Maybe you've got a more basic or an older unit or something but AFAIK most half-decent modern in-car GPS systems support bluetooth phone coupling and can play MP3s from an SD card. Only thing that might be missing is the radio.
Actually, it turns out that these things (automotive entertainment systems0 actually have to be extremely reliable. Windows Embedded for Automotive has to be way more robust than regular Windows Embedded, for example. The pressure comes from the car manufacturers themselves, not the public.
We wish... I have spent many happy hours in QANTAS business class watching the Windows CE based in-flight entertainment system rebooting, and rebooting, and rebooting - You got to see a Windows CE error screen for minutes at a time. The last occasion it it happened to us, they gave us refunds/gifts to the value of $700 as the system was out all the way from Singapore to Frankfurt. One of the nice cabin crew told me that it happened regularly, and that the experience had put her off Windows - She had just bought an Apple Mac. Anecdotal, but still frightening.
I saw something similar on a Virgin flight - only their in-flight entertainment system was Linux based. The kernel kept on booting, failing to start anything useful, giving up and rebooting.
From what I could tell, it looked like there was some sort of corruption to the root fs and the designers had failed to account for the possibility that power to non-essential systems (eg. entertainment) might be cut at awkward moments. Which we all know NEVER happens on an aircraft.
Bottom line: I don't care whether it's Windows Embedded, Linux or what, if the system isn't designed and implemented properly it's going to screw up.
There is probably some holes in my reasoning, but I am sure smarter souls will be more than happy to correct me.
This has been the case for over 10 or 15 years at least.
UK prices have almost always been set as $1=£1. Historically, this affected hardware as well as software, but by the time you account for taxes that's not really the case any more. Software, however, has actually become more expensive - and if you're a business (and therefore not too keen on getting bent over by the BSA), you'll note that the EULA of most software packages clearly states "only for distribution within ${AREA}." Hence even if you're a US company with a UK subsidiary, the US parent company can't take out a site license for (say) Microsoft products across its entire business and all subsidiaries and take advantage of US prices.
Like let's say a new movie like "Sex and the City" is out, and you're half interested for whatever reason, but you would never pay $9 for a ticket to see it. Obviously, nobody sees that movie for the graphics, right?
Actually, I'm not sure what anyone would watch it for. It's certainly not the high-quality acting, the fantastic script or the deep intellectual plot.
This isn't a smart-assed comment, but what does this version do?
I've got an enclosure of 10 PowerEdge 1955s that I have ~ 6months to play with until I need to make them production servers. I'm sorely tempted to use this, but I'm unfamiliar with the ESX product line. What does this ESXi do for me?
Not sure I follow you.
Virtualisation is very well covered in Wikipedia and I won't waste time explaining it again now.
This offers a few features which are absent from VMWare Server:
1. Runs directly on bare metal. So you have to dedicate less disk space to a full-blown OS. 2. Should perform better. 3. Easy upgrade path to the paid version. The paid version is where things get really interesting - for instance, you can set up high-availability on a per-VM basis, effectively bringing HA to applications which don't natively have any sort of HA support.
However, not all audio players support it. I'm pretty sure the iPod doesn't, nor does iTunes. (For some reason iTunes does have a "normalise levels on all selected tunes" option but that works by decoding/re-encoding the audio, which is a lot slower because in addition to the audio analysis you have to re-encode the file and is likely to introduce further interference to the stream).
Having said that, I've only got a fourth gen ipod. For all I know, more recent models do make use of this tag and furthermore, for all I know if iTunes knows that it's being synced with an ipod which does support the tag then that's what it uses to adjust the gain.
They lie about their product, they should be taken out and shot in the head.
Fuck them and the horse they rode on.
This may be true, but the person answering the email is a human being and is more likely to pass on a request to the relevant people if it's worded as "I think I've found a bug in your implementation - yes I know it will work in Windows, it'll work in Linux too if you make these minor changes" than if it's worded in the fashion the poster in the Ubuntu forums did.
