"By the time 2038 comes around, nobody will be using 32 bit CPUs".:)
Seriously though, I think that statement's probably true. People are still using programs from the 60's and 70's but I don't know of anyone using hardware from then. (Excluding that old computer they're trying to restore in Melbourne of course).
You _do_ have the source to everything, right?:)
I stopped using this and changed all my time variables to doubles.
Doesn't that break all the libc interfaces which use time_t? Also, you change CPUs, you'll have to change it back again.
certainly it can't be too hard for someone to do a check
Actually, I'd disagree with that. If it hadn't been a slashdot-related article, I wouldn't have been sure whether I saw it on slashdot, LWN, or somewhere else.
When I've submitted stories, time differences mean that its generally just before the slashdot crew start going through their submissions in the morning. There are about 300-350 submissions, of which 5-10 are accepted. Noone can keep up with that, especially a story from February. Someone else might have posted that one, and, as part of quickies, it wouldn't have been as noticeable.
I checked, and Slashdot search turns up this as part of some quickies.
Re:Still no Mailing Lists archives on Redhat.com!
on
redhat.com Redone
·
· Score: 2
This has been annoying me too. The quote on archive.redhat.com (which is now a link back to the main redhat site) is:
Wondering what happened to the mailing list archive search function?
Now that the quiet period is over, it will be returning soon! Thanks for your patience.
which is a bit out of date.:) (Or redhat has a different definition of "soon" than I do)
The Moongroup one above is only the redhat-list, not the others, and doesn't appear to be arranged in any logical way. (Archive[1-5], of vastly varying sizes). The "how do we do this" page is the same as the original page as well.
Maybe someone should subscribe mail-archive.com to the lists...
I'm not talking about 2 line comments, but the long paragraph type things, when adding an extra sentence means wrapping the text around and then having to fix all the// comments.
To get slightly back on topic, I'm sure GNU Emacs can do this automatically, if I could work out how.:)
Re the nesting, the C9X rationale, which I found while looking for this, says something along the lines of the C89 group considered allowing nested comments, but decided that it was easier on the compiler writers if we all just used #if 0 instead. They also suggest using if (0) {...}, mentioning that most compilers optimise that away to nothing.
The beta keyserver is beta.dcti.org. www.dcti.org gives the same page as www.distributed.net. When was the change made (or is this just another test thing)?
Every player -- including consoles from Sony, Toshiba, and other consumer electronics vendors, as well as software vendors for PCs like WinDVD and ATI DVD -- has its own unique unlock key. Every DVD disc, in turn, has 400 of these 5-byte keys stamped onto the disc. That way, the unlock key from every licensee, be it WinDVD or a Pioneer DV-525 unit, will read the disc.
So does this mean that there can only ever be 400 DVD drive manufacturers? What happens if XYZ Corp starts manufacturing these tomorrow, and I buy one. Can I then not play a DVD that was printed yesterday?
I know one of the honours students has a debian mirror on his machine, and theres a local redhat disssstricution with isos, so most of us get someone in college to burn it for us.
There's an ongoing discussion on news:aus.legal about this. I don't know if that spreads to the US though. Deja gets it. You can ignore the republic thread though.
Sydney Uni's CS department has its first year students use a teaching language called Blue. Theres a sunos version (for the uni servers) and a linux version.
All CS students get given a CD containing the full version of RedHat, plus an installer for a UMSDOS partiaion of linux. The CD also has stuff like Java, Mesa, an updated python and a few other things for 2nd and third year students.
If you run install, it uncompresses from a self extracting zip file ionto c:\linnx, and puts an icon onto the desktop. The first time you run it, It runs a command line mouse/cd inerface (this was before redaht had mouseconfig/linuxconf). Only a few packages are installed, but you can easily install any others from the CD.
It hasn't been updated for a while, so its currently RedHat 5.0 + some updates + a CD you can borrow from the Dept with updated XServers, but I believe that it will be 6.1 next year.
Well, Win'98 has an option to select its Y2K window in the date control panel.
Actually, I went to a talk once where it was pointed out that because of this, you can get strange data corruption. Someone with one setting can create (say) a spreadsheet with 2 digit years, and it all works fine. If this is sent to another person (even in the same organisation) with different settings, then the date will look like it came from another century. If noone else can see it, then you get data corruption. It gets wors if the second person then saves in in aa 4 digit date format...
