True, but it could be a lazily programmed first test - i.e. if it doesn't mark itself as genuineintel it can't have SSE2/3 (whichever was just out at the time), so why bother looking for it?
a reasonable argument, except that if a user is downloading his porn from an isp's newsserver, its basically at no cost to the isp ( no upstream bandwidth is used ).
If they remove the newsserver, and the user then switches to bt/http/whatever, that *will* cost the isp upstream bandwidth costs, so from that economic point of view, its cheaper to maintain the servers.
A more likely reason is culpability and fear of lawsuits / criminal investigations for hosting illegal stuff- yes, I know safe harbour provisions, but there are arguements that could be made based on individual groups as seperate channels. I don't know if it would stand up, but I bet the ISP wouldn't want to pay their defence attourneys the cost of finding out...
as an aside, I suppose an economic argument could be made based around the usage patterns of heavy downloaders, the typical retention of ISP newsservers ( and the subsequent need for upstream fill servers ), and the relative inefficiency of nntp as a binary distribution protocol ( even with yenc ), but that is really just an argument against having a *bad* isp news server - disc space is cheap.
I fully expect to see someone at AVG go to jail, anyway
Hmmm. expecting someone corporate to be held properly accountable for misdeeds these days? Optimistic at best ( although not impossible ).
but hten again I'm a cynic. Maybe I've just been here too long.
Re:RAID5 is stupid, RAID 10 or no RAID
on
What NAS To Buy?
·
· Score: 1
I qualify my above statement.
The comparison is 'enterprise grade' (not that is much different from commercial these days apart from the warranty and cost) sata-1 (the single channel type) with sas.
I work in the storage industry and have to deal with both types daily.
I see failure rates on the sata based controllers typically 5-6x as often. I suspect this is due to the drive trying to fake having a dual controller chip, so the new sata-2 dual controller drives may be better.
oh and a number of the above controllers can take both sata and sas, and we have a mix of each for testing purposes, so I don't think its down to controller, unless there are drive-type specific defects in the controller firmware ( not that I am willing to rule it out).
I have to ask though, aren't there professionals who administrate these websites? Why aren't they dealing with this problem, as it is their job?
How? This is something entirely outside the webmasters control. Take/. (as we have some actual figures from above to play with). A 6% increase in traffic means a 6% increase in costs*, and for no gain to the site ( a loss in fact, as viewing metrics are now screwed up as well ).
I've seen two suggestions:
1. vigilantism - embed a miniture iframe with a google search for 100 pages from their site in your site's footer and hurt them back ( they have to pay someone else ). I find this is a rather distasteful approach, ethically.
2.create some form of RBL list of user ip's that are using AVG and bounce them to a page with instructions on why and how to disable it. Which is just silly, really.
*Yes, I realise that assumes a linear cost/bandwitch pricing structure which is relatively inaccurate, but fundamentally more bandwidth will cost more money.
Re:RAID5 is stupid, RAID 10 or no RAID
on
What NAS To Buy?
·
· Score: 1
Personally, I wouldn't touch RAID5 with big SATA drives with a 10 foot pole.
Personally, for a file server I wouldn't touch sata drives at all if I could help it. Ruddy unreliable pieces of metal. Sas is a much better option. Shame its so damn pricey (yes, I know its cheap compared to FC, but that is even more outside of the home users budget). I guess you get what you pay for.
Re:RAID5 is stupid, RAID 10 or no RAID
on
What NAS To Buy?
·
· Score: 1
(raid 5), mirrored. You can lose any 3 drives and it stays up.
(raid 6), mirrored lets you lose 5 drives.
to be fair, the D201GLY is also 64bit.
What is likely to be a concern is the size of the bandwidth on the backplane ( the D201GLY's is ok, but not stellar ). Still, the 945G chipset is a mainstream intel chipset, so it ought to be pretty good in that regard.
Looks like a nice box, but unfortunately, is somewhat outclassed by D945GCLF - 1/3rd of the cpu speed, lower ram capacity, higher power draw, fewer drive ports. Although the builtin dc-dc is nice, and the mini-pci makes adding wifi reasonable.
I'm guessing performance will be a tad better than that of my other nas - a nslu2, which is acceptable for samba/nfs, but not much more.
