Because the RIAA is disputing whether the defendant's legal fees are reasonable. So the defendant said "OK, let's see what you consider reasonable then.", and the judge said "Yes, let's.". The basic argument behind that is that if the plaintiff spent 500 hours on the case, then the defense claiming 550 hours on the case isn't unreasonable. Similarly, if the plaintiff's lawyers are billing $500/hour, a defense lawyer billing $550/hour isn't unreasonable. After all, if plaintiff really thought times and rates of roughly the magnitudes it itself spent really were unreasonable, why did they spend that?
It's only one rule, though. There's another that looks at the rates the majority of lawyers in the jurisdiction would charge and the number of hours they'd take on that type of case, so if you found a really really cheap lawyer you couldn't stop the defendant from claiming fees at a more normal rate.
OK, I appear to have mis-read. It's not the spammer complaining, it's someone who supports spammers.
Yes, I stand by that. peacefire.org is hosted at a provider that also hosts a fair number of spammers. The owner's been complaining about this for a long time. He knows what kind of people he's standing with. He declines to move. Those spammers are blocked by the RBL, which means all sites on the same machines are blocked. Including peacefire.org.
Which means it still has nothing to do with network neutrality. The blocks are still based on abuse of the network, and I still see nowhere where network neutrality requires putting up with abuse of the network.
Translation: MAPS put a persistent spammer's machines in the RBL. AboveNet and Teleglobe black-hole things in the RBL at the router level. Spammer doesn't like this.
Spammer's fallacy: that network neutrality includes the idea that networks cannot act to limit abuse of their resources. Sorry, but that's wrong. Network neutrality means ISPs aren't allowed to filter based on the source or destination of the traffic. The filter here isn't based on source or destination, just on volume. Analogy: UPS charges everyone the same rates and takes anyone's packages, but they won't take any packages weighing more than 1000lbs. When the spammer shows up with a 10,000lb package and UPS refuses it, they aren't refusing it because it's from the spammer, they're refusing it because it's over-weight. The spammer may not like it that UPS is applying the same rules to him as to everyone else, but that doesn't mean that UPS is giving him special (disadvantageous) treatment.
Because it's not piracy that's the threat to the RIAA/MPAA. It's legitimate electronic distribution.
Yes, I said legitimate.
With VHS tapes and DVDs, you need a big checkbook to finance physical duplication and packaging. You need connections to retailers to have shelf space for the product. You need a big operation to move physical objects from the duplication plant to the retail outlets, and to warehouse them while awaiting distribution. You need conventional advertisement and PR to get consumers to notice the product. That's where the labels come in. They don't produce the material, they just provide the conduit to get it from the people who do create it to the people who want to buy it (while taking a nice hefty cut of the profit in the process).
Electronic distribution over the Internet simply destroys that niche. You need absolutely nothing to make as many copies of a data file as you need. Setting up a Web site and on-line store is easy and relatively cheap (at least compared to physical outlets) and there's companies that specialize in doing it for you. Google and other on-line resources make it easy to get yourself noticed by people who might be interested in your product. So, if I'm an artist producing music and I'm doing legitimate on-line distribution myself, exactly why again do I need to give 95% of my revenue to a label? That is what has the labels terrified of on-line distribution, and why they seem to go out of their way to make even legitimate distribution as cumbersome and unusable as possible: because legitimate on-line distribution is a lethal threat to their business model.
That's also why the RIAA's first target way back when wasn't the pirate FTP sites. It was MyMP3.com, the only site in operation that made any attempt to limit downloads to people who already had legitimate copies.
Except that it may backfire on Viacom. One of the things Google/YouTube depend on to control legal exposure is the DMCA's safe-harbor provision. By attacking that directly, Viacom may force Google to conclude it can't afford not to fight simply to insure it retains that protection. And it's not necessarily just about YouTube, any weakening of the safe-harbor protection impacts almost all of Google's other business. Google may decide this is one fight they can't afford not to fight, and they've surely got the money to do it.
And then there's the iFilm matter. That also weighs into it, and may make Google more confident that they can win.
Oh, and on follow-up, those outside e-mail addresses benefit the company too. When I'm travelling, I often can't reach the company mail system because it's not accessible outside the company network and the local firewalls and access setup at hotels often won't permit the VPN to connect properly. But almost always I can manage to get an SSH connection to my home machine through, and when I can't I can still use a Web browser to get at Web-mail, which means my bosses can reach me via my personal e-mail even if I can't be reached through company e-mail. So that Web-mail may not be without any benefit to the company.
