"New PC with x86-64"? The last mainstream Intel CPU that didn't support 64-bit instructions was the original Core. (Not the Core 2, which was a rather different beast). This was a bit of an anomaly - Intel already had 64 bit processors out in 2005 though the Core was released at the beginning of 2006. It only ever made it into mobile chips. It's still available, though I wonder how many Intel sell - they often have processors available for purchase long after they've gone out of mainstream use.
AGP similarly was being phased out in 2004.
I get that Linux has a huge hardware compatibility list, but you know something? I don't really care about hardware that hasn't been generally available in five years and hasn't been seen in the wild in two.
It doesn't need to be effective against everyone. It just needs to be enough of an inconvenience that those who can't figure their way around it will instead sign up for something like Netflix or Spotify.
DataBlast, a small magazine that delivered pages of text at 5 per second (I think) during the titles of Bad Influence - a TV programme in the UK devoted to computer games - was probably inspired by Ceefax/Oracle. You needed to record the section on video (remember them?) and then use pause to read the content.
I remember that. Didn't work very well for a couple of reasons:
- There were so many games out there on such a wide variety of platforms at the time that the likelihood of seeing anything particularly interesting was slim.
- Our video didn't do a particularly good job of pausing. There was so much noise on the screen when paused that the Datablast was unreadable.
Probably by providing specific versions of the libraries they depend on.
A quick look at Firefox for Linux suggests I'm right:
- libmozsqlite3.so - Oooh, I wonder what that could be? Not SQLite, by any chance?
- libnss3.so - Network Security Service libraries. Most Linux distributions would already include this anyway.
- libssl3.so - More SSL stuff. Would usually be in the same package as libnss3.so.
There's nothing intrinsically wrong with this - indeed, it's exactly what applications on Windows and OS X have been doing for years. But lots of Linux distributions aren't really devised with that in mind, mainly because it becomes an absolute pig to keep your software up to date if an issue is found in a library.
This is fairly obviously a re-hash of the old "dumb terminal" idea that does the rounds every ten or fifteen years.
In the past the big issue has been "we'd need to re-structure an awful lot of backend IT in order to actually use these dumb terminals, and they're not that much cheaper". This probably remains an issue for large businesses, but for smaller organisations that are buying in most of their IT (and quite often buying it in in the form of web-based systems that they pay a monthly fee for), I wonder if this makes more sense.
In the past you'd probably sell them a machine running Small Business Server, add all their PCs to the domain and charge for ongoing support, but as SBS is basically being retired this leaves the door open for Google. After all, if the server's on its last legs and the replacement will necessitate moving some or all of the infrastructure to an online service anyway, why does it have to be Microsoft's?
Compared with an individual PC and a copy of Office, I'd agree that SaaS is a "pay as you breathe" model. But that's not where it becomes interesting.
Where it becomes interesting is if you're looking for industry-specific software. This quite often costs four or five figures for even a relatively small business and can come with annual maintenance costs that you pretty much have to pay whether you want to or not.
That's a lot of money to find once a year. But your customers don't pay in one big lump once a year. So paying a monthly fee - even if that monthly fee works out slightly dearer - often makes far more business sense.
If your bank offers an app that stores your login details so you don't have to re-enter them, I would question where else your bank operates terrible security.
It does, but AFAICT last time I checked, TrueCrypt makes it relatively easy for the end user to change the encryption key that's used and you can't stop the user from doing this. As soon as they do, the backup key block is useless.
I accept that commercial products that implement other key recovery tools are by definition less secure; what I don't accept is that they are so much less secure you may as well not bother with them in the first place.
Every single time I've heard about a large fine like this being imposed for breach of data protection law, there's been background information - usually aggravating circumstances that make the transgression rather worse.
And so it is here:
The ICO found that a number of officers across the force regularly used unencrypted memory sticks, which may also have been used to copy data from police computers to access away from the office. Despite a similar security breach in September 2010, the force had not put restrictions on downloading information, and staff were not sufficiently trained in data protection.
