I think the lubricant is probably intended to be used on the spindle, which is where the bulk of the friction in a hard drive will be. I doubt that using it on the head actuators as well is going to make seek times noticably faster, although it may be easier for manufacturing to use one lube throughout the drive.
what's the longest english word that only uses each letter once?
The longest I (OK, egrep and the words file) can come up with is "uncopyrightable" at 15 letters. I'd have posted the command line too, but the Slashdot lameness filter is eating it for lunch...:)
I don't know your education, but in mine learning the difference between a worm and a caterpillar came a good few years before any option of learning Latin. If the choice of the brand new icon for articles involving worms is anything to go by, I doubt we are going to be seeing accurate Latin in stories anytime soon...;)
I submit my #9 Imagine128 graphics card, which I never did get to work under RH9, despite it being in the list of supported cards.
That touches on a problem that is probably going to make this project a lame duck. There are far more people out there who will give up or accept a compromise after repeated failures than there are those who keep going until they get things working. I suspect a large number of "x doesn't work" entries are more likely to represent "I couldn't get x to work". Clearly the latter doesn't necessarily mean that the device is incompatible with Linux, although it certainly implies there is room for improvement.
I'm afraid it's not that simple. Not by a long shot, which is why these micropayment/hashcash schemes are not getting all that much support.
Sure it shouldn't penalise senders of solicited email, but the problem I pointed out was that the system has to accomodate all levels of user. Some of those users may not understand how to add a whitelist entry, and you can't provide them with detailed instructions unless you know what email client they are using. Consider the scenario of a clueless user trying to subscribe to a legitimate emailing list; what process can be used to establish that:
The user really wanted the emails (subscription confirmation)
The mailing list is whitelisted
No charges are incurred by either party. At a pinch, you could have an initial charge being refunded, but bear in mind the effect that a large number of bogus subscriptions could have.
all the while of course, ensuring that the system cannot be abused for fun or profit... What about handling some of the other scenarios email is used in? Not so easy is it?
Micropayments are also an issue which is more complex once you start looking at it in more depth with regard to actually doing it. For instance, are you just going to require everyone to use one monopolistic micropayment provider which would then have a strangehold over email and be a huge target for scams? Or how about how two systems using different micropayment providers might interact if there are a non-trivial number of micropayment service providers? You also seem to use "email provider" in the context of "ISP" or "Webmail provider", but what about entities that want to run their own SMTP server? A large corporate might shrug off any costs of doing so, but what about an SME, SOHO or even a technically inclined end user?
Even if the issue of the micropayment system is resolved and open to all from Hotmail to "Joe on DSL", how do you deal with the change over period when only a fraction of users are using the system? ESMTP has been around since 1995, yet I'm still seeing servers using the older SMTP "HELO" nearly a decade later, I think it's a given adoption would be similarly slow here. So you need to be able to handle emails between those who have adopted the system and those that have not, preferably without Joe-jobbing everyone whose email address got faked in a spam with a solicitation to get added to your whitelist...
If someone can resolve those issues (and numerous others I haven't touched on), then the email using world may well beat a path to their door with large sums of cash. Until then however, I think it's safe to say that micropayment based email is a non-starter, more is the pity.
While Penny Black, or something like it, would certainly help make spamming less economical there are a couple of major problems with it that need resolving. Firstly, it will penalise legitimate mailing lists like the LKML and so on. Sure, you can implement a whitelist mechanism to waive the charge, but it only takes so many users to overlook this, either through ignorance or forgetfullness, and the costs start to add up. You could possibly build this waiving into the sign-up process - "click here to confirm your subscription and waive all Penny Black costs". The trick is doing so in a manner that could not be automatically invoked by a spammer, because if they figure that out then they've not only validated your email address but they've got carte blanche to spam you for free. Needless to say, this was not part of the Penny Black proposal and nor would I trust Microsoft to come up with a secure implementation of such as mechanism.
Secondly, and this is the show-stopper at the moment, it relies on there being an effective micropayment system that can be easily integrated into SMTP, so far there isn't really a viable micropayment system, let alone one that works with SMTP. Hopefully the likes of iTunes etc. will change that, but Penny Black would need to handle several orders of magnitude more transactions than iTunes, which might pose problems. The vast majority of spammers also don't care much for the law, so the payment system would need to be proof against stolen credit card numbers, abuse of compromised PCs, faked domain names...
