You may be thinking of CLISP. It was non-GPL and used libreadline, Stallman pointed out the GPL issue, a public e-mail exchanged ensued, and Haible eventually agreed with Stallman and decided to go GPL with CLISP.
I work for a major financial company, and our policy is: you give out your card details, your problem.
If you are from the US, you must be very confused.
According to the FTC, if your credit card details are used without your approval, you are liable for absolutely nothing, and you have 60 days to report the fraud charges. From the FTC website,
Also, if the loss involves your credit card number, but not the card itself, you have no liability for unauthorized use.
I imagine it works in a similar a way in many other countries. I quick search wasn't giving me any non-US info though, so I don't have any links.
I completely agree. I was about to quote the YouTube TOS, but I now see that they have actually made a major change in the last couple weeks. Here is what it used to say,
the data is intended for real-time viewing and not intended to be copied, stored, permanently downloaded, or redistributed by the user.
Which made it entirely inappropriate for government public domain videos. These videos should be redistributable by anyone. However, check out their new wording in their new TOS,
[...]
Content on the Website is provided to you AS IS for your information and personal use only and may not be downloaded, copied, reproduced, distributed, transmitted, broadcast, displayed, sold, licensed, or otherwise exploited for any other purposes whatsoever without the prior written consent of the respective owners.
[...]
You shall not copy or download any User Submission unless you see a "download" or similar link displayed by YouTube on the YouTube Website for that User Submission.
[...]
So if you have "written" permission of the author or the author has explicitly told YouTube they want a download button, you are allowed to download the video, making YouTube more like file distribution than don't-touch-this video streaming. As long as government videos automatically have this download button appear, I feel a lot more comfortable about it.
But, as you said, they still should not be making a special exception for YouTube. The ideal thing would be the government self-hosting downloadable, public domain videos in a patent unencumbered format, and then let individuals go put them up on YouTube or whatever.
Most wikipedia articles on Film and TV contain stolen content from IMDB, or from labors-of-love fan sites that depend on page views and micro-ad revenue for survival.
I call bullshit on this. Find us an example if one even exists. Wikipedia is very paranoid about these things, so I doubt you can find one. And no, getting information like dates and names are just facts, not subject to copyright.
In fact, I have always seen it the other way around. Ever search for something and find the Wikipedia echo chamber? All kinds of websites copy-paste Wikipedia articles, and then don't indicate the GFDL license its under, which is copyright infringement.
By choosing wikipedia over them you are NOT helping the shows you love.
Bullshit. People probably go to the Wikipedia article because the fan sites suck ass.
Plus, most film and TV wikipedia pages contain spoilers without any warnings
Wikipedia doesn't do disclaimers, not even spoiler disclaimers. If you are worried about spoilers don't read past the first paragraph in the article.
as well as synopses that breach the rights of the creators.
Bullshit.
Joss Wheedon in particular has suffered a lot of cancellations of his work
Right, that's Wikipedia's fault. Bullshit.
By using wikpedia you are not, as well as simply encouraging people to steal from him.
Go read some Encyclopedia Britannica articles. They aren't any better than Wikipedia. For example, Britannica's article on Wikipedia contains some factual errors.
because it only reads as many/dev/random bytes as it needs, preserving entropy.
Actually, your "% 64" operation needlessly tosses 2 bits. So for every 4 bytes you read you are wasting one byte of entropy: at least 2 bytes of entropy per password.
This program prints one of 2^47.6 possible 8 character passwords
Because you don't allow numbers as the first character, its 2^47.4.:-P
And, you can shorten that a bit (I have no idea how to preserve indenting so use your imagination),
import sys
if ( __name__ == "__main__" ): sel = range(65, 91) + range(97, 123) + range(48, 58) r = file('/dev/urandom', 'rb') for i in xrange(8): ch = 62 while (( ch >= 62 ) or (( i == 0 ) and ( ch >= 52 ))): ch = ord(r.read(1)) % 64 sys.stdout.write(chr(sel[ch])) print
This is functionally equivalent. (I don't know any Python, but this much was obvious to me.)
The parent's version is base 74, so the passwords it generates are a bit denser in terms of entropy, i.e. the passwords can be shorter. Plus his is a one-liner.;-)
There used to be a Gaim plugin for Blizzard's Battle.net that I used several years ago. I could talk to people on Warcraft 3, Starcraft, Diablo, and even World of Warcraft (I think).
However, I somehow appeared strangely in the chat lists because people were always IMing me, out of nowhere, thinking I was some kind of official Blizzard tech support. Sometimes I had fun messing with them by telling them go through futile exercises: "Ok, try unplugging your monitor for 30 seconds. No? Ok now try..."
Unfortunately, Blizzard shut down the protocol this used, so it's a dead project.
A possible solution to this mess would be some sort of live streaming P2P protocol, BitTorrent style. A couple people with access to the stream seed the torrent and thousands of people can watch it from anywhere (many with a bit of a delay I imagine).