This wasn't one rogue officer breaching policy, this was a complete failure by management to implement a policy some two years after it had become pretty obvious that such a policy needed to exist.
Not really. Ideally you need a system which marries some degree of security with a mechanism to recover lost keys. Few organisations will accept "you lost the password to your encrypted drive? Then you're stuffed. Not even MI5/NSA/FBI/B&Q can help."
Most commercial encryption products include one or more "user has forgotten their password" recovery mechanisms for exactly this reason.
Interesting to note that precisely nothing has changed since 1999 - the last time I worked in a school.
Back then, the problems were:
Absurd licensing requirements: "You must buy one license for every PC you want to use this software on. We've written this software so it's only really useful if each pupil can use it independently or as part of a 2-person group. We've conceived this software to be useful for one very specific part of the curriculum at a very particular age, so it's only useful for maybe a week or two per year. We've priced this software so that the only way you'll ever be able to justify it to the people who sign the cheques is if it can be useful every single day of the year. We don't offer discounts if you want lots of licenses."
Absurd technical demands: "Visit every PC you wish to install this software on, insert the CD, go Start.., Run.., D:\setup.exe. Our installer eschews such niceties as MSI and installs slightly differently depending on what's currently installed on your PC so you can't easily capture what it does. There isn't a/silent switch or anything equivalent - come on, it's not that difficult! What do you mean, you've got 100 PCs, each PC takes two and a half minutes and so that's four hours work which you somehow have to schedule around classes?"
Even more absurd technical demands: "Our installer only puts a stub on your hard disk; you need the actual CD inserted into the PC in order to run the program. Our program expects to find everything in the root path of the drive you point it at. You can fudge things with a mapped drive letter but you've got another 27-odd programs to install with similar requirements so you're going to have to get creative somehow - either by bootstrapping our program with a script you write yourself (something we won't support) or doing clever things based on group membership."
Sane technical demands, absurd teaching demands: "Or you could copy the files to a network share and run directly from there, doesn't really matter. Unfortunately we don't know the first thing about how kids learn or even what they're expected to learn, so we've had to fudge that bit. Hope you don't mind."
I dedicated my final year project to my housemate for providing moral support - he provided moral support by dropping out of college and spending £27,000 on crack cocaine and underage prostitutes.
Whenever I had trouble finding motivation, I thought of him. This was remarkably effective in motivating me to do a good job of it - I didn't want to wind up in the same boat.
It's patented by the Invention Science Fund - which, as far as I can gather, is a glorified patent troll.
The thing is, adding this sort of functionality to a 3D printer is an awful lot of development for what benefit? Why would someone who's looking to buy any sort of manufacturing machine, whether it's a 3D printer, a CNC lathe or whatever demand a model that will explicitly look at what's fed into it and refuse to print if it looks too much like some other product in a vast database somewhere? It doesn't make any sense whatsoever for manufacturers to do this.
Unless, as you say, at some point in the future they're legally obliged to.
It's amazing how terrible support in the mainline Linux kernel is for ADSL chips and PPPoA when you consider that the great majority of ADSL routers on the market today run Linux.
In plain English, an ARM processor isn't a chip you can go out and buy from ARM Ltd. It's a processor design (or rather a family of processor designs) you can license from ARM Ltd, re-engineer it to suit your needs if you so choose then fabricate.
If you want a ready-to-go chip, you have to buy it from someone else who's already done that. Broadcom have done so, and it's one of their chips in the Raspberry Pi, but so have lots of other companies.
As a result, there's a whole lot of subtly different variants out there. Not all of them are 100% binary compatible with each other. I haven't been able to find out exactly which variant is used in the raspberry pi.
Possibly, but the thing is it's honour based. Short of legislation being enacted, there is no earthly way Do Not Track will be anything other than a tick box in the preferences window that has precisely zero real-world impact - and Microsoft enabling it by default will only speed up the process of Do Not Track becoming useless.
Just a thought - and I appreciate it's probably giving far too much credit to Microsoft for joined-up thinking.
