Or it could be done in the filesystem layer of course, like ZFS (not directly available under Linux ATM due to licensing issues, though there are ongoing efforts to port it in a license compatible manner) can.
Maybe the existence of ZFS is why it hasn't appeared at the block device level: the interested+capable parties are working on a ZFS port (or something similar) instead.
Another issue with the OCZ product: What problem does it actually solve which cannot also be solved by a good OS, a competent admin, an SSD, and a spinning disk?
It solves the "competent admin" bit somewhat, and removes the need to rearrange things when the data most commonly in use varies over time.
I'm surprised I've not seen talk of a Linux block device that merges two block devices (one an SSD and the other linked to larger spinning disks) and uses one as cache for the other. Such a software solution would have the disadvantage of having no non-volatile RAM to keep indexing structures safe through power fails so they would have to be synced to the cache block device very regularly (and in a journalled manner so these writes were safe from power-fail initiated corruption) adding a little extra inefficiency (unless these cards also have no faster NVR and are relying on the SSD to always hold the latest indexes of what is cached where on the SSD, of course), but should be quite practical to implement and certainly not beyond the abilities of many a kernel coder.
My first thought was Google (it would fit with their annual conference of that name and they have cash to play with), but shutting down services like that when they take over isn't their style.
There is a TV provider going by that name, it could also be them.
Depending on how much it went for it could just be prospectors hoping to make something out of the two-letter.com name before the new bubble bursts, of course.
There is for many Trek episodes, and the fans communicate inn these acronyms. It is no different from fans (like myself) of "classic" Doctor Who referring to significant episodes in short form (if you say "genesis" other fans know you mean "Genesis of the Daleks", "frontier" for "Frontier in space" and so forth). It happens in any situation where you have people who probably know too much about a subject talking to each other (cars, trains and facets of certain sports being three other common examples) - large parts of the subject get rendered as some form of short-hand.
I'm thinking Amy's child, which is somehow fathered (or entirely created) from the Flesh's remaining imprint of the ganger Doctor. By killing the doctor proper and taking the energy from his regeneration this being can then regenerate, either "back" into Smith (as he is apparently sign-on for at least another series) or the next Doctor (if the plot line encompasses the next series too). This might be their way of getting around the 12 regenerations limit for the Doctor, assuming the resulting being is starting with a fresh count, rather than saving the decision of how to deal with that until a few series hence.
There is also timelord DNA floating about the universe as a result of the "The Doctor's Daughter" episode though. They could bring that into it too/instead.
They appeared fairly regularly in the "classic" series, but not quite as frequently as peoples' memories would lead them to believe.
Indeed, Doctors 5 through 7 only met them once each on-screen. 2 & 4 encountered them in two televised stories each, 3 bumped into them in 3 stories and 1 holds the record (if you count modern two-parters as single stories) at 5 televised meetings. I'm ignoring short "guest" appearances here, like the few minutes in the 5 doctors, and counting the last segment of Frontier in Space as a run-on to the full Dalek story that followed.
The fact that they didn't appear often heightened the excitement for the fans when they did and I agree with many that they have perhaps been overused in recent series, so giving their narrative a break for a bit certainly makes good sense to me.
Have you seen the specs required for the latest Worms game? See the bottom of http://store.steampowered.com/app/22600/ for the list. 1Gb RAM and a little over 2Gb of disk space required?! The backgrounds and sprites can't be *that* much more numerous and higher resolution can they? Needing a beefy machine to run Doom via C-compiled-to-JS-running-via-JIT-compile-bytecode doesn't seem so ridiculous when you consider that "Worms Reloaded" is running native.
which may void your warranty (on the hardware, not just the OS)
You simply restore to factory OS before taking it in for hardware support.
Because if you jailbreak you have a clue. Remember?
Could you please explain, for I am obviously clueless on this matter, how one would go about restoring the factory default OS on hardware that isn't currently working and hence needs to be sent in for warranty repair/replacement?
OK so if they just replace the device they'll not notice. But you can't guarantee that will be the case.
I doubt they would have an easy way to enforce this given that if you've performed a factory reset on the device they probably can't tell it has been jailbroken (but then again if you are sending it out for repair/replacement under warranty you might not be in a position to perform such a reset).
