What's missing is a protocol to determine where your ISP's borders are and sticking to peers within those borders. Essentially what you have are P2P supernodes (after a fashion) at each major ISP (Akamai already does this for normal HTTP traffic, after all) and they're the only hosts that actually establish connections outside of the ISP's network.
You could do it with BitTorrent--in a controlled situation--by serving different.torrents to users of different ISPs, and making the supernode a tracker as well as a peer on the wider network.
Realistically, though, working along these lines is something that P2P protocols are going to have to do sooner or later, and kludges like serving different.torrents wouldn't scale well. If your BT (or other P2P) client can figure out a metric for a peer, it can prioritise the traffic appropriately: peers from the same ISP as you get prioritised highest, falling back to peers from other ISPs if they don't have what you need.
None if this is rocket science, conceptually, but figuring out whether host A is from the same ISP as host B is a bit trickier. I guess you could do it by fetching the AS number from WHOIS and then caching the result across the swarms (so WHOIS servers didn't get continually beaten). Alternatively, you could add TXT records to reverse DNS, though it suffers from inflexibility, and I'm not sure ISPs would find it sufficiently in their interests to add them (they'd rather just charge you extortionate fees for excess bandwidth).
The bottom line is that ISPs don't want to support P2P. They want P2P users to go away and use other providers. Persuading them to do anything that will make life easier for P2P users--even if it saves them money--is nigh-on impossible, so solutions must be sought at a protocol level.
Nope, for two reasons: one is that we were both situated in the UK (it was a DMCA notice because LiveJournal itself is in the US, and they were hosting the image), and the other was that I really couldn't deal with the headache at the time.
I once had a weird situation where I received a notice that somebody was claiming copyright of a photo I'd taken (a self-portrait of all things), on LiveJournal, by way of a DMCA notice.
LiveJournal told me that I could file a counterclaim, and if the original claimant didn't follow it up, I was free to reinstate the photograph. I did, they didn't, so I put it back up.
"Can you guarantee that if Yahoo had refused to comply, its China-based employees would have been safe? Can you guarantee that the layoffs necessitated by Yahoo's withdrawal from the Chinese market would do less cumulative harm than the jailing that happened because they stayed and played ball?"
No, I can't, but then it was fairly a predictable consequence of them getting into China in the first place.
Was there significant shareholder pressure for Yahoo! to set up shop in China? Was it a make-or-break moment?
Guess what? It wasn't. It hasn't been for any of the big-name companies that's got itself involved in the Chinese market. Yes, it's extremely difficult to pull out now, but at least one positive aspect will result: other companies will be far more wary of joining in.
(Investors also don't look particularly fondly upon companies who fail to effectively juggle the legal responsibilities of the various markets in which it operates, because it becomes a volatile investment--even moreso when the market in question [in this case, China] is itself notoriously volatile).
So do I, until I remember that they're in China through choice.
All of these western companies set up shop in China and then say "well, we have to abide by local laws" when somebody complains about them colluding with the Chinese authorities. There's an easy solution: don't set up shop in China. You won't win anyway.
If all of the western corporations steered well clear of China (and other questionable regimes), and indeed Chinese companies, it would send a far stronger message than anything any human rights organisation would do, and shed an extremely favourable light upon the western corporations. Call it a voluntary trade sanction if you will.
As it stands, human rights laws are flouted the world over because corporations and governments get away with it. If everybody stopped doing business with the companies and regimes responsible, the world would be a slightly nicer place.
Nothing says "fuck you and your oppressive dictatorial policies" than the rest of the world refusing to take part in your GDP growth exercise: China's capital reserves wouldn't last forever, after all.
If you use strncpy(), you make sure the string has a terminator if you're going to need one.
It's really that simple. Every specification which explains strncpy() says as much.
Using strncpy() as specified is infinitely safer than using a function which blindly copies characters forever irrespective of your buffer size.