It's a nice idea, and I might even use it if it were to happen, but somehow I just can't see something like Penny Black ever getting off the ground.
Exactly my view; it's all about the deterance factor and I very much doubt that their lawyer's costs run to less than $3000 per case unless they are on-staff. The RIAA has already shown that they are not afraid to take their blood money from 12 year old girls and grandmothers, just to emphasise that no one is expempt as well. True, the former's costs were actually met by a third party, but that does not change the fact that the RIAA still expected payment.
What I'm curious about though, is that all the cases I've read about do seem to be very much canted in favour of the RIAA. The defendents are almost always financial unable to fight the case and there is also almost always clear cut copyright infringement. Is this merely media bias, or does the RIAA get to pick and choose its cases once they know who the mark is to better meet their goals of deterrence?
As I understand it, the RIAA usually files a John/Jane Doe case to subpoena the evidence needed to establish their victim's identity. Supposing that J. Doe turns out to be a very wealthy and outspoken proponent of fair use and realistic copyright laws who quite probably would be prepared to fight them in court. Issues of whether they would or not aside, could the RIAA make some excuses and "opt out" of the case at that point, or not?
I've also noticed that my pet dog has demonstrated a fair grasp of basic mathematics as it pertains to food recently. She's been put onto a diet to keep her weight down after pulling a leg muscle, and so her treats are being rationed - hence the developing grasp of math.
She definitely understands the concept of "I've had zero after-dinner chews" and doesn't let up until she has had the two she is allowed. One won't do, nor will one and a half, one and two half biscuits is OK, but she inspects the two halves quite closely as if to make sure she isn't being conned. A more impressive display awaits in the evening when she has her choc drops; she's allowed six and once again complains if you try and put the jar away before she's had all six.
It's probably just driven by the requirements of circumstance of course, but I'm tempted to try and work some more advanced math and larger numbers into the scenario and see how well she does.
That is, of course, assuming that they ever really intended to go with an open source solution. Newham is currently a Microsoft shop, but as the article says some of thier software is not up to date. So, provided that the cost of the Linux study was less than that of the discount offered by Microsoft to stay with them, then they come out ahead.
Whether that's as far ahead as dumping Microsoft and going with open souce is an entirely different matter. It'll sure be interesting to revisit this in a few years time and compare and contrast Newham with Munich though...
That's a rather sweeping and largely inaccurate statement you just made. Sure some routers are only "firewalls" by virtue of utilising NAT, but more because that's the way they have been configured, not because of a design limitation. I'd class my Draytek DSL router as a "home router", but it's definitely something I'd class as a firewall as well - as far as I can ascertain it's using BSD's PF - and NAT is optional; mine is routing real IPs very nicely.
A user with your NAT setup and packet filter (a firewall is a dedicated PC/router as far as I am concerned) that does ingres filtering only can still run a trojan. If they do, then that trojan can obtain outbound access without the user being prompted to allow it permission. If it can open an outbound TCP channel, then it can retrieve and act upon commands from an IRC channel, website or whatever, to propagate itself, spam, DDoS other hosts...
There's a difference between being able to be switched off by the user with confirmation via interaction with onscreen menus etc. and being able to be switched off via by software. And by "software", read "the very next trojan horse that comes along", especially since trojans are already trying to disable known AV and local packet filters. Abuse of a monopoly has nothing to do with it, this is looking very much like standard Microsoft; a poor product propped up with first rate marketing.
I've been playing with SP2 for a couple of betas, and initial impressions that while it is certainly an improvement, it's still a *long* way from being secure, even when configured properly. With admin priviledges and a command prompt I can disable all the new security bells and whistles in less than 10 seconds; how long is a trojan going to take? The only way forward that I can see is that "admin by default" has got to go, but that isn't likely to happen until Longhorn at the earliest.
Looks like we're not done with the incessant script kiddie probes for vulnerable MS network ports and trojan backdoors just yet.:(
Actually, Microsoft has published an MD5 sum for SP2 (or one version of it anyway), although they do not seem to be advertising the fact and I only stumbled across it. You can find it in the last paragraph of the article Top 10 Reasons to Deploy Windows XP Service Pack 2, and maybe elsewhere on Microsoft's site.