I submitted this as a story, but its late in America now, so I doubt anyone will get to it while its still Halloween. (Its actually November in AU now, but thats a different matter...)
b) I'm bothered by "off-topic" and "redundant" moderating deep down in the discussions. There is no harm in people straying from the topic when its not in the main thread.
Yeah, me too. What annoys me more is that often the stuff marked redundant has a CID _before_ the other post, which is in the first level, with a score of generally +4 or +5. I tend to ignore posts marked redundant when doing meta-moderation, because of this.
From the posting page:
try to reply to other people comments instead of starting new threads
I really think that redundant shouldn't be an option for moderation. Unless you read every post in posted order (I tend to read in threaded mode), its impossible to work out who said what first, especially if the posts were made within a couple of minutes of each other. Also, saying the same thing twice can be useful, especially if both comments are buried at the bottom of a large thread.
just like although you can't patent gravity you can patent useful descriptions of gravity.
Can you though? Assume that Newton discovered gravity and his laws of motion today. What could he have patented?
1. Gravity itself - (hopefully) obviously not. As others have mentioned, there's always been gravity, etc, etc.
2. The descriptions (Gravity is...) - Do the same arguements apply to this as above?
3. His laws of motion. (Every action has an equal and opposite reaction, etc) - Can something like this be patented? Its a "useful description", which is (was) non obvious, but how could you infringe against this? Would someone who made, say, shock absorbers for cars be infringing this patent?
4. The mathematical equations - Since (IIRC) algorithms can be patented, could this be patented? ("A method for determining the force due to gravity between two objects"). Would this mean that anyone who uses this equation have to licence it in the same way RSA needs to be licenced?
5. Newton's cradle - this is obviously a physical thing, which probably could be patented. It can be used to "describe" gravity/forces in an easy-to-understand manner, so could this be patented?
Won't this just lead to the end of the Encyclopedia Brittanica?
Maybe. But, remember the speed of the internet, and the number of people likely to be using this at a time, and the speed of most people's access, etc, etc. Think large movies/pictures/etc.
If its just for occasional access, then I'd use this. But if I was a {uni,library,masters/phd student}, then I'd buy the CD (on a site license - I seem to recall they have these), and get faster access for a reasonably minimal price.
If you do not know your administrator account password, you will have to completely reinstall Windows NT because eventually you'll need to have access to this account.
Actually, this isn't true. A linux boot disk can be used to change the administrator password. Do read the warnings though.
The best way to avoid this dilemma is to immediately add your personal user account to the administrators local group of the system. This will make your main user account an administrator of the system, sparing you from heartaches and time later.
No comment on this one....
Can you get an equivalent of su for nt, and run the GUI apps by typing in a console?
Keyword ratings just don't work. If you specify the keywords with too much detail, each web page becomes 10 K of keywords, and 1 K of content. If the list of keywords is not ridgidly fixed, you also end up with "the chicken problem", where a hard core sex site is rated the same as a cooking site, because the both have the keyword "breast". The fact that one of them refers to chicken breasts is not an issue to someone blocking keywords.
If you don't permit enough detail, then things which shouldn't get through do.
For a set of good examples of this, using RSAC to prove the point, see here, but specifically this link, which rates both Alex's Haley's Roots and a pornographic, racist novel using RSACi, and finds that they both have to be given almost the same rating.
I think that this sort of thing is a good idea, but there are some important issues:
1. What about people connecting through IPMasq/Proxies, or using dynamic IPs? I connect to the net through my uni (forced proxy, and a dynamic IP to the outside world anyway), and this means that one idiot could cause everyone's posts for a period of time (maybe just the time the article stays on the front page, but that's a lot of time for a dialup dynamic IP address).
What about displaying the ip address in the form 1.2.3.xxx? Maybe (although this would probably load the server), when a moderator tries to "SuperTroll", this would be disallowed if the reverse DNS lookup contains "cache","proxy", or something similar. This could possibly be abused though.
2. People may decide to do stupid DoS attacks against the displayed IP, which may affect lots of other people.
Maybe there should be an option "Ignore -1 posts from this IP", which would still have the problems in 1, but stop people doing DoS attacks.
2. There's still the potential for abuse. If something like this was done, then setting "SuperTroll" should require _all 5_ of the moderator points to be used, and require two moderators. The person who uses this option won't know if they're the first or second moderator to set this, and so would have to be really sure that they want this.
AFAIR, the legislation required that the censorship was "technically feasable". If studies like the ifilter one here are done on the proposed "approved" filters, could it be reasonable (even though reasonable has nothing to do with it) to show that the current stuff doesn't work?