Speaking purely as a sysadmin, I'd block those sites utterly. Web-browser components are the biggest target of malware out there, it's bad enough when targeted at an e-mail client that can lock down scripting and such but Web-mail sites let that stuff through to a browser that has to allow scripting in a corporate environment. And if you're a business you've got your own e-mail system, no company e-mail should be going through a Web-mail system in the first place.
As a techie, no decision would affect me. I deal with my personal e-mail by SSHing to my home machine and reading my personal mail there via mutt. Call me a bigot, but the only protocols a mail client should be using are IMAP (for reading) and SMTP (for sending) and the only acceptable interpretation of the message body is as plain text. Anything else just ends in tears these days.
OTOH, as an employee I'd have to think you've an obligation to provide that access at least for some employees. Think about your IT staff, for example. They're probably expected to work extended or odd hours, usually without extra compensation since they're salaried. In effect the company's asking the employees to give it a big chunk of their personal, outside-of-work, "I have a life" time, for the company's benefit, for free. To me it's then only fair that the company has some obligation to let employees take a certain amount of company time, for free, to deal with all the things they'd've otherwise dealt with during that time the company's wanting from them. If you don't find some way to accommodate them, you're likely to end up with employees who're dissatisfied, frustrated and actively looking to ways to get access to those services. They'll succeed eventually and then you'll have the worst of both worlds. At least if you provide some authorized way to access those services you've got some ability to control the situation, eg. adding specialty filters on the Web proxy for the worst problems.
Unfortunately, if Viacom wins it means the end of Web hosting. Period. You know the Web space your ISP provides? Gone. Cheap basic Web hosting? Gone. Livejournal? Gone. A win for Viacom means any of those would be heavily liable for anything their users did, and they're a much easier target than the individuals responsible for the infringement. The only people who could get Web sites would be those with the wherewithal to own and run their own servers. I'm sure this is exactly what Viacom et. al. would love to see happen, but it's not what the DMCA intended.
Well, on the first one, the counter-argument is that my own Web-host does the same thing to my Web pages. They're uncompressed when I upload them, but if the requesting browser supports it the Web server will gzip-compress them for download. That would, by what you present, disqualify my Web host from the safe-harbor provision. Yet the case of my uploading static Web pages to my Web host to be downloaded by requesting browsers, where the Web host doesn't actively approve content, is exactly the circumstance envisioned by the authors of the safe-harbor provision.
For the third, yes, if Google/YouTube is actively looking for material and taking action based on what they find then they get excluded from the safe-harbor provision. But mere searches aren't quite enough to show that, since they can be done unofficially by employees not acting in their official capacity, or they might have been done as a consequence of a notification (in which case they clearly wouldn't disqualify Google from the safe-harbor provision, it's simply not reasonable to hold that by complying with it's obligations under that section Google sacrifices the protection granted by that self-same section if it complies).
The second one is the only one where Viacom has any solid arguments. And even there, the advertising and such isn't specifically attributable to the infringing activity. It's global to all activity on YouTube, including all of the vast sea of non-infringing material put up there by creators. I guess the question then is, is it reasonable to say that the safe-harbor provision was intended to block all financial benefit to any entity wishing to take advantage of it? If it wasn't then Viacom seems to end up having to show that the majority of content on YouTube is infringing somehow, that YouTube knows this and that YouTube encourages it at the expense of encouraging legitimate uses. That's the Grokster standard, IIRC, and I don't think Viacom can do that without a devastating response from Google.
I'm not on the side of either Viacom or Google. I'm on the side of the law. The law, specifically the DMCA, spells out what responsibilities Google has, and what Viacom has. Viacom's argument here is that, while Google lives up to it's responsibilities, Viacom finds living up to theirs inconvenient and therefore Google should be saddled with Viacom's responsibilities too. Sorry, Viacom, but that's a matter for you to take up with Congress (who wrote the law).
Well, what Microsoft holds the a (purportedly valid) patent on isn't FAT or FAT32. It's on the particular algorithm for mapping long filenames into an 8.3 format and (I think) storing the long filename where it can be found. What the German court found was that a) the idea of doing such a mapping isn't original enough to be potentially patentable, and b) even if it was, the Rock Ridge extensions to the ISO-9660 filesystem (specifically the parts that allow mapping of Unix long filenames to the 8.3 upper-case-only native 9660 names) are similar enough to be prior art and invalidate MS's patent (as it would be simply an obvious extension of that prior art).