Ultimately, there ought to be some law requiring that companies that claim to support a particular spec or standard actually fulfill that obligation or at least have a good faith effort to implement it.
This is what logo programmes are meant to achieve. The USB 2 logo has a specific meaning and while lots of kit claims "USB 2 compatible", the manufacturer can't display the logo without licensing it and one of the conditions of licensing is "your product passes a whole bunch of tests".
I don't think there is a logo programme for ACPI.http://linux.slashdot.org/article.pl?sid=08/07/25/1150218#
The groundwork's already in place for TPM to become pervasive - Vista supports it, virtually every new PC leaves the factory with such a chip. The interesting thing will be to see if anyone ever starts demanding it.
The nightmare scenario (which is almost always unrealistic) is that you won't even be able to connect to the Internet. I don't think this is likely, but I could certainly see a few banks demanding your PC is trusted - if not now then after the first person in government gets their account raided because their PC was inundated with spyware and proposes legislation to force financial institutions to verify that the customer's PC is secure.
Had you asked me 18 months ago, I could also see the RIAA demanding that any online music store require it as a condition of use as soon as the technology's sufficiently mature - but with Amazon's MP3 download store I'm now not so sure.
Let me put it like this - I think things could go either way. But I'm not above pointing out that if things go down the "TPM everywhere" route then the approach "meh. it'll be easy to crack" is a tad optimistic.
So you're going to preempt the people who were going to correct the parent by spreading mis-information in advance?
The TPM doesn't stop you from using your computer as a general purpose system. It simply allows third party software to choose not to work if certain conditions aren't met. You can still do "whatever you want" with your computer, as long as "whatever you want" doesn't include running such software.
There are extensions already implemented by Cisco to deny routing to anything that can't certify itself as trusted. So it can also stop you from communicating with third parties.
The entire system can verify itself as being trusted, and operating "untrusted" software can alter how the system verifies itself (and hence identifies itself) as being trusted. Third party software can quite happily refuse to communicate with you if your computer can't verify itself as being fully trusted.
It's a fantastic solution for businesses - I can see lots of businesses liking the idea of being able to guarantee that their staff are only able to run approved software and verify this remotely - but I don't fancy the idea of my bank dictating that I use something that they deem "trusted".
So open hardware will become more prevalent. There's obviously demand for unrestricted hardware, so somebody will make it.
Ahem:
Or, if you do, you'll create as many problems as you solve because online banking, shopping and even Internet access can and quite possibly will demand that your 'computer' prove it's fully "Trusted" before they have anything to do with it."
Nothing closed about TPM apart from the encryption keys themselves. You just can't claim to be "Trusted" without it.
Its not fraud to close a branch of a company. Sure its annoying, but its perfectly legal.
This is something which the law as it stands is not able to deal with.
There's plenty of laws in most countries regarding selling tangible products which protect the buyer at least until such time as they receive the product, and generally for a few months (even a few years) afterwards. Granted, some of these laws are unenforceable if the company goes bust but the only people who'll be affected are those who have paid but not yet received their items. I can't imagine they'd be unenforceable if a parent company said "Right, we're closing this subsidiary down, anyone who's placed and paid for an order but not received it is SOL". Generally they'd honour outstanding orders and stop accepting new ones.
There are also laws regarding selling services which protect the customer. Again, they're not much use if the supplier goes bust but if a parent company said "Right, we're closing this subsidiary and anyone who's got a service contract is SOL" would surely be sued to kingdom come.
But goods (tangible or otherwise) which can lose some or all of their functionality on the whim of the supplier long after they were purchased? Hmmm.... IANAL but I can't think of anything legally wrong. Ethically, yes it stinks.
This reminds me of a news article a year or two back where some fancy computer controlled car park was remotely disabled locking all the cars in because the supplier hadn't been paid. What next? The computer controlling the car park phones home every week and if that process fails for three weeks running it locks everything in?