Good idea, but skip the D201GLY2A and get a D945GCLF instead - afact its the same board approach, but its the atom version so only ~2.5W and a proper intel mobo chipset instead of the via chipset on hte D201GLY2A for the same sort of money. Not that the D201 is bad ( I have one at home and its great ), but it seems the D945 should be a better choice.
Not quite the same since there is an active response from the system granting authorisation (not that I entirely like their metaphor, but hey) - for your analogy to match I suppose the wifi stealer would have had set up a static ip and route which happened to match that which was in use, except that still isn't quite right ( arp for a start)
I suppose a closer one might be if the door has a lock of some kind, but any key fits it (or a more accurate one a fingerprint reader rigged up to accept all prints).
Now I suppose it could be a defence if you could argue that you had reason to believe that the system had been setup so that some people should be let in and others kept out. In which case the fact it let you in could be construed as authorisation. maybe.
The SPU display code is not nearly as advanced as you imply (unfortunately) - whilst it can display at that rate / resolution, it is severaly limited in what data it can currently render. See http://wiki.ps2dev.org/ps3:spu-medialib here for the current status
Well, if that addon requires something that is only availiable on OSX then banning OS-dependant addons will either mean:
a. those addons will break other installs or
b. those addons will go away, denying the people who are capable of using it from doing so.
Neither is particularily good.
So I can pick between a precise but trivial definition of no use, or the more generally used definition which is simply wrong. Welcome to the world of formal logic (seriously).
The door analogy is actually quite interesting because it highlights a number of the issues we face with this sort of thing:
1. addative objects - do we consider the door and the door-with-a-chip as seperate objects (naive approach, leading to the stupidity you outlined above), or do we consider all doors doors, but then also have the fact that this one has a chip in it (better, but means our world view is now non-monotonic, a fact which breaks 95%+ of all computer reasoning).
2. heap 'paradox' - a door with a single chip in it is still structurally sound ( assuming small chips). The act of adding a small chip to a structurally sound door does not make it structurally unsound, yet eventually the door fails. How many chips are required? (the classic version is how many grains of rice do you need to make a heap?)
As for what is and isn't sanity, well lets just say I disagree with the original quote entirely (sanity is a subjective measurement, and hence not measurable against a specific belief / pattern of actions)
Sorry, but that was bad logic:
If I swing an axe at a door and put a chip in it, then I would expect that when I next swing an axe at a door I would also put a chip in that too.
However the first door is no longer a door, its a door with a chip in it.
What will happen when I swing my axe at a door with a chip in it? I don't know.
mercenaries? you expect the RIAA to have to pay to deal with this problem? No, this is an issue for the FBI,CIA and Marines. Its clearly the responsibility of the American tax-payers to fund this as just reprobation for all the copyright-infringers they have knowingly harboured.
Adobe have been attempting to remedy this with their dng raw format (Digital NeGative). Its a good idea, but I haven't noticed it getting anywhere really - I don't know of any cameras that take dng natively (although most slr's raw format can be converted to it), and equally I don't know of any programs which support dng which can't also read the other raw formats, making it only really useful if you have multiple cameras which take different raw formats and you want to store them all in the same format (not quite as crazy as it sounds, as certainly canon at least has two different raw formats in use (crw and crw2)).
indeed, but I think you will agree, its possibly not quite something for people "who are on the lookout for a new fs to entrust with keeping their precious data" [parent] - well, unless you don't intend to be doing the entrusting until a few years down the line. - it may well end up superior to ext4, but it isn't *currently*
I agree btrfs looks nice, but its somewhat behind ext4 in terms of implmentation and stability (which is saying something) - theres the small matter of not yet handling E_NOSPACE, for instance
Ok, Have now rtfa'd. Basically, they have built a shiny controller/enclosure (the enclosure is the frame that contains the drives and the controller the circuit that interfaces, although to be confusing controllers often are built into enclosures (especially on the lower end) and still referred to as a controller)
This controller is a sealed unit (read: better heat/vibration support, but not a user servicable component) with excess disks inside (multiple hot-spares, so even if several drives fail over time it keeps going), combined with the knowledge san techs across the globe know: most errors are transient, and if you pull the disk out and stick it back in, it will probably work again. They have just built a controller that does that for you automatically. Definately on the evolution rather than revolution side of things, and I have to admit I fail to see the disruption here, although I could well be missing something ( the whitepaper is somewhat light on details shall we say ).
