Probably then Internet Watch Foundation, a completely unaccountable non-government organisation that maintains the current child porn lists (including things like Wikipedia).
Correct me if I am wrong here, but the black text on black background will just be one large rectangle filled uniformly with black color? Meaning, the printer itself won't print it in a such a way that you could tell what the text was at all.
Nope. In PostScript and in PDF, they will be saved as text data or, at the very least, sequences of bezier paths. The will only become a black rectangle when the PDF / PS is rasterised. This happens in the printer for a PostScript printer. If you are using a PDF printer driver, it will not happen at all.
You're assuming that the redaction was done in the PDF. It's equally likely that it was done in the MS Word document and then converted to a PDF, preserving the black text on a black background. It's been a long time since I did any work for the MoD, but I wouldn't be at all surprised if this were the case. Back then, MS Office was installed on everyone's PC as was Acrobat Reader (which tells you how long ago it was), but no PDF editing tools.
Use a PDF printer driver to print the document all over again
That probably won't work. Most of these work by converting the PDF to PostScript for the printer and then back again. In both the PDF and the PostScript, the text will be represented as black text on a black background.
Export it out as a graphic and then put that up on the website.
This removes the ability to search the text.
The correct solution is to replace the object in the PDF file that represents the string of text with an object that draws a black rectangle. Even this is quite tricky. Most PDF editors will do this by just writing a new object and then a new version table, so the original text will still be present, it just won't be visible unless you regenerate the PDF (the PDF format is designed to be edited by appending data to the file, so you can modify it without needing to keep rewriting the entire file).
Some people cherry pick the good parts of a religion and live their lives accordingly.
Some people cherry pick parts of a religion to justify being an asshat.
As far as I can tell, no major religion has a monopoly on people in either category, and every major religion has a lot of people from both. It seems that the teachings of the religion are largely irrelevant.
If your email client doesn't suck, then it will be constructing full-text search indexes in the background, so searching is still relatively quick. I have about 4GB of (mostly plain text) emails on my machine, and I can search them in a few seconds. It isn't having to search everything, it's having to search a smallish data structure that maps character sequences to emails (or to groups of emails that can then be searched properly). If your mail client is still doing the equivalent of running grep on your mail spool, then you need a better mail client. If you host your own mail server, Dovecot can do its own full text indexing, so any client that can issue IMAP search commands can use it.
Folders are still useful for filtering the result. My mail is sorted into folders by some rules as it arrives. If I search for a term and I know the email is in a specific folder, then I have fewer search results to go through than if I do the search globally.
Europe was less affected by the increase in the price of oil than the USA because of taxes. We pay 2-3 times as much as the USA for petrol, and that difference is almost all tax. When the price of crude oil doubles, the price at the pump in the USA almost doubles, while the price in Europe goes up by 10-20%. It's much easier to adapt to a 20% increase in a regular expense than a 100% increase.
You see something similar with the recent increases in global food prices. In most of the western world, the price of food bought in a supermarket is a lot more than the raw material costs - there's packaging, distribution, and often taxes on top. When the price of wheat goes up by 50%, you may see a small increase in the price of bread, but someone living in an agricultural economy will see their price of food double. We see a slight increase in our grocery bills, they starve in large numbers.
I guess the moral of this story is that it's good to have the price of goods depend on a large number of mostly independent factors, so that when one suddenly increases the others act as a buffer, but I'm not really convinced by that argument.
Which is a good reason to try to avoid working for them at all. If companies that don't show any loyalty to their staff find it difficult to recruit competent employees, then that gives them an incentive to change. Otherwise you end up with a feedback cycle where employers and employees both treat each other worse and worse, because they expect the other to screw them over sooner or later.
Ok 9,990 dollars for a book as an income? That is a joke!
That is more than most people make from a book. There's a reason that the world isn't full of writers - it's not a very profitable profession. If you're not in the top 1,000, then you're very unlikely to see that much, especially in fiction. Getting 10,000 to pay full paperback price for a book is very difficult, and the original author is luck to see $1 from each sale in the traditional publishing model.