Posting five examples of "the author doesn't understand C arrays or strncpy()" isn't an argument for strncpy() being horrifically unsafe, it's an argument that for every single programming construct, there are five programmers out there who are guaranteed to fuck them up.
The worst thing is, this is first-grade C programming. If you don't understand this stuff, you need to go back and learn how arrays and strings work.
Next week: why memcpy() on overlapping buffer regions can eat your cat!
1. If the proportion of FF users is so tiny, the effect upon the bottom line is rapidly approaching zero--especially if you only consider the proportion of those users who use AdBlock.
2. Why do you think users are blocking your ads? Could it be because they are, in fact, shit?
3. Good ads spread. People like good ads. 99.9% of ads aren't, though.
4. If your target audience is blocking ads, you need to ask your ad-serving network why they're serving ads not suited to your demographic.
No no, nobody said anything about giving it out. It's still company property, yes? Held within the company? Nobody's (supposed to be) taking anything home. The company isn't giving its employees the copies, because that would be distribution. All of the copies are produced, owned and still held by the company.
A company is a single legal entity, right? (Okay, we'll ignore subsidaries and parents for the moment, let's just "yes").
If this was an individual, buying a magazine and then photocopying pages--but not distributing them--there wouldn't be a snowball in hell's chance of a case.
The parallel applies: the company wasn't redistributing articles; despite the spurious use of "distributing", "making copies available only within the company" is not distribution in the copyright sense. It created copies, and gave them to itself.
Even in the software world there's nothing legally preventing you from this (you might be prevented through license terms from using--or even installing--more than one copy simultaneously, but there's nothing stopping you from making the copy in the first place).
Had this gone to court, I'd be surprised if the outcome had gone the same way.
No, if you read the stuff he says, it's clear that he does get it, he just disagrees.
He doesn't think people should have some god-given right to roll their own kernels on Tivos, for example, and he doesn't want to be the one to dictate that right to the people producing the Tivos, either.
On the other hand, though, he's perfectly happy with other people building their own alternatives to Tivos, complete with custom Linux kernels, if that's what they want to do.
Really, though, I get the impression his focus is really on x86 servers, where the software freedom thing isn't remotely as complicated and doesn't extend past the kernel into the firmware.
In all honesty, I can't help but wonder if Linus would've been happier with the MIT license.
It's worth noting that Workgroup Manager is a handy tool to run on your own Mac, even without an Open Directory domain, as it's a bit more flexible than Accounts.prefpane, especially for (for example) configuring limited accounts for family members.
[q]I recently saw an ad for an embedded game developer.... by apple. Requiring many years of experience etc yadda yadda yadda.[/q]
Not particularly surprising: iPod games have existed since the last generation was released (not counting the iPhone), and we know from Jobs that the next-gen iPods will be more like the iPhone architecturally (running [not Mac] OS X). I'd imagine it's a whole heap cheaper to hire somebody with experience rather than license games if they want to distribute any with the device without bumping the price: given the number of iPods they shift, any royalties would be fairly immense.
...parents (and I am one, incidentally) could exercise some intelligence and keep an eye on what their kids are doing with the TV/Internet/phone/whatever.
E-mail gets wrapped--typically--by the receiving client. With format=flowed (which Outlook does support, kinda), it doesn't matter how long the lines you send are.
78 characters is pretty much solely an issue with non-flowed plain-text e-mail, and the vast majority of clients out there send flowed mail by default (because it removes the hard limit altogether).
They can't (legally) describe Leopard-on-PPC as "Unix" because it's not been certified, though I suspect Apple would avoid doing that--they'll talk about how "Leopard is Unix" without going into the specifics.
This is, of course, unless they've done a special deal with the Open Group to get trademark rights without having to do certification, but IBM/HP/Sun/Fujitsu would go nutso if they did that.
Carbon and Cocoa aren't covered by the certification. They're extensions (just as every conformant OS has: you won't find a great many Solaris APIs and userland tools in the specification, for example) which you can choose to ignore if you want to produce software that's guaranteed to compile and run across compliant systems.