/. - Don't be in such a hurry to mod this "Funny". If the so called "Trusted Computing Initiative" goes through as planned, then indeed your Linux distro may well turn out to be illegal, especially if you have added or removed stuff and recompiled. In these cases, it will not be "approved" software as the hash will have changed.
Just to pre-empt the inevitable shower of folk who have neither glanced at nor fully understood the implications of Trusted Computing saying "It's my computer, how will they stop me?"
In answer to those people, "Very simple. Your computer will no longer be a general purpose computer, it'll be a device like your Tivo or your DVD player. And, like your Tivo, it'll be more or less impossible to change the software on. Or, if you do, you'll create as many problems as you solve because online banking, shopping and even Internet access can and quite possibly will demand that your 'computer' prove it's fully "Trusted" before they have anything to do with it."
The technology has all been thought through very carefully and virtually every counter-argument (particularly the "it'll never work" arguments) has been dealt with in hardware. AFAICT, the only way you'll break it wholesale is by infiltrating a chip fab and maintaining the breakage for so long that it's not practical for the manufacturer to revoke all the compromised keys.
That's why you run an ACID database and not a toy database if you care about your data. If the database says "yes, it's committed" then the data damnwell better be committed (unless your hardware or OS is lying about when data hits non-volatile storage, in which case there's nothing the database can do).
The ACID aspect guarantees that you won't wind up rebooting after a crash to discover that a transaction is half-committed and the numbers in table A don't correspond with the numbers in table B. One obvious way to implement that is that you mark the start and end of each transaction directly to the filesystem as you're going along, and if you come back from a reboot to find a bunch of transactions with start markers but no end markers, you know they didn't commit so you roll them back. With a large database, the commit process could involve updating lots of tables in lots of different files so it's not possible to do them with one OS call.
Furthermore, sync is an operating system command which will have the operating system process any outstanding filesystem synchronisation tasks. It doesn't contact the database and say "hey, are you through with committing any outstanding transactions?"
legions of poor, music hungry teenagers: 3 hired guns of the music industry: 0
you're dying music industry. please just get dead already please
Speaking purely theoretically here, where are the music hungry teenagers going to find out about the next Britney Fucking Spears (seriously, "Fucking"'s her middle name) from if the music industry does get dead already?
Not that I'm particularly enamoured with the music industry as it stands, but someone is going to find a way of bringing musicians and fans together and that someone is going to want paying somehow.
Filtering/inspecting traffic implies taking responsibility implies getting lawsuits directed at ISPs for users' content.
That's exactly what I was thinking. Doesn't this strip them of their "safe harbor" status?
Firstly, you could at least have read the headline. They're UK ISPs, "safe harbor" (or even "harbour") has no meaning in the UK.
Secondly, and something which wasn't mentioned here, the UK government has discovered something which has proven remarkably effective in getting things done quickly and efficiently without having to mess around with passing legislation. They tell an industry "There's nothing illegal about what you're doing right now but we don't like it and if you don't stop doing it we'll make it illegal".
I suspect the ISPs would rather be seen to do something now than risk getting fined by the government.
That will sync the disks, but it won't stop the database from accepting incoming data. It won't stop cron jobs which might be just about to trigger. It won't deal with tasks that are in the middle of a big operation which involves a lot of writing to disk.
How about this?
killall5 -19; sync
killall5 sends a signal to all running processes other than the current session's. Signal 19 is a SIGSTOP, it immediately freezes a process and cannot be caught or ignored.
Doesn't work.
Postgres, for instance (and I doubt it's the only program to which this applies - I suspect the same is true of most databases) doesn't guarantee that its own data files are consistent unless it's been cleanly shut down. While you will flush any writes which are in progress and prevent any further writes, you won't guarantee that the end result will contain consistent data.
Any sensible database will have a mechanism to recover from this, but that mechanism may result in you losing the last transaction which wasn't fully committed. All the atomicity in ACID means is "you won't start up the database after a crash and find your data doesn't make any sense because only half a transaction was committed".
It doesn't mean "As soon as you hit enter a transaction is guaranteed to commit successfully even if the power fails during the commit process".