Yes, before you ask, a significant fraction of my income does derive from book sales.
It takes time. There's a good argument for wanting to get rid of it, because it can waste 5 minutes at the start of a lesson. My school only did them first thing in the morning and after lunch, not sure if American schools do them more or less frequently, but that's still ten minutes of the school day wasted. It always seemed a bit pointless to me, because if you wanted to skip school you could go to those two lessons, skip everything else, and still appear on the register. I can see why you'd want to save this time, but I'm not convinced that teaching students that it's okay to be monitored that closely is a good idea.
Well, he's been selling it as a brand to put on movies and computer games for decades, so I suppose that's not such a great leap. He's not that rare either. Arthur C. Clarke has 'co-written' a lot of books since the late '80s that are a sketch of an idea by him turned into a complete work by someone else, but a book by Arthur C. Clarke and Gentry Lee sells a lot better than a book by Gentry Lee.
If they really wanted to make their machines secure, they'd remove the users. Unfortunately, IT is never told to make the network secure, they are told to make it secure and usable. Uninstalling Windows typically means rewriting a large amount of Windows-only code, or at least doing exhaustive testing of it in WINE, and that's beyond the budget of most IT departments.
Amortising costs over decades works for a telephone system. They transitioned the exchanges from loop-disconect to DTMF, but aside from that the wires laid in the '20s still work today. They've had almost a century to make back the initial (taxpayer subsidised) investment. The demands of an analogue voice channel have remained constant for that entire time.
When it comes to Internet access, the demand changes much more rapidly. In 2001, I had a 1Mb/s Internet connection, and it was the fastest that my ISP provided. Now I have a 10Mb/s connection and they've just finished deploying infrastructure to support 100Mb/s in my area. That's the third major upgrade that they've done in that time. They can't amortise the cost over decades, because the new infrastructure is obsolete after just one decade. I pay less for 10Mb/s as they were charging for 512Kb/s back then (much less if you include inflation), and for the cost of my 1Mb/s connection in 2001, they'll now sell me a 50Mb/s connection today.
That figure is the amount that I pay for data from my colo, but that assumes that the infrastructure already exists. If you have a cable network with 100Mb/s of bandwidth, then you can sell 10Mb/s connections to 10 people. If you've sold them to 5 people, then the cost of adding another customer is basically zero. You can probably get away with selling 10Mb/s connections to 100 or even 200 people if they have typical modest usage patterns, because each one will still be able to get 10Mb/s for the short periods that they saturate the line. If they start all using the connection at the same time, then you have no choice but to increase your overall network capacity. This means laying more fibre. The cost may still be under three eurocents per gigabyte, but that's amortised over the entire life of the new cable, which may be a decade (or more): the ISP has to pay for it all up front. This is where the increase in costs comes from. They have to make significant capital investments, they don't have a significant change in their operating expenses.
It's difficult to determine which of Google's services are actually making money. Google is in the market of harvesting huge amounts of data about people and using that to present them with targeted adverts. Anything that gives the more data about users or more places to put adverts contributes to this, but the profits will most likely show up elsewhere. This is especially true for some of their mobile services. Google Maps Mobile, for example, lets them track where a user goes, but doesn't show any ads directly. This improves the information that they use for targeting adverts, but the revenue will show up in the place where the advert is displayed.
I think the grandparent was talking specifically about the search market. Last figures I saw showed that Bing was still losing money. Microsoft makes insane amounts on Windows and Office (and probably quite a lot on Android now, since they get $15 of profit for every Android handset sold...) which lets them make a loss in a lot of other markets without it making a dent in their bottom line. A few years ago, MSN was making a loss of $20m annually on their UK portal alone, but they were making something like $8bn each on Windows and Office, so the shareholders didn't care. They made a loss of about $700m in their online division in the last quarter. That's tiny compared to the amount of profit they made from Windows and Office, but it's still a lot of cash to be throwing away just to try to reduce Google's market share.
