Next time one of our PHBs refuses to let Linux into the computer room, using the all time classic "When this weird open source stuff breaks, who can I sue?" argument to defend his halucinations, we can reply along the lines of: "Well, actually Mr. Boss, that sueing business has been done before by a well repected law firm. So please feel right at ease with us using Linux now. Look, here's the hard evidence..."
While I love Linux a lot (cfr. my sig), and have a quite stable SMP machine currently running 2.4.0 at home, I indeed have some reservations about recomending for the kind of setup under discussion here. Maybe I'm just overconservative, but a premature move to Linux would close a lot of doors. Keep in mind that in a professional environment everything counts: it doesn't matter if teh Linux machine gives very good performance (especially for the buck), if the sysadmins end up having to deal with some obscure NFS problem (it's just an example, OK?) every other day.
On the other hand, I use HP-UX machinery at work that is very similar to what is called for here, and can assure you that:
Except maybe for really floating point dominated work, nowadays PC processors are quite a bit faster than currrent HP processors when compared on a 1-1 basis.
Not all industrial strength unix SMP machines scale very well. The expensive ones do, but not all of them.
One of our HP-UX machines goes down every so many days due to SMP locking/racing problems. It's been doing that for months now, and none of many OS patches that we have installed has solved it, despite claiming to address problems that look exactly like what we're seeing. Seems to be a rather buggy OS (at least on that specific type of hardware, our other HP SMP machines don't have this problem), industrial strength or not.
Yes, I remember that post from Alan (some 2 or 3 months ago, IIRC). I just did a Google search, but can't find it archived right now. However, here's an equivalent quote from the January 16, 1996 version (yes, this whole confusion is that old already) of the Linux NET-2/3-HOWTO:
2. Disclaimer.
The Linux networking code is a brand new implementation of kernel
based tcp/ip networking. It has been developed from scratch and is not
a port of any existing kernel networking code. ...
NOTE: While its name may appear similar to the Berkeley Software
Distribution NET-2 release, the Linux network code actually has
nothing at all to do with it. Please don't confuse them.
Well, looking at the date on that page (01 October 1998) it seems to me that this info just might be just a tad out of date. Have you actually looked at what's in the 2.4 kernel? Maybe things haven't changed, but it sure wouldn't harm to have a look before fudding.
As much as I'd like not to have to write this, but 4.0 was extremely unstable for me (daily server crashes due to SIGSEGV, no 3D involved at all, only an old Matrox Millenium). 4.0.1 is a lot better, but still goes down once in a while with the same problem (same stack trace even). Can't wait to get 4.0.2 installed...
Well, I've been using Linux ever since the good old days when one had to attack a freshly compiled kernel with a binary editor to tell it how to get at its root file system (2 bytes somewhere in the vicinity of 508 IIRC) and other "heroic" things like that. It goes without saying that I've moved things over to another box since then (although the old one still works just fine under DOS for my father, even after 9 years).
Well, I've built my entire box from source. Oh wait, I tell a lie. Actually, it's everything except XFree86, because I simply don't have the diskspace needed to compile that "monster".
So yes, binaries can have use even for non-average users who theoretically are perfectly capable and willing to work with the source.
Indeed, it's not an option now. But it should be considered very seriously before the time of the next election.
The current mess has clearly shown that vote counting is inaccurate, and will always be so, no matter what you do. A difference of 500 or 2000 votes on a total of 6000000 obviously is smaller than the margin of error of the (current) counting process. This fact of science should be taken into account in the law: either the law should provide for the possibility of a tie (not counting down to the very last vote), or it should be such that an error this small has an effect that is equally small. A proportional system would provide the latter.
Unfortunately (oh, well, as I'm European, it doesn't matter that much to me), I have strong doubts that anything like this will be put into practice any time soon. While individual states might decide to go for a (more or less) proportional system (I doubt that any of them will, though), the stuff you propose will in the end require a modification of the US consitution, even if one would try to avoid this by changing the rules on a state by state basis. This is because making proportional systems work well in small states may imply giving these states more votes in the college. And then, in order to maintain the current balances between states, the larger states need to get more votes too. And then,
for sure, some other issue that also needs to be rebalanced will be raised by some interested party that feels it might loose in the deal...