Why 5 minutes? It usually takes less than a second to run a sync on the disks depending on how active they are. A couple seconds of runtime should be enough to do an "emergency shutdown" and avoid data corruption.
####@johncash:~$ time sync
real 0m0.004s user 0m0.004s sys 0m0.000s
That will sync the disks, but it won't stop the database from accepting incoming data. It won't stop cron jobs which might be just about to trigger. It won't deal with tasks that are in the middle of a big operation which involves a lot of writing to disk.
The case hinges on No Service Password Recovery commands Childs allegedly configured onto several Cisco devices, as well as dial-up and DSL modems the SFPD has discovered that would allow unauthorized connections to the FiberWAN.
Mr. Paranoid Admin with a God complex had big freakin' huge vulnerabilities on his precious network?
It sounds to me like Mr. Paranoid Admin was so paranoid that people had started to do what they tend to do when Mr. Paranoid Admin is so paranoid they can't get anything they need done.
They'd started to work around him.
Net result: All sorts of little unauthorised connections popping up.
In being too paranoid, you wind up creating exactly the situation you fear the most: a network with lots of uncontrolled, unknown systems appearing creating security holes where none previously existed. Doesn't matter how many fancy "no unauthorised access" features your infrastructure has, sooner or later someone's going to succeed in working around them. The last thing you need to do is give them an incentive.
"It's too risky for anybody to translate that [The Bible] into other languages. Mistakes can creep in... and that can lead to heresy. True Christians should only read English."
"If your original Hebrew disagrees with my original King James --- your original Hebrew is wrong. If your original Hebrew agrees with my original King James, your original Hebrew is right."
Like I said, I think you are confusing the software assurance program with the open and site licensing programs. The software assurance used to be separate but now it comes with the site licenses and open licenses except they don't apply to some OEM software.. You can however, get the software assurance separately. Or at least you could in the past.
The software assurance is a program that does what you mention, it give an upgrade path to ensure that your software purchases aren't outdated during the useful life to you. A site and open licenses cover any use of specified products company wide (and beyond in some cases) They cover the OS, server products, CALs, office, and any other product specified in the agreements. You build a configuration and each product lists points and after certain levels, you get a certain discount for that software. But if you have a site license for 15 OSes and 15 office programs, it doesn't matter which computers they are on as long as the company owns or controls them.
I know what the various bits of licensing are; the PDF states that operating systems are available through the volume licensing schemes only as upgrades (in other words: it's equivalent to the "Windows XP Upgrade Edition" box you can get from a retail store which is cheaper than the "Full version" box but you're only allowed to use on a system which already has an older version of Windows on).
Software Assurance is just an add-on to the license scheme that allows you to upgrade from the version you originally licensed if/when a new version is released. Sucks to be you if they don't release an upgrade during the course of the license agreement.
This isn't just my understanding; I've confirmed it with account managers from a number of Microsoft resellers in the UK on a couple of occasions.
I also remember reading a story once about a guy who makes giant works of art, using Legos like pixels. I believe they said that if you want to buy like 10,000 blue bricks, you can get bulk prices straight from Lego.
If you want to buy 10,000 of most things you can get a bulk price straight from the manufacturer.
Seriously, though, you make a good point. If they wanted to be evil, I guess they could make them less durable, but I think in the end it's better that everyone knows them as the best.
There is also one thing to bear in mind - as soon as a company like Lego races for the bottom (in terms of price/quality), they're going to lose the customers who valued the quality and then be beaten to the customers who don't by other organisations who 10 or 20 years ago perfected the art of "juuuuusst enough quality that it won't fall apart within 15 minutes of the box opening. Probably."
Slashdot Mirror
Maybe you've got a more basic or an older unit or something but AFAIK most half-decent modern in-car GPS systems support bluetooth phone coupling and can play MP3s from an SD card. Only thing that might be missing is the radio.
