creating PostScript. (Cups also has some support for converting other common formats, such as plain text or raster images, but PostScript (or equivalently PDF) is the platform's preferred format.)
The application can Query CUPS about the printers capabilities, and produce optimized PostScript, but producing generic PostScript is also completely acceptable. One of the filter passes in CUPS should read the PPD file, and make some small tweaks to the file to help optimize it for the printer, such as cutting down the resolution of embedded raster graphics where they would significantly exceed the printer's DPI, utilizing of well known PostScript extensions if the printer supports them, and by embedding any fonts needed that the printer does not already have.
At this point if the printer actually supports PostScript, CUPS simply sends it. If not, it goes on to use one of many PostScript to other format converters. These are are the equivalent of printer drivers.
---
Now you might call CUPS a printer driver itself. That is fairly misleading though. If a Linux application knew the printer spoke PostScript, it could simply send the generic PostScript directly to the printer. Not being optimized for the printer means it would take a bit longer to send and longer for the printer to process, but it would work[1]. That is clearly driver-less, since the app simply sent its normal print format directly to the printer.
Footnote:
[1] Unless the PostScript required fonts not present on the printer, which were not embedded in the PostScript.
---
Now one could argue that the PostScript interpreter in a printer is effectively a printer driver. But who cares. The printer is a black box. All I care about is that if I regularly print to 20 different printers that I don't need to have 20 print drivers on my computer, or that I don't need a full CUPS install on my phone, but can get away with a simple PostScript print queue, and perhaps a quick optimizing pass using a PDD file obtained by asking the printer directly.
I'm emphatically not a FPS enthusiast. The only FPSs I've ever really played are the half-life series, the Portal series (hardly a FPS at all anyway), and the metroid prime series, which are primarily a 3d platformer and only secondarily a FPS.
So why have I played and enjoyed the Half life series (including the original HL, HL:OF, HL:BS, HL2, HL3:EP1, HL3:EP2)? (I refuse to call the episodes HL2, since even Valve admits that was a mistake.)
I've enjoyed the story and atmosphere. Sure I've heard claim that the original HL hardly had a story. That is not true even in the slightest. It had a very rich story. Consider the excellent introduction sequence. Quite a bit of the story is present there. If you stop and listen to the scientists and other characters you get quite a bit more. Then the sequence of locations and events were carefully chosen to add more to the story. As long as you include the implied portions of the story it had substantially more than the average FPS of its day.
With HL2 the plot was even more explicitly story driven, with a bit less being only implied, but if you carefully check out some of the textures you'll see even more attention to detail. HL3 has so far been rather similar, which is why I await the release of HL3:EP3.
As for what is taking valve so long, consider that the companies attention is now diviided between licensing the source engine, managing the steam platform, and quite a few distinct game series. Its little surprise that HL3:EP3 is on the back burner.
Not entirely off the rails.
With a stolen car, the owner cannot utilize nearly any their rights.
Without a copy of the copyrighted work, the copyright holder cannot utilize some of their rights. For example without a copy, the copyright holder can still go after infringes, but would experience great difficulty in licensing the work to others, since in addition to a license, most licensees require an initial copy of the work, and telling them to buy it from one of the other licensees is unlikely to go over very well, and may not even be possible. Not to mention that if you are suing for infringement you basically need to register the work, which requires a copy. For works of US origins it is officially a requirement, although it is plausible that a court would waive the requirement if registration were not possible, but you would face a greatly increased burden of proof, since you would need to prove that you created the work, purchased the rights, or commissioned the creation of the work under the works-for-hire doctrine in order to establish holding copyright on the work, plus the normal burden of showing infringement.
Both cases (stolen car and lack of a copy) can be resolved by re-obtaining the item in question, either the stolen car, or a copy of the work.
They are by no means an exact match, but there is a distinct similarity between them.
Oddly though, I think that
This could happen if a person were to fraudulently convince the state agency that administers copyrights that the owner of the copyright has assigned it to him.
sounds a lot more like registering a fraudulent title to the car.
Pure theft of the car would be far more like destroying all copies of the work the copyright owner has. The copyright owner is still lawfully regarded as the owner in that case, just like with the car theft case, but down must track down a copy in order to fully utilize their ownership rights.