Actually it's the two things. The first one that springs to mind is that when an application unmaps pages in Linux they do not go into "free" memory. They go into the dirty-page cache and stay there until some other application really needs the memory. If the original application maps that page back in and the original copy is still in the dirty-page cache, it can be recovered without ever having to page to/from disk. Windows, by contrast, accounts for all released pages as free, essentially combining the "Free" and "Cache" entries from the Linux free command into a single total where Linux seperates them.
The usual bugaboo in these studies is training. They usually assume no training costs for Windows vs. having to train every user on everything from scratch for Linux. This means things like assuming you'll need to train them on how to use OpenOffice before they can get any work done, even though they've been using MS Office for years and already know 80+% of what they need (and 99+% of the basic stuff like how to save files, change fonts, right-align text and the like). Basically the studies count initial set-up costs for Linux but omit them for Windows. That's like saying a Ford GT ($65K) costs less than a Focus ($15K) because this month's payment on the GT is only $500 while the Focus costs nearly $3000 to drive out of the show-room ($2K down-payment, $200 first month's payment, plus tax, license, registration, insurance, etc.).
Another bugaboo is in how they calculate system administration and maintenance costs. They usually calculate it assuming people administer Linux systems like they do Windows: by physically going to each computer that needs work and working on it. Good Unix admins don't do that, though, they set up SSH and rsync and the like so they can get into any box on the network and do their work without ever leaving their desks (which is a lot cheaper than having admins walking around to desks all day). Even the work that requires GUI tools, since X11 works as readily over the network as on the local console. That's also how you get arguments like "Linux doesn't have a standardized desktop, so the admins will have more work trying to figure out each different desktop.". Well, if the sysadmin's logging in to his own account, he'll get his desktop which'll be the same on every machine independent of what the user has picked. Desktop support needs to deal with the user's desktop to answer questions and help with configuration of that, but administrative work doesn't happen in the user's account.
They don't need a new law to deal with this. Reckless driving is already on the books in every jurisdiction. All they need is a law, regulation or ruling saying that failure to pay attention to the road, regardless of why you're not paying attention, is indeed reckless driving. Then it doesn't matter whether the guy's texting on his Blackberry, gabbing on his cel phone, has his head down fiddling with his overly-complicated stereo system or is turned around yelling at his kid in the back seat, he can get pulled over and ticketed. In California this is already the case, see California Vehicle Code sections 23103 and 23104. Besides the fine and possible jail time, it's also a 2-point violation and 4 points in 12 months or 6 points in 24 months is a suspended license which will really put a crimp in these people's lives.
This sounds like a basic surveying technique: nail down one or more reference locations, then to locate any spot on the site shoot the reference points from it. If all you've got is a station that can tell you azimuth and inclination, you need to shoot two reference points and basic trigonometry gives you your exact location. Add in range and you only need one reference point (but two are usually used for error-checking). I was taught this way back in community college, including the use of GPS stations as reference points (much easier than trying to shoot in your reference points based on other known reference points), close to 20 years ago, and except for the GPS addition it was an ancient technique then. I'm not even "skilled in the art", and their method is the second thing I thought of as soon as I read that it was supposed to involve locating yourself underwater. The first thing of course was simply having a good sonar reflector on the buoy and letting the sub's active sonar determine the range and bearing.
All expired entries in the subscription list must be in the result list.
The third and fourth are simply a summary of the function's definition. The first lacks sense in a lot of languages because if the current object was null the method call itself would've generated an error. The second is an invalid requirement, the subscription list can very well be empty and the result should be an empty list (exactly equivalent to having no expired subscriptions in the subscription list).
The two post-conditions, though, show why DbC never took off: they're utterly redundant code. First you scan the subscription list to find all expired subscriptions. Then you scan the list you just built to ensure all subscriptions in it are expired. Well, if they aren't, why were they added in the first place? If you made a mistake in the test for expiration in the result-building code, you're going to make the same mistake in the post-condition. Then you scan the subscription list and, for each expired item, make sure it's in the result list. Um, isn't this what you just did building the result list? So you have 3 iterations of exactly the same thing. And what you're checking isn't even a run-time check, it's a coding-time check on whether the expired() method actually returns the correct results. Worse, the DbC stuff doesn't even eliminate the need for that desk-check, since if the expired() method is wrong the post-condition checks will still succeed even though the results are wrong.