A group of ISPs could then put a tender out for a "working" filtering program, but have a penalty clause if the program gets it wrong. If noone bids, then that proves that its unfeasable. If someone does bid, then someone makes lot of money....
Also, the article I read in Tuesday's Australian seemed to imply that the $5 charge was what they thought it would cost the user if an ISP bought the software in bulk. It still has to run on the subscriber's computer, which leads to the question of Linux, and other OSs...
Bradley
PS Why is this shown in the wrong day by slashdot? I didn't see this yesterday, and only 4 comments so far...
The CSIRO (The government science body), along with almost everyone else, told the Senate inquiry that it was impossible to do.
The full amendment is available from the EFA web site, with comments. Links to reviews of filtering software can also be found.
There was basically no Aus news about this - the first I knew that it had actually been passed into law was from UserFriendly (although there was no doubt that the lower house would pass it anyway).
What I think should happen is that a group of ISPs shouyld put out a tender for filtering software, but specify a penalty cloause for anything which gets through that shouldn't, or anything which doesn't get through but should - including SSL, other langauges, and aything else pepole can come up with. Anyone which any sense won't take that up, so the ISPs can take the excuse to the bill along the lines that they don't have to do anything if its technically unfeasable.
Currently, this law only applies to Aus sites, but it must be reviewed within 3 years to see if the bill can be made to cover overseas sites as well. The big problem is that once an ISP has been told to remove something, they then have to stop anything similar from appearing on their site. Private email is not covered, but the status of mailing lists, newsgroups, and similar stuff is not clear.
Slashdot Mirror
On a more serious note, how do you handle time_t?
:)
:)
"By the time 2038 comes around, nobody will be using 32 bit CPUs".
Seriously though, I think that statement's probably true. People are still using programs from the 60's and 70's but I don't know of anyone using hardware from then. (Excluding that old computer they're trying to restore in Melbourne of course).
You _do_ have the source to everything, right?
I stopped using this and changed all my time variables to doubles.
Doesn't that break all the libc interfaces which use time_t? Also, you change CPUs, you'll have to change it back again.
certainly it can't be too hard for someone to do a check
Actually, I'd disagree with that. If it hadn't been a slashdot-related article, I wouldn't have been sure whether I saw it on slashdot, LWN, or somewhere else.
When I've submitted stories, time differences mean that its generally just before the slashdot crew start going through their submissions in the morning. There are about 300-350 submissions, of which 5-10 are accepted. Noone can keep up with that, especially a story from February. Someone else might have posted that one, and, as part of quickies, it wouldn't have been as noticeable.
I'll do a "Me too" to that.
And I'll do a "me too" to the "me too".
I checked, and Slashdot search turns up this as part of some quickies.
This has been annoying me too. The quote on archive.redhat.com (which is now a link back to the main redhat site) is:
:) (Or redhat has a different definition of "soon" than I do)
Wondering what happened to the mailing list archive search function?
Now that the quiet period is over, it will be returning soon! Thanks for your patience.
which is a bit out of date.
The Moongroup one above is only the redhat-list, not the others, and doesn't appear to be arranged in any logical way. (Archive[1-5], of vastly varying sizes). The "how do we do this" page is the same as the original page as well.
Maybe someone should subscribe mail-archive.com to the lists...
We did sign Wassner (sp) though, so some alterations to encryption laws (I don't think we have any now, but IANAL) are possibly on the way.
I'm not talking about 2 line comments, but the long paragraph type things, when adding an extra sentence means wrapping the text around and then having to fix all the // comments.
:)
To get slightly back on topic, I'm sure GNU Emacs can do this automatically, if I could work out how.
Re the nesting, the C9X rationale, which I found while looking for this, says something along the lines of the C89 group considered allowing nested comments, but decided that it was easier on the compiler writers if we all just used #if 0 instead. They also suggest using if (0) {...}, mentioning that most compilers optimise that away to nothing.
Perhaps you might be so kind as to please remind me which revision of the standard added those silly C++ comments to C.
// as comments. OTOH, people using // for multi-line comments probably shouldn't be.
C9X added them. See here for the summary of changes, or here for the full standard (section 6.4.9).
I don't think there's anything wrong with
Creative did release drivers.
See http://opensource.creative.com.
The beta keyserver is beta.dcti.org. www.dcti.org gives the same page as www.distributed.net. When was the change made (or is this just another test thing)?