That said, I still wouldn't compare a built-in feature with warnings about possible consequences, that will always be available, that does not affect the device's warranty to a "feature" that is only available by exploiting bugs in the OS, may be disabled completely if future OS revision fill in all the relevant holes, and may (according to statements made by the manufacturer) invalidate your warranty.
My parent's microwave interferes with their video sender but not their wireless, though the video sender *does* interfere with the wireless network (no matter what channel the AP is using), pushing it down from 5Mbit+ in the chair nearest to the sender to ~1Mbit.
I suspect that most people who have trouble with microwaves are people who have very bad reception anyway - their laptop or other machine being in a position (either due to distance, materials between them and the router, other sources of interference, bad kit, or some combination of the above) where the wireless connection barely works. That last tiny bit of interference lifts the noise level above the useful signal level, but until that point the users don't notice the bad signal reception as they are only checking email and facebook (or performing other tasks that generally work fine despite low bandwidth and high latency) or are blaming slow responses on the sites or the ISP rather than the local network leg.
I actually found it cheaper to run a colour laser printer. $350 3 years ago and it still has 80% toner. It's a network printer and installs quicker on OSX/Linux than Windows.
Be careful of some particularly cheap colour laser printers if you are likely to print a lot though - the price per page (what-ever page you are talking about: b/w business text, colour charts or full-page photos,...) is steadily rising at that part of the market especially when it come time to replace the drum(s) and such, and some of the inkjet tricks have crept in (the first set of carts that come with the machine being half full or less, for instance). Also I've seen really cheap models that produce noticable intermittent banding in some colour output.
My colour laser (which replaced a B/W laser and InkJet pairing just over three years ago) works out noticeably more expensive per page in consumables than the ones at work, but they cost five times a much to buy initially. It is still considerably cheaper than running an inkjet, for any amount of output, but the way the market is going some manufacturers might find a way to "fix" that on their cheap models - some are already noticeably worse than mine. Essentially avoid anything explicitly sold as a "home" printer and you'll get much better value in the long run (and unlike many inkjets that break down completely fairly quickly, "long term" tends to be a valid period to consider for laser/LED machines). Mine is targeted at "small workgroup/office" and didn't cost much more then the "home" models on the market at the time. The chunky toys like we have at the office are what you want if you have extensive printing needs (but for home users the initial expense will probably far outweigh the reduced running costs).
Though even if the real cheapies do start to approach parity with the dreaded inkjet market on consumables (which they might in a couple of years, unless inkjet ink costs are rising by the same factors of course: it is a while since I've priced up anything in that market), the other advantages would still exist: good quality output without paying 15+pence per page for high quality coated paper, usable duplex without even more expensive paper, fast operation with more-or-less honest speed claims in marketing (my laser claims 16ppm for b/w output, over a long run I clocked it at more like 13ppm, mainly as it pauses for a short while every now and then, my last inkjet claimed 20ppm and couldn't even feed blank papar through that fast and even "fast/draft" output was more like 4ppm), being able top leave it idle for a few weeks without having to waste 20% of each cart running the head clean operation, and so forth.
There's no reason to waste your time with an inkjet printer at all, unless you're just buying it for the scanner or faxing facilities, even then you can get a decent scanner cheap enough.
There are a few good inexpensive (again: check for reviews first and generally go for "inexpensive" over "dirt cheap") all-in-one scanner/printer/copier/fax machines that use toner-plus-laser/LED printing rather than spitting ink. Though I still tend to recommend separate units unless the user is going to make much use of the "operate as a fax/copier without needing the computer powered on" option - you tend to get a better scanner and a better printer for not much more money that way and if one breaks down you don't need to replace both.
I'm aware of how hashing works, but as SHA1 is currently a "cryptographically secure" hash the chance of an accidental collision between two non-identical sets of content (even small commits - a change to a small line plus what-ever context needs to be stored to mark where the change is to be made) is very small [approaching the order of 1/(2^160)] - so even given how many people there are potentially pushing commits (directly or otherwise) towards the core kernel repositories I wouldn't expect a collision to be commonplace at all (aside from the occasional completely identical commits when two or more people fix a small typo and change nothing else in the commits).
