Huh? Every implementation of PNG I've ever seen has worked like that - it's the zlib compression level. Never saved a PNG in the gimp?
No. I've used multiple versions of photoshop, paint shop pro (some years ago, now, so may not remember it clearly), and Java ImageIO, none of which seem to provide such an option. I assume they just use the maximum, as really on modern hardware there isn't any reason at all not to use maximum zlib compression in just about every imaginable scenario. I'm quite surprised GIMP does offer such an option, for this reason and the fact that people appear to be confusing it with a JPEG-like quality setting, and are therefore presumably setting a lower compression level in the believe they're getting a better quality image out.
What happened to CP/M and GEM, MSX and Unix which were licensed to multiple manufacturers?
CP/M was effectively cloned by Microsoft and became MS-DOS (DOS 1 was designed to be API-compatible with CP/M); current versions of Windows still include a backwards-compatible CP/M API (NTVDM.exe), so CP/M is essentially still widely available. Windows still ships with CMD.exe which is mostly backwards-compatible with the commands of the CP/M shell.
GEM is still available as an open-source project, although it isn't popular. Its popularity was mostly killed by an Apple "look-and-feel" lawsuit against its developers, forcing them to remove functionality. Given that such lawsuits aren't really fair, IMO, I don't see this as a fair comparison.
MSX was a hardware platform, not an operating system, and an 8-bit one at that. It was, therefore, doomed to failure.
Unix & its various clones are still widely deployed, including in Apple OS/X, so I'm not sure what it's doing on this list.
No, it isn't. Qt is LGPL, which means there is a requirement in distribution that the end user should be able to relink with an updated version of the library. This required freedom is simply incompatible with the closed-platform app store policies of both iOS and Windows Mobile. I think you'd be on legal thin ice developing an app for either of these platforms with Qt, and as the copyright holder is now Nokia, who might like to put a dent in availability of apps on rival platforms, don't think for a moment it won't happen.
Not sure why not. S13 of the DPA gives the data subject the right to sue for damages arising from any breach of the act. That would include damage to reputation, which could be argued to exist in most cases where somebody is accused of participating in an illegal activity.
DPA also requires that the data subject has consented to the processing
There are a number of exemptions to this requirement; see schedule 2 of the Act. The most relevant one is probably this one:
6. The processing is necessary for the purposes of legitimate interests pursued by the data controller or by the third party or parties to whom the data are disclosed, except where the processing is unwarranted in any particular case by reason of prejudice to the rights and freedoms or legitimate interests of the data subject.
(1)An individual who suffers damage by reason of any contravention by a data controller of any of the requirements of this Act[1] is entitled to compensation from the data controller for that damage.
[1]: this would include the requirements (a) not to disclose the data to third parties without authorisation, and (b) to take reasonable steps to ensure the data is secure.
Why is that? Because you don't get why a Turing-complete language with internet access could be a security threat?
It's only a security threat if you can't trust the site that the programs are originating from. Sure, this search engine *may* be able to dump a tracking code into their output and therefore break the TOR privacy[1], but you have to ask how likely to happen is this? And my answer: very unlikely.
[1] Doing so is, however, hard, an not even obviously possible: the identifying information javascript can access is rather limited, and it doesn't provide any means of bypassing your outgoing proxy server, so I don't *think* there's anything it can get access to that's damaging. It can identify your web browser and operating system, which is info you might be stripping out of HTTP requests with (e.g.) privoxy. It is also be able to list your browser plugins and their versions. Except in very unusual circumstances, I don't think we're talking enough information to identify an individual. And people in those circumstances ought to understand them well enough to be able to avoid the problem (we're basically talking people with custom browser builds, or private browser plugins).
Well, unlike other TOR servers, where anonymity hides the server's location, it's pretty obvious who's hosting this server. The Government would just be like "Hey...stop it."
That's how.
And all they have to do is leak their database and TOR configuration files, and suddenly anyone can run it, and seamlessly for anyone using the TOR service URL. And as long as they lay the groundwork in advance, they can even deny having done it: an appropriate rootkit on their server should suffice to cast doubt on any evidence against them.
Civilization 1 may not have included DRM, but it did have a form a copy protection that would quiz you about certain game concepts that was included with the manual. These were the days before hard drives were common so you were quizzed each time your played the game. While definitely not as insidious as current DRM schemes, it was still pretty irritating If you failed, I believe you were dropped to a DOS prompt without having the option to save.
Not as serious as that: it wiped out all your military units but carried on playing. I didn't find it too problematic. After you'd played the game through 4 or 5 times you pretty much knew the answers to all the questions anyway.