Slashdot Mirror
True, but it could be a lazily programmed first test - i.e. if it doesn't mark itself as genuineintel it can't have SSE2/3 (whichever was just out at the time), so why bother looking for it?
Unfortunatly, intel/amd chips don't let you set cpuid
a reasonable argument, except that if a user is downloading his porn from an isp's newsserver, its basically at no cost to the isp ( no upstream bandwidth is used ).
If they remove the newsserver, and the user then switches to bt/http/whatever, that *will* cost the isp upstream bandwidth costs, so from that economic point of view, its cheaper to maintain the servers.
A more likely reason is culpability and fear of lawsuits / criminal investigations for hosting illegal stuff- yes, I know safe harbour provisions, but there are arguements that could be made based on individual groups as seperate channels. I don't know if it would stand up, but I bet the ISP wouldn't want to pay their defence attourneys the cost of finding out...
as an aside, I suppose an economic argument could be made based around the usage patterns of heavy downloaders, the typical retention of ISP newsservers ( and the subsequent need for upstream fill servers ), and the relative inefficiency of nntp as a binary distribution protocol ( even with yenc ), but that is really just an argument against having a *bad* isp news server - disc space is cheap.
I fully expect to see someone at AVG go to jail, anyway
Hmmm. expecting someone corporate to be held properly accountable for misdeeds these days? Optimistic at best ( although not impossible ).
but hten again I'm a cynic. Maybe I've just been here too long.
I qualify my above statement. The comparison is 'enterprise grade' (not that is much different from commercial these days apart from the warranty and cost) sata-1 (the single channel type) with sas.
I work in the storage industry and have to deal with both types daily.
I see failure rates on the sata based controllers typically 5-6x as often. I suspect this is due to the drive trying to fake having a dual controller chip, so the new sata-2 dual controller drives may be better.
oh and a number of the above controllers can take both sata and sas, and we have a mix of each for testing purposes, so I don't think its down to controller, unless there are drive-type specific defects in the controller firmware ( not that I am willing to rule it out).
I have to ask though, aren't there professionals who administrate these websites? Why aren't they dealing with this problem, as it is their job?
How? This is something entirely outside the webmasters control. Take /. (as we have some actual figures from above to play with). A 6% increase in traffic means a 6% increase in costs*, and for no gain to the site ( a loss in fact, as viewing metrics are now screwed up as well ).
I've seen two suggestions:
1. vigilantism - embed a miniture iframe with a google search for 100 pages from their site in your site's footer and hurt them back ( they have to pay someone else ). I find this is a rather distasteful approach, ethically.
2.create some form of RBL list of user ip's that are using AVG and bounce them to a page with instructions on why and how to disable it. Which is just silly, really.
*Yes, I realise that assumes a linear cost/bandwitch pricing structure which is relatively inaccurate, but fundamentally more bandwidth will cost more money.
Personally, I wouldn't touch RAID5 with big SATA drives with a 10 foot pole.
Personally, for a file server I wouldn't touch sata drives at all if I could help it. Ruddy unreliable pieces of metal. Sas is a much better option. Shame its so damn pricey (yes, I know its cheap compared to FC, but that is even more outside of the home users budget). I guess you get what you pay for.
(raid 5), mirrored. You can lose any 3 drives and it stays up. (raid 6), mirrored lets you lose 5 drives.
to be fair, the D201GLY is also 64bit. What is likely to be a concern is the size of the bandwidth on the backplane ( the D201GLY's is ok, but not stellar ). Still, the 945G chipset is a mainstream intel chipset, so it ought to be pretty good in that regard.
Looks like a nice box, but unfortunately, is somewhat outclassed by D945GCLF - 1/3rd of the cpu speed, lower ram capacity, higher power draw, fewer drive ports. Although the builtin dc-dc is nice, and the mini-pci makes adding wifi reasonable. I'm guessing performance will be a tad better than that of my other nas - a nslu2, which is acceptable for samba/nfs, but not much more.
Good idea, but skip the D201GLY2A and get a D945GCLF instead - afact its the same board approach, but its the atom version so only ~2.5W and a proper intel mobo chipset instead of the via chipset on hte D201GLY2A for the same sort of money. Not that the D201 is bad ( I have one at home and its great ), but it seems the D945 should be a better choice.