I might have to grab the whole repo as sipper suggests and see if the history can enlighten me as to how common it actually is (once I've learnt enough to be able to dig around the data with some degree of assurance that I know what I'm looking at!).
Not really, or at least it is of a different scale. The "only 23 people needed for a 50% chance of a shared birthday" is due to the low variance involved (there are only 366 possible birthdays). Is it common enough that the chance of two developers submitting exactly the same commit (making exactly the same small change and submitting it with exactly the same commit comment) that there are lots of duplicate commit hashes in there? Or am I misunderstanding what goes into the pool from which the hash is made?
so many that there are several commits that have exactly the same SHA1 hash, so we're hitting SHA1 collisions.
Really? That seems somewhat wrong to me. I assumed that the hash associated with a commit was based on its content and the commit comment, so for there to be a collision more than one user must have pushed *exactly* the same commit a repo with exactly the same comment. Otherwise a SHA1 collision is mathematically very unlikely. Unless you are referring to the shortened forms often used (I've seen people quoting as little as the first six hex digits of a commit when talking about the history or a small repo)?
Caveat: I've not used Git in anger as yet, so might be completely and utterly misunderstanding it or your point or both. Please feel free to educate me if that is the case!
I don't understand why so many people put "Free Book" on the web, but put it in an HTML page with links to the various chapters. Is it too much to ask for the convenience of a single PDF, MOBI, or EPUB I can download to an eReader?
I don't understand why so many people, when given good content for free, moan that it is not in the format they prefer. Is it too much to ask that they pony up for the print version, or make a modicum of effort to convert the content if they so desire?
On a less factious note: this is how I sometimes prefer to read such content. A good PDF (by "good" I mean "properly formatted as pages", not just the HTML hastily thrown at a PDF generating virtual printer) is nice for printing sections or the whole document, but HTML is usually better for on-screen reading IMO. Different stroke an' all that.
You need to not use wireless at all in that case, aside from known trusted networks that you are sure contain only trusted clients. Unless you are using WPA-Enterprise all clients on the same AP are using the same encryption key so can decode each others packets (intercepted simply by putting your network adaptor into promiscuous mode) easily.
So public wireless is a no-no even if it is not working "plain" (no authentication/encryption), and private wireless is out too unless you have audited every device that has access.
You could get around this by using some for of VPN setup of course, but that option is not open to non-technical users.
To be fair, Windows 7 doesn't run too badly on a netbook with 1Gb RAM and a decent drive (i.e. not one of the slow SSDs that came with a lot of models a year or two ago), at least if you are using it mainly for web browsing and a little bit of other stuff. No worse than XP anyway (unlike Vista).
Though Ubuntu (and no doubt any other distro, but that is the one I have significant personal experience of on netbooks) does work better, both the latest release (which a friend of mine uses in dual-boot with W7 on her netbook) and the last LTS release (which I run, with a couple of updates (FF4 being the main one) from PPAs, in a similar manner though I very very rarely touch the W7 setup). It starts and logs in noticeably faster and feels nippier in general operation. Ubuntu+OO.o is certainly faster than either W7+OO.o or W7+MSOffice on these little machines in my subjective opinion (I've not done any scientific tests, so add salt to taste). And that is with Windows be on the faster end of the drive. But I'd not go nearly as far as describing Windows7 as slower than congealed shit.
It could mean that the tools have been updated slightly since they were actually used to create the Portal levels. The last work on the levels involving those tools might have been done a couple of months before the release of the game so there could be some bug fixes and improvements made.
Also if they are tools they use for a range of titles they might have removed features that are not relevant to Portal 2, either because they are not implemented at all in that game or would cause problems specific to it (perhaps some geometry options cause excess hassle for the through-portal rendering, for instance). Internally you can expect people to know (or find out from documentation when they hit a problem) what features are not relevant/useful to a given game engine (or version of that engine), when you hit the real world you will instead get an infinite number of angry entitled idiots posting in forums how crap the editing tools are because feature X doesn't do what that person interprets from what it says on the tin.
They may also have had to remove any in-jokes (that t hey don't want the rest of the world knowing about) and curse words (or other things that may cause offense, which might include in-jokes again!) that somehow got as far as to be present on the interface or in the documentation.