SuS specifies APIs, behaviours and tools. If you care about portability, you'll be sticking to the standard APIs. If you don't care, the fact that Leopard's certified Unix03 isn't going to matter to you.
It isn't just money, though. You can't just pay your way into getting a UNIX03 cert--you actually have to be compliant with the definition.
Being certified Unix is a big deal in a lot of environments.
Linux is Linux; Linux itself can't be Unix: though a specific version of a specific distribution on a specific platform could be certified as such if somebody were willing to pony up the cash, and pay for the necessary tweaks to be made (glibc, for example, aims for SuS conformance, but I'd place money on there being something that didn't pass the testsuite, and that's without getting into all of the userland utilities).
As unpopular as it might be to mention it in a comment attached to *this* article, Apple's laptop batteries have such a date. I suspect many PC batteries do too, but getting the information out of them might be tricky.
My MacBook's, apparently, is 13 months old and holds 93% of its original charge (4886mAh as opposed to 5000mAh). I use my MacBook every day--pretty much without fail--and in that 13 months it's had 126 full charge cycles.
Excepting faults, Apple tends to utilise pretty good battery tech.
Yeah, nothing stopping you. Just drag them into the right bit of your iTunes library and let it sync with the iPhone. Same with any other MP3/AAC files from anywhere else.
Slashdot Mirror
...he's read Slashdot, now, then?
What's missing is a protocol to determine where your ISP's borders are and sticking to peers within those borders. Essentially what you have are P2P supernodes (after a fashion) at each major ISP (Akamai already does this for normal HTTP traffic, after all) and they're the only hosts that actually establish connections outside of the ISP's network.
.torrents to users of different ISPs, and making the supernode a tracker as well as a peer on the wider network.
.torrents wouldn't scale well. If your BT (or other P2P) client can figure out a metric for a peer, it can prioritise the traffic appropriately: peers from the same ISP as you get prioritised highest, falling back to peers from other ISPs if they don't have what you need.
You could do it with BitTorrent--in a controlled situation--by serving different
Realistically, though, working along these lines is something that P2P protocols are going to have to do sooner or later, and kludges like serving different
None if this is rocket science, conceptually, but figuring out whether host A is from the same ISP as host B is a bit trickier. I guess you could do it by fetching the AS number from WHOIS and then caching the result across the swarms (so WHOIS servers didn't get continually beaten). Alternatively, you could add TXT records to reverse DNS, though it suffers from inflexibility, and I'm not sure ISPs would find it sufficiently in their interests to add them (they'd rather just charge you extortionate fees for excess bandwidth).
The bottom line is that ISPs don't want to support P2P. They want P2P users to go away and use other providers. Persuading them to do anything that will make life easier for P2P users--even if it saves them money--is nigh-on impossible, so solutions must be sought at a protocol level.
Nope, for two reasons: one is that we were both situated in the UK (it was a DMCA notice because LiveJournal itself is in the US, and they were hosting the image), and the other was that I really couldn't deal with the headache at the time.
I once had a weird situation where I received a notice that somebody was claiming copyright of a photo I'd taken (a self-portrait of all things), on LiveJournal, by way of a DMCA notice.
LiveJournal told me that I could file a counterclaim, and if the original claimant didn't follow it up, I was free to reinstate the photograph. I did, they didn't, so I put it back up.
"Can you guarantee that if Yahoo had refused to comply, its China-based employees would have been safe? Can you guarantee that the layoffs necessitated by Yahoo's withdrawal from the Chinese market would do less cumulative harm than the jailing that happened because they stayed and played ball?"
No, I can't, but then it was fairly a predictable consequence of them getting into China in the first place.
Was there significant shareholder pressure for Yahoo! to set up shop in China? Was it a make-or-break moment?
Guess what? It wasn't. It hasn't been for any of the big-name companies that's got itself involved in the Chinese market. Yes, it's extremely difficult to pull out now, but at least one positive aspect will result: other companies will be far more wary of joining in.
