I always wondered why so many of their apps (MS Word needs write access to win32/ ???) require that you let them touch (not just read) files outside of your home directory
I've never had a problem with an MS app. I'm currently running Win2K using a user with no admin priveleges, and other than installing updates all the MS apps I have on my system are able to work just fine. MS Word does _not_ need write access to any specific directory. I'm not sure what exactly you mean by "win32/" as there's no such directory, maybe you mean %windir% (aka "c:\winnt" on most installations) or possibly %windir%\system32, but whichever you mean, Word does not need to be able to write to it.
That's not to say I have no problems at all. I need to run several programs as administrator to make them work. CD writing software seems to be the main culprit (and yes, I do have the local policy set that's supposed to allow that, it didn't work). I also have a parallel port scanner that won't work properly from a user account (it needs to write to "%windir%\twain32" for some reason). But all of these problems involve non-MS software, so I'm not sure whether I can blame MS for them or not.
That's a good idea. Another good one along similar lines, is if you have a sentence with "who", try to replace "who" with "he". You might have to restructure it a little.
So, "who did it?" becomes "he did it" and "who's there" becomes "he is there."
Now, you might find you have a sentence that becomes completely wrong. "I should send this to who?" becomes "I should send this to he?". Obviously you want "I should send this to him?" instead. This is a dead giveaway that you should have used "whom" instead of "who".
Re:I don't think this will work
on
Lego Logic Gates
·
· Score: 1
So, how does he address the issue of fan out? As far as I can tell, if changing an input will cause a computer built from mechanical lego to do N transitions, then the mechanical force for all these N transitions would have to be applied at the input.
This doesn't scale and will hit the limits very soon.
He doesn't. But it would be trivial to add force amplifiers.
This page is about torque amps, which aren't quite what you want here, but the design might serve as a starting point.:)
Not linking. Citing in offline sources. It's too difficult for most users with the current process, first you have to find a version of the page that seems authoritative (thus introducing your own bias into the selection process), then quote it with its version ID; they want to make a collection of vetted pages that would be suitable as a base to make citations from, and which anyone can easily access by clicking a big "citable version" link near the top of the page.
It's a use of the already existing features of wikis in general, yes, but a desirable one.
Well, cygwin normally needs to be installed with registry entries to tell it where to find its root directory (and potentially other "mounted fs"'s, too). Presumably this CD has a hacked version that doesn't need those entries (perhaps by just using the path to cygwin1.dll returned by GetModulePath() to derive it automatically for each new program started?).
There are two drivers, the colinux driver (linux.sys) and the tap-win32 driver (for the virtual network interface to the guest system).
I don't think any reboots are necessary, however it needs to have write access to %windir%\system32 (to install the driver files) and to the HKEY_LOCAL_MACHINE registry section, which typically means you need to be logged in as administrator.
You also need a disk image of the root to be booted, although I guess this could be a file on the CD.
For me, it seems to actually be faster to download movies from bittorrent (I often get about 30K/sec on my connection, which tends to come out at about 8-12 hours downloading for a feature length movie) than to rip & reencode (which seems to take 14-18 hours for a feature length movie on my 450MHz celeron).
Not to mention I don't have the inconvenience of having to wait around for the disks to be delivered, etc.
Plus there's the chance to download screeners before the rental DVD is released.
So even if netflix did operate in my area, I don't think I'd use them.
I've replied to and discussed this suggestion before. It doesn't work, and here's why:
1. You need to have copyright on some of the content that has been decrypted to sue under the DMCA. The workaround is to add some of your own content to the file.
2. Only the copyright holder can sue, so anyone else sharing the file you modified is out of luck.
3. Adding your own content establishes that you intended to do this, rather than that you're sharing as a side effect of having downloaded. This potentially makes any charges against you more serious.
4. Even if you do succeed, what would happen is damages would be awarded based on the commercial value of the two infringed copyrights. The MPAA'd get some huge award, probably in the region of $10,000, whereas you'd get something tiny. Maybe $100. It doesn't help you very much, does it?
It's easy to show that suprnova's administrators intend the site to be used for copyright material, seeing it is they who created the categories on the site, into which visitors subsequently post links.