Unfortunately Mr. Meyer runs up against the halting problem. If the definition of a function, or the contract, says that a parameter must never be the NULL pointer, there are two choices: the code must check whether that pointer is NULL, or it must be proven that that pointer can never possibly be NULL. The second is, with the current state of the art, impossible no matter what language constructs are around to help. That means the check has to be done, the only question is whether it's done by the caller or the called function. And the first rule I learned is to eliminate redundancy, which means that given a choice of doing a check in one place or doing it in a large number of places you do it in the one place.
Design by contract is many things, a large number of them good, but it is not a replacement for error-checking.
As has been noted, most programmers already do design-by-contract, they just don't call it that. They call it argument checking. The first thing most routines do is validate their arguments, and return an error if any of them are invalid. The last thing done is to check the results and return an error if the results aren't valid. The calling code then checks for error codes or invalid results (eg. a search function returning a null pointer indicating the item wasn't found).
In the real world I often skip this overhead when the conditions are enforced elsewhere. For example, a data structure needed by an internal function may not have to be checked for existence since if it hadn't been created my initialization function would've detected this and signaled an error and the program would've exited. In cases like that, I either omit the check or wrap it in an ifdef so it's only done during development and ignored by the compiler during the release build.
Don't make the mistake of confusing the name of a concept with the concept itself. You'll find quite often that that nifty shiny-new concept someone's presenting as their own has actually been around for 30-40 years and they've just added some chrome, filed off the serial numbers and changed the name to keep you from noticing this.
There's one big difference, Mr. Blumenthal, between putting a man on the Moon and verifying someone's age over the Internet: when you're trying to put a man on the Moon, the laws of physics aren't lying to you at every turn.
"As part of this agreement, Microsoft will provide a covenant not to assert its patent rights against customers who have purchased SUSE Linux Enterprise Server or other covered products from Novell, and Novell will provide an identical covenant to customers who have a licensed version of Windows or other covered products from Microsoft."
Now, why would Microsoft have to offer any protection from suit to users of anything that they didn't believe contained any of their patents?
Not quite. The deal alleges that things like Samba already include Microsoft's IP, and that through the deal Novell acquires the right to distribute them. But the GPLv3 would require Novell, if it wants to distribute Samba for example and they need a license from Microsoft for that IP, to sub-license anyone for that IP (including people who didn't get Samba from Novell). The Novell/MS deal prohibits that sub-licensing. Catch-22.
Now, whether Samba actually does include any of Microsoft's IP is a completely separate matter. It matters to us, but we're not bound by the terms of the MS/Novell deal. Novell is, and I'm sure Microsoft included language in there where Novell agrees that it does.
Actually it was practicality. It's the same disruption going from 4 bytes to either 8 or 16, so it doesn't matter which you go to. With 8, you have the same issues of subnetting you do with IPv4: the ISP assigns a network number, the network has to assign hosts within the network. With 16 bytes you can have the ISP assign an 8-byte network number (using whatever subnetting scheme they need within their network to group interfaces to downstream networks) and still hand you out an 8-byte local portion. 8 bytes for the local part was chosen to be big enough to hold a standard 48-bit (6 byte) MAC (eg. Ethernet MACs) and have a bit left over for other uses like having a flag bit to distinguish MAC-based from hand-assigned addresses and allowing a bit of local subnetting.
That requires you to have physical access to the local Ethernet segment, though. This means you're already in my building and on my network, which means I've got bigger problems. So forget finding MAC addresses, the worm's on the other side of a hardware router and only sees the router's MAC address regardless of what it tries. Packets to or from broadcast addresses are subject to heavy filtering at the border and on individual hosts, since there's very very few legitimate reasons for such packets in the first place (eg. an ICMP ECHO REQUEST to a broadcast address should never occur and should never be responded to, to give an ancient and obvious case).
The address space doesn't make worm propagation impossible, but it does make finding machines via scanning a nigh-impossible job.
Not quite. An IPv6 address is 128 bits, a 64-bit network number (assigned by your ISP) and a 64-bit local portion (generated by the machine). If a worm knows the network you're on, it still has to scan the entire 64-bit local space to find your actual address. Current IPv4 addresses are 32 bits, so the worm has 4 billion current Internets to scan to find your host. That's going to take a while.
Is it wrong and/or illegal to sit around in a coffee-shop talking about your opinions of other people by name?Is letting people do that letting free speech go too far? It seems to me that this question and the one raised must have one and the same answer.