I don't really think there will ever be more than 400 DVD drive manufacturers.
And nobody will be using this software from the 60s when we have to worry about date rollovers...
Presumably that means that whoever's in charge of the whole process has these keys already autogenerated, and pressed onto each disk?
Every player -- including consoles from Sony, Toshiba, and other consumer electronics vendors, as well as software vendors for PCs like WinDVD and ATI DVD -- has its own unique unlock key. Every DVD disc, in turn, has 400 of these 5-byte keys stamped onto the disc. That way, the unlock key from every licensee, be it WinDVD or a Pioneer DV-525 unit, will read the disc.
So does this mean that there can only ever be 400 DVD drive manufacturers? What happens if XYZ Corp starts manufacturing these tomorrow, and I buy one. Can I then not play a DVD that was printed yesterday?
Am I missing something?
The major problem was that is was openly user settable, not even hidden in the registry.
"Oh, lets see what this one does!"
Send one report to two people, and get different results.
OTOH, anyone suing Microsoft for patent infrigement isn't likely to have them just pay up.
We can't do that :(
I know one of the honours students has a debian mirror on his machine, and theres a local redhat disssstricution with isos, so most of us get someone in college to burn it for us.
There's an ongoing discussion on news:aus.legal about this. I don't know if that spreads to the US though. Deja gets it. You can ignore the republic thread though.
Sydney Uni's CS department has its first year students use a teaching language called Blue.
Theres a sunos version (for the uni servers) and a linux version.
All CS students get given a CD containing the full version of RedHat, plus an installer for a UMSDOS partiaion of linux. The CD also has stuff like Java, Mesa, an updated python and a few other things for 2nd and third year students.
If you run install, it uncompresses from a self extracting zip file ionto c:\linnx, and puts an icon onto the desktop. The first time you run it, It runs a command line mouse/cd inerface (this was before redaht had mouseconfig/linuxconf). Only a few packages are installed, but you can easily install any others from the CD.
It hasn't been updated for a while, so its currently RedHat 5.0 + some updates + a CD you can borrow from the Dept with updated XServers, but I believe that it will be 6.1 next year.
Well, Win'98 has an option to select its Y2K window in the date control panel.
Actually, I went to a talk once where it was pointed out that because of this, you can get strange data corruption.
Someone with one setting can create (say) a spreadsheet with 2 digit years, and it all works fine. If this is sent to another person (even in the same organisation) with different settings, then the date will look like it came from another century. If noone else can see it, then you get data corruption. It gets wors if the second person then saves in in aa 4 digit date format...
Bradley
I submitted this as a story, but its late in America now, so I doubt anyone will get to it while its still Halloween. (Its actually November in AU now, but thats a different matter...)
Go to Google, and look twice at the logo.
b) I'm bothered by "off-topic" and "redundant" moderating deep down in the discussions. There is no harm in people straying from the topic when its not in the main thread.
Yeah, me too. What annoys me more is that often the stuff marked redundant has a CID _before_ the other post, which is in the first level, with a score of generally +4 or +5. I tend to ignore posts marked redundant when doing meta-moderation, because of this.
From the posting page:
try to reply to other people comments instead of starting new threads
I really think that redundant shouldn't be an option for moderation. Unless you read every post in posted order (I tend to read in threaded mode), its impossible to work out who said what first, especially if the posts were made within a couple of minutes of each other. Also, saying the same thing twice can be useful, especially if both comments are buried at the bottom of a large thread.
just like although you can't patent gravity you can patent useful descriptions of gravity.
Can you though? Assume that Newton discovered gravity and his laws of motion today. What could he have patented?
1. Gravity itself - (hopefully) obviously not. As others have mentioned, there's always been gravity, etc, etc.
2. The descriptions (Gravity is...) - Do the same arguements apply to this as above?
3. His laws of motion. (Every action has an equal and opposite reaction, etc) - Can something like this be patented? Its a "useful description", which is (was) non obvious, but how could you infringe against this? Would someone who made, say, shock absorbers for cars be infringing this patent?
4. The mathematical equations - Since (IIRC) algorithms can be patented, could this be patented? ("A method for determining the force due to gravity between two objects"). Would this mean that anyone who uses this equation have to licence it in the same way RSA needs to be licenced?
5. Newton's cradle - this is obviously a physical thing, which probably could be patented. It can be used to "describe" gravity/forces in an easy-to-understand manner, so could this be patented?
