Exactly the same thing has been said about technologies like e-books and print-on-demand presses allowing writers to get their content to the public without going through traditional presses.
I don't see the public queueing up to buy these products.
There has to be a filter on quality, and the record labels / presses have traditionally provided this. Of course, other organisations could step in to provide a similar framework for these 'new channels', but without there being a viable business model for them, they will be unstable at best. Building the public's trust in your filtering abilities will take a long time.
Actually, he's almost certainly right. Ask any patent attorney - they'll tell you that a patent that contains no actual invention (i.e. something that is possible using the method they suggest that wasn't before) cannot be enforced. If what is said is true (I read the Unisys patent a while ago, but haven't read the IBM one), then they don't have an enforceable patent.
He's free to say that there is a concensus without providing links. The consensus could well have been between his two lawyer friends.
No. JPEG at max quality looks perfect to the human eye, but it still has differences from the original image. Lossy compression should be avoided in situations where images are going to be decoded and recoded many times, as these errors build up to the point where they can become noticeable.
Also: make sure your PNG encoder is configured correctly. In most cases you want to be using the 'adaptive' filter.
Just make the software decode GIFs and replace them with PNGs. The user doesn't notice anything except that the site gets faster.
The user (of graphics editing software that did this) would then have to fix all the references in their documents to the new filename, which they would see as an annoying pointless task. I strongly recommend against doing this to them, because they will see your software as broken and switch to something else.
Compatibility. A huge number of existing web sites still use GIF as their primary image format. We need to be able to produce software that can manipulate these images if we want any hope of penetrating the web authoring market. This has prompted many workarounds in the past (such as libungif, a piece of software that produces GIF files without using the patented algorithm -- but unfortunately this means not having any compression) which will become obsolete once all patent issues have been cleared up.
There's actually a solution to this. You require users to certify that they own the copyright to the content they distribute and grant permission to the software to do so before allowing it to be distributed.
As this is as close to being sure that any software is going to be able to get (how on Earth can software check that a random file whose format it doesn't understand is the property of its user?), it should be permissible.
Unfortunately, having contributed to this discussion, I can't. I like to point out good articles when I see them, because they can often be buried under the pile of crap that is the average slashdot comment.
Those graphs are for the memory usage of a web cache. It delays cleaning up expired objects until it is close to running out of storage space for them, so this is the kind of graph you would expect. I don't believe those troughs correspond to a restart, there's certainly no evidence of that on this page.
So if one contributer feels 2+2=4 and another feels 2+2=6
So then according to the "neutral viewpoint" on the issue, the entry should be 2+2=5. "Neutral viewpoint" is just meaningless jargon, what matters is being accurate, knowledgeable and correct, not establishing some phantom neutrality.
The use of a "neutral viewpoint" (a literary convention which is, unfortunately, unattainable in the real world) is appropriate where there are two contradictory viewpoints which are either both true, or it cannot be established which is true.
Your example is rather extreme, but if the case was that nobody could determine with any authority whether 2+2 was 4 or 6, then obviously wikepedia should include both possible answers along with the most important arguments for and against each one. This is what they're talking about. They wouldn't decide that 2+2=5 was more appropriate and use that, and I can't see any single case where they've done this kind of thing.
Unless you can point me to examples where they've included incorrect statements on uncontroversial topics that haven't been quickly corrected, I'll take your viewpoint as unnecessarily biased against them.
I've not spent much time reading Everything2, but I've never been as impressed by its authority as I have by wikipedia.
For instance, compare the Everything2 page on Water (I can't link to it, for some reason the site uses HTTP POST for identifying which article you want) to the the wikepedia one.
I find the wikipedia article much more clearly structured, more informative, and I think more authoratitive. Although only the Everything2 article contains an ASCII-art rendering of the Kanji character for water.
Only Americans call that a pound sign. In Britain, a pound sign is a fancy L with a loop on the top and a crossbar approximately one third the way up. It has the ISO-8859-1 character code 0xA3.
I've always thought this was sensible. The benefits are clear:
* Less CPU power is required per person -- an average user has about 80% CPU idle time. Serving 4 users probably only requires about 50% more power than serving 1 user.
* Less memory is required, as most of the time users will all be running the same software, enabling shared memory usage and shared disk cache (n.b., memory is currently the most expensive item in a low-cost computer system).
* Fewer machines -> less administration time to keep them all running.