Slashdot Mirror
Because the RIAA is disputing whether the defendant's legal fees are reasonable. So the defendant said "OK, let's see what you consider reasonable then.", and the judge said "Yes, let's.". The basic argument behind that is that if the plaintiff spent 500 hours on the case, then the defense claiming 550 hours on the case isn't unreasonable. Similarly, if the plaintiff's lawyers are billing $500/hour, a defense lawyer billing $550/hour isn't unreasonable. After all, if plaintiff really thought times and rates of roughly the magnitudes it itself spent really were unreasonable, why did they spend that?
It's only one rule, though. There's another that looks at the rates the majority of lawyers in the jurisdiction would charge and the number of hours they'd take on that type of case, so if you found a really really cheap lawyer you couldn't stop the defendant from claiming fees at a more normal rate.
OK, I appear to have mis-read. It's not the spammer complaining, it's someone who supports spammers.
Yes, I stand by that. peacefire.org is hosted at a provider that also hosts a fair number of spammers. The owner's been complaining about this for a long time. He knows what kind of people he's standing with. He declines to move. Those spammers are blocked by the RBL, which means all sites on the same machines are blocked. Including peacefire.org.
Which means it still has nothing to do with network neutrality. The blocks are still based on abuse of the network, and I still see nowhere where network neutrality requires putting up with abuse of the network.Translation: MAPS put a persistent spammer's machines in the RBL. AboveNet and Teleglobe black-hole things in the RBL at the router level. Spammer doesn't like this.
Spammer's fallacy: that network neutrality includes the idea that networks cannot act to limit abuse of their resources. Sorry, but that's wrong. Network neutrality means ISPs aren't allowed to filter based on the source or destination of the traffic. The filter here isn't based on source or destination, just on volume. Analogy: UPS charges everyone the same rates and takes anyone's packages, but they won't take any packages weighing more than 1000lbs. When the spammer shows up with a 10,000lb package and UPS refuses it, they aren't refusing it because it's from the spammer, they're refusing it because it's over-weight. The spammer may not like it that UPS is applying the same rules to him as to everyone else, but that doesn't mean that UPS is giving him special (disadvantageous) treatment.
Because it's not piracy that's the threat to the RIAA/MPAA. It's legitimate electronic distribution.
Yes, I said legitimate.
With VHS tapes and DVDs, you need a big checkbook to finance physical duplication and packaging. You need connections to retailers to have shelf space for the product. You need a big operation to move physical objects from the duplication plant to the retail outlets, and to warehouse them while awaiting distribution. You need conventional advertisement and PR to get consumers to notice the product. That's where the labels come in. They don't produce the material, they just provide the conduit to get it from the people who do create it to the people who want to buy it (while taking a nice hefty cut of the profit in the process).
Electronic distribution over the Internet simply destroys that niche. You need absolutely nothing to make as many copies of a data file as you need. Setting up a Web site and on-line store is easy and relatively cheap (at least compared to physical outlets) and there's companies that specialize in doing it for you. Google and other on-line resources make it easy to get yourself noticed by people who might be interested in your product. So, if I'm an artist producing music and I'm doing legitimate on-line distribution myself, exactly why again do I need to give 95% of my revenue to a label? That is what has the labels terrified of on-line distribution, and why they seem to go out of their way to make even legitimate distribution as cumbersome and unusable as possible: because legitimate on-line distribution is a lethal threat to their business model.
That's also why the RIAA's first target way back when wasn't the pirate FTP sites. It was MyMP3.com, the only site in operation that made any attempt to limit downloads to people who already had legitimate copies.
Except that it may backfire on Viacom. One of the things Google/YouTube depend on to control legal exposure is the DMCA's safe-harbor provision. By attacking that directly, Viacom may force Google to conclude it can't afford not to fight simply to insure it retains that protection. And it's not necessarily just about YouTube, any weakening of the safe-harbor protection impacts almost all of Google's other business. Google may decide this is one fight they can't afford not to fight, and they've surely got the money to do it.
And then there's the iFilm matter. That also weighs into it, and may make Google more confident that they can win.
Oh, and on follow-up, those outside e-mail addresses benefit the company too. When I'm travelling, I often can't reach the company mail system because it's not accessible outside the company network and the local firewalls and access setup at hotels often won't permit the VPN to connect properly. But almost always I can manage to get an SSH connection to my home machine through, and when I can't I can still use a Web browser to get at Web-mail, which means my bosses can reach me via my personal e-mail even if I can't be reached through company e-mail. So that Web-mail may not be without any benefit to the company.