With a country that's divided 50-50 along party lines, and that just went through the current mess, I'd be extremely surprised to see anyone even propose for real to change the constitution.
Especially if one considers that, had the rules
been changed before this year's elections, the
(likely) winner would likely have been the looser and vice versa.
The electoral college serves to normalize the power distribution. With a straight popular vote,
smaller states would get glossed over, with presidential candidates focusing on the population
centers. What the electoral college does is guarantees that those smaller states have a voice.
That's what I keep reading from people who defend
the idea, but IMHO this reasoning is flawed. It mixes up two unrelated issues: keeping the big states under control, and keeping the citizens under control (which also was one of the original considerations, even if people don't like to say that aloud anymore). One of the things that are being overlooked in this way, is the goal of coming to a fair representation of what the citizens of each individual state want.
So why don't you guys (I'm not American) do this instead:
Keep the electoral college, adapting the number of representatives from each state as you see fit to protect the smaller ones.
Within each state, use a proportional representation within the college, such that it cannot happen that 50% of the people in any given
state are not represented at all in the college.
That way, one can have the best of both worlds: a
college that reflects the propular vote more closely, but that at the same time protects those who feel threatened by the urban centres.
Most of our servers (I'm ignoring the little crap
like print servers for the sake of simplicity)
have between 1 and 4 gigs.
The 1 gig machines are
pseudo-desktops, actually, in the sense that they
provide desktop functionality for multiple people.
The 2 and 4 gig ones are mostly used for chip development, and while we have not seen it over here yet, we have project partners that have had single ECAD processes grow past the 4 gig boundary (and no, these are not leaking memory like mad).
We also have a machine on which we develop ECAD software of our own, and guess what: we wanted to put 2 gigs in it, but were told by the vendor (HP) that 4 was the minimum. Yes, you read that right: they wouldn't sell us less than 4. What's more, even though we use it as server, HP officially call it a (technical) workstation.
Please -- open your eyes. You're dealing with people here. In any given sample, you'll have a
certain number of misguided kooks who don't have a clue. Does the fact that it was a stupid thing to
do automatically mean that it a Linux user ? Nope. Sorry. The non-Linux community has its fair share of kooks and idiots too.
Hence: please -- open your eyes and read what you reply to before replying. The previous poster
did not say that this virus could not be a by a Linux zealot, or that it probably was by an MS advocate. He only said that it the latter is possible. Which it is.
As the subject says: don't jump to conclusions. It could just as well be an immature Linux hater who thinks that he has found a nice way to cause some harm to The Enemy.
For completeness: there are 3 separate parts to the ArsTechnica story. Here's the entire list:
part 1 covers SRAM and DRAM basics,
part 2 covers asynchronous and synchronous DRAM, and
part 3 covers DDR DRAM and the RAMBUS thingie. It's good reading indeed.
all illegal transport of copyrighted material across these servers is strictly
prohibited and will constitute a violation of the terms of this agreement
That's more or less what they ended up doing in the end. Can't check the final wording anymore, though, because they've since been bought by uunet.
Well, sometimes you can block them. For instance, there's a forum that I post to every so often that has a webbug on its entry form. So I saved the thing as HTML, edited it to my taste, and nowadays always post via that local page. Problem solved. Well, in that particular case, at least...
I basically have cookie support enabled (only from the site that the originating page came from, by the way, but that doesn't make much difference in practice) because some sites that I visit regularly use them for good purposes and I want things to work transparantly when I find a new site with cookies worth keeping.
To compensate for this, I have a cookie eating script that gets run every night from my crontab. It reads in a cookie filter file, containing commands such as:
retain ^slashdot\.org
remove ads\..*
and then acts accordingly: a cookie that matches a
"retain" line is, well, retained. One that matches a "remove" line is... you guessed it. Cookies that match neither are reported to me, so that I can decide what to do with them next time round. Initially I ran this script once a week, but as I kept accumulating more "remove" directives, the number of manual interventions went down quite rapidly, so I increased the frequency. I'd move on to once every 6 hours, if it weren't for the fact that it's no use editing the cookie file while Netscape is running.