Also: before hard drives were common? Err... no. Civ1 was released in 1991. Most PCs of the era were supplied with either Windows 2 or Windows 3, both of which required a hard drive. A typical '91 PC would have somewhere between 32MB (low end, sometimes used for DOS 3 compatibility) and 200MB of hard disk space. The machine I had at the time was a 4-year-old 286 that had a 45MB hard disk, which was considered small by most people I knew (although my school's machines were 186s with 20MB disks). I believe the version of DOS available at the time, DOS 5, also required a hard disk (or a RAM disk larger than a floppy, at the very least), but have been unable to confirm this.
I think it's been a while since I read so much rubbish modded informative.
PNG can be lossless, if you choose 0 for compression level - then the image is basically a bitmap when it comes to file-size (except png inherently supports Alpha-channel transparency).
Err... no. PNG is always lossless. PNG is also always zlib-compressed, so is almost always smaller than the equivalent.BMP file. I've also never seen an implementation of PNG that supports numeric setting of compression level: the choices are normally between different strategies, although most modern implementations simply autodetect the best strategy for each line of the image.
Similiar to Video - every time you change format you will almost-always introduce artifacts.
Depends on the format of course; there are few lossless video formats out there, but there are some (HuffYUV being the most common, but there are others)
Also, JPG image quality can be greatly improved by choosing the option to disable color subsampling.
This depends heavily on the image. Chroma subsampling is very useful in many situations, but it cannot represent very small areas of colour in a reasonable way. It does reduce the amount of data that needs to be compressed by half (6 values per 2x2 block rather than 12) and results in images that usually are indistinguishable from the source without enlargement, which really means that you can use a higher quality level for the rest of the process, which in many cases results in a better overall image quality. If you're targetting very high quality and don't much care about file size, then disabling it is an obvious win, otherwise you may want to try both ways.
JPG's tend to be smaller than PNG when you start with a BMP, but not always -- depends on the content of the image.
Yes, some images compress particularly well with PNG (e.g. line art, uniform fills, smooth gradient fills) but not very well at all with JPG. JPG is suitable only for photographic or similar images.
BMP->JPG: Smallish File size. BMP->PNG: Smallish File size, may be smaller than previous JPG. YET...Take that JPG and save as a level 6 compression PNG and it will be bigger than it's parent JPG and bigger than the BMP->PNG.
Yes, because JPG is lossy and PNG isn't. JPG compression artifacts are very hard for PNG to encode as they appear essentially as random noise to PNG's compressor. Therefore smooth BMP -> noisy JPG -> PNG will result in a larger file than direct BMP->PNG conversion. This is pretty much guaranteed to happen in any case where the resulting PNG is smaller than the JPG. There's probably an information-theoretic way to explain this, but off the top of my head I can't think of it.
How on earth would a regular person have any idea how to do all that crap?
The free advisor at his local citizens advice bureau should be able to tell him this.
He shouldn't need to hire a lawyer to tell him how to get his stolen dog back.
Everything required should be possible without a lawyer, unless the dog is worth more than £5,000 which would put it outside of small claims court channels.
Why wasn't this treated as a criminal (or even civil property) matter? Aren't the new owners guilty of receiving stolen property? I mean, even if they didn't know it before (assuming they bought the dog from the thief and didn't realize it was stolen), they obviously do now.
Under UK law property can't be confiscated from its new owners unless they had reason to believe it was stolen at the time they acquired it. The article says the police investigated and concluded there wasn't a criminal matter, so presumably they felt they had no reason to believe the dog was stolen.
And even if the dog wasn't stolen, it's still the original owner's property, no?
No. The aphorism "posession is nine tenths of the law" holds true: somebody who has done no wrong and is in posession of property that they had a right to believe was theirs is the owner.
We all know the girl was a bit stupid ("I didn't know it was illegal"? Seriously? That's your defense?)
It appears to be what the judge interpreted her defense as ("Harper cannot rely on her purported legal naivety," he's quoted as saying). But FWICS her defence was *actually* "I didn't know I was doing it" (as in "the software did it without making it clear exactly what it would do" -- she apparently said she thought Kazaa was like a streaming radio program).
There's a substantial difference: the first isn't a legal defence at all (hence the appeal judge dismissing it), the latter is (hence the original trial judge allocating reduced damages because of it).
The big WTF is that according to every principle of fairness, there shouldn't be statutory damages in such a case at all: let them claim actual damages and be done with it, I reckon. There should be no statutory penalties for doing something that you aren't aware of doing, as if you aren't aware of them there's no realistic way of avoiding doing them.