I don't know of any such technology that exists in a usable form yet. There are some experiments going on, like SwarmPlayer, but I haven't seen anything actually functional.
Now, where does flashy eye candy come in to that picture? It doesn't.
I would say that fancy eye candy leads to less usability. For example, I remember trying out Compiz a couple years ago, but after a few minutes I noticed that all the animation effects were a bit annoying, making me wait for them to complete before continuing. The fancy 3D virtual desktop switching looks really cool, but a boring, instant switch between desktops is much more practical.
it's the act of having to queue at this final gatekeeper and wait for their OK before I can walk past.
As a couple others said, don't even acknowledge the employee at the exit. Just walk out confidently. If they say something, ignore it and keep walking. They don't want to deal with an awkward encounter, so they'll let you go. I do this myself and have not yet had a problem.
From a technical perspective, being decentralized is a major feature for a system. Decentralized systems are generally robust, reliable, and scalable. There is no single point of failure, and the system will be much more resistant to attacks than a centralized system. The cost of the system is also shared by its nodes, so there is no expensive central point, like SSL authorities.
From a human perspective, a web of trust is very similar to human social behavior. For most people, there are a handful of people we absolutely trust. They might be parents, family, spouse, or close friends. This is how the web of trust works in PGP, but you only need to trust them as far as being competent with PGP (properly checking and exchanging keys, etc). The web of trust maps perfectly to our normal behavior and mindset.
Yes, there are trade-offs. For example, centralized systems, at certain scales, will be more efficient than decentralized ones. Each will be vulnerable to different kinds of attacks, like the ones you mention. However, many people think the trade-off favors the web of trust over the alternative.
This is almost what Debian already is. If you do a net install and then only install the base system, you are left with a tiny little system with almost nothing on it. Not even ssh. Then you can apt-get anything you want from there, with apt-get automatically handling (almost) all the dependencies. (This is the way I do it.)
If someone just set up some large meta-packages, like your "media-system" example, that depended on all the desired packages, it would be precisely what you are talking about.
Slashdot Mirror
You may be thinking of CLISP. It was non-GPL and used libreadline, Stallman pointed out the GPL issue, a public e-mail exchanged ensued, and Haible eventually agreed with Stallman and decided to go GPL with CLISP.
You just need to find a buddy and give each other tug jobs
We're still talking about BitTorrent, right?
Don't forget about Freenet. ;-)
It is *not* your right to take them for free anyway
Yes it is. The copyright contract has been violated, so I feel that I have no need to respect my side of the bargain either.
Have you considered dressing up as a minstrel and playing some music?
This should also work if there are children chewing on your wires too.
I work for a major financial company, and our policy is: you give out your card details, your problem.
If you are from the US, you must be very confused.
According to the FTC, if your credit card details are used without your approval, you are liable for absolutely nothing, and you have 60 days to report the fraud charges. From the FTC website,
Also, if the loss involves your credit card number, but not the card itself, you have no liability for unauthorized use.
I imagine it works in a similar a way in many other countries. I quick search wasn't giving me any non-US info though, so I don't have any links.
I completely agree. I was about to quote the YouTube TOS, but I now see that they have actually made a major change in the last couple weeks. Here is what it used to say,
the data is intended for real-time viewing and not intended to be copied, stored, permanently downloaded, or redistributed by the user.
Which made it entirely inappropriate for government public domain videos. These videos should be redistributable by anyone. However, check out their new wording in their new TOS,
[...]
Content on the Website is provided to you AS IS for your information and personal use only and may not be downloaded, copied, reproduced, distributed, transmitted, broadcast, displayed, sold, licensed, or otherwise exploited for any other purposes whatsoever without the prior written consent of the respective owners.
[...]
You shall not copy or download any User Submission unless you see a "download" or similar link displayed by YouTube on the YouTube Website for that User Submission.
[...]
So if you have "written" permission of the author or the author has explicitly told YouTube they want a download button, you are allowed to download the video, making YouTube more like file distribution than don't-touch-this video streaming. As long as government videos automatically have this download button appear, I feel a lot more comfortable about it.
But, as you said, they still should not be making a special exception for YouTube. The ideal thing would be the government self-hosting downloadable, public domain videos in a patent unencumbered format, and then let individuals go put them up on YouTube or whatever.
Most wikipedia articles on Film and TV contain stolen content from IMDB, or from labors-of-love fan sites that depend on page views and micro-ad revenue for survival.
I call bullshit on this. Find us an example if one even exists. Wikipedia is very paranoid about these things, so I doubt you can find one. And no, getting information like dates and names are just facts, not subject to copyright.
In fact, I have always seen it the other way around. Ever search for something and find the Wikipedia echo chamber? All kinds of websites copy-paste Wikipedia articles, and then don't indicate the GFDL license its under, which is copyright infringement.