It's hard to see how categories like "TV Shows/24" (to pick a category from the top of the list) are intended to be used other than for copyright infringing purposes.
I believe this is called "contributory infringement" in many jurisdictions. However, I'd be unsurprised (to put it mildly) to find that suprnova is hosted in a country that doesn't have such a legal concept.
You subversive, you. You're obviously planning to write a web site that will encourage the use of magic to send people to the other side of the galaxy where they'll have interesting adventures with hot alien chicks.
THIS MUST BE STOPPED. NOW.
We have your IP address. Step away from the computer and wait for the thought police to arrive.
Actually, in the European release of the movie all references to Taco Bell were changed to Pizza Hut. Guess because Pizza Hut is so much bigger than Taco Bell here.
Not in the version I saw (on UK television, not sure about DVD releases, etc.) it wasn't.
I still suspect VNC on a USB key or CD might be easier, and the difference between forwarding X and using VNC isn't that much in my experience.
That's not my experience. As part of a recent OS upgrade to my office's server, I installed KDE3 over XF86 with TightVNC using a 100Mbit LAN (hubbed rather than switched, but with only 3 workstations using it primarily for internet access and IMAP e-mail it doesn't make a huge difference). The performance was abysmal. Redraws when new applications were opened typically took in excess of a second. I scrapped it and installed Cygwin & X on my desktop and the performance is now substantially better. And I'm not even using anything like NX, this is just straightforward traditional X11.
Sure, but (a) I rounded up the length of the collection by a not insignificant amount, (b) it's more usually 80x10^6x1024 (I've no idea why, but that's what I've been seeing the most of lately), and (c) we're talking about a portable music player, here. I reckon on your average set of headphones, 99% of people wouldn't be able to tell 160kbit from ~128kbit VBR. Maybe you're one of the few who can. I don't think I am, at least.
[Sorry, hit post before I was ready. Here's the rest]
It also has to do with firefox's cross platform XUL interface, IE has a much lighter interface that just calls OS functions, firefox can't do this so easily as it would become far less portable..
I've seen plenty of lightweight, platform independent user interfaces. For instance, applications written using FLTK frequently have memory usages of less than 2Mb on my system, and are capable of being built without modification on Windows, UNIX/X11 and OSX. Admittedly I don't think FLTK is mature enough yet to support an app as complicated as Firefox, but it exemplifies an approach which I believe the firefox authors should have taken.
XUL is interesting, but I'm not sure it was a wise idea.
Also, firefox has a far more complete rendering engine, including proper support for png and much better css/css2 support,
I've added proper PNG support to one of my applications lately. It added about 400K to the code size, which wasn't loaded into memory unless a PNG image was loaded. Runtime memory usage did not increase at all, except in the case where an image that was previously 24 bits per pixel became 32 due to the presence of an alpha layer. Which hardly accounts for such a large difference.
The better CSS support is interesting, but comparing current firefox memory usage to that of older versions of mozilla before this support was added, I'm pretty sure most of the memory usage is going on XUL.
Which is a waste. It provides features that most of us don't need.
This has a lot to do with large chunks of IE already being resident in memory as part of the OS.
No, actually that's not true. While I'll grant that this argument works in favour of why IE seems to start so much faster than Moz-based browsers do, if anything, this probably reduces the amount of memory that IE is using, as those 'part of the OS' bits are probably shared with other processes than "iexplore.exe" ones -- with the result that the actual increase in memory usage from having IE running compared to not running is probably less than that.
But that doesn't change this fact: the figure I quoted INCLUDED the shared sections. So even if IE were the only process you were running, this is the worst case usage.
I note that he has linked to a site "247news.net" that looks at first glance like an at least vaguely official news site, but if you look more closely is run by Sollog himself. He has links back to his other sites at the top.
His second postscript is headed "PSS". And incites people to file a blatantly false accusation of criminal behaviour. I suspect this is a crime in itself.
My understanding is that this in fact a planned feature of wikipedia for the near future; it is intended to get around the "inability to cite" problem by allowing people to reference an article in, say, "Wikipedia 1.0", which will be available with prominent links, whereas general visitors to the site will see "Wikipedia Current" unless they request otherwise.