Actually, it turns out that these things (automotive entertainment systems0 actually have to be extremely reliable. Windows Embedded for Automotive has to be way more robust than regular Windows Embedded, for example. The pressure comes from the car manufacturers themselves, not the public.
We wish... I have spent many happy hours in QANTAS business class watching the Windows CE based in-flight entertainment system rebooting, and rebooting, and rebooting - You got to see a Windows CE error screen for minutes at a time. The last occasion it it happened to us, they gave us refunds/gifts to the value of $700 as the system was out all the way from Singapore to Frankfurt. One of the nice cabin crew told me that it happened regularly, and that the experience had put her off Windows - She had just bought an Apple Mac. Anecdotal, but still frightening.
I saw something similar on a Virgin flight - only their in-flight entertainment system was Linux based. The kernel kept on booting, failing to start anything useful, giving up and rebooting.
From what I could tell, it looked like there was some sort of corruption to the root fs and the designers had failed to account for the possibility that power to non-essential systems (eg. entertainment) might be cut at awkward moments. Which we all know NEVER happens on an aircraft.
Bottom line: I don't care whether it's Windows Embedded, Linux or what, if the system isn't designed and implemented properly it's going to screw up.
There is probably some holes in my reasoning, but I am sure smarter souls will be more than happy to correct me.
This has been the case for over 10 or 15 years at least.
UK prices have almost always been set as $1=£1. Historically, this affected hardware as well as software, but by the time you account for taxes that's not really the case any more. Software, however, has actually become more expensive - and if you're a business (and therefore not too keen on getting bent over by the BSA), you'll note that the EULA of most software packages clearly states "only for distribution within ${AREA}." Hence even if you're a US company with a UK subsidiary, the US parent company can't take out a site license for (say) Microsoft products across its entire business and all subsidiaries and take advantage of US prices.
Like let's say a new movie like "Sex and the City" is out, and you're half interested for whatever reason, but you would never pay $9 for a ticket to see it. Obviously, nobody sees that movie for the graphics, right?
Actually, I'm not sure what anyone would watch it for. It's certainly not the high-quality acting, the fantastic script or the deep intellectual plot.
This isn't a smart-assed comment, but what does this version do?
I've got an enclosure of 10 PowerEdge 1955s that I have ~ 6months to play with until I need to make them production servers. I'm sorely tempted to use this, but I'm unfamiliar with the ESX product line. What does this ESXi do for me?
Not sure I follow you.
Virtualisation is very well covered in Wikipedia and I won't waste time explaining it again now.
This offers a few features which are absent from VMWare Server:
1. Runs directly on bare metal. So you have to dedicate less disk space to a full-blown OS.
2. Should perform better.
3. Easy upgrade path to the paid version. The paid version is where things get really interesting - for instance, you can set up high-availability on a per-VM basis, effectively bringing HA to applications which don't natively have any sort of HA support.
There is, in MP3 at least. It's used by mp3gain:
http://mp3gain.sourceforge.net/faq.php
However, not all audio players support it. I'm pretty sure the iPod doesn't, nor does iTunes. (For some reason iTunes does have a "normalise levels on all selected tunes" option but that works by decoding/re-encoding the audio, which is a lot slower because in addition to the audio analysis you have to re-encode the file and is likely to introduce further interference to the stream).
Having said that, I've only got a fourth gen ipod. For all I know, more recent models do make use of this tag and furthermore, for all I know if iTunes knows that it's being synced with an ipod which does support the tag then that's what it uses to adjust the gain.
http://slashdot.org/~Zaphod+Beeblebrox/
They lie about their product, they should be taken out and shot in the head.
Fuck them and the horse they rode on.
This may be true, but the person answering the email is a human being and is more likely to pass on a request to the relevant people if it's worded as "I think I've found a bug in your implementation - yes I know it will work in Windows, it'll work in Linux too if you make these minor changes" than if it's worded in the fashion the poster in the Ubuntu forums did.
Ultimately, there ought to be some law requiring that companies that claim to support a particular spec or standard actually fulfill that obligation or at least have a good faith effort to implement it.