Speaking purely as a sysadmin, I'd block those sites utterly. Web-browser components are the biggest target of malware out there, it's bad enough when targeted at an e-mail client that can lock down scripting and such but Web-mail sites let that stuff through to a browser that has to allow scripting in a corporate environment. And if you're a business you've got your own e-mail system, no company e-mail should be going through a Web-mail system in the first place.
As a techie, no decision would affect me. I deal with my personal e-mail by SSHing to my home machine and reading my personal mail there via mutt. Call me a bigot, but the only protocols a mail client should be using are IMAP (for reading) and SMTP (for sending) and the only acceptable interpretation of the message body is as plain text. Anything else just ends in tears these days.
OTOH, as an employee I'd have to think you've an obligation to provide that access at least for some employees. Think about your IT staff, for example. They're probably expected to work extended or odd hours, usually without extra compensation since they're salaried. In effect the company's asking the employees to give it a big chunk of their personal, outside-of-work, "I have a life" time, for the company's benefit, for free. To me it's then only fair that the company has some obligation to let employees take a certain amount of company time, for free, to deal with all the things they'd've otherwise dealt with during that time the company's wanting from them. If you don't find some way to accommodate them, you're likely to end up with employees who're dissatisfied, frustrated and actively looking to ways to get access to those services. They'll succeed eventually and then you'll have the worst of both worlds. At least if you provide some authorized way to access those services you've got some ability to control the situation, eg. adding specialty filters on the Web proxy for the worst problems.
Unfortunately, if Viacom wins it means the end of Web hosting. Period. You know the Web space your ISP provides? Gone. Cheap basic Web hosting? Gone. Livejournal? Gone. A win for Viacom means any of those would be heavily liable for anything their users did, and they're a much easier target than the individuals responsible for the infringement. The only people who could get Web sites would be those with the wherewithal to own and run their own servers. I'm sure this is exactly what Viacom et. al. would love to see happen, but it's not what the DMCA intended.
Well, on the first one, the counter-argument is that my own Web-host does the same thing to my Web pages. They're uncompressed when I upload them, but if the requesting browser supports it the Web server will gzip-compress them for download. That would, by what you present, disqualify my Web host from the safe-harbor provision. Yet the case of my uploading static Web pages to my Web host to be downloaded by requesting browsers, where the Web host doesn't actively approve content, is exactly the circumstance envisioned by the authors of the safe-harbor provision.
For the third, yes, if Google/YouTube is actively looking for material and taking action based on what they find then they get excluded from the safe-harbor provision. But mere searches aren't quite enough to show that, since they can be done unofficially by employees not acting in their official capacity, or they might have been done as a consequence of a notification (in which case they clearly wouldn't disqualify Google from the safe-harbor provision, it's simply not reasonable to hold that by complying with it's obligations under that section Google sacrifices the protection granted by that self-same section if it complies).
The second one is the only one where Viacom has any solid arguments. And even there, the advertising and such isn't specifically attributable to the infringing activity. It's global to all activity on YouTube, including all of the vast sea of non-infringing material put up there by creators. I guess the question then is, is it reasonable to say that the safe-harbor provision was intended to block all financial benefit to any entity wishing to take advantage of it? If it wasn't then Viacom seems to end up having to show that the majority of content on YouTube is infringing somehow, that YouTube knows this and that YouTube encourages it at the expense of encouraging legitimate uses. That's the Grokster standard, IIRC, and I don't think Viacom can do that without a devastating response from Google.
I'm not on the side of either Viacom or Google. I'm on the side of the law. The law, specifically the DMCA, spells out what responsibilities Google has, and what Viacom has. Viacom's argument here is that, while Google lives up to it's responsibilities, Viacom finds living up to theirs inconvenient and therefore Google should be saddled with Viacom's responsibilities too. Sorry, Viacom, but that's a matter for you to take up with Congress (who wrote the law).
Well, what Microsoft holds the a (purportedly valid) patent on isn't FAT or FAT32. It's on the particular algorithm for mapping long filenames into an 8.3 format and (I think) storing the long filename where it can be found. What the German court found was that a) the idea of doing such a mapping isn't original enough to be potentially patentable, and b) even if it was, the Rock Ridge extensions to the ISO-9660 filesystem (specifically the parts that allow mapping of Unix long filenames to the 8.3 upper-case-only native 9660 names) are similar enough to be prior art and invalidate MS's patent (as it would be simply an obvious extension of that prior art).