But it occurs to me that Microsoft own Bing, which (like any search engine) is paid for through advertising. And if the advertising can be tightly targeted, it's possible to charge a lot more for it. It follows that at least one business unit within Microsoft wants Do Not Track to be a complete disaster.
However, the days when Microsoft could simply not bother to implement something - or implement it so badly as to make it pointless - are over. Particularly as regards web-based technologies.
So, how to deal with this? Do Not Track is based on an honour system that was only ever going to work if a relatively small percentage of people took advantage of it. By making it a default, that honour system breaks down almost immediately. I honestly can't see very many businesses even bothering to install such a function, much less enable it.
The beauty of doing it this way is it gives Microsoft the opportunity to kill Do Not Track while at the same time getting positive publicity from tech-illiterate journalists for being "the only browser to ask websites to respect your publicity by default". Win-win.
This sort of mess doesn't happen as a result of careful planning. It happens as a result of shortsighted knee-jerk management decisions. Those management decisions probably work out OK in a strong economy, particularly if you are able to predict how many of each model you'll sell.
The problem arises when you face a downturn. You've got an entire chain from building through to distribution devised around this idea of shipping 2,000 different printers (with, let's say 50 basic printers and 40 variants on each one). Which means your driver team is put together based on that assumption, your factories are tooled up based on that assumption, your warehouse processes are based around that assumption and your management team is built around that assumption.
It'd make far more sense to have maybe 10 or 15 basic printers and a whole lot of optional extras - which is precisely what everyone else in the industry does. But in order to get your processes down to that level, you need to drastically cut staff, warehouse space, re-engineer your factories (or pay your contract manufacturers to do so) - and in so doing, an awful lot of middle managers who have been merrily building up their own little empire will push back. They won't do so obviously - well, some might but they can be dealt with very easily - they'll do so insidiously. Terrified for their own job, they'll do everything in their power to avoid making any change that might ultimately mean their team (and hence their empire) is no longer needed.
You really need someone at the top who has the strength to push through this sort of mess and sort it all out - you can't trust the entire business to work with you to achieve it because in so many people's case, it goes against their best interests. Even then it's famously difficult to get right - there is a damn good reason why people who've succeeded in turning around massive companies are greatly respected, and it's nothing to do with their enormous salary.
Slashdot Mirror
You're not forced. There's still plenty of lightweight window managers available in the Ubuntu repositories.
Granted, Canonical could detect old hardware and automatically install such things by default. But it's hardly the end of the world.
Hate to break it to you, but your processor was released circa 2005/2006, as was your graphics card.
Meaning it's knocking on for six years old now.
Yes I accept it's a perfectly adequate computer - far more so than a six year old PC would have been in 2005. But it's still getting on.
Give me a break.
"New PC with x86-64"? The last mainstream Intel CPU that didn't support 64-bit instructions was the original Core. (Not the Core 2, which was a rather different beast). This was a bit of an anomaly - Intel already had 64 bit processors out in 2005 though the Core was released at the beginning of 2006. It only ever made it into mobile chips. It's still available, though I wonder how many Intel sell - they often have processors available for purchase long after they've gone out of mainstream use.
AGP similarly was being phased out in 2004.
I get that Linux has a huge hardware compatibility list, but you know something? I don't really care about hardware that hasn't been generally available in five years and hasn't been seen in the wild in two.
It doesn't need to be effective against everyone. It just needs to be enough of an inconvenience that those who can't figure their way around it will instead sign up for something like Netflix or Spotify.
DataBlast, a small magazine that delivered pages of text at 5 per second (I think) during the titles of Bad Influence - a TV programme in the UK devoted to computer games - was probably inspired by Ceefax/Oracle. You needed to record the section on video (remember them?) and then use pause to read the content.
I remember that. Didn't work very well for a couple of reasons:
- There were so many games out there on such a wide variety of platforms at the time that the likelihood of seeing anything particularly interesting was slim.
- Our video didn't do a particularly good job of pausing. There was so much noise on the screen when paused that the Datablast was unreadable.
Probably by providing specific versions of the libraries they depend on.