By choosing wikipedia over them you are NOT helping the shows you love.
Bullshit. People probably go to the Wikipedia article because the fan sites suck ass.
Plus, most film and TV wikipedia pages contain spoilers without any warnings
Wikipedia doesn't do disclaimers, not even spoiler disclaimers. If you are worried about spoilers don't read past the first paragraph in the article.
as well as synopses that breach the rights of the creators.
Bullshit.
Joss Wheedon in particular has suffered a lot of cancellations of his work
Right, that's Wikipedia's fault. Bullshit.
By using wikpedia you are not, as well as simply encouraging people to steal from him.
That doesn't even make sense.
Go read some Encyclopedia Britannica articles. They aren't any better than Wikipedia. For example, Britannica's article on Wikipedia contains some factual errors.
Yes, it's "stably challenged". ;-)
This should have been put in idle, where I would have never seen it.
Judging by the fact that nearly everyone does it so poorly, I would say it's pretty tough and warrants a book on the subject.
because it only reads as many /dev/random bytes as it needs, preserving entropy.
Actually, your "% 64" operation needlessly tosses 2 bits. So for every 4 bytes you read you are wasting one byte of entropy: at least 2 bytes of entropy per password.
This program prints one of 2^47.6 possible 8 character passwords
Because you don't allow numbers as the first character, its 2^47.4. :-P
And, you can shorten that a bit (I have no idea how to preserve indenting so use your imagination),
This is functionally equivalent. (I don't know any Python, but this much was obvious to me.)
Ah, I missed that. Clever! ;-)
The parent's version is base 74, so the passwords it generates are a bit denser in terms of entropy, i.e. the passwords can be shorter. Plus his is a one-liner. ;-)
The parent's version is base 74, so the passwords it generates are a bit denser in terms of entropy, i.e. the passwords can be shorter.
There used to be a Gaim plugin for Blizzard's Battle.net that I used several years ago. I could talk to people on Warcraft 3, Starcraft, Diablo, and even World of Warcraft (I think).
However, I somehow appeared strangely in the chat lists because people were always IMing me, out of nowhere, thinking I was some kind of official Blizzard tech support. Sometimes I had fun messing with them by telling them go through futile exercises: "Ok, try unplugging your monitor for 30 seconds. No? Ok now try ..."
Unfortunately, Blizzard shut down the protocol this used, so it's a dead project.
Burma Shave
A possible solution to this mess would be some sort of live streaming P2P protocol, BitTorrent style. A couple people with access to the stream seed the torrent and thousands of people can watch it from anywhere (many with a bit of a delay I imagine).
I don't know of any such technology that exists in a usable form yet. There are some experiments going on, like SwarmPlayer, but I haven't seen anything actually functional.
There is also Coral Cache: http://go-dl.eve-files.com.nyud.net/media/0902/mittani.mp3
Now, where does flashy eye candy come in to that picture? It doesn't.
I would say that fancy eye candy leads to less usability. For example, I remember trying out Compiz a couple years ago, but after a few minutes I noticed that all the animation effects were a bit annoying, making me wait for them to complete before continuing. The fancy 3D virtual desktop switching looks really cool, but a boring, instant switch between desktops is much more practical.
it's the act of having to queue at this final gatekeeper and wait for their OK before I can walk past.
As a couple others said, don't even acknowledge the employee at the exit. Just walk out confidently. If they say something, ignore it and keep walking. They don't want to deal with an awkward encounter, so they'll let you go. I do this myself and have not yet had a problem.
You are talking about two different subjects that have nothing to do with each other: copyright and privacy.
Apples to oranges. It's nonsense.
From a technical perspective, being decentralized is a major feature for a system. Decentralized systems are generally robust, reliable, and scalable. There is no single point of failure, and the system will be much more resistant to attacks than a centralized system. The cost of the system is also shared by its nodes, so there is no expensive central point, like SSL authorities.
From a human perspective, a web of trust is very similar to human social behavior. For most people, there are a handful of people we absolutely trust. They might be parents, family, spouse, or close friends. This is how the web of trust works in PGP, but you only need to trust them as far as being competent with PGP (properly checking and exchanging keys, etc). The web of trust maps perfectly to our normal behavior and mindset.
Yes, there are trade-offs. For example, centralized systems, at certain scales, will be more efficient than decentralized ones. Each will be vulnerable to different kinds of attacks, like the ones you mention. However, many people think the trade-off favors the web of trust over the alternative.
This is almost what Debian already is. If you do a net install and then only install the base system, you are left with a tiny little system with almost nothing on it. Not even ssh. Then you can apt-get anything you want from there, with apt-get automatically handling (almost) all the dependencies. (This is the way I do it.)
If someone just set up some large meta-packages, like your "media-system" example, that depended on all the desired packages, it would be precisely what you are talking about.