I think what you saw is more general: it is big city behaviour, not unique to NYC or America. While it is worse in some countries, it is a general facet of the way our behaviour seems to be tweaked by high density urban living.
It is probably powered via HDMI which has a 5V pin
Probably not. IIRC that pin can only supply 50mA rather than the 500mA USB can supply (1800mA for USB3 in power-only mode), which is a significant difference. I can't see 50mA being enough to drive all that.
The extra wires connected at the top in the official pic presumably carry power. They are going to a USB hub so presumably it'll run of the standard USB supply (up to 500mA at 5v) so could easily run off batteries too. It might be wanting more (such as the the up-to 1800mA that USB3 and fast device chargers offer) if it intends to offer power to a device plugged into the USB port, but that is still doable on batteries too.
Apparently it supports composite video out too, so an old analogue TV with the right input would do in place of a fancier HDMI capable model. Though you are right in that the comparison with the OLPC is a bit unfair to the OLPC.
You could network them with USB ethernet controllers, but communication latency with that arrangement would limit what tasks a cluster made this way could do well.
Never mind the fact that not a single cause of fraud has been associated with the intrusion, and with 12 million cards you would think a pattern of fraud would start to appear pretty quickly.
It could be that the negotiations for fencing the data along the chain are not going smoothly. Large scale crime is often a quite compartmentalised business these days and the people taking large amounts of personal data or financial information are often not the same people who end up using it for nefarious means.
If the exploited flaw allowed arbitrary commands to be sent via Apache but did not result in output from Apache that was useful to the hacker (and needed to hack in further to get the target data, they would need to send the output by some other means. In this instance a firewall would be able to help by blocking outgoing connections that were not to a set whitelist of destinations.
While closing off all unneeded services does not protect you from many attack vectors without the need for a firewall, it is conceivable that there are a number that it would not necessarily block and a well configured firewall would. Single level security is more likely to fail, particularly in the presence of a previously unknown flaw though that is not the case here, than security in depth.
Slashdot Mirror
Or it could be done in the filesystem layer of course, like ZFS (not directly available under Linux ATM due to licensing issues, though there are ongoing efforts to port it in a license compatible manner) can.
Maybe the existence of ZFS is why it hasn't appeared at the block device level: the interested+capable parties are working on a ZFS port (or something similar) instead.
Another issue with the OCZ product: What problem does it actually solve which cannot also be solved by a good OS, a competent admin, an SSD, and a spinning disk?
It solves the "competent admin" bit somewhat, and removes the need to rearrange things when the data most commonly in use varies over time.
I'm surprised I've not seen talk of a Linux block device that merges two block devices (one an SSD and the other linked to larger spinning disks) and uses one as cache for the other. Such a software solution would have the disadvantage of having no non-volatile RAM to keep indexing structures safe through power fails so they would have to be synced to the cache block device very regularly (and in a journalled manner so these writes were safe from power-fail initiated corruption) adding a little extra inefficiency (unless these cards also have no faster NVR and are relying on the SSD to always hold the latest indexes of what is cached where on the SSD, of course), but should be quite practical to implement and certainly not beyond the abilities of many a kernel coder.
My first thought was Google (it would fit with their annual conference of that name and they have cash to play with), but shutting down services like that when they take over isn't their style.
.com name before the new bubble bursts, of course.
There is a TV provider going by that name, it could also be them.
Depending on how much it went for it could just be prospectors hoping to make something out of the two-letter
There is for many Trek episodes, and the fans communicate inn these acronyms. It is no different from fans (like myself) of "classic" Doctor Who referring to significant episodes in short form (if you say "genesis" other fans know you mean "Genesis of the Daleks", "frontier" for "Frontier in space" and so forth). It happens in any situation where you have people who probably know too much about a subject talking to each other (cars, trains and facets of certain sports being three other common examples) - large parts of the subject get rendered as some form of short-hand.
(* My money is on Amy )
I'm thinking Amy's child, which is somehow fathered (or entirely created) from the Flesh's remaining imprint of the ganger Doctor. By killing the doctor proper and taking the energy from his regeneration this being can then regenerate, either "back" into Smith (as he is apparently sign-on for at least another series) or the next Doctor (if the plot line encompasses the next series too). This might be their way of getting around the 12 regenerations limit for the Doctor, assuming the resulting being is starting with a fresh count, rather than saving the decision of how to deal with that until a few series hence.