Any sites that are doing more than linking to the official download sources are probably going to be getting nastygrams though; check out the second to last paragarph. There are some pretty useful links for those involved in largescale rollouts at the very bottom as well.
Is it just me, or does that "Aurora" on the web page at abovetopsecret you linked to look suspiciously like the Scramjet NASA has been very publically testing recently? The one they have a photo of a few planes down (Hyper-X)... I think they are probably one and the same somehow.
That's not to say there isn't some other cool shit at A51 we won't be seeing at airshows for a few decades of course.:)
Lawyers or law firms are not allowed to sue independently. A client is required.
My understanding of the law from the KIllustrator thing was that there is indeed a client, which in that particular case was Adobe. However, there is a quirk in German trademark law that allows lawyers to initiate a case on behalf of a client without that client being aware of the fact. So, technically there is a "client", and I assume that at some point in the proceedings that they would have to be informed of the fact. Were this be applied to the meta tag instance, then the client would presumably always be the state.
Who is going to search every web page to find incorrect meta tags
Since we are talking about Germany here, I have a good idea how this is actually going to "work". Does anyone remember the whole Adobe Illustrator / KIllustrator trademark fiasco? Germany has an odd system of trademark enforcement whereby lawfirms look for trademark infringements and bring suits, telling the entity whose trademark was infringed is optional - as happened with Adobe. Naturally the real winners are the lawfirms, many of whom are bottom feeders with dubious case selection and billing practices.
You could quite easily employ the same scheme here; the lawfirms "police" the sites and bring suits against those they believe infringe. If the court decides that they are right or the defendent settles before trial, then they make money. The problem is obviously that you are going to attract the same type of ambulance chasing lawyers to the business as the trademark example, but modern legislation hardly has a good track record for being sensible, does it?
True, 34dB isn't quiet, but this is only a first stab at this and I suspect it'll become the default soon enough. Hopefully the other (quieter) PSU makers will release their own take on this pretty soon and we'll get the best of both worlds. A *standard PSU connector* between vendors would be nice too, if any of the designers are out there.:)
As to the noise, I've recently upgraded a couple of my systems and while I was prepping them I just left the PSU cables loose in the case. Once everything was done I tied them all up nice and neat and tucked away the unused PSU cables out of the airflow. The idling temperature of the AMD CPU dropped a full two degrees C with this simple change, but more importantly the case and CPU fan dropped by nearly 500 RPMs apiece. I know airflow matters, but not being a serious modder I had know idea it could be that much!
Re:Sounds Like...
on
Broken Angels
·
· Score: 2, Informative
Although, from the sounds of it, this doesn't go off on the same religious bent.
Well, actually, both books do touch on this. There is a religion (can't remember the name - I read the books over a year ago) that believes that the process of downloading to a cortical stack leaves the soul behind and it is therefore something to be avoided. I thought it was more like the Jehovah's Witness prohibition against receiving blood transfusions than Catholism's view of contraception though.
Agreed, it's essential that they get this right before they release it because every WinXP using man and his dog with a iota of a clue about patching is going to be installing it. The other problem with SP2 is the people that are eagerly awaiting this patch are probably not responsible for the bulk of compromised PCs that are hammering my firewall with probes to ports 135 and 445.
If Microsoft is now as serious about security as they claim then they should get this patch right *and* rescind their earlier edict against magazines carrying patches on their cover CDs. Combine that with an article that focuses more on the new features than the importance of security and you just might get the thing more widely deployed. More Windows users with fewer ports exposed to the Internet, even if it's only via ICF or whatever it's called now, has got to have an impact on the size of the BotNets out there, right?
That's because the lawyers for IBM and the rest have a knack for demonstrating that the gun is indeed smoking, but that it is pointed at SCO's foot and they pulled the trigger themselves.
Nope, you didn't miss anything as SCP has no incremental option. You could work around that by using a data compression tool to create an archive of just the files that have been changed, then SCPing that and uncompressing over the data mirror on the remote server. That doesn't help much if part of your data set is a big database file of which only a couple of records have changed though.