What can be patented about an idea such as this?
Won't this just lead to the end of the Encyclopedia Brittanica?
Maybe. But, remember the speed of the internet, and the number of people likely to be using this at a time, and the speed of most people's access, etc, etc. Think large movies/pictures/etc.
If its just for occasional access, then I'd use this. But if I was a {uni,library,masters/phd student}, then I'd buy the CD (on a site license - I seem to recall they have these), and get faster access for a reasonably minimal price.
Bradley
If you do not know your administrator account password, you will have to completely reinstall Windows NT because eventually you'll need to have access to this account.
Actually, this isn't true. A linux boot disk can be used to change the administrator password. Do read the warnings though.
The best way to avoid this dilemma is to immediately add your personal user account to the administrators local group of the system. This will make your main user account an administrator of the system, sparing you from heartaches and time later.
No comment on this one....
Can you get an equivalent of su for nt, and run the GUI apps by typing in a console?
Keyword ratings just don't work. If you specify the keywords with too much detail, each web page becomes 10 K of keywords, and 1 K of content. If the list of keywords is not ridgidly fixed, you also end up with "the chicken problem", where a hard core sex site is rated the same as a cooking site, because the both have the keyword "breast". The fact that one of them refers to chicken breasts is not an issue to someone blocking keywords.
If you don't permit enough detail, then things which shouldn't get through do.
For a set of good examples of this, using RSAC to prove the point, see here, but specifically this link, which rates both Alex's Haley's Roots and a pornographic, racist novel using RSACi, and finds that they both have to be given almost the same rating.
Bradley
I think that this sort of thing is a good idea, but there are some important issues:
1. What about people connecting through IPMasq/Proxies, or using dynamic IPs? I connect to the net through my uni (forced proxy, and a dynamic IP to the outside world anyway), and this means that one idiot could cause everyone's posts for a period of time (maybe just the time the article stays on the front page, but that's a lot of time for a dialup dynamic IP address).
What about displaying the ip address in the form 1.2.3.xxx? Maybe (although this would probably load the server), when a moderator tries to "SuperTroll", this would be disallowed if the reverse DNS lookup contains "cache","proxy", or something similar. This could possibly be abused though.
2. People may decide to do stupid DoS attacks against the displayed IP, which may affect lots of other people.
Maybe there should be an option "Ignore -1 posts from this IP", which would still have the problems in 1, but stop people doing DoS attacks.
2. There's still the potential for abuse. If something like this was done, then setting "SuperTroll" should require _all 5_ of the moderator points to be used, and require two moderators. The person who uses this option won't know if they're the first or second moderator to set this, and so would have to be really sure that they want this.
Just some thoughts.
AFAIR, the legislation required that the censorship was "technically feasable". If studies like the ifilter one here are done on the proposed "approved" filters, could it be reasonable (even though reasonable has nothing to do with it) to show that the current stuff doesn't work?
A group of ISPs could then put a tender out for a "working" filtering program, but have a penalty clause if the program gets it wrong. If noone bids, then that proves that its unfeasable. If someone does bid, then someone makes lot of money....
Also, the article I read in Tuesday's Australian seemed to imply that the $5 charge was what they thought it would cost the user if an ISP bought the software in bulk. It still has to run on the subscriber's computer, which leads to the question of Linux, and other OSs...
Bradley
PS Why is this shown in the wrong day by slashdot? I didn't see this yesterday, and only 4 comments so far...
The CSIRO (The government science body), along with almost everyone else, told the Senate inquiry that it was impossible to do.
The full amendment is available from the EFA web site, with comments. Links to reviews of filtering software can also be found.
There was basically no Aus news about this - the first I knew that it had actually been passed into law was from UserFriendly (although there was no doubt that the lower house would pass it anyway).
What I think should happen is that a group of ISPs shouyld put out a tender for filtering software, but specify a penalty cloause for anything which gets through that shouldn't, or anything which doesn't get through but should - including SSL, other langauges, and aything else pepole can come up with. Anyone which any sense won't take that up, so the ISPs can take the excuse to the bill along the lines that they don't have to do anything if its technically unfeasable.
Currently, this law only applies to Aus sites, but it must be reviewed within 3 years to see if the bill can be made to cover overseas sites as well. The big problem is that once an ISP has been told to remove something, they then have to stop anything similar from appearing on their site. Private email is not covered, but the status of mailing lists, newsgroups, and similar stuff is not clear.