A nice side effect of what I'm doing is that each time that I visit one of these tracking companies, they actually think that they've found another victim. Let them...:-)
By the way, originally, I used the possibility to opt out whenever possible. But I found that, while my DoubleClick opt out cookie would indeed be kept around, the similar one from prefences.com tends to disappear automatically after some time. So nowadays I bluntly remove anything that's not absolutely required.
Everybody seems to be (redundantly) jumping on the "too bad, he should have read his contract" bandwaggon. The question, however, was whether the ISP should have the right to define their TOS like this one did.
Now, in order to shortcut a lot of useless discussion: they have that right (a.k.a the right to behave in a stupid and/or silly way), I think they should have it too, and know nobody who would actually disagree with that. It's their operation, after all. Conclusion so far: why was this topic posted on Ask Slashdot in the first place? Seems like there's nothing much to talk about.
On the other hand, an ISP that's not totally braindead will not invent this kind of rule, and will certainly not enforce it in this braindead way. Still,
they're not alone in their sillyness. When I was shopping for an ISP some 5 years ago, I actually read the AUPs of all the Belgian ISPs that I could find. To my surprise, the AUP of the then biggest Belgian ISP had several holes in it, and also contained the rule that transfering copyright material over their network was prohibited. It took me nearly six months to convince them that there was copyright material that could be legally copied. The fact that their legal department hadn't yet heard of Linux and the GPL surely didn't help. I guess I should have picked a more mainstream example. Anyway, I ended up choosing a competitor long before they saw the light (but went on trying to convince them, and finally did).
Nope. All it says is that I'm a diehard Linux user who used it when (and since, but especially when) it was not even barely usable yet. I use this sig because I tend to defend non-linux points of view when I think they are right. The sig then cuts down on the bullshit replies of the "you ignorant MS lover" kind.
I hate to break the news to you, but you got some crucial bits wrong. We've already witnessed the release of Linus 2.0 some time ago. So he might want to go for 3.0, 3.1 and/or 3.12 next. However, that has been done already by some one else.
Hence is blatantly obvious that to be an effective marketing measure, the next generation cannot called be anything other than Linus NT, skipping the 3.x series alltogether, and thereby proving that this stuff matures more rapidly than the competing product.
It's not so much Linux/bsd that are the problem
(OK, sometimes these are missing too), it's that other platforms. Such as all (or most) of the commercial Unices. Despite what the FUD says, there are still plenty of people using one of those on their desktop. E.g. I myself am on HP/UX when not at home.
One wonders why companies such as Sun and HP don't go and talk to plugin providers such as Macromedia to talk them into porting their stuff. One almost gets the impression that Sun and Co. don't really believe they can hold on to their desktop share anyway. Sort of a self-fullfilling prophecy, if you ask me. How do these guys expect long time (power) users like me to convince management into not switching to NT if all that flashy stuff that PHBs like so much doesn't run on our good old Unix? (OK, OK, I actually want us to switch to Linux, but if I can't have that (yet)...)
Yes, they broke some codes. For instance, the Germans read British naval traffic for most, if not all, of the war.
But it's not as black-and-white as people usually put it. For instance, some German codes were never broken, despite Ultra. For more information, see
this page at uboat.net. No doubt something like this will also have been true in the other direction.
It sure feels silly to follow up on my own post, but the Uboat.net Enigma pages seem to be working at least somewhat (not 100 percent, though), even if much of the rest of the site isn't in good shape at the moment. They are available here. Their list of other Enigma pages on the net is here.
Well, there can be no definite proof that this method is what they used (unless some contemporary text shows up in which the actual method is described), but there are valid data points that support the theory. Like how the (minor) alignment errors of the various pyramids shift over time.
However, if they did not use this method, they must have used another one instead. Theory so far was that they needed (and thus had) a lot more knowledge of astronomy than what is required to use this "new" method. If they actually did have that knowledge, then surely they will also have known about the trick discussed in the article.