A far more efficient method would be to look at using RSYNC with SSH as a few others have pointed out. There's a pretty good HOWTO for Windows here, which avoids the overhead of a full Cygwin install. You'll probably want to check out the link to cwRSYNC at the top of that page too.
Also, shut down all that extra
crud that Microsoft enabled by default for the few users that might think about using it some day. You'll have more free memory (or less junk in the pagefile) and be less vulnerable to remote attacks as well. These packages might do this kind of thing for you, but most likely they are just snake-oil relying on the placebo effect and a "no-refund" clause.
Actually, it does make some sense, although I don't think it's necessary just at the moment and hope that it never will be. I'm assuming that it will function in the same way as car/home insurance if you have cause to make a claim, and also that OSRM itself will be under-written to protect its clients against OSRM going bust.
So, when Foo Corp claims that by using Linux you are infringing their patents you simply remain noncommittal (just as you should never accept blame at a car accident) and call OSRM. If OSRM is as good as my car insurer that's pretty much the end of the matter as far as you are concerned. They deal with Foo Corp and their lawyers and resolve the issue as best they can, whether that be getting the case dismissed or negotiating and paying your license fees. You'll probably get a letter every now and then letting you know of any developments, requesting information they might need and so on, but that's all. For many CTOs paying a company like OSRM some money each year might just be worth the removal of one less thing to fret and loose sleep over.
It sure sounds like a false economy going with S754 now to me too; saving up a little more and getting a S939 mobo and the 3500+ CPU would be much better. S939 is the way AMD is going with its mainstream CPUs, and there is the dual-core "Toledo" chip due late next year according to their roadmap to give a sweet upgrade path that's pretty much guaranteed to work.
I've just upgraded two of my boxes to the 3500+ and 3800+ S939 chips and couldn't be happier with the results. Both Linux (FC2 x86_64 on the 3500+) and Windows (XP on the 3800+) motor along far faster than I was expecting, and I have that dual core upgrade to fall back on when games require that much grunt. As far as I'm concerned it's "Intel Inside" alright - "Inside the store, covered in dust on the shelf".:)
Slashdot Mirror
I think the lubricant is probably intended to be used on the spindle, which is where the bulk of the friction in a hard drive will be. I doubt that using it on the head actuators as well is going to make seek times noticably faster, although it may be easier for manufacturing to use one lube throughout the drive.
what's the longest english word that only uses each letter once? The longest I (OK, egrep and the words file) can come up with is "uncopyrightable" at 15 letters. I'd have posted the command line too, but the Slashdot lameness filter is eating it for lunch... :)
I don't know your education, but in mine learning the difference between a worm and a caterpillar came a good few years before any option of learning Latin. If the choice of the brand new icon for articles involving worms is anything to go by, I doubt we are going to be seeing accurate Latin in stories anytime soon... ;)
That touches on a problem that is probably going to make this project a lame duck. There are far more people out there who will give up or accept a compromise after repeated failures than there are those who keep going until they get things working. I suspect a large number of "x doesn't work" entries are more likely to represent "I couldn't get x to work". Clearly the latter doesn't necessarily mean that the device is incompatible with Linux, although it certainly implies there is room for improvement.
No harm in trying though. ;)
Sure it shouldn't penalise senders of solicited email, but the problem I pointed out was that the system has to accomodate all levels of user. Some of those users may not understand how to add a whitelist entry, and you can't provide them with detailed instructions unless you know what email client they are using. Consider the scenario of a clueless user trying to subscribe to a legitimate emailing list; what process can be used to establish that:
- The user really wanted the emails (subscription confirmation)
- The mailing list is whitelisted
- No charges are incurred by either party. At a pinch, you could have an initial charge being refunded, but bear in mind the effect that a large number of bogus subscriptions could have.
all the while of course, ensuring that the system cannot be abused for fun or profit... What about handling some of the other scenarios email is used in? Not so easy is it?Micropayments are also an issue which is more complex once you start looking at it in more depth with regard to actually doing it. For instance, are you just going to require everyone to use one monopolistic micropayment provider which would then have a strangehold over email and be a huge target for scams? Or how about how two systems using different micropayment providers might interact if there are a non-trivial number of micropayment service providers? You also seem to use "email provider" in the context of "ISP" or "Webmail provider", but what about entities that want to run their own SMTP server? A large corporate might shrug off any costs of doing so, but what about an SME, SOHO or even a technically inclined end user?