All that was needed was mainstream hardware & OS support. I hadn't noticed that USB keyboards & mice were the final thing to fit into place to enable this with readily available hardware, nor that Linux supported the necessary software configuration (or is this using a modified kernel?)
Isn't that, like, a little extreme? They could get nearly 40 hours worth of video on that. I wouldn't have thought that much footage would have been shot during the making of 3 films.
ondon police cars are fitted with cameras that automatically scan car registrations and will notify the PC if it spots a vehicle with either outstanding tickets or no insurance
They might tell you this, but this is actually impossible as there is no existent database that lists all insured vehicles. There cannot be, as "drive any vehicle" insurance is available, and you then don't have to notify them of the registration details of any vehicles you drive but do not own yourself.
If I wanted to be taken seriously I would disclose my method.
Exactly my point: they haven't answered the questions of which applications were included; they haven't disclosed a method (and the statistics that are described as 'easily accessible on their web page' in the article seem to be buried so deeply that I can't find them), so the chances are the method they used is one that wouldn't be taken seriously if it were known.
Cyber is an idiot prefix/word for soundbites, the fear mongers at DHS, and William Gibson
Amen. Now, please tell CERT to stop using it in their advisories. It makes them sound incompetent.
It also doesn't mean what they think it means:
Cybernetics
n. The theoretical study of communication and control processes in biological, mechanical, and electronic systems, especially the comparison of these processes in biological and artificial systems.
[From Greek kubernts, governor, from kubernn, to govern.]
Until telnetd is totally removed (not just turned off) from Linux, Linux will not be secure. There are just too many exploits involving telnet to take Linux seriously.
What's wrong with having insecure features that are disabled by default? Many people operate in secure environments where such features (which they need for interoperability reasons) offer a "good enough" degree of security. There's no point in making these people's life harder.
Slashdot Mirror
Exactly the same thing has been said about technologies like e-books and print-on-demand presses allowing writers to get their content to the public without going through traditional presses.
I don't see the public queueing up to buy these products.
There has to be a filter on quality, and the record labels / presses have traditionally provided this. Of course, other organisations could step in to provide a similar framework for these 'new channels', but without there being a viable business model for them, they will be unstable at best. Building the public's trust in your filtering abilities will take a long time.
Actually, he's almost certainly right. Ask any patent attorney - they'll tell you that a patent that contains no actual invention (i.e. something that is possible using the method they suggest that wasn't before) cannot be enforced. If what is said is true (I read the Unisys patent a while ago, but haven't read the IBM one), then they don't have an enforceable patent.
He's free to say that there is a concensus without providing links. The consensus could well have been between his two lawyer friends.
Because, frankly, there was never any reason to do so. Can you point out one practical benefit that would have been gained had they done this?
No. JPEG at max quality looks perfect to the human eye, but it still has differences from the original image. Lossy compression should be avoided in situations where images are going to be decoded and recoded many times, as these errors build up to the point where they can become noticeable.
Also: make sure your PNG encoder is configured correctly. In most cases you want to be using the 'adaptive' filter.
You could've done this anyway. You're free to use patent-encumbered technologies for "research purposes".
BTW: LZW is a very simple and 'nice' algorithm, and there are a lot of web pages that give very clear instructions on how to implement it. Google it.
Because Unisys only ever threatened users of the GIF format with legal action; they considered other uses of LZW as acceptable.
Just make the software decode GIFs and replace them with PNGs. The user doesn't notice anything except that the site gets faster.
The user (of graphics editing software that did this) would then have to fix all the references in their documents to the new filename, which they would see as an annoying pointless task. I strongly recommend against doing this to them, because they will see your software as broken and switch to something else.
IE's support for PNGs only lacks in features that GIFs don't support anyway, so there's no disadvantage to using PNG over GIF from that front.
Compatibility. A huge number of existing web sites still use GIF as their primary image format. We need to be able to produce software that can manipulate these images if we want any hope of penetrating the web authoring market. This has prompted many workarounds in the past (such as libungif, a piece of software that produces GIF files without using the patented algorithm -- but unfortunately this means not having any compression) which will become obsolete once all patent issues have been cleared up.
There's actually a solution to this. You require users to certify that they own the copyright to the content they distribute and grant permission to the software to do so before allowing it to be distributed.
As this is as close to being sure that any software is going to be able to get (how on Earth can software check that a random file whose format it doesn't understand is the property of its user?), it should be permissible.