OpenSolaris may be dead, but FreeBSD now ships with version 28 of ZFS, which includes nice things like deduplication. iXSystems, which sells storage appliances based on FreeBSD, is funding ongoing development work, so ZFS on FreeBSD is going to stay actively maintained irrespective of whether Oracle makes future versions of the code available. FreeBSD also got the ability to boot from ZFS before Solaris and integrates ZFS very nicely into the existing storage stack (it works as a GEOM consumer and provider, so you can easily do things like put a FAT partition in a ZVOL and have transactional I/O and deduplication on the filesystem used by a Windows 95 virtual machine).
The worst thing about the first complaint is that ZFS actually does have very clean layering. At the bottom you have the storage pool allocator at the bottom is basically malloc() for persistent storage (equivalent to the block device layer, but with a more convenient interface), the data management unit sits on top of this and provides transactional I/O to the underlying storage, and the ZFS POSIX layer sits on top of this and provides POSIX filesystem semantics. You could replace the ZPL with something that provides things that look like raw block devices (ZVOLS - used for iSCSI shares, among other things), and you could also fairly easily add something that provided an SQL interface to storage, since the transactional support is all done a layer lower down the stack.
The second complaint is really a matter of semantics. Scrubbing a ZFS partition provides a superset of the functionality of fsck. Not only does it validate and repair the metadata, it also validates and repairs the data on the disks. Most ZFS users don't need to run it, because all it really does is try to read every file on the disk - the underlying filesystem code performs the repairs automatically any time some read data fails the checksums.
I think the real problem with ZFS is the name. Calling it a filesystem makes people mentally class it as something like UFS or ext2fs, when it's actually a complete storage stack.
Slashdot Mirror
Probably then Internet Watch Foundation, a completely unaccountable non-government organisation that maintains the current child porn lists (including things like Wikipedia).
Not sure what tileset you're using, but last time I played NetHack it wasn't first person...
Too much latency?
Thunderbolt is designed by Intel (Though they call it Light Peak)
Nope, they call it Thunderbolt. Light Peak was the development code name.
Correct me if I am wrong here, but the black text on black background will just be one large rectangle filled uniformly with black color? Meaning, the printer itself won't print it in a such a way that you could tell what the text was at all.
Nope. In PostScript and in PDF, they will be saved as text data or, at the very least, sequences of bezier paths. The will only become a black rectangle when the PDF / PS is rasterised. This happens in the printer for a PostScript printer. If you are using a PDF printer driver, it will not happen at all.
You're assuming that the redaction was done in the PDF. It's equally likely that it was done in the MS Word document and then converted to a PDF, preserving the black text on a black background. It's been a long time since I did any work for the MoD, but I wouldn't be at all surprised if this were the case. Back then, MS Office was installed on everyone's PC as was Acrobat Reader (which tells you how long ago it was), but no PDF editing tools.
Use a PDF printer driver to print the document all over again
That probably won't work. Most of these work by converting the PDF to PostScript for the printer and then back again. In both the PDF and the PostScript, the text will be represented as black text on a black background.
Export it out as a graphic and then put that up on the website.
This removes the ability to search the text.
The correct solution is to replace the object in the PDF file that represents the string of text with an object that draws a black rectangle. Even this is quite tricky. Most PDF editors will do this by just writing a new object and then a new version table, so the original text will still be present, it just won't be visible unless you regenerate the PDF (the PDF format is designed to be edited by appending data to the file, so you can modify it without needing to keep rewriting the entire file).
Some people cherry pick the good parts of a religion and live their lives accordingly.
Some people cherry pick parts of a religion to justify being an asshat.
As far as I can tell, no major religion has a monopoly on people in either category, and every major religion has a lot of people from both. It seems that the teachings of the religion are largely irrelevant.