A quick look at Firefox for Linux suggests I'm right:
- libmozsqlite3.so - Oooh, I wonder what that could be? Not SQLite, by any chance?
- libnss3.so - Network Security Service libraries. Most Linux distributions would already include this anyway.
- libssl3.so - More SSL stuff. Would usually be in the same package as libnss3.so.
There's nothing intrinsically wrong with this - indeed, it's exactly what applications on Windows and OS X have been doing for years. But lots of Linux distributions aren't really devised with that in mind, mainly because it becomes an absolute pig to keep your software up to date if an issue is found in a library.
No, that's assonance you're thinking of.
Maybe they bought it in from an outside company that had been selling similar systems in other parts of the world for years?
This is fairly obviously a re-hash of the old "dumb terminal" idea that does the rounds every ten or fifteen years.
In the past the big issue has been "we'd need to re-structure an awful lot of backend IT in order to actually use these dumb terminals, and they're not that much cheaper". This probably remains an issue for large businesses, but for smaller organisations that are buying in most of their IT (and quite often buying it in in the form of web-based systems that they pay a monthly fee for), I wonder if this makes more sense.
In the past you'd probably sell them a machine running Small Business Server, add all their PCs to the domain and charge for ongoing support, but as SBS is basically being retired this leaves the door open for Google. After all, if the server's on its last legs and the replacement will necessitate moving some or all of the infrastructure to an online service anyway, why does it have to be Microsoft's?
Compared with an individual PC and a copy of Office, I'd agree that SaaS is a "pay as you breathe" model. But that's not where it becomes interesting.
Where it becomes interesting is if you're looking for industry-specific software. This quite often costs four or five figures for even a relatively small business and can come with annual maintenance costs that you pretty much have to pay whether you want to or not.
That's a lot of money to find once a year. But your customers don't pay in one big lump once a year. So paying a monthly fee - even if that monthly fee works out slightly dearer - often makes far more business sense.
If your bank offers an app that stores your login details so you don't have to re-enter them, I would question where else your bank operates terrible security.
It does, but AFAICT last time I checked, TrueCrypt makes it relatively easy for the end user to change the encryption key that's used and you can't stop the user from doing this. As soon as they do, the backup key block is useless.
I accept that commercial products that implement other key recovery tools are by definition less secure; what I don't accept is that they are so much less secure you may as well not bother with them in the first place.
Every single time I've heard about a large fine like this being imposed for breach of data protection law, there's been background information - usually aggravating circumstances that make the transgression rather worse.
And so it is here:
The ICO found that a number of officers across the force regularly used unencrypted memory sticks, which may also have been used to copy data from police computers to access away from the office. Despite a similar security breach in September 2010, the force had not put restrictions on downloading information, and staff were not sufficiently trained in data protection.
This wasn't one rogue officer breaching policy, this was a complete failure by management to implement a policy some two years after it had become pretty obvious that such a policy needed to exist.
There's hoops in earlier versions of Windows, but Server 2008 introduces a group policy object that makes it pretty easy:
http://www.techrepublic.com/blog/datacenter/disable-removable-media-through-windows-server-2008s-group-policy-configuration/452
Not really. Ideally you need a system which marries some degree of security with a mechanism to recover lost keys. Few organisations will accept "you lost the password to your encrypted drive? Then you're stuffed. Not even MI5/NSA/FBI/B&Q can help."
Most commercial encryption products include one or more "user has forgotten their password" recovery mechanisms for exactly this reason.
Interesting to note that precisely nothing has changed since 1999 - the last time I worked in a school.
Back then, the problems were:
I dedicated my final year project to my housemate for providing moral support - he provided moral support by dropping out of college and spending £27,000 on crack cocaine and underage prostitutes.
Whenever I had trouble finding motivation, I thought of him. This was remarkably effective in motivating me to do a good job of it - I didn't want to wind up in the same boat.
It's patented by the Invention Science Fund - which, as far as I can gather, is a glorified patent troll.