The big difference is that with a car, there is only one that could be tracked down, while with a copyrighted work, there are potentially many, and the owner may be able to use any of them.
Considering that in all likelihood when you remade the processor, you used a smaller process size than the original[1], and that when you use a smaller process size you generally scale the elements to fit[2], the maximum stable clock rate of your remake should be higher substantially greater than the clock rate of the original If the turbo button was a simple clock rate switch, it could speed things up.
The only particularly tricky parts are sound and video, since if you don't compensate for the increased clock rate, these output would be garbled. (I'm assuming that this hypothetical USB device includes the audio video hardware, connected to a a modern video/audio capture chip for display/reproduction on the computer.
[1] Since it can be hard to find a fab for the original process size.
[2] While you could use the smaller size to make the chip at the original scale, that is wasteful.
To quote wikipedia:
In the mathematical community, the word modulo is often used informally. Generally, to say "A is the same as B modulo C" means, more-or-less, "A and B are the same except for differences accounted for or explained by C".
With that in mind, the summary is saying this article and and the previous one cover the same topic except for differences accounted for by "kernel debugging abilities".
In this case that does mean pretty much the same thing as the word minus would have, and the word minus would probably be have a better choice.
The only way to obtain the private key other than breaking the key directly, is to be given access by the key's owner, or given access by somebody the owner gave access to, (recursively). Giving access in this sense means the person in question failed to protect the certificate sufficiently to prevent access. There is no further chain.
It is trivially the case that modified code on the client side could lie to the user. It is also the case that compromised client code could permit a man in the middle to obtain the session key, and then pretend to be the server.
However from a security model view that is equivalent to the recipient conspiring with the MITM to allow the MITM access.
In the US with the exception of tests/exams, it is almost unheard of to not have your name directly on an essay. A few Universities or programs may do that, but the vast majority do not.
On standardized tests/exams the essays never have the authors name.
On professor generated exams, the grader usually can see the name, but professors setting up exams such that graders do not know the student's name, while rare, is not unheard of. If such an exam contains an essay, then the essay would be graded without knowing the student's identity, (unless the identity can be inferred from the essay's content of course).
I think the artist's idea is that in the future we will have fully separate lanes on major roadways with cars driving automatically at really high velocities. With fully separate lanes an automatic car need only worry about not hitting the car ahead of them, since the car would be unable to leave the lane. So no worrying about the car in the next lane cutting you off, etc.
Obviously in practice the there would be the on and off ramps, and periodic opportunities to change lanes, but even so, the cars would basically be on rails, and it is far, far simpler to create a safe automatic driving system for vehicles on rails than free-driving vehicles.
Nevertheless I'm not at all sure if that is what the author meant, since that video was just terrible.
You are correct. A cert does not prove identity per se. Nothing can. Fundamentally the concept of identity is only a psychological/sociological construct.
There is no physical notion of identity. Since we insist on pretending there is such a concept (it is rather convenient, after all), we must accept the fact hat there is always some level of doubt.
Now what do SSL certs have to do with identity? Well lets ask some questions.
Does an SSL cert for Yahoo.com prove that the the other end is controlled by Yahoo! Inc? No it definitely does not.What it does indicate is that the machine is controlled by somebody who the owner of the cert in question gave access to the private key. [1] It gives a presumption that the person controlling the machine is authorized to speak on behalf of the key owner, but that is not absolute. But who is the key owner? Without considering who the CA is we don't have any real idea.
Sometimes this does not matter. In the case of an SSL Cert for http://somerandomsite.example/, we may not be making any assumptions about who is behind the domain. All we might want to know is when we come back later if we are still talking to machines run by the same entity. If the key is still the same, then we can be fairly confident that we are indeed talking to machines run by the same entity.
Or consider the case of being a reporter receiving a signed email from some anonymous tipster. We don't know who is at the other end, but over time the person can establish reliability. We don't necessarily even care who it is in the "real world". What we do care about is knowing if this latest Tip came from the same entity as the previous messages. If they were all valid and all used the same certificate we can be reasonably confident in that.