Aha, thanks. When I found it the address in my URL bar was just 'everything2.com/index.pl'.
Unfortunately, having contributed to this discussion, I can't. I like to point out good articles when I see them, because they can often be buried under the pile of crap that is the average slashdot comment.
Those graphs are for the memory usage of a web cache. It delays cleaning up expired objects until it is close to running out of storage space for them, so this is the kind of graph you would expect. I don't believe those troughs correspond to a restart, there's certainly no evidence of that on this page.
So if one contributer feels 2+2=4
and another feels 2+2=6
So then according to the "neutral viewpoint" on the issue, the entry should be 2+2=5. "Neutral viewpoint" is just meaningless jargon, what matters is being accurate, knowledgeable and correct, not establishing some phantom neutrality.
The use of a "neutral viewpoint" (a literary convention which is, unfortunately, unattainable in the real world) is appropriate where there are two contradictory viewpoints which are either both true, or it cannot be established which is true.
Your example is rather extreme, but if the case was that nobody could determine with any authority whether 2+2 was 4 or 6, then obviously wikepedia should include both possible answers along with the most important arguments for and against each one. This is what they're talking about. They wouldn't decide that 2+2=5 was more appropriate and use that, and I can't see any single case where they've done this kind of thing.
Unless you can point me to examples where they've included incorrect statements on uncontroversial topics that haven't been quickly corrected, I'll take your viewpoint as unnecessarily biased against them.
I've not spent much time reading Everything2, but I've never been as impressed by its authority as I have by wikipedia.
For instance, compare the Everything2 page on Water (I can't link to it, for some reason the site uses HTTP POST for identifying which article you want) to the the wikepedia one.
I find the wikipedia article much more clearly structured, more informative, and I think more authoratitive. Although only the Everything2 article contains an ASCII-art rendering of the Kanji character for water.
He speaks the truth! :)
Only Americans call that a pound sign. In Britain, a pound sign is a fancy L with a loop on the top and a crossbar approximately one third the way up. It has the ISO-8859-1 character code 0xA3.
I've always thought this was sensible. The benefits are clear:
* Less CPU power is required per person -- an average user has about 80% CPU idle time. Serving 4 users probably only requires about 50% more power than serving 1 user.
* Less memory is required, as most of the time users will all be running the same software, enabling shared memory usage and shared disk cache (n.b., memory is currently the most expensive item in a low-cost computer system).
* Fewer machines -> less administration time to keep them all running.
All that was needed was mainstream hardware & OS support. I hadn't noticed that USB keyboards & mice were the final thing to fit into place to enable this with readily available hardware, nor that Linux supported the necessary software configuration (or is this using a modified kernel?)
Do you mean GBP 44.99? (Slashdot eats pound signs)
Isn't that, like, a little extreme? They could get nearly 40 hours worth of video on that. I wouldn't have thought that much footage would have been shot during the making of 3 films.
There'll be 300 virus infested copies in half an hour. Post a magnet link, next time.
ondon police cars are fitted with cameras that automatically scan car registrations and will notify the PC if it spots a vehicle with either outstanding tickets or no insurance
They might tell you this, but this is actually impossible as there is no existent database that lists all insured vehicles. There cannot be, as "drive any vehicle" insurance is available, and you then don't have to notify them of the registration details of any vehicles you drive but do not own yourself.
If I wanted to be taken seriously I would disclose my method.
Exactly my point: they haven't answered the questions of which applications were included; they haven't disclosed a method (and the statistics that are described as 'easily accessible on their web page' in the article seem to be buried so deeply that I can't find them), so the chances are the method they used is one that wouldn't be taken seriously if it were known.
Cyber is an idiot prefix/word for soundbites, the fear mongers at DHS, and William Gibson
Amen. Now, please tell CERT to stop using it in their advisories. It makes them sound incompetent.
It also doesn't mean what they think it means:
Cybernetics
n. The theoretical study of communication and control processes in biological, mechanical, and electronic systems, especially the comparison of these processes in biological and artificial systems.
[From Greek kubernts, governor, from kubernn, to govern.]
(Source: dictionary.com)
Until telnetd is totally removed (not just turned off) from Linux, Linux will not be secure. There are just too many exploits involving telnet to take Linux seriously.
What's wrong with having insecure features that are disabled by default? Many people operate in secure environments where such features (which they need for interoperability reasons) offer a "good enough" degree of security. There's no point in making these people's life harder.