The thing is, adding this sort of functionality to a 3D printer is an awful lot of development for what benefit? Why would someone who's looking to buy any sort of manufacturing machine, whether it's a 3D printer, a CNC lathe or whatever demand a model that will explicitly look at what's fed into it and refuse to print if it looks too much like some other product in a vast database somewhere? It doesn't make any sense whatsoever for manufacturers to do this.
Unless, as you say, at some point in the future they're legally obliged to.
That's pretty endemic in the embedded industry.
It's amazing how terrible support in the mainline Linux kernel is for ADSL chips and PPPoA when you consider that the great majority of ADSL routers on the market today run Linux.
They spent most of those 10 minutes saying "It looks like a blackberry". So I don't think it's fair to say they didn't know it was a smartphone.
Yeah, I got that far. I just couldn't figure out whether that made it a highbank, a vexpress, a mvebu, a picoxcell or something else entirely.
In plain English, an ARM processor isn't a chip you can go out and buy from ARM Ltd. It's a processor design (or rather a family of processor designs) you can license from ARM Ltd, re-engineer it to suit your needs if you so choose then fabricate.
If you want a ready-to-go chip, you have to buy it from someone else who's already done that. Broadcom have done so, and it's one of their chips in the Raspberry Pi, but so have lots of other companies.
As a result, there's a whole lot of subtly different variants out there. Not all of them are 100% binary compatible with each other. I haven't been able to find out exactly which variant is used in the raspberry pi.
Possibly, but the thing is it's honour based. Short of legislation being enacted, there is no earthly way Do Not Track will be anything other than a tick box in the preferences window that has precisely zero real-world impact - and Microsoft enabling it by default will only speed up the process of Do Not Track becoming useless.
Just a thought - and I appreciate it's probably giving far too much credit to Microsoft for joined-up thinking.
But it occurs to me that Microsoft own Bing, which (like any search engine) is paid for through advertising. And if the advertising can be tightly targeted, it's possible to charge a lot more for it. It follows that at least one business unit within Microsoft wants Do Not Track to be a complete disaster.
However, the days when Microsoft could simply not bother to implement something - or implement it so badly as to make it pointless - are over. Particularly as regards web-based technologies.
So, how to deal with this? Do Not Track is based on an honour system that was only ever going to work if a relatively small percentage of people took advantage of it. By making it a default, that honour system breaks down almost immediately. I honestly can't see very many businesses even bothering to install such a function, much less enable it.
The beauty of doing it this way is it gives Microsoft the opportunity to kill Do Not Track while at the same time getting positive publicity from tech-illiterate journalists for being "the only browser to ask websites to respect your publicity by default". Win-win.
WTF were they thinking?
This sort of mess doesn't happen as a result of careful planning. It happens as a result of shortsighted knee-jerk management decisions. Those management decisions probably work out OK in a strong economy, particularly if you are able to predict how many of each model you'll sell.
The problem arises when you face a downturn. You've got an entire chain from building through to distribution devised around this idea of shipping 2,000 different printers (with, let's say 50 basic printers and 40 variants on each one). Which means your driver team is put together based on that assumption, your factories are tooled up based on that assumption, your warehouse processes are based around that assumption and your management team is built around that assumption.
It'd make far more sense to have maybe 10 or 15 basic printers and a whole lot of optional extras - which is precisely what everyone else in the industry does. But in order to get your processes down to that level, you need to drastically cut staff, warehouse space, re-engineer your factories (or pay your contract manufacturers to do so) - and in so doing, an awful lot of middle managers who have been merrily building up their own little empire will push back. They won't do so obviously - well, some might but they can be dealt with very easily - they'll do so insidiously. Terrified for their own job, they'll do everything in their power to avoid making any change that might ultimately mean their team (and hence their empire) is no longer needed.
You really need someone at the top who has the strength to push through this sort of mess and sort it all out - you can't trust the entire business to work with you to achieve it because in so many people's case, it goes against their best interests. Even then it's famously difficult to get right - there is a damn good reason why people who've succeeded in turning around massive companies are greatly respected, and it's nothing to do with their enormous salary.