This is what logo programmes are meant to achieve. The USB 2 logo has a specific meaning and while lots of kit claims "USB 2 compatible", the manufacturer can't display the logo without licensing it and one of the conditions of licensing is "your product passes a whole bunch of tests".
I don't think there is a logo programme for ACPI.http://linux.slashdot.org/article.pl?sid=08/07/25/1150218#
No offence taken ;)
The groundwork's already in place for TPM to become pervasive - Vista supports it, virtually every new PC leaves the factory with such a chip. The interesting thing will be to see if anyone ever starts demanding it.
The nightmare scenario (which is almost always unrealistic) is that you won't even be able to connect to the Internet. I don't think this is likely, but I could certainly see a few banks demanding your PC is trusted - if not now then after the first person in government gets their account raided because their PC was inundated with spyware and proposes legislation to force financial institutions to verify that the customer's PC is secure.
Had you asked me 18 months ago, I could also see the RIAA demanding that any online music store require it as a condition of use as soon as the technology's sufficiently mature - but with Amazon's MP3 download store I'm now not so sure.
Let me put it like this - I think things could go either way. But I'm not above pointing out that if things go down the "TPM everywhere" route then the approach "meh. it'll be easy to crack" is a tad optimistic.
So you're going to preempt the people who were going to correct the parent by spreading mis-information in advance?
The TPM doesn't stop you from using your computer as a general purpose system. It simply allows third party software to choose not to work if certain conditions aren't met. You can still do "whatever you want" with your computer, as long as "whatever you want" doesn't include running such software.
There are extensions already implemented by Cisco to deny routing to anything that can't certify itself as trusted. So it can also stop you from communicating with third parties.
The entire system can verify itself as being trusted, and operating "untrusted" software can alter how the system verifies itself (and hence identifies itself) as being trusted. Third party software can quite happily refuse to communicate with you if your computer can't verify itself as being fully trusted.
It's a fantastic solution for businesses - I can see lots of businesses liking the idea of being able to guarantee that their staff are only able to run approved software and verify this remotely - but I don't fancy the idea of my bank dictating that I use something that they deem "trusted".
So open hardware will become more prevalent. There's obviously demand for unrestricted hardware, so somebody will make it.
Ahem:
Or, if you do, you'll create as many problems as you solve because online banking, shopping and even Internet access can and quite possibly will demand that your 'computer' prove it's fully "Trusted" before they have anything to do with it."
Nothing closed about TPM apart from the encryption keys themselves. You just can't claim to be "Trusted" without it.
Its not fraud to close a branch of a company.
Sure its annoying, but its perfectly legal.
This is something which the law as it stands is not able to deal with.
There's plenty of laws in most countries regarding selling tangible products which protect the buyer at least until such time as they receive the product, and generally for a few months (even a few years) afterwards. Granted, some of these laws are unenforceable if the company goes bust but the only people who'll be affected are those who have paid but not yet received their items. I can't imagine they'd be unenforceable if a parent company said "Right, we're closing this subsidiary down, anyone who's placed and paid for an order but not received it is SOL". Generally they'd honour outstanding orders and stop accepting new ones.
There are also laws regarding selling services which protect the customer. Again, they're not much use if the supplier goes bust but if a parent company said "Right, we're closing this subsidiary and anyone who's got a service contract is SOL" would surely be sued to kingdom come.
But goods (tangible or otherwise) which can lose some or all of their functionality on the whim of the supplier long after they were purchased? Hmmm.... IANAL but I can't think of anything legally wrong. Ethically, yes it stinks.
This reminds me of a news article a year or two back where some fancy computer controlled car park was remotely disabled locking all the cars in because the supplier hadn't been paid. What next? The computer controlling the car park phones home every week and if that process fails for three weeks running it locks everything in?
/. - Don't be in such a hurry to mod this "Funny". If the so called "Trusted Computing Initiative" goes through as planned, then indeed your Linux distro may well turn out to be illegal, especially if you have added or removed stuff and recompiled. In these cases, it will not be "approved" software as the hash will have changed.