Not quite the same since there is an active response from the system granting authorisation (not that I entirely like their metaphor, but hey) - for your analogy to match I suppose the wifi stealer would have had set up a static ip and route which happened to match that which was in use, except that still isn't quite right ( arp for a start)
I suppose a closer one might be if the door has a lock of some kind, but any key fits it (or a more accurate one a fingerprint reader rigged up to accept all prints).
Now I suppose it could be a defence if you could argue that you had reason to believe that the system had been setup so that some people should be let in and others kept out. In which case the fact it let you in could be construed as authorisation. maybe.
Oh? where on irc/cvs?
The SPU display code is not nearly as advanced as you imply (unfortunately) - whilst it can display at that rate / resolution, it is severaly limited in what data it can currently render. See http://wiki.ps2dev.org/ps3:spu-medialib here for the current status
Well, if that addon requires something that is only availiable on OSX then banning OS-dependant addons will either mean:
a. those addons will break other installs or
b. those addons will go away, denying the people who are capable of using it from doing so.
Neither is particularily good.
fanless and draws (slightly) less power. Its really quite spiffy, actually
it didn't drop all the way, and you wouldn't expect it to, since it was pretty likely (but not certain) that MS would be back.
The door analogy is actually quite interesting because it highlights a number of the issues we face with this sort of thing:
1. addative objects - do we consider the door and the door-with-a-chip as seperate objects (naive approach, leading to the stupidity you outlined above), or do we consider all doors doors, but then also have the fact that this one has a chip in it (better, but means our world view is now non-monotonic, a fact which breaks 95%+ of all computer reasoning).
2. heap 'paradox' - a door with a single chip in it is still structurally sound ( assuming small chips). The act of adding a small chip to a structurally sound door does not make it structurally unsound, yet eventually the door fails. How many chips are required? (the classic version is how many grains of rice do you need to make a heap?)
As for what is and isn't sanity, well lets just say I disagree with the original quote entirely (sanity is a subjective measurement, and hence not measurable against a specific belief / pattern of actions)
Sorry, but that was bad logic: If I swing an axe at a door and put a chip in it, then I would expect that when I next swing an axe at a door I would also put a chip in that too. However the first door is no longer a door, its a door with a chip in it. What will happen when I swing my axe at a door with a chip in it? I don't know.
mercenaries? you expect the RIAA to have to pay to deal with this problem? No, this is an issue for the FBI,CIA and Marines. Its clearly the responsibility of the American tax-payers to fund this as just reprobation for all the copyright-infringers they have knowingly harboured.
Adobe have been attempting to remedy this with their dng raw format (Digital NeGative). Its a good idea, but I haven't noticed it getting anywhere really - I don't know of any cameras that take dng natively (although most slr's raw format can be converted to it), and equally I don't know of any programs which support dng which can't also read the other raw formats, making it only really useful if you have multiple cameras which take different raw formats and you want to store them all in the same format (not quite as crazy as it sounds, as certainly canon at least has two different raw formats in use (crw and crw2)).
indeed, but I think you will agree, its possibly not quite something for people "who are on the lookout for a new fs to entrust with keeping their precious data" [parent] - well, unless you don't intend to be doing the entrusting until a few years down the line. - it may well end up superior to ext4, but it isn't *currently*
I agree btrfs looks nice, but its somewhat behind ext4 in terms of implmentation and stability (which is saying something) - theres the small matter of not yet handling E_NOSPACE, for instance
Ok, Have now rtfa'd. Basically, they have built a shiny controller/enclosure (the enclosure is the frame that contains the drives and the controller the circuit that interfaces, although to be confusing controllers often are built into enclosures (especially on the lower end) and still referred to as a controller)
This controller is a sealed unit (read: better heat/vibration support, but not a user servicable component) with excess disks inside (multiple hot-spares, so even if several drives fail over time it keeps going), combined with the knowledge san techs across the globe know: most errors are transient, and if you pull the disk out and stick it back in, it will probably work again. They have just built a controller that does that for you automatically. Definately on the evolution rather than revolution side of things, and I have to admit I fail to see the disruption here, although I could well be missing something ( the whitepaper is somewhat light on details shall we say ).