What I find astonishing is that you can perform an act which *you have no reason to know is illegal*, perhaps even *without knowing you are doing it*, and which the court is aware you didn't know was illegal, and you're *still* expect to pay the statutory damages (i.e. a charge that is intended to discourage you from doing it).
Here in the UK, you wouldn't be able to get any more than actual damages. Which is perhaps why the BPI isn't so litigation-happy.
If she left bittorent running till the share ratio hit 2.0, maybe she should even pay &1.98 per song
RTFA. It appears she left Kazaa running for multiple years, without realising it. The songs *could* have been downloaded hundreds of times each, but we have no way of knowing.
The biggest problem (and I witnessed this 1st hand) is that the people running government IT seem to lack focus on what they want to have delivered, so projects run on and on.
Yes, but part of the problem here is what the other half of this is addressing: the fact that most government IT contracts go to one or two large companies (I'm primarily thinking Capita here), who don't really have to compete on quality because there's a minimum turnover requirement in the tender that eliminates almost all of the potential competition.
Yes. The news here is (1) the reduction in multi-million-pound contracts in favour of more, smaller contracts. This means that the bidders themselves can be smaller (UK government tenders have, in the past, often had existing turnover requirements that mean most IT consultancies can't bid on them, leaving the field open to just a few large specialist companies, with most contracts apparently going to Capita), and (2) an apparent _preference_ for open-source solutions, rather than just (as the last government did) a requirement to evaluate open-source solutions as well as closed-source ones.
Bring in alligators to eat the opossum, and then in the winter, they'll all freeze to death.
A friend keeps singing a song to her kid about an alligator going snap. I keep telling her she'll need liquid oxygen to achieve that, but I don't think she's got the message.
While conducting a medical lecture in Japan at the Tokyo Convention Center, Dr. Allen explained that the human brain acts like an "organic computer."
Dr. Allen pointed out that a prime example of the brain-computer relationship was shown in Rainman, a very popular movie about Kim Peek, a Savant, whose brain works exactly like a computer.
Slashdot Mirror
Add in the LEGO logo on each nub, as raised text, and you've added a lot more polygons.
This is what bumpmaps are for. You shouldn't be looking at them closely enough to see they aren't actually raised.
Huh? Every implementation of PNG I've ever seen has worked like that - it's the zlib compression level. Never saved a PNG in the gimp?
No. I've used multiple versions of photoshop, paint shop pro (some years ago, now, so may not remember it clearly), and Java ImageIO, none of which seem to provide such an option. I assume they just use the maximum, as really on modern hardware there isn't any reason at all not to use maximum zlib compression in just about every imaginable scenario. I'm quite surprised GIMP does offer such an option, for this reason and the fact that people appear to be confusing it with a JPEG-like quality setting, and are therefore presumably setting a lower compression level in the believe they're getting a better quality image out.
Ever heard about dynamic libraries?
How does an iPhone user update the dynamic libraries used by an app they acquired in the app store on their phone?
What happened to CP/M and GEM, MSX and Unix which were licensed to multiple manufacturers?
CP/M was effectively cloned by Microsoft and became MS-DOS (DOS 1 was designed to be API-compatible with CP/M); current versions of Windows still include a backwards-compatible CP/M API (NTVDM.exe), so CP/M is essentially still widely available. Windows still ships with CMD.exe which is mostly backwards-compatible with the commands of the CP/M shell.
GEM is still available as an open-source project, although it isn't popular. Its popularity was mostly killed by an Apple "look-and-feel" lawsuit against its developers, forcing them to remove functionality. Given that such lawsuits aren't really fair, IMO, I don't see this as a fair comparison.
MSX was a hardware platform, not an operating system, and an 8-bit one at that. It was, therefore, doomed to failure.
Unix & its various clones are still widely deployed, including in Apple OS/X, so I'm not sure what it's doing on this list.
[Qt is] also free for everybody.
No, it isn't. Qt is LGPL, which means there is a requirement in distribution that the end user should be able to relink with an updated version of the library. This required freedom is simply incompatible with the closed-platform app store policies of both iOS and Windows Mobile. I think you'd be on legal thin ice developing an app for either of these platforms with Qt, and as the copyright holder is now Nokia, who might like to put a dent in availability of apps on rival platforms, don't think for a moment it won't happen.
(Not a lawyer; not legal advice.)
Incorrect. Apple requires Mac OS X for development. This software can be virtualized on most modern PC configurations.
And where can I legally buy a copy without also buying Apple hardware?
Sued - probably not.
Not sure why not. S13 of the DPA gives the data subject the right to sue for damages arising from any breach of the act. That would include damage to reputation, which could be argued to exist in most cases where somebody is accused of participating in an illegal activity.