At this point, what the article describes looks like being the simplest theory that explains the experimental data. That in itself makes it a very good choice for an explanation, especially in a case like this.
Slashdot Mirror
--
On the other hand, I use HP-UX machinery at work that is very similar to what is called for here, and can assure you that:
--
2. Disclaimer.
...
The Linux networking code is a brand new implementation of kernel
based tcp/ip networking. It has been developed from scratch and is not
a port of any existing kernel networking code.
NOTE: While its name may appear similar to the Berkeley Software
Distribution NET-2 release, the Linux network code actually has
nothing at all to do with it. Please don't confuse them.
--
--
--
Well, I've been using Linux ever since the good old days when one had to attack a freshly compiled kernel with a binary editor to tell it how to get at its root file system (2 bytes somewhere in the vicinity of 508 IIRC) and other "heroic" things like that. It goes without saying that I've moved things over to another box since then (although the old one still works just fine under DOS for my father, even after 9 years).
--
So yes, binaries can have use even for non-average users who theoretically are perfectly capable and willing to work with the source.
--
The current mess has clearly shown that vote counting is inaccurate, and will always be so, no matter what you do. A difference of 500 or 2000 votes on a total of 6000000 obviously is smaller than the margin of error of the (current) counting process. This fact of science should be taken into account in the law: either the law should provide for the possibility of a tie (not counting down to the very last vote), or it should be such that an error this small has an effect that is equally small. A proportional system would provide the latter.
--
Unfortunately (oh, well, as I'm European, it doesn't matter that much to me), I have strong doubts that anything like this will be put into practice any time soon. While individual states might decide to go for a (more or less) proportional system (I doubt that any of them will, though), the stuff you propose will in the end require a modification of the US consitution, even if one would try to avoid this by changing the rules on a state by state basis. This is because making proportional systems work well in small states may imply giving these states more votes in the college. And then, in order to maintain the current balances between states, the larger states need to get more votes too. And then, for sure, some other issue that also needs to be rebalanced will be raised by some interested party that feels it might loose in the deal...
With a country that's divided 50-50 along party lines, and that just went through the current mess, I'd be extremely surprised to see anyone even propose for real to change the constitution. Especially if one considers that, had the rules been changed before this year's elections, the (likely) winner would likely have been the looser and vice versa.
Just my 2 Eurocents.
--
That's what I keep reading from people who defend the idea, but IMHO this reasoning is flawed. It mixes up two unrelated issues: keeping the big states under control, and keeping the citizens under control (which also was one of the original considerations, even if people don't like to say that aloud anymore). One of the things that are being overlooked in this way, is the goal of coming to a fair representation of what the citizens of each individual state want.
So why don't you guys (I'm not American) do this instead:
- Keep the electoral college, adapting the number of representatives from each state as you see fit to protect the smaller ones.
- Within each state, use a proportional representation within the college, such that it cannot happen that 50% of the people in any given
state are not represented at all in the college.
That way, one can have the best of both worlds: a college that reflects the propular vote more closely, but that at the same time protects those who feel threatened by the urban centres.--
The 1 gig machines are pseudo-desktops, actually, in the sense that they provide desktop functionality for multiple people. The 2 and 4 gig ones are mostly used for chip development, and while we have not seen it over here yet, we have project partners that have had single ECAD processes grow past the 4 gig boundary (and no, these are not leaking memory like mad).
We also have a machine on which we develop ECAD software of our own, and guess what: we wanted to put 2 gigs in it, but were told by the vendor (HP) that 4 was the minimum. Yes, you read that right: they wouldn't sell us less than 4. What's more, even though we use it as server, HP officially call it a (technical) workstation.
--
Please -- open your eyes. You're dealing with people here. In any given sample, you'll have a certain number of misguided kooks who don't have a clue. Does the fact that it was a stupid thing to do automatically mean that it a Linux user ? Nope. Sorry. The non-Linux community has its fair share of kooks and idiots too.