Even if the issue of the micropayment system is resolved and open to all from Hotmail to "Joe on DSL", how do you deal with the change over period when only a fraction of users are using the system? ESMTP has been around since 1995, yet I'm still seeing servers using the older SMTP "HELO" nearly a decade later, I think it's a given adoption would be similarly slow here. So you need to be able to handle emails between those who have adopted the system and those that have not, preferably without Joe-jobbing everyone whose email address got faked in a spam with a solicitation to get added to your whitelist...
If someone can resolve those issues (and numerous others I haven't touched on), then the email using world may well beat a path to their door with large sums of cash. Until then however, I think it's safe to say that micropayment based email is a non-starter, more is the pity.
Secondly, and this is the show-stopper at the moment, it relies on there being an effective micropayment system that can be easily integrated into SMTP, so far there isn't really a viable micropayment system, let alone one that works with SMTP. Hopefully the likes of iTunes etc. will change that, but Penny Black would need to handle several orders of magnitude more transactions than iTunes, which might pose problems. The vast majority of spammers also don't care much for the law, so the payment system would need to be proof against stolen credit card numbers, abuse of compromised PCs, faked domain names...
It's a nice idea, and I might even use it if it were to happen, but somehow I just can't see something like Penny Black ever getting off the ground.
What I'm curious about though, is that all the cases I've read about do seem to be very much canted in favour of the RIAA. The defendents are almost always financial unable to fight the case and there is also almost always clear cut copyright infringement. Is this merely media bias, or does the RIAA get to pick and choose its cases once they know who the mark is to better meet their goals of deterrence?
As I understand it, the RIAA usually files a John/Jane Doe case to subpoena the evidence needed to establish their victim's identity. Supposing that J. Doe turns out to be a very wealthy and outspoken proponent of fair use and realistic copyright laws who quite probably would be prepared to fight them in court. Issues of whether they would or not aside, could the RIAA make some excuses and "opt out" of the case at that point, or not?
She definitely understands the concept of "I've had zero after-dinner chews" and doesn't let up until she has had the two she is allowed. One won't do, nor will one and a half, one and two half biscuits is OK, but she inspects the two halves quite closely as if to make sure she isn't being conned. A more impressive display awaits in the evening when she has her choc drops; she's allowed six and once again complains if you try and put the jar away before she's had all six.
It's probably just driven by the requirements of circumstance of course, but I'm tempted to try and work some more advanced math and larger numbers into the scenario and see how well she does.
Whether that's as far ahead as dumping Microsoft and going with open souce is an entirely different matter. It'll sure be interesting to revisit this in a few years time and compare and contrast Newham with Munich though...
That's a rather sweeping and largely inaccurate statement you just made. Sure some routers are only "firewalls" by virtue of utilising NAT, but more because that's the way they have been configured, not because of a design limitation. I'd class my Draytek DSL router as a "home router", but it's definitely something I'd class as a firewall as well - as far as I can ascertain it's using BSD's PF - and NAT is optional; mine is routing real IPs very nicely.
A user with your NAT setup and packet filter (a firewall is a dedicated PC/router as far as I am concerned) that does ingres filtering only can still run a trojan. If they do, then that trojan can obtain outbound access without the user being prompted to allow it permission. If it can open an outbound TCP channel, then it can retrieve and act upon commands from an IRC channel, website or whatever, to propagate itself, spam, DDoS other hosts...
I've been playing with SP2 for a couple of betas, and initial impressions that while it is certainly an improvement, it's still a *long* way from being secure, even when configured properly. With admin priviledges and a command prompt I can disable all the new security bells and whistles in less than 10 seconds; how long is a trojan going to take? The only way forward that I can see is that "admin by default" has got to go, but that isn't likely to happen until Longhorn at the earliest.
Looks like we're not done with the incessant script kiddie probes for vulnerable MS network ports and trojan backdoors just yet. :(
Any sites that are doing more than linking to the official download sources are probably going to be getting nastygrams though; check out the second to last paragarph. There are some pretty useful links for those involved in largescale rollouts at the very bottom as well.