If your email client doesn't suck, then it will be constructing full-text search indexes in the background, so searching is still relatively quick. I have about 4GB of (mostly plain text) emails on my machine, and I can search them in a few seconds. It isn't having to search everything, it's having to search a smallish data structure that maps character sequences to emails (or to groups of emails that can then be searched properly). If your mail client is still doing the equivalent of running grep on your mail spool, then you need a better mail client. If you host your own mail server, Dovecot can do its own full text indexing, so any client that can issue IMAP search commands can use it.
Folders are still useful for filtering the result. My mail is sorted into folders by some rules as it arrives. If I search for a term and I know the email is in a specific folder, then I have fewer search results to go through than if I do the search globally.
Europe was less affected by the increase in the price of oil than the USA because of taxes. We pay 2-3 times as much as the USA for petrol, and that difference is almost all tax. When the price of crude oil doubles, the price at the pump in the USA almost doubles, while the price in Europe goes up by 10-20%. It's much easier to adapt to a 20% increase in a regular expense than a 100% increase.
You see something similar with the recent increases in global food prices. In most of the western world, the price of food bought in a supermarket is a lot more than the raw material costs - there's packaging, distribution, and often taxes on top. When the price of wheat goes up by 50%, you may see a small increase in the price of bread, but someone living in an agricultural economy will see their price of food double. We see a slight increase in our grocery bills, they starve in large numbers.
I guess the moral of this story is that it's good to have the price of goods depend on a large number of mostly independent factors, so that when one suddenly increases the others act as a buffer, but I'm not really convinced by that argument.
Which is a good reason to try to avoid working for them at all. If companies that don't show any loyalty to their staff find it difficult to recruit competent employees, then that gives them an incentive to change. Otherwise you end up with a feedback cycle where employers and employees both treat each other worse and worse, because they expect the other to screw them over sooner or later.
I don't remember anyone voting for Vint Cerf or Tim Berners-Lee either, but they seemed to do a good job...
Ok 9,990 dollars for a book as an income? That is a joke!
That is more than most people make from a book. There's a reason that the world isn't full of writers - it's not a very profitable profession. If you're not in the top 1,000, then you're very unlikely to see that much, especially in fiction. Getting 10,000 to pay full paperback price for a book is very difficult, and the original author is luck to see $1 from each sale in the traditional publishing model.
Yes, before you ask, a significant fraction of my income does derive from book sales.
Very appropriate, linking to a video that I can't watch watch unless I opt in to Google's tracking...
It takes time. There's a good argument for wanting to get rid of it, because it can waste 5 minutes at the start of a lesson. My school only did them first thing in the morning and after lunch, not sure if American schools do them more or less frequently, but that's still ten minutes of the school day wasted. It always seemed a bit pointless to me, because if you wanted to skip school you could go to those two lessons, skip everything else, and still appear on the register. I can see why you'd want to save this time, but I'm not convinced that teaching students that it's okay to be monitored that closely is a good idea.
Well, he's been selling it as a brand to put on movies and computer games for decades, so I suppose that's not such a great leap. He's not that rare either. Arthur C. Clarke has 'co-written' a lot of books since the late '80s that are a sketch of an idea by him turned into a complete work by someone else, but a book by Arthur C. Clarke and Gentry Lee sells a lot better than a book by Gentry Lee.
If they really wanted to make their machines secure, they'd remove the users. Unfortunately, IT is never told to make the network secure, they are told to make it secure and usable. Uninstalling Windows typically means rewriting a large amount of Windows-only code, or at least doing exhaustive testing of it in WINE, and that's beyond the budget of most IT departments.
10,000 downloads doesn't sound too outrageous, and that's still a lot more than most authors make per book in the traditional model...
Don't worry writers, Jeffery Archer, Dan Brown, and Tom Clancy have been successfully using this AI for years...
Amortising costs over decades works for a telephone system. They transitioned the exchanges from loop-disconect to DTMF, but aside from that the wires laid in the '20s still work today. They've had almost a century to make back the initial (taxpayer subsidised) investment. The demands of an analogue voice channel have remained constant for that entire time.