There is also timelord DNA floating about the universe as a result of the "The Doctor's Daughter" episode though. They could bring that into it too/instead.
They appeared fairly regularly in the "classic" series, but not quite as frequently as peoples' memories would lead them to believe.
Indeed, Doctors 5 through 7 only met them once each on-screen. 2 & 4 encountered them in two televised stories each, 3 bumped into them in 3 stories and 1 holds the record (if you count modern two-parters as single stories) at 5 televised meetings. I'm ignoring short "guest" appearances here, like the few minutes in the 5 doctors, and counting the last segment of Frontier in Space as a run-on to the full Dalek story that followed.
The fact that they didn't appear often heightened the excitement for the fans when they did and I agree with many that they have perhaps been overused in recent series, so giving their narrative a break for a bit certainly makes good sense to me.
Have you seen the specs required for the latest Worms game? See the bottom of http://store.steampowered.com/app/22600/ for the list. 1Gb RAM and a little over 2Gb of disk space required?! The backgrounds and sprites can't be *that* much more numerous and higher resolution can they? Needing a beefy machine to run Doom via C-compiled-to-JS-running-via-JIT-compile-bytecode doesn't seem so ridiculous when you consider that "Worms Reloaded" is running native.
which may void your warranty (on the hardware, not just the OS)
You simply restore to factory OS before taking it in for hardware support.
Because if you jailbreak you have a clue. Remember?
Could you please explain, for I am obviously clueless on this matter, how one would go about restoring the factory default OS on hardware that isn't currently working and hence needs to be sent in for warranty repair/replacement?
OK so if they just replace the device they'll not notice. But you can't guarantee that will be the case.
Where is this "Open mode, I am not a moron" button for the iOS devices?
It's called jailbreaking.
... which may void your warranty (on the hardware, not just the OS) apparently: http://www.bbc.co.uk/news/technology-10836692
I doubt they would have an easy way to enforce this given that if you've performed a factory reset on the device they probably can't tell it has been jailbroken (but then again if you are sending it out for repair/replacement under warranty you might not be in a position to perform such a reset).
That said, I still wouldn't compare a built-in feature with warnings about possible consequences, that will always be available, that does not affect the device's warranty to a "feature" that is only available by exploiting bugs in the OS, may be disabled completely if future OS revision fill in all the relevant holes, and may (according to statements made by the manufacturer) invalidate your warranty.
My parent's microwave interferes with their video sender but not their wireless, though the video sender *does* interfere with the wireless network (no matter what channel the AP is using), pushing it down from 5Mbit+ in the chair nearest to the sender to ~1Mbit.
I suspect that most people who have trouble with microwaves are people who have very bad reception anyway - their laptop or other machine being in a position (either due to distance, materials between them and the router, other sources of interference, bad kit, or some combination of the above) where the wireless connection barely works. That last tiny bit of interference lifts the noise level above the useful signal level, but until that point the users don't notice the bad signal reception as they are only checking email and facebook (or performing other tasks that generally work fine despite low bandwidth and high latency) or are blaming slow responses on the sites or the ISP rather than the local network leg.
I actually found it cheaper to run a colour laser printer. $350 3 years ago and it still has 80% toner. It's a network printer and installs quicker on OSX/Linux than Windows.
Be careful of some particularly cheap colour laser printers if you are likely to print a lot though - the price per page (what-ever page you are talking about: b/w business text, colour charts or full-page photos, ...) is steadily rising at that part of the market especially when it come time to replace the drum(s) and such, and some of the inkjet tricks have crept in (the first set of carts that come with the machine being half full or less, for instance). Also I've seen really cheap models that produce noticable intermittent banding in some colour output.
My colour laser (which replaced a B/W laser and InkJet pairing just over three years ago) works out noticeably more expensive per page in consumables than the ones at work, but they cost five times a much to buy initially. It is still considerably cheaper than running an inkjet, for any amount of output, but the way the market is going some manufacturers might find a way to "fix" that on their cheap models - some are already noticeably worse than mine. Essentially avoid anything explicitly sold as a "home" printer and you'll get much better value in the long run (and unlike many inkjets that break down completely fairly quickly, "long term" tends to be a valid period to consider for laser/LED machines). Mine is targeted at "small workgroup/office" and didn't cost much more then the "home" models on the market at the time. The chunky toys like we have at the office are what you want if you have extensive printing needs (but for home users the initial expense will probably far outweigh the reduced running costs).