Just to pre-empt the inevitable shower of folk who have neither glanced at nor fully understood the implications of Trusted Computing saying "It's my computer, how will they stop me?"
In answer to those people, "Very simple. Your computer will no longer be a general purpose computer, it'll be a device like your Tivo or your DVD player. And, like your Tivo, it'll be more or less impossible to change the software on. Or, if you do, you'll create as many problems as you solve because online banking, shopping and even Internet access can and quite possibly will demand that your 'computer' prove it's fully "Trusted" before they have anything to do with it."
The technology has all been thought through very carefully and virtually every counter-argument (particularly the "it'll never work" arguments) has been dealt with in hardware. AFAICT, the only way you'll break it wholesale is by infiltrating a chip fab and maintaining the breakage for so long that it's not practical for the manufacturer to revoke all the compromised keys.
That's why you run an ACID database and not a toy database if you care about your data. If the database says "yes, it's committed" then the data damnwell better be committed (unless your hardware or OS is lying about when data hits non-volatile storage, in which case there's nothing the database can do).
The ACID aspect guarantees that you won't wind up rebooting after a crash to discover that a transaction is half-committed and the numbers in table A don't correspond with the numbers in table B. One obvious way to implement that is that you mark the start and end of each transaction directly to the filesystem as you're going along, and if you come back from a reboot to find a bunch of transactions with start markers but no end markers, you know they didn't commit so you roll them back. With a large database, the commit process could involve updating lots of tables in lots of different files so it's not possible to do them with one OS call.
Furthermore, sync is an operating system command which will have the operating system process any outstanding filesystem synchronisation tasks. It doesn't contact the database and say "hey, are you through with committing any outstanding transactions?"
you lose. you just don't know it yet
legions of poor, music hungry teenagers: 3
hired guns of the music industry: 0
you're dying music industry. please just get dead already please
Speaking purely theoretically here, where are the music hungry teenagers going to find out about the next Britney Fucking Spears (seriously, "Fucking"'s her middle name) from if the music industry does get dead already?
Not that I'm particularly enamoured with the music industry as it stands, but someone is going to find a way of bringing musicians and fans together and that someone is going to want paying somehow.
Filtering/inspecting traffic implies taking responsibility implies getting lawsuits directed at ISPs for users' content.
That's exactly what I was thinking. Doesn't this strip them of their "safe harbor" status?
Firstly, you could at least have read the headline. They're UK ISPs, "safe harbor" (or even "harbour") has no meaning in the UK.
Secondly, and something which wasn't mentioned here, the UK government has discovered something which has proven remarkably effective in getting things done quickly and efficiently without having to mess around with passing legislation. They tell an industry "There's nothing illegal about what you're doing right now but we don't like it and if you don't stop doing it we'll make it illegal".
I suspect the ISPs would rather be seen to do something now than risk getting fined by the government.
That will sync the disks, but it won't stop the database from accepting incoming data. It won't stop cron jobs which might be just about to trigger. It won't deal with tasks that are in the middle of a big operation which involves a lot of writing to disk.
How about this?
killall5 -19; sync
killall5 sends a signal to all running processes other than the current session's. Signal 19 is a SIGSTOP, it immediately freezes a process and cannot be caught or ignored.
Doesn't work.
Postgres, for instance (and I doubt it's the only program to which this applies - I suspect the same is true of most databases) doesn't guarantee that its own data files are consistent unless it's been cleanly shut down. While you will flush any writes which are in progress and prevent any further writes, you won't guarantee that the end result will contain consistent data.
Any sensible database will have a mechanism to recover from this, but that mechanism may result in you losing the last transaction which wasn't fully committed. All the atomicity in ACID means is "you won't start up the database after a crash and find your data doesn't make any sense because only half a transaction was committed".
It doesn't mean "As soon as you hit enter a transaction is guaranteed to commit successfully even if the power fails during the commit process".