When it comes to Internet access, the demand changes much more rapidly. In 2001, I had a 1Mb/s Internet connection, and it was the fastest that my ISP provided. Now I have a 10Mb/s connection and they've just finished deploying infrastructure to support 100Mb/s in my area. That's the third major upgrade that they've done in that time. They can't amortise the cost over decades, because the new infrastructure is obsolete after just one decade. I pay less for 10Mb/s as they were charging for 512Kb/s back then (much less if you include inflation), and for the cost of my 1Mb/s connection in 2001, they'll now sell me a 50Mb/s connection today.
That figure is the amount that I pay for data from my colo, but that assumes that the infrastructure already exists. If you have a cable network with 100Mb/s of bandwidth, then you can sell 10Mb/s connections to 10 people. If you've sold them to 5 people, then the cost of adding another customer is basically zero. You can probably get away with selling 10Mb/s connections to 100 or even 200 people if they have typical modest usage patterns, because each one will still be able to get 10Mb/s for the short periods that they saturate the line. If they start all using the connection at the same time, then you have no choice but to increase your overall network capacity. This means laying more fibre. The cost may still be under three eurocents per gigabyte, but that's amortised over the entire life of the new cable, which may be a decade (or more): the ISP has to pay for it all up front. This is where the increase in costs comes from. They have to make significant capital investments, they don't have a significant change in their operating expenses.
It's difficult to determine which of Google's services are actually making money. Google is in the market of harvesting huge amounts of data about people and using that to present them with targeted adverts. Anything that gives the more data about users or more places to put adverts contributes to this, but the profits will most likely show up elsewhere. This is especially true for some of their mobile services. Google Maps Mobile, for example, lets them track where a user goes, but doesn't show any ads directly. This improves the information that they use for targeting adverts, but the revenue will show up in the place where the advert is displayed.
I think the grandparent was talking specifically about the search market. Last figures I saw showed that Bing was still losing money. Microsoft makes insane amounts on Windows and Office (and probably quite a lot on Android now, since they get $15 of profit for every Android handset sold...) which lets them make a loss in a lot of other markets without it making a dent in their bottom line. A few years ago, MSN was making a loss of $20m annually on their UK portal alone, but they were making something like $8bn each on Windows and Office, so the shareholders didn't care. They made a loss of about $700m in their online division in the last quarter. That's tiny compared to the amount of profit they made from Windows and Office, but it's still a lot of cash to be throwing away just to try to reduce Google's market share.
OpenSolaris may be dead, but FreeBSD now ships with version 28 of ZFS, which includes nice things like deduplication. iXSystems, which sells storage appliances based on FreeBSD, is funding ongoing development work, so ZFS on FreeBSD is going to stay actively maintained irrespective of whether Oracle makes future versions of the code available. FreeBSD also got the ability to boot from ZFS before Solaris and integrates ZFS very nicely into the existing storage stack (it works as a GEOM consumer and provider, so you can easily do things like put a FAT partition in a ZVOL and have transactional I/O and deduplication on the filesystem used by a Windows 95 virtual machine).
The worst thing about the first complaint is that ZFS actually does have very clean layering. At the bottom you have the storage pool allocator at the bottom is basically malloc() for persistent storage (equivalent to the block device layer, but with a more convenient interface), the data management unit sits on top of this and provides transactional I/O to the underlying storage, and the ZFS POSIX layer sits on top of this and provides POSIX filesystem semantics. You could replace the ZPL with something that provides things that look like raw block devices (ZVOLS - used for iSCSI shares, among other things), and you could also fairly easily add something that provided an SQL interface to storage, since the transactional support is all done a layer lower down the stack.
The second complaint is really a matter of semantics. Scrubbing a ZFS partition provides a superset of the functionality of fsck. Not only does it validate and repair the metadata, it also validates and repairs the data on the disks. Most ZFS users don't need to run it, because all it really does is try to read every file on the disk - the underlying filesystem code performs the repairs automatically any time some read data fails the checksums.
I think the real problem with ZFS is the name. Calling it a filesystem makes people mentally class it as something like UFS or ext2fs, when it's actually a complete storage stack.