DPA also requires that the data subject has consented to the processing
There are a number of exemptions to this requirement; see schedule 2 of the Act. The most relevant one is probably this one:
Yes. Data Protection Act 1988, section 13:
[1]: this would include the requirements (a) not to disclose the data to third parties without authorisation, and (b) to take reasonable steps to ensure the data is secure.
That document is about criminal prosecutions, not civil actions for damages. I think you're looking for this.
Why is that? Because you don't get why a Turing-complete language with internet access could be a security threat?
It's only a security threat if you can't trust the site that the programs are originating from. Sure, this search engine *may* be able to dump a tracking code into their output and therefore break the TOR privacy[1], but you have to ask how likely to happen is this? And my answer: very unlikely.
[1] Doing so is, however, hard, an not even obviously possible: the identifying information javascript can access is rather limited, and it doesn't provide any means of bypassing your outgoing proxy server, so I don't *think* there's anything it can get access to that's damaging. It can identify your web browser and operating system, which is info you might be stripping out of HTTP requests with (e.g.) privoxy. It is also be able to list your browser plugins and their versions. Except in very unusual circumstances, I don't think we're talking enough information to identify an individual. And people in those circumstances ought to understand them well enough to be able to avoid the problem (we're basically talking people with custom browser builds, or private browser plugins).
Well, unlike other TOR servers, where anonymity hides the server's location, it's pretty obvious who's hosting this server. The Government would just be like "Hey...stop it."
That's how.
And all they have to do is leak their database and TOR configuration files, and suddenly anyone can run it, and seamlessly for anyone using the TOR service URL. And as long as they lay the groundwork in advance, they can even deny having done it: an appropriate rootkit on their server should suffice to cast doubt on any evidence against them.
How long until it's going to be shut down because you can find nasty bits with it?
How would somebody go about shutting it down?
Civilization 1 may not have included DRM, but it did have a form a copy protection that would quiz you about certain game concepts that was included with the manual. These were the days before hard drives were common so you were quizzed each time your played the game. While definitely not as insidious as current DRM schemes, it was still pretty irritating If you failed, I believe you were dropped to a DOS prompt without having the option to save.
Not as serious as that: it wiped out all your military units but carried on playing. I didn't find it too problematic. After you'd played the game through 4 or 5 times you pretty much knew the answers to all the questions anyway.
Also: before hard drives were common? Err... no. Civ1 was released in 1991. Most PCs of the era were supplied with either Windows 2 or Windows 3, both of which required a hard drive. A typical '91 PC would have somewhere between 32MB (low end, sometimes used for DOS 3 compatibility) and 200MB of hard disk space. The machine I had at the time was a 4-year-old 286 that had a 45MB hard disk, which was considered small by most people I knew (although my school's machines were 186s with 20MB disks). I believe the version of DOS available at the time, DOS 5, also required a hard disk (or a RAM disk larger than a floppy, at the very least), but have been unable to confirm this.
I think it's been a while since I read so much rubbish modded informative.
PNG can be lossless, if you choose 0 for compression level - then the image is basically a bitmap when it comes to file-size (except png inherently supports Alpha-channel transparency).
Err... no. PNG is always lossless. PNG is also always zlib-compressed, so is almost always smaller than the equivalent .BMP file. I've also never seen an implementation of PNG that supports numeric setting of compression level: the choices are normally between different strategies, although most modern implementations simply autodetect the best strategy for each line of the image.
Similiar to Video - every time you change format you will almost-always introduce artifacts.
Depends on the format of course; there are few lossless video formats out there, but there are some (HuffYUV being the most common, but there are others)
Also, JPG image quality can be greatly improved by choosing the option to disable color subsampling.
This depends heavily on the image. Chroma subsampling is very useful in many situations, but it cannot represent very small areas of colour in a reasonable way. It does reduce the amount of data that needs to be compressed by half (6 values per 2x2 block rather than 12) and results in images that usually are indistinguishable from the source without enlargement, which really means that you can use a higher quality level for the rest of the process, which in many cases results in a better overall image quality. If you're targetting very high quality and don't much care about file size, then disabling it is an obvious win, otherwise you may want to try both ways.
JPG's tend to be smaller than PNG when you start with a BMP, but not always -- depends on the content of the image.
Yes, some images compress particularly well with PNG (e.g. line art, uniform fills, smooth gradient fills) but not very well at all with JPG. JPG is suitable only for photographic or similar images.
BMP->JPG: Smallish File size.
BMP->PNG: Smallish File size, may be smaller than previous JPG.