(Investors also don't look particularly fondly upon companies who fail to effectively juggle the legal responsibilities of the various markets in which it operates, because it becomes a volatile investment--even moreso when the market in question [in this case, China] is itself notoriously volatile).
I actually feel bad for Yahoo in a way.
So do I, until I remember that they're in China through choice.
All of these western companies set up shop in China and then say "well, we have to abide by local laws" when somebody complains about them colluding with the Chinese authorities. There's an easy solution: don't set up shop in China. You won't win anyway.
If all of the western corporations steered well clear of China (and other questionable regimes), and indeed Chinese companies, it would send a far stronger message than anything any human rights organisation would do, and shed an extremely favourable light upon the western corporations. Call it a voluntary trade sanction if you will.
As it stands, human rights laws are flouted the world over because corporations and governments get away with it. If everybody stopped doing business with the companies and regimes responsible, the world would be a slightly nicer place.
Nothing says "fuck you and your oppressive dictatorial policies" than the rest of the world refusing to take part in your GDP growth exercise: China's capital reserves wouldn't last forever, after all.
And this got moderated "Troll"!?
If you use strncpy(), you make sure the string has a terminator if you're going to need one.
It's really that simple. Every specification which explains strncpy() says as much.
Using strncpy() as specified is infinitely safer than using a function which blindly copies characters forever irrespective of your buffer size.
Posting five examples of "the author doesn't understand C arrays or strncpy()" isn't an argument for strncpy() being horrifically unsafe, it's an argument that for every single programming construct, there are five programmers out there who are guaranteed to fuck them up.
The worst thing is, this is first-grade C programming. If you don't understand this stuff, you need to go back and learn how arrays and strings work.
Next week: why memcpy() on overlapping buffer regions can eat your cat!
1. If the proportion of FF users is so tiny, the effect upon the bottom line is rapidly approaching zero--especially if you only consider the proportion of those users who use AdBlock.
2. Why do you think users are blocking your ads? Could it be because they are, in fact, shit?
3. Good ads spread. People like good ads. 99.9% of ads aren't, though.
4. If your target audience is blocking ads, you need to ask your ad-serving network why they're serving ads not suited to your demographic.
It's all a troll, though, really.
No no, nobody said anything about giving it out. It's still company property, yes? Held within the company? Nobody's (supposed to be) taking anything home. The company isn't giving its employees the copies, because that would be distribution. All of the copies are produced, owned and still held by the company.
I'm not sure how this could be illegal.
A company is a single legal entity, right? (Okay, we'll ignore subsidaries and parents for the moment, let's just "yes").
If this was an individual, buying a magazine and then photocopying pages--but not distributing them--there wouldn't be a snowball in hell's chance of a case.
The parallel applies: the company wasn't redistributing articles; despite the spurious use of "distributing", "making copies available only within the company" is not distribution in the copyright sense. It created copies, and gave them to itself.
Even in the software world there's nothing legally preventing you from this (you might be prevented through license terms from using--or even installing--more than one copy simultaneously, but there's nothing stopping you from making the copy in the first place).
Had this gone to court, I'd be surprised if the outcome had gone the same way.
No, if you read the stuff he says, it's clear that he does get it, he just disagrees.
He doesn't think people should have some god-given right to roll their own kernels on Tivos, for example, and he doesn't want to be the one to dictate that right to the people producing the Tivos, either.
On the other hand, though, he's perfectly happy with other people building their own alternatives to Tivos, complete with custom Linux kernels, if that's what they want to do.
Really, though, I get the impression his focus is really on x86 servers, where the software freedom thing isn't remotely as complicated and doesn't extend past the kernel into the firmware.
In all honesty, I can't help but wonder if Linus would've been happier with the MIT license.
...but WTF does Oracle have to do with *Novell* releasing the source to YaST?
I've skim-RTFA (the one related to YaST), but nothing leapt out as being anything whatsoever to do with Oracle.