Now so far we have been ignoring the CA. All the above holds even with self-signed certificates. But the above is pretty weak. the above gives us absolutely no confidence that the certificate belongs to the company or person named in it.
However, if you check who the CA is for the certificate you may be able to establish some confidence that the certificate is owned by by the entity named in it. This requires trusting the CA to verify the identity of the entity requesting a certificate. I do trust that a Verisign extended Validation certificate does indeed belong to the named entity. Given that trust I have high confidence that the pages I recently received from a server using the certificate with serial number "2E:33:87:4F:6F:E2:D4:1E:D3:FF:FF:35:F6:A4:C9:18" did indeed come from PayPal Inc.
I also trust that a non-EV Verisign cert was indeed issued to somebody who once controlled web server or DNS servers of the indicated domain.
I don't however place any trust in some of the CA's found in my browser.'s trusted list. I agree that the current system is unworkable since it requires trusting the CA's, and yet even I cannot be bothered to remove the CA's from the browser that I don't trust. The average Joe definitely won't.
Turning it around to make it opt in would not work either. The average Joe does not even know who Verisign is, much less whether to trust them, and there is no way we can educate the masses enough to allow for this.
So I fully agree we need a change. I also agree that the browser accepting a certificate says nothing about the identity of the entity on the other end. But I disagree that certificates cannot establish identity with reasonably high certainty.
I'm kind of partial to a web of trust based solution in which the browser trusts the Browser writer's key by default (you probably trust them). To ensure better results the browser would also come with a set of certificates of large companies that people may trust, (Google, Microsoft, AT&T, Yahoo, Comcast, IBM, SONY, etc)[2], all of which the browser maker have verified belongs to the company in question. The browser would give the opportunity to chose which if any of those certificates to also m
It would be very interesting to see this redesigned today. Take the basic design, like the use of composite plastics for crash saftey, replace the old automatically controlled manual transmission with a automatic DSG, and add some front seatbelts!
Replace the engine with a modern design of similar weight and power, but which meets modern emissions standards. The improvements of a modern engine and an automatic DSG might be enough to bring the vehicle up to 40 mpg or so. The car should trivially get a 5 star rating on all standard crash tests. The maker may even be able to lobby for having higher speed testing available to earn a 6 star rating.
Now that would be something!
But you don't need to do that. You do need to actively protect your trademarks. However you only need to take actual action if likely confusion would result.
Since Notch's game is very unlikely to cause confusion, a cease and desist is not only not warranted, but might even open up Bethesda to potential liability for one of several torts. (I've never heard of anybody successfully suing for an overzealous C&D letter, but that does not make it impossible.)
What may been prudent is to send a letter reminding Notch that they own a trademark on "The Elder Scrolls", and that care should be taken in ensuring that the logo, gameplay, etc are sufficiently distinct to avoid any possible consumer confusion. The letter properly written would make it clear that they have no objection to the name as long as the rest of the game is sufficiently distinct.
That is all they need to do. Unfortunately most lawyers are not trained in writing such non-threatening letters, and they are very afraid of trying, since the wrong wording could be construed as a grant of rights to use the Trademark, which could result in being fired by their client, and possible a malpractice suit being filed against them.
On the other hand if it is changing on a weekly basis, them many times there will be no new features to familiarize yourself with, or there will be only one or two small features. One person could spend an hour or two testing in those cases. (If nobody can afford that, then the company should have hired another person for IT prior to rolling out the chrome-book, especially since the rollout process could easily monopolize the time of a member of IT for several weeks.
Educating support staff on an invisible bug fix consists of doing exactly nothing. For small features, it may also consist of extremely little.
Communicating changes to end users should always be even less involved than educating the support staff. Basically an extremely high level overview like "You can now plug your Android phones/tablets into the chrome-book and see the files on it" should be enough. Those afraid to experiment can ask support for more details.
Well if you wanted to write a Doom 3 like game, this would be an excellent place to start. You would need to develop graphics and levels, and make some changes for any unique features, but you would need to do that anyway even if starting from scratch, and at least here you already have a working engine. The communities that develop around these source drops often focus on making the code portable, and making minor changes here and there so that it can work well on modern hardware.