Somehow I get the feeling that this entire thing is some kind of satirical setup. From the guy's web site (www.patenting-art.com, headline: "Artists, entertainers, lawyers, executives and investors: learn how to use the patent system to protect your art and entertainment creations and methods") to the blatant ridiculousness of the idea that patents can be an effective way of protecting software by themselves.
By my calculation (based on average CD length of 55 minutes, don't know if this is accurate for your collection) it should fit.
1200 x 55 x 60 is a little under 4 million seconds x 160 KBps = 640 Gbits
= 80 Gbytes
It'll be tight though. You might have to drop some of your least-favourite tracks (with 1200 CDs, I'm sure there are some on there that you actually don't like, right?)
So far, only perpetual motion machines and immortality have shown to be impossible. And even then, we might prove otherwise someday.
Actually, no, there are plenty of other things that have been shown to be impossible. For instance, I have seen and understood the proofs that neither of the following can exist:
- a computer program which can take an arbitrary computer program and input and tell whether it will ever terminate
- a formal mathematical system that can prove every true statement to be true without also proving some false statements to be true
I'm pretty sure (although I haven't seen a proof) that it can also be proved that no such thing as a secure P2P network can be developed (at least not without architectural changes to the way the Intenet works, and that isn't likely at this point in time), because of the fundamental fact which keeps getting in the way that you need to be able to connect to a source and download data from it and then transform that data (somehow) into the data you wanted to download. And if you can do it, so can the *AA. And if they can do it, they can tell who they downloaded it from and prosecute them, because you cannot transfer data on the Internet without both a source and destination address, and if they can find your source address they can find you.
Legal loopholes are the only way to avoid prosecution, if you want to play this game. That, or just don't get noticed, which is going to get tougher all the time.
Slashdot Mirror
I always wondered why so many of their apps (MS Word needs write access to win32/ ???) require that you let them touch (not just read) files outside of your home directory
I've never had a problem with an MS app. I'm currently running Win2K using a user with no admin priveleges, and other than installing updates all the MS apps I have on my system are able to work just fine. MS Word does _not_ need write access to any specific directory. I'm not sure what exactly you mean by "win32/" as there's no such directory, maybe you mean %windir% (aka "c:\winnt" on most installations) or possibly %windir%\system32, but whichever you mean, Word does not need to be able to write to it.
That's not to say I have no problems at all. I need to run several programs as administrator to make them work. CD writing software seems to be the main culprit (and yes, I do have the local policy set that's supposed to allow that, it didn't work). I also have a parallel port scanner that won't work properly from a user account (it needs to write to "%windir%\twain32" for some reason). But all of these problems involve non-MS software, so I'm not sure whether I can blame MS for them or not.
That's a good idea. Another good one along similar lines, is if you have a sentence with "who", try to replace "who" with "he". You might have to restructure it a little.
So, "who did it?" becomes "he did it" and "who's there" becomes "he is there."
Now, you might find you have a sentence that becomes completely wrong. "I should send this to who?" becomes "I should send this to he?". Obviously you want "I should send this to him?" instead. This is a dead giveaway that you should have used "whom" instead of "who".
So, how does he address the issue of fan out? As far as I can tell, if changing an input will cause a computer built from mechanical lego to do N transitions, then the mechanical force for all these N transitions would have to be applied at the input.
:)
This doesn't scale and will hit the limits very soon.
He doesn't. But it would be trivial to add force amplifiers.
This page is about torque amps, which aren't quite what you want here, but the design might serve as a starting point.
...LegoLAN ..and imagine a beowulf cluster of those!
What, you mean like this one ?
Not linking. Citing in offline sources. It's too difficult for most users with the current process, first you have to find a version of the page that seems authoritative (thus introducing your own bias into the selection process), then quote it with its version ID; they want to make a collection of vetted pages that would be suitable as a base to make citations from, and which anyone can easily access by clicking a big "citable version" link near the top of the page.
It's a use of the already existing features of wikis in general, yes, but a desirable one.