It's worth noting that Workgroup Manager is a handy tool to run on your own Mac, even without an Open Directory domain, as it's a bit more flexible than Accounts.prefpane, especially for (for example) configuring limited accounts for family members.
n tools104.html
It's part in the Server Admin Tools: http://www.apple.com/support/downloads/serveradmi
I don't know if the license/installer says you have to have a Mac OS X Server installation to use them, because I haven't looked.
[q]I recently saw an ad for an embedded game developer.... by apple. Requiring many years of experience etc yadda yadda yadda.[/q]
Not particularly surprising: iPod games have existed since the last generation was released (not counting the iPhone), and we know from Jobs that the next-gen iPods will be more like the iPhone architecturally (running [not Mac] OS X). I'd imagine it's a whole heap cheaper to hire somebody with experience rather than license games if they want to distribute any with the device without bumping the price: given the number of iPods they shift, any royalties would be fairly immense.
...parents (and I am one, incidentally) could exercise some intelligence and keep an eye on what their kids are doing with the TV/Internet/phone/whatever.
E-mail gets wrapped--typically--by the receiving client. With format=flowed (which Outlook does support, kinda), it doesn't matter how long the lines you send are.
78 characters is pretty much solely an issue with non-flowed plain-text e-mail, and the vast majority of clients out there send flowed mail by default (because it removes the hard limit altogether).
No; one of the BSDs or Linux distributions (on a specific hardware platform) could be certified.
It's probably easier if you've got SysV roots, but not essential.
Mac OS X's userland borrowed heavily from BSD--I could be wrong but I'm pretty sure there's no SysV in there at all.
Technically? No.
They can't (legally) describe Leopard-on-PPC as "Unix" because it's not been certified, though I suspect Apple would avoid doing that--they'll talk about how "Leopard is Unix" without going into the specifics.
This is, of course, unless they've done a special deal with the Open Group to get trademark rights without having to do certification, but IBM/HP/Sun/Fujitsu would go nutso if they did that.
Er, whut?
Carbon and Cocoa aren't covered by the certification. They're extensions (just as every conformant OS has: you won't find a great many Solaris APIs and userland tools in the specification, for example) which you can choose to ignore if you want to produce software that's guaranteed to compile and run across compliant systems.
SuS specifies APIs, behaviours and tools. If you care about portability, you'll be sticking to the standard APIs. If you don't care, the fact that Leopard's certified Unix03 isn't going to matter to you.
It isn't just money, though. You can't just pay your way into getting a UNIX03 cert--you actually have to be compliant with the definition.
Being certified Unix is a big deal in a lot of environments.
Linux is Linux; Linux itself can't be Unix: though a specific version of a specific distribution on a specific platform could be certified as such if somebody were willing to pony up the cash, and pay for the necessary tweaks to be made (glibc, for example, aims for SuS conformance, but I'd place money on there being something that didn't pass the testsuite, and that's without getting into all of the userland utilities).
That very much depends on where in the world the poster might live--in the UK, OTA HD is still very much in its infancy.
As unpopular as it might be to mention it in a comment attached to *this* article, Apple's laptop batteries have such a date. I suspect many PC batteries do too, but getting the information out of them might be tricky.
My MacBook's, apparently, is 13 months old and holds 93% of its original charge (4886mAh as opposed to 5000mAh). I use my MacBook every day--pretty much without fail--and in that 13 months it's had 126 full charge cycles.
Excepting faults, Apple tends to utilise pretty good battery tech.
Yeah, nothing stopping you. Just drag them into the right bit of your iTunes library and let it sync with the iPhone. Same with any other MP3/AAC files from anywhere else.
OOXML is crap and ODF works. That's important and I didn't know it.
That pretty much sums up the whole problem. Nobody who actually makes the decisions is aware of it.
In actual fact, ODF does have its fair share of problems, but it's vendor-neutral. The OOXML spec is unimplementable unless you're Microsoft.