Why 5 minutes? It usually takes less than a second to run a sync on the disks depending on how active they are. A couple seconds of runtime should be enough to do an "emergency shutdown" and avoid data corruption.
####@johncash:~$ time sync
real 0m0.004s
user 0m0.004s
sys 0m0.000s
That will sync the disks, but it won't stop the database from accepting incoming data. It won't stop cron jobs which might be just about to trigger. It won't deal with tasks that are in the middle of a big operation which involves a lot of writing to disk.
The case hinges on No Service Password Recovery commands Childs allegedly configured onto several Cisco devices, as well as dial-up and DSL modems the SFPD has discovered that would allow unauthorized connections to the FiberWAN.
Mr. Paranoid Admin with a God complex had big freakin' huge vulnerabilities on his precious network?
It sounds to me like Mr. Paranoid Admin was so paranoid that people had started to do what they tend to do when Mr. Paranoid Admin is so paranoid they can't get anything they need done.
They'd started to work around him.
Net result: All sorts of little unauthorised connections popping up.
In being too paranoid, you wind up creating exactly the situation you fear the most: a network with lots of uncontrolled, unknown systems appearing creating security holes where none previously existed. Doesn't matter how many fancy "no unauthorised access" features your infrastructure has, sooner or later someone's going to succeed in working around them. The last thing you need to do is give them an incentive.
"It's too risky for anybody to translate that [The Bible] into other languages. Mistakes can creep in... and that can lead to heresy. True Christians should only read English."
"If your original Hebrew disagrees with my original King James --- your original Hebrew is wrong. If your original Hebrew agrees with my original King James, your original Hebrew is right."
http://wanusmaximus.livejournal.com/1131751.html
At least a few of those quotes I recognise as having come from the Landover Baptist Church forum:
http://landoverbaptist.net/
(or Forth if you had a Jupiter Ace... you lonely lonely soul...)
I had a Jupiter Ace, you insensitive clod!
(No kidding. I did. And a crappy little machine it was too. Think "ZX spectrum without the software library").
Like I said, I think you are confusing the software assurance program with the open and site licensing programs. The software assurance used to be separate but now it comes with the site licenses and open licenses except they don't apply to some OEM software.. You can however, get the software assurance separately. Or at least you could in the past.
The software assurance is a program that does what you mention, it give an upgrade path to ensure that your software purchases aren't outdated during the useful life to you. A site and open licenses cover any use of specified products company wide (and beyond in some cases) They cover the OS, server products, CALs, office, and any other product specified in the agreements. You build a configuration and each product lists points and after certain levels, you get a certain discount for that software. But if you have a site license for 15 OSes and 15 office programs, it doesn't matter which computers they are on as long as the company owns or controls them.
I know what the various bits of licensing are; the PDF states that operating systems are available through the volume licensing schemes only as upgrades (in other words: it's equivalent to the "Windows XP Upgrade Edition" box you can get from a retail store which is cheaper than the "Full version" box but you're only allowed to use on a system which already has an older version of Windows on).
Software Assurance is just an add-on to the license scheme that allows you to upgrade from the version you originally licensed if/when a new version is released. Sucks to be you if they don't release an upgrade during the course of the license agreement.
This isn't just my understanding; I've confirmed it with account managers from a number of Microsoft resellers in the UK on a couple of occasions.
I also remember reading a story once about a guy who makes giant works of art, using Legos like pixels. I believe they said that if you want to buy like 10,000 blue bricks, you can get bulk prices straight from Lego.
If you want to buy 10,000 of most things you can get a bulk price straight from the manufacturer.
Seriously, though, you make a good point. If they wanted to be evil, I guess they could make them less durable, but I think in the end it's better that everyone knows them as the best.
There is also one thing to bear in mind - as soon as a company like Lego races for the bottom (in terms of price/quality), they're going to lose the customers who valued the quality and then be beaten to the customers who don't by other organisations who 10 or 20 years ago perfected the art of "juuuuusst enough quality that it won't fall apart within 15 minutes of the box opening. Probably."