However, an X server behaves as client to an XDM server. :)
Well, cygwin normally needs to be installed with registry entries to tell it where to find its root directory (and potentially other "mounted fs"'s, too). Presumably this CD has a hacked version that doesn't need those entries (perhaps by just using the path to cygwin1.dll returned by GetModulePath() to derive it automatically for each new program started?).
There are two drivers, the colinux driver (linux.sys) and the tap-win32 driver (for the virtual network interface to the guest system).
I don't think any reboots are necessary, however it needs to have write access to %windir%\system32 (to install the driver files) and to the HKEY_LOCAL_MACHINE registry section, which typically means you need to be logged in as administrator.
You also need a disk image of the root to be booted, although I guess this could be a file on the CD.
For me, it seems to actually be faster to download movies from bittorrent (I often get about 30K/sec on my connection, which tends to come out at about 8-12 hours downloading for a feature length movie) than to rip & reencode (which seems to take 14-18 hours for a feature length movie on my 450MHz celeron).
Not to mention I don't have the inconvenience of having to wait around for the disks to be delivered, etc.
Plus there's the chance to download screeners before the rental DVD is released.
So even if netflix did operate in my area, I don't think I'd use them.
I've replied to and discussed this suggestion before. It doesn't work, and here's why:
1. You need to have copyright on some of the content that has been decrypted to sue under the DMCA. The workaround is to add some of your own content to the file.
2. Only the copyright holder can sue, so anyone else sharing the file you modified is out of luck.
3. Adding your own content establishes that you intended to do this, rather than that you're sharing as a side effect of having downloaded. This potentially makes any charges against you more serious.
4. Even if you do succeed, what would happen is damages would be awarded based on the commercial value of the two infringed copyrights. The MPAA'd get some huge award, probably in the region of $10,000, whereas you'd get something tiny. Maybe $100. It doesn't help you very much, does it?
It's easy to show that suprnova's administrators intend the site to be used for copyright material, seeing it is they who created the categories on the site, into which visitors subsequently post links.
It's hard to see how categories like "TV Shows/24" (to pick a category from the top of the list) are intended to be used other than for copyright infringing purposes.
I believe this is called "contributory infringement" in many jurisdictions. However, I'd be unsurprised (to put it mildly) to find that suprnova is hosted in a country that doesn't have such a legal concept.
You subversive, you. You're obviously planning to write a web site that will encourage the use of magic to send people to the other side of the galaxy where they'll have interesting adventures with hot alien chicks.
THIS MUST BE STOPPED. NOW.
We have your IP address. Step away from the computer and wait for the thought police to arrive.
Do you think Taco Bell actually paid to have S.B. announce that in the bland and flavourless future all restaurants would be run by them?
If I'd seen that script, there's no way I'd have paid for a placement.
Actually, in the European release of the movie all references to Taco Bell were changed to Pizza Hut. Guess because Pizza Hut is so much bigger than Taco Bell here.
Not in the version I saw (on UK television, not sure about DVD releases, etc.) it wasn't.
I still suspect VNC on a USB key or CD might be easier, and the difference between forwarding X and using VNC isn't that much in my experience.
That's not my experience. As part of a recent OS upgrade to my office's server, I installed KDE3 over XF86 with TightVNC using a 100Mbit LAN (hubbed rather than switched, but with only 3 workstations using it primarily for internet access and IMAP e-mail it doesn't make a huge difference). The performance was abysmal. Redraws when new applications were opened typically took in excess of a second. I scrapped it and installed Cygwin & X on my desktop and the performance is now substantially better. And I'm not even using anything like NX, this is just straightforward traditional X11.
Sure, but (a) I rounded up the length of the collection by a not insignificant amount, (b) it's more usually 80x10^6x1024 (I've no idea why, but that's what I've been seeing the most of lately), and (c) we're talking about a portable music player, here. I reckon on your average set of headphones, 99% of people wouldn't be able to tell 160kbit from ~128kbit VBR. Maybe you're one of the few who can. I don't think I am, at least.
[Sorry, hit post before I was ready. Here's the rest]
It also has to do with firefox's cross platform XUL interface, IE has a much lighter interface that just calls OS functions, firefox can't do this so easily as it would become far less portable..