Actually it's the two things. The first one that springs to mind is that when an application unmaps pages in Linux they do not go into "free" memory. They go into the dirty-page cache and stay there until some other application really needs the memory. If the original application maps that page back in and the original copy is still in the dirty-page cache, it can be recovered without ever having to page to/from disk. Windows, by contrast, accounts for all released pages as free, essentially combining the "Free" and "Cache" entries from the Linux free command into a single total where Linux seperates them.
The usual bugaboo in these studies is training. They usually assume no training costs for Windows vs. having to train every user on everything from scratch for Linux. This means things like assuming you'll need to train them on how to use OpenOffice before they can get any work done, even though they've been using MS Office for years and already know 80+% of what they need (and 99+% of the basic stuff like how to save files, change fonts, right-align text and the like). Basically the studies count initial set-up costs for Linux but omit them for Windows. That's like saying a Ford GT ($65K) costs less than a Focus ($15K) because this month's payment on the GT is only $500 while the Focus costs nearly $3000 to drive out of the show-room ($2K down-payment, $200 first month's payment, plus tax, license, registration, insurance, etc.).
Another bugaboo is in how they calculate system administration and maintenance costs. They usually calculate it assuming people administer Linux systems like they do Windows: by physically going to each computer that needs work and working on it. Good Unix admins don't do that, though, they set up SSH and rsync and the like so they can get into any box on the network and do their work without ever leaving their desks (which is a lot cheaper than having admins walking around to desks all day). Even the work that requires GUI tools, since X11 works as readily over the network as on the local console. That's also how you get arguments like "Linux doesn't have a standardized desktop, so the admins will have more work trying to figure out each different desktop.". Well, if the sysadmin's logging in to his own account, he'll get his desktop which'll be the same on every machine independent of what the user has picked. Desktop support needs to deal with the user's desktop to answer questions and help with configuration of that, but administrative work doesn't happen in the user's account.
They don't need a new law to deal with this. Reckless driving is already on the books in every jurisdiction. All they need is a law, regulation or ruling saying that failure to pay attention to the road, regardless of why you're not paying attention, is indeed reckless driving. Then it doesn't matter whether the guy's texting on his Blackberry, gabbing on his cel phone, has his head down fiddling with his overly-complicated stereo system or is turned around yelling at his kid in the back seat, he can get pulled over and ticketed. In California this is already the case, see California Vehicle Code sections 23103 and 23104. Besides the fine and possible jail time, it's also a 2-point violation and 4 points in 12 months or 6 points in 24 months is a suspended license which will really put a crimp in these people's lives.
This sounds like a basic surveying technique: nail down one or more reference locations, then to locate any spot on the site shoot the reference points from it. If all you've got is a station that can tell you azimuth and inclination, you need to shoot two reference points and basic trigonometry gives you your exact location. Add in range and you only need one reference point (but two are usually used for error-checking). I was taught this way back in community college, including the use of GPS stations as reference points (much easier than trying to shoot in your reference points based on other known reference points), close to 20 years ago, and except for the GPS addition it was an ancient technique then. I'm not even "skilled in the art", and their method is the second thing I thought of as soon as I read that it was supposed to involve locating yourself underwater. The first thing of course was simply having a good sonar reflector on the buoy and letting the sub's active sonar determine the range and bearing.
Translating the contract:
- We can't be a null object.
- The subscription list can't be empty.
- All entries in the result list must be expired.
- All expired entries in the subscription list must be in the result list.
The third and fourth are simply a summary of the function's definition. The first lacks sense in a lot of languages because if the current object was null the method call itself would've generated an error. The second is an invalid requirement, the subscription list can very well be empty and the result should be an empty list (exactly equivalent to having no expired subscriptions in the subscription list).The two post-conditions, though, show why DbC never took off: they're utterly redundant code. First you scan the subscription list to find all expired subscriptions. Then you scan the list you just built to ensure all subscriptions in it are expired. Well, if they aren't, why were they added in the first place? If you made a mistake in the test for expiration in the result-building code, you're going to make the same mistake in the post-condition. Then you scan the subscription list and, for each expired item, make sure it's in the result list. Um, isn't this what you just did building the result list? So you have 3 iterations of exactly the same thing. And what you're checking isn't even a run-time check, it's a coding-time check on whether the expired() method actually returns the correct results. Worse, the DbC stuff doesn't even eliminate the need for that desk-check, since if the expired() method is wrong the post-condition checks will still succeed even though the results are wrong.