Though even if the real cheapies do start to approach parity with the dreaded inkjet market on consumables (which they might in a couple of years, unless inkjet ink costs are rising by the same factors of course: it is a while since I've priced up anything in that market), the other advantages would still exist: good quality output without paying 15+pence per page for high quality coated paper, usable duplex without even more expensive paper, fast operation with more-or-less honest speed claims in marketing (my laser claims 16ppm for b/w output, over a long run I clocked it at more like 13ppm, mainly as it pauses for a short while every now and then, my last inkjet claimed 20ppm and couldn't even feed blank papar through that fast and even "fast/draft" output was more like 4ppm), being able top leave it idle for a few weeks without having to waste 20% of each cart running the head clean operation, and so forth.
There's no reason to waste your time with an inkjet printer at all, unless you're just buying it for the scanner or faxing facilities, even then you can get a decent scanner cheap enough.
There are a few good inexpensive (again: check for reviews first and generally go for "inexpensive" over "dirt cheap") all-in-one scanner/printer/copier/fax machines that use toner-plus-laser/LED printing rather than spitting ink. Though I still tend to recommend separate units unless the user is going to make much use of the "operate as a fax/copier without needing the computer powered on" option - you tend to get a better scanner and a better printer for not much more money that way and if one breaks down you don't need to replace both.
If you think that is a trait unique to slashdot, then you have a higher opinion than I of human nature in general!
I'm aware of how hashing works, but as SHA1 is currently a "cryptographically secure" hash the chance of an accidental collision between two non-identical sets of content (even small commits - a change to a small line plus what-ever context needs to be stored to mark where the change is to be made) is very small [approaching the order of 1/(2^160)] - so even given how many people there are potentially pushing commits (directly or otherwise) towards the core kernel repositories I wouldn't expect a collision to be commonplace at all (aside from the occasional completely identical commits when two or more people fix a small typo and change nothing else in the commits).
I might have to grab the whole repo as sipper suggests and see if the history can enlighten me as to how common it actually is (once I've learnt enough to be able to dig around the data with some degree of assurance that I know what I'm looking at!).
Not really, or at least it is of a different scale. The "only 23 people needed for a 50% chance of a shared birthday" is due to the low variance involved (there are only 366 possible birthdays). Is it common enough that the chance of two developers submitting exactly the same commit (making exactly the same small change and submitting it with exactly the same commit comment) that there are lots of duplicate commit hashes in there? Or am I misunderstanding what goes into the pool from which the hash is made?
so many that there are several commits that have exactly the same SHA1 hash, so we're hitting SHA1 collisions.
Really? That seems somewhat wrong to me. I assumed that the hash associated with a commit was based on its content and the commit comment, so for there to be a collision more than one user must have pushed *exactly* the same commit a repo with exactly the same comment. Otherwise a SHA1 collision is mathematically very unlikely. Unless you are referring to the shortened forms often used (I've seen people quoting as little as the first six hex digits of a commit when talking about the history or a small repo)?
Caveat: I've not used Git in anger as yet, so might be completely and utterly misunderstanding it or your point or both. Please feel free to educate me if that is the case!
I don't understand why so many people put "Free Book" on the web, but put it in an HTML page with links to the various chapters. Is it too much to ask for the convenience of a single PDF, MOBI, or EPUB I can download to an eReader?
I don't understand why so many people, when given good content for free, moan that it is not in the format they prefer. Is it too much to ask that they pony up for the print version, or make a modicum of effort to convert the content if they so desire?
On a less factious note: this is how I sometimes prefer to read such content. A good PDF (by "good" I mean "properly formatted as pages", not just the HTML hastily thrown at a PDF generating virtual printer) is nice for printing sections or the whole document, but HTML is usually better for on-screen reading IMO. Different stroke an' all that.