I've seen plenty of lightweight, platform independent user interfaces. For instance, applications written using FLTK frequently have memory usages of less than 2Mb on my system, and are capable of being built without modification on Windows, UNIX/X11 and OSX. Admittedly I don't think FLTK is mature enough yet to support an app as complicated as Firefox, but it exemplifies an approach which I believe the firefox authors should have taken.
XUL is interesting, but I'm not sure it was a wise idea.
Also, firefox has a far more complete rendering engine, including proper support for png and much better css/css2 support,
I've added proper PNG support to one of my applications lately. It added about 400K to the code size, which wasn't loaded into memory unless a PNG image was loaded. Runtime memory usage did not increase at all, except in the case where an image that was previously 24 bits per pixel became 32 due to the presence of an alpha layer. Which hardly accounts for such a large difference.
The better CSS support is interesting, but comparing current firefox memory usage to that of older versions of mozilla before this support was added, I'm pretty sure most of the memory usage is going on XUL.
Which is a waste. It provides features that most of us don't need.
This has a lot to do with large chunks of IE already being resident in memory as part of the OS.
No, actually that's not true. While I'll grant that this argument works in favour of why IE seems to start so much faster than Moz-based browsers do, if anything, this probably reduces the amount of memory that IE is using, as those 'part of the OS' bits are probably shared with other processes than "iexplore.exe" ones -- with the result that the actual increase in memory usage from having IE running compared to not running is probably less than that.
But that doesn't change this fact: the figure I quoted INCLUDED the shared sections. So even if IE were the only process you were running, this is the worst case usage.
I note that he has linked to a site "247news.net" that looks at first glance like an at least vaguely official news site, but if you look more closely is run by Sollog himself. He has links back to his other sites at the top.
D'oh!
His second postscript is headed "PSS". And incites people to file a blatantly false accusation of criminal behaviour. I suspect this is a crime in itself.
My understanding is that this in fact a planned feature of wikipedia for the near future; it is intended to get around the "inability to cite" problem by allowing people to reference an article in, say, "Wikipedia 1.0", which will be available with prominent links, whereas general visitors to the site will see "Wikipedia Current" unless they request otherwise.
Somehow I get the feeling that this entire thing is some kind of satirical setup. From the guy's web site (www.patenting-art.com, headline: "Artists, entertainers, lawyers, executives and investors: learn how to use the patent system to protect your art and entertainment creations and methods") to the blatant ridiculousness of the idea that patents can be an effective way of protecting software by themselves.
Sorry, I don't buy it.
This guy's just making a point.
Sounds like Windows to me! :-)
:)
;-) <--- NOTE SMILEY. :) :)
It's a joke, laugh.
It's a sure sign that the moderators are getting stupid when you have to point out the jokes to them.
By my calculation (based on average CD length of 55 minutes, don't know if this is accurate for your collection) it should fit.
1200 x 55 x 60 is a little under 4 million seconds
x 160 KBps = 640 Gbits
= 80 Gbytes
It'll be tight though. You might have to drop some of your least-favourite tracks (with 1200 CDs, I'm sure there are some on there that you actually don't like, right?)
So far, only perpetual motion machines and immortality have shown to be impossible. And even then, we might prove otherwise someday.
Actually, no, there are plenty of other things that have been shown to be impossible. For instance, I have seen and understood the proofs that neither of the following can exist:
- a computer program which can take an arbitrary computer program and input and tell whether it will ever terminate
- a formal mathematical system that can prove every true statement to be true without also proving some false statements to be true
I'm pretty sure (although I haven't seen a proof) that it can also be proved that no such thing as a secure P2P network can be developed (at least not without architectural changes to the way the Intenet works, and that isn't likely at this point in time), because of the fundamental fact which keeps getting in the way that you need to be able to connect to a source and download data from it and then transform that data (somehow) into the data you wanted to download. And if you can do it, so can the *AA. And if they can do it, they can tell who they downloaded it from and prosecute them, because you cannot transfer data on the Internet without both a source and destination address, and if they can find your source address they can find you.
Legal loopholes are the only way to avoid prosecution, if you want to play this game. That, or just don't get noticed, which is going to get tougher all the time.