Unfortunately Mr. Meyer runs up against the halting problem. If the definition of a function, or the contract, says that a parameter must never be the NULL pointer, there are two choices: the code must check whether that pointer is NULL, or it must be proven that that pointer can never possibly be NULL. The second is, with the current state of the art, impossible no matter what language constructs are around to help. That means the check has to be done, the only question is whether it's done by the caller or the called function. And the first rule I learned is to eliminate redundancy, which means that given a choice of doing a check in one place or doing it in a large number of places you do it in the one place.
Design by contract is many things, a large number of them good, but it is not a replacement for error-checking.
As has been noted, most programmers already do design-by-contract, they just don't call it that. They call it argument checking. The first thing most routines do is validate their arguments, and return an error if any of them are invalid. The last thing done is to check the results and return an error if the results aren't valid. The calling code then checks for error codes or invalid results (eg. a search function returning a null pointer indicating the item wasn't found).
In the real world I often skip this overhead when the conditions are enforced elsewhere. For example, a data structure needed by an internal function may not have to be checked for existence since if it hadn't been created my initialization function would've detected this and signaled an error and the program would've exited. In cases like that, I either omit the check or wrap it in an ifdef so it's only done during development and ignored by the compiler during the release build.
Don't make the mistake of confusing the name of a concept with the concept itself. You'll find quite often that that nifty shiny-new concept someone's presenting as their own has actually been around for 30-40 years and they've just added some chrome, filed off the serial numbers and changed the name to keep you from noticing this.
There's one big difference, Mr. Blumenthal, between putting a man on the Moon and verifying someone's age over the Internet: when you're trying to put a man on the Moon, the laws of physics aren't lying to you at every turn.
I don't believe the agreement's publicly available, but from the Novell press release about it:
"As part of this agreement, Microsoft will provide a covenant not to assert its patent rights against customers who have purchased SUSE Linux Enterprise Server or other covered products from Novell, and Novell will provide an identical covenant to customers who have a licensed version of Windows or other covered products from Microsoft."
Now, why would Microsoft have to offer any protection from suit to users of anything that they didn't believe contained any of their patents?
Not quite. The deal alleges that things like Samba already include Microsoft's IP, and that through the deal Novell acquires the right to distribute them. But the GPLv3 would require Novell, if it wants to distribute Samba for example and they need a license from Microsoft for that IP, to sub-license anyone for that IP (including people who didn't get Samba from Novell). The Novell/MS deal prohibits that sub-licensing. Catch-22.
Now, whether Samba actually does include any of Microsoft's IP is a completely separate matter. It matters to us, but we're not bound by the terms of the MS/Novell deal. Novell is, and I'm sure Microsoft included language in there where Novell agrees that it does.
Actually it was practicality. It's the same disruption going from 4 bytes to either 8 or 16, so it doesn't matter which you go to. With 8, you have the same issues of subnetting you do with IPv4: the ISP assigns a network number, the network has to assign hosts within the network. With 16 bytes you can have the ISP assign an 8-byte network number (using whatever subnetting scheme they need within their network to group interfaces to downstream networks) and still hand you out an 8-byte local portion. 8 bytes for the local part was chosen to be big enough to hold a standard 48-bit (6 byte) MAC (eg. Ethernet MACs) and have a bit left over for other uses like having a flag bit to distinguish MAC-based from hand-assigned addresses and allowing a bit of local subnetting.
That requires you to have physical access to the local Ethernet segment, though. This means you're already in my building and on my network, which means I've got bigger problems. So forget finding MAC addresses, the worm's on the other side of a hardware router and only sees the router's MAC address regardless of what it tries. Packets to or from broadcast addresses are subject to heavy filtering at the border and on individual hosts, since there's very very few legitimate reasons for such packets in the first place (eg. an ICMP ECHO REQUEST to a broadcast address should never occur and should never be responded to, to give an ancient and obvious case).
The address space doesn't make worm propagation impossible, but it does make finding machines via scanning a nigh-impossible job.
Not quite. An IPv6 address is 128 bits, a 64-bit network number (assigned by your ISP) and a 64-bit local portion (generated by the machine). If a worm knows the network you're on, it still has to scan the entire 64-bit local space to find your actual address. Current IPv4 addresses are 32 bits, so the worm has 4 billion current Internets to scan to find your host. That's going to take a while.
Is it wrong and/or illegal to sit around in a coffee-shop talking about your opinions of other people by name?Is letting people do that letting free speech go too far? It seems to me that this question and the one raised must have one and the same answer.