So you get to start with a portablized version of a proven game engine. The existing engines are designed with performance on relatively slow machines (by today's standards) in mind. That sounds like a reasonably good fit for something like a modern cell phone. Rewriting the graphics code to take advantage of a mobile's platform's limited 3d acceleration is often not nearly as much work as it sounds.
A properly written balanced budget amendment would only require that the original budget be balanced. Any sane version would permit Congress to authorize emergency deficit spending, and may even permit deficit spending in an executive declared state-of-emergency if Congress is not in session and an emergency session cannot be called quickly enough.
Whether the amendments being proposed are sane or not, I have absolutely no idea. But sane ones are possible.
Sure, but only up to a point. If for example, you limit trades to once per minute, but break ties by who submitted a buy order/sell order first, then there would still be emphasis on getting your offers in as quickly as possible, (i.e. within the first few microseconds). Some, but not all of of the existing HFT practices would change. the HFT would become every-block traders, and would still still try to be first to the transaction.
Breaking ties by last to submit wins would produce something similar. Randomly breaking ties would be best, but people who accuse the system of being based, even if it were not.
Sure, but a such an adjustment is not binding on the actual investor. It cannot be, since no contract at the adjusted price was ever made by the trader. The trader is free to refuse to settle at the adjusted price, and no sane court would compel payment.
It will degrade the signal of downstream recipients. So does absolutely every radio receiver, with no exceptions.
However, please consider that the only downstream recipient may well be the earth or space, considering that the path between a transmitter and receiver often does not pass particularly close to another receiver. How much one of these would impact the downstream signal quality anyway depends on just how much power this is extracting, and just how weak the signal would have been at the downstream receiver without this being present.
Also keep in mind that radio waves can be rather fickle. Placing these devices in certain locations may actually increase the received signal strength downstream, perhaps by absorbing an interference source, or by attenuating a secondary path of the signal which would have interfered with the primary signal.
You don't have access to integrated profiles, but the last I knew there was nothing preventing you from using your apps email address to create a standard Google Account (just like a Yahoo mail user would do), and create a profile using the Google Account, rather than the Google Apps Account.
Is that no longer possible?
The site is no longer invite only. I know this for an absolute fact, since I have a Google+ account but never received an invite.
As far as I can tell, they are turning on and off the ability to register several times a day, or perhaps it has something to so with which server you get connected to. Regardless, I did need to try several times to avoid the "temporarily exceeded our capacity" message.
The security regulations apply to repeating scheduled flights. If you own an airplane you are quite free to offer charter flights with reduced security. (Such as only a metal detector, or even no security at all.)
On the other hand, the average Linksys box actually uses a switch that supports multiple VLANs. The Wan port is actually on the same switch as the 4 LAN ports, but just on a separate VLAN. 6th port (to the router proper) is the only one configured by default to be in multiple VLANs and to see tagged frames.
As you might be able to guess, it is relatively straightforward to merge all ports onto a single VLAN, or to sacrifice one of the LAN ports to create a second WAN port. Obviously the default firmware for the routers don't supply this option, but custom firmware can do this.
"Que" is most certainly a word. It is a rather common French pronoun. It is just not an English word.
Do you even know what office 365 is?
Office 365 is:
That is literally all there is to it. They have tweaked the Web-app UIs to be branded Office 365 and put the whole thing behind a single sign-in portal, but you could set up the exact same service locally if you were a large business.
The more expensive plans also include subscription based access to the Office 2010 "Professional Plus" (previously known as "Enterprise").
Microsoft knows that many small businesses will end up buying full versions of the desktop apps. Since the "Cloud" is merely Exchange and SharePoint, there is no real problem using the desktop apps in conjunction with the web-based apps.
True, but the artifacts skips and drops only happen if you are right near the point of losing signal entirely, or you experienced exceptionally strong interference. However those are quite noticeable unlike quite a bit of analog degradation.
Jitter on the other hand can be far more insidious. If the cable was properly constructed and tested it should not be an issue. If it is very short (6 feet or less) it is not likely to be much of an issue either. But for long runs it can definitely be a problem.