YET...Take that JPG and save as a level 6 compression PNG and it will be bigger than it's parent JPG and bigger than the BMP->PNG.
Yes, because JPG is lossy and PNG isn't. JPG compression artifacts are very hard for PNG to encode as they appear essentially as random noise to PNG's compressor. Therefore smooth BMP -> noisy JPG -> PNG will result in a larger file than direct BMP->PNG conversion. This is pretty much guaranteed to happen in any case where the resulting PNG is smaller than the JPG. There's probably an information-theoretic way to explain this, but off the top of my head I can't think of it.
How on earth would a regular person have any idea how to do all that crap?
The free advisor at his local citizens advice bureau should be able to tell him this.
He shouldn't need to hire a lawyer to tell him how to get his stolen dog back.
Everything required should be possible without a lawyer, unless the dog is worth more than £5,000 which would put it outside of small claims court channels.
Why wasn't this treated as a criminal (or even civil property) matter? Aren't the new owners guilty of receiving stolen property? I mean, even if they didn't know it before (assuming they bought the dog from the thief and didn't realize it was stolen), they obviously do now.
Under UK law property can't be confiscated from its new owners unless they had reason to believe it was stolen at the time they acquired it. The article says the police investigated and concluded there wasn't a criminal matter, so presumably they felt they had no reason to believe the dog was stolen.
And even if the dog wasn't stolen, it's still the original owner's property, no?
No. The aphorism "posession is nine tenths of the law" holds true: somebody who has done no wrong and is in posession of property that they had a right to believe was theirs is the owner.
We all know the girl was a bit stupid ("I didn't know it was illegal"? Seriously? That's your defense?)
It appears to be what the judge interpreted her defense as ("Harper cannot rely on her purported legal naivety," he's quoted as saying). But FWICS her defence was *actually* "I didn't know I was doing it" (as in "the software did it without making it clear exactly what it would do" -- she apparently said she thought Kazaa was like a streaming radio program).
There's a substantial difference: the first isn't a legal defence at all (hence the appeal judge dismissing it), the latter is (hence the original trial judge allocating reduced damages because of it).
The big WTF is that according to every principle of fairness, there shouldn't be statutory damages in such a case at all: let them claim actual damages and be done with it, I reckon. There should be no statutory penalties for doing something that you aren't aware of doing, as if you aren't aware of them there's no realistic way of avoiding doing them.
What I find astonishing is that you can perform an act which *you have no reason to know is illegal*, perhaps even *without knowing you are doing it*, and which the court is aware you didn't know was illegal, and you're *still* expect to pay the statutory damages (i.e. a charge that is intended to discourage you from doing it).
Here in the UK, you wouldn't be able to get any more than actual damages. Which is perhaps why the BPI isn't so litigation-happy.
If she left bittorent running till the share ratio hit 2.0, maybe she should even pay &1.98 per song
RTFA. It appears she left Kazaa running for multiple years, without realising it. The songs *could* have been downloaded hundreds of times each, but we have no way of knowing.
The biggest problem (and I witnessed this 1st hand) is that the people running government IT seem to lack focus on what they want to have delivered, so projects run on and on.
Yes, but part of the problem here is what the other half of this is addressing: the fact that most government IT contracts go to one or two large companies (I'm primarily thinking Capita here), who don't really have to compete on quality because there's a minimum turnover requirement in the tender that eliminates almost all of the potential competition.
Yes. The news here is (1) the reduction in multi-million-pound contracts in favour of more, smaller contracts. This means that the bidders themselves can be smaller (UK government tenders have, in the past, often had existing turnover requirements that mean most IT consultancies can't bid on them, leaving the field open to just a few large specialist companies, with most contracts apparently going to Capita), and (2) an apparent _preference_ for open-source solutions, rather than just (as the last government did) a requirement to evaluate open-source solutions as well as closed-source ones.
I'm the least surprised that it was Terry Pratchett that made himself a sword.
Oh, I dunno. If you'd asked me without me seeing the story, I'd probably have gone for John Scalzi.
Bring in alligators to eat the opossum, and then in the winter, they'll all freeze to death.
A friend keeps singing a song to her kid about an alligator going snap. I keep telling her she'll need liquid oxygen to achieve that, but I don't think she's got the message.
Has anyone else noticed that at the end of the meta keywords for her page is "Edible Computer Chips"?
Well, that was a mistake. She means "Edible Computer Chips(R)"
Choice quote:
Webwasher flags it as something I should not be looking at.
I wonder what Dr. Allen is up to...
Said website is perfectly SFW. Webwasher is clearly a heap of shit.