You need to not use wireless at all in that case, aside from known trusted networks that you are sure contain only trusted clients. Unless you are using WPA-Enterprise all clients on the same AP are using the same encryption key so can decode each others packets (intercepted simply by putting your network adaptor into promiscuous mode) easily.
So public wireless is a no-no even if it is not working "plain" (no authentication/encryption), and private wireless is out too unless you have audited every device that has access.
You could get around this by using some for of VPN setup of course, but that option is not open to non-technical users.
To be fair, Windows 7 doesn't run too badly on a netbook with 1Gb RAM and a decent drive (i.e. not one of the slow SSDs that came with a lot of models a year or two ago), at least if you are using it mainly for web browsing and a little bit of other stuff. No worse than XP anyway (unlike Vista).
Though Ubuntu (and no doubt any other distro, but that is the one I have significant personal experience of on netbooks) does work better, both the latest release (which a friend of mine uses in dual-boot with W7 on her netbook) and the last LTS release (which I run, with a couple of updates (FF4 being the main one) from PPAs, in a similar manner though I very very rarely touch the W7 setup). It starts and logs in noticeably faster and feels nippier in general operation. Ubuntu+OO.o is certainly faster than either W7+OO.o or W7+MSOffice on these little machines in my subjective opinion (I've not done any scientific tests, so add salt to taste). And that is with Windows be on the faster end of the drive. But I'd not go nearly as far as describing Windows7 as slower than congealed shit.
It could mean that the tools have been updated slightly since they were actually used to create the Portal levels. The last work on the levels involving those tools might have been done a couple of months before the release of the game so there could be some bug fixes and improvements made.
Also if they are tools they use for a range of titles they might have removed features that are not relevant to Portal 2, either because they are not implemented at all in that game or would cause problems specific to it (perhaps some geometry options cause excess hassle for the through-portal rendering, for instance). Internally you can expect people to know (or find out from documentation when they hit a problem) what features are not relevant/useful to a given game engine (or version of that engine), when you hit the real world you will instead get an infinite number of angry entitled idiots posting in forums how crap the editing tools are because feature X doesn't do what that person interprets from what it says on the tin.
They may also have had to remove any in-jokes (that t hey don't want the rest of the world knowing about) and curse words (or other things that may cause offense, which might include in-jokes again!) that somehow got as far as to be present on the interface or in the documentation.
I think what you saw is more general: it is big city behaviour, not unique to NYC or America. While it is worse in some countries, it is a general facet of the way our behaviour seems to be tweaked by high density urban living.
It is probably powered via HDMI which has a 5V pin
Probably not. IIRC that pin can only supply 50mA rather than the 500mA USB can supply (1800mA for USB3 in power-only mode), which is a significant difference. I can't see 50mA being enough to drive all that.
The extra wires connected at the top in the official pic presumably carry power. They are going to a USB hub so presumably it'll run of the standard USB supply (up to 500mA at 5v) so could easily run off batteries too. It might be wanting more (such as the the up-to 1800mA that USB3 and fast device chargers offer) if it intends to offer power to a device plugged into the USB port, but that is still doable on batteries too.
Apparently it supports composite video out too, so an old analogue TV with the right input would do in place of a fancier HDMI capable model. Though you are right in that the comparison with the OLPC is a bit unfair to the OLPC.
You could network them with USB ethernet controllers, but communication latency with that arrangement would limit what tasks a cluster made this way could do well.
Never mind the fact that not a single cause of fraud has been associated with the intrusion, and with 12 million cards you would think a pattern of fraud would start to appear pretty quickly.
It could be that the negotiations for fencing the data along the chain are not going smoothly. Large scale crime is often a quite compartmentalised business these days and the people taking large amounts of personal data or financial information are often not the same people who end up using it for nefarious means.
If the exploited flaw allowed arbitrary commands to be sent via Apache but did not result in output from Apache that was useful to the hacker (and needed to hack in further to get the target data, they would need to send the output by some other means. In this instance a firewall would be able to help by blocking outgoing connections that were not to a set whitelist of destinations.
While closing off all unneeded services does not protect you from many attack vectors without the need for a firewall, it is conceivable that there are a number that it would not necessarily block and a well configured firewall would. Single level security is more likely to fail, particularly in the presence of a previously unknown flaw though that is not the case here, than security in depth.