Hence: please -- open your eyes and read what you reply to before replying. The previous poster did not say that this virus could not be a by a Linux zealot, or that it probably was by an MS advocate. He only said that it the latter is possible. Which it is.
--
--
--
That's more or less what they ended up doing in the end. Can't check the final wording anymore, though, because they've since been bought by uunet.
--
--
To compensate for this, I have a cookie eating script that gets run every night from my crontab. It reads in a cookie filter file, containing commands such as:
retain ^slashdot\.org
remove ads\..*
and then acts accordingly: a cookie that matches a "retain" line is, well, retained. One that matches a "remove" line is... you guessed it. Cookies that match neither are reported to me, so that I can decide what to do with them next time round. Initially I ran this script once a week, but as I kept accumulating more "remove" directives, the number of manual interventions went down quite rapidly, so I increased the frequency. I'd move on to once every 6 hours, if it weren't for the fact that it's no use editing the cookie file while Netscape is running.
A nice side effect of what I'm doing is that each time that I visit one of these tracking companies, they actually think that they've found another victim. Let them... :-)
By the way, originally, I used the possibility to opt out whenever possible. But I found that, while my DoubleClick opt out cookie would indeed be kept around, the similar one from prefences.com tends to disappear automatically after some time. So nowadays I bluntly remove anything that's not absolutely required.
--
Now, in order to shortcut a lot of useless discussion: they have that right (a.k.a the right to behave in a stupid and/or silly way), I think they should have it too, and know nobody who would actually disagree with that. It's their operation, after all. Conclusion so far: why was this topic posted on Ask Slashdot in the first place? Seems like there's nothing much to talk about.
On the other hand, an ISP that's not totally braindead will not invent this kind of rule, and will certainly not enforce it in this braindead way. Still, they're not alone in their sillyness. When I was shopping for an ISP some 5 years ago, I actually read the AUPs of all the Belgian ISPs that I could find. To my surprise, the AUP of the then biggest Belgian ISP had several holes in it, and also contained the rule that transfering copyright material over their network was prohibited. It took me nearly six months to convince them that there was copyright material that could be legally copied. The fact that their legal department hadn't yet heard of Linux and the GPL surely didn't help. I guess I should have picked a more mainstream example. Anyway, I ended up choosing a competitor long before they saw the light (but went on trying to convince them, and finally did).
--
Nope. All it says is that I'm a diehard Linux user who used it when (and since, but especially when) it was not even barely usable yet. I use this sig because I tend to defend non-linux points of view when I think they are right. The sig then cuts down on the bullshit replies of the "you ignorant MS lover" kind.
--
Hence is blatantly obvious that to be an effective marketing measure, the next generation cannot called be anything other than Linus NT, skipping the 3.x series alltogether, and thereby proving that this stuff matures more rapidly than the competing product.
--
One wonders why companies such as Sun and HP don't go and talk to plugin providers such as Macromedia to talk them into porting their stuff. One almost gets the impression that Sun and Co. don't really believe they can hold on to their desktop share anyway. Sort of a self-fullfilling prophecy, if you ask me. How do these guys expect long time (power) users like me to convince management into not switching to NT if all that flashy stuff that PHBs like so much doesn't run on our good old Unix? (OK, OK, I actually want us to switch to Linux, but if I can't have that (yet)...)
--
Yes, they broke some codes. For instance, the Germans read British naval traffic for most, if not all, of the war.
But it's not as black-and-white as people usually put it. For instance, some German codes were never broken, despite Ultra. For more information, see this page at uboat.net. No doubt something like this will also have been true in the other direction.
--
--
There are also some interesting Enigma pages at uboat.net. Unfortunately, that site seems to be having some database problems at the moment.
--
However, if they did not use this method, they must have used another one instead. Theory so far was that they needed (and thus had) a lot more knowledge of astronomy than what is required to use this "new" method. If they actually did have that knowledge, then surely they will also have known about the trick discussed in the article.
At this point, what the article describes looks like being the simplest theory that explains the experimental data. That in itself makes it a very good choice for an explanation, especially in a case like this.
--