"Ripper". Ah, sweet irony. :)
That's not to say there isn't some other cool shit at A51 we won't be seeing at airshows for a few decades of course. :)
My understanding of the law from the KIllustrator thing was that there is indeed a client, which in that particular case was Adobe. However, there is a quirk in German trademark law that allows lawyers to initiate a case on behalf of a client without that client being aware of the fact. So, technically there is a "client", and I assume that at some point in the proceedings that they would have to be informed of the fact. Were this be applied to the meta tag instance, then the client would presumably always be the state.
Since we are talking about Germany here, I have a good idea how this is actually going to "work". Does anyone remember the whole Adobe Illustrator / KIllustrator trademark fiasco? Germany has an odd system of trademark enforcement whereby lawfirms look for trademark infringements and bring suits, telling the entity whose trademark was infringed is optional - as happened with Adobe. Naturally the real winners are the lawfirms, many of whom are bottom feeders with dubious case selection and billing practices.
You could quite easily employ the same scheme here; the lawfirms "police" the sites and bring suits against those they believe infringe. If the court decides that they are right or the defendent settles before trial, then they make money. The problem is obviously that you are going to attract the same type of ambulance chasing lawyers to the business as the trademark example, but modern legislation hardly has a good track record for being sensible, does it?
As to the noise, I've recently upgraded a couple of my systems and while I was prepping them I just left the PSU cables loose in the case. Once everything was done I tied them all up nice and neat and tucked away the unused PSU cables out of the airflow. The idling temperature of the AMD CPU dropped a full two degrees C with this simple change, but more importantly the case and CPU fan dropped by nearly 500 RPMs apiece. I know airflow matters, but not being a serious modder I had know idea it could be that much!
Well, actually, both books do touch on this. There is a religion (can't remember the name - I read the books over a year ago) that believes that the process of downloading to a cortical stack leaves the soul behind and it is therefore something to be avoided. I thought it was more like the Jehovah's Witness prohibition against receiving blood transfusions than Catholism's view of contraception though.
If Microsoft is now as serious about security as they claim then they should get this patch right *and* rescind their earlier edict against magazines carrying patches on their cover CDs. Combine that with an article that focuses more on the new features than the importance of security and you just might get the thing more widely deployed. More Windows users with fewer ports exposed to the Internet, even if it's only via ICF or whatever it's called now, has got to have an impact on the size of the BotNets out there, right?
That's because the lawyers for IBM and the rest have a knack for demonstrating that the gun is indeed smoking, but that it is pointed at SCO's foot and they pulled the trigger themselves.
A far more efficient method would be to look at using RSYNC with SSH as a few others have pointed out. There's a pretty good HOWTO for Windows here, which avoids the overhead of a full Cygwin install. You'll probably want to check out the link to cwRSYNC at the top of that page too.
Also, shut down all that extra crud that Microsoft enabled by default for the few users that might think about using it some day. You'll have more free memory (or less junk in the pagefile) and be less vulnerable to remote attacks as well. These packages might do this kind of thing for you, but most likely they are just snake-oil relying on the placebo effect and a "no-refund" clause.
So, when Foo Corp claims that by using Linux you are infringing their patents you simply remain noncommittal (just as you should never accept blame at a car accident) and call OSRM. If OSRM is as good as my car insurer that's pretty much the end of the matter as far as you are concerned. They deal with Foo Corp and their lawyers and resolve the issue as best they can, whether that be getting the case dismissed or negotiating and paying your license fees. You'll probably get a letter every now and then letting you know of any developments, requesting information they might need and so on, but that's all. For many CTOs paying a company like OSRM some money each year might just be worth the removal of one less thing to fret and loose sleep over.
I've just upgraded two of my boxes to the 3500+ and 3800+ S939 chips and couldn't be happier with the results. Both Linux (FC2 x86_64 on the 3500+) and Windows (XP on the 3800+) motor along far faster than I was expecting, and I have that dual core upgrade to fall back on when games require that much grunt. As far as I'm concerned it's "Intel Inside" alright - "Inside the store, covered in dust on the shelf". :)