I've never understood why people found operator overloading to be hidden or anything. When you see
a-b
you should think
operator-(a,b)
, unless a and b a native/primitive/whats-it-called-in-python types. Easy, I find.
What makes operator overloading hard (sometimes) is the operator precedence. E.g., in C++, the following does not do what you would expect:
cout << mybool ? "true" : "false" << endl;
This is, of course, because somebody thought it a good idea to take the bit shift operator and use it for stream operations...
If Perl, Python and C++ can be criticized re operators, it must be the (exceedingly) long hierachi. I believe C++ is the shallowest with only 11 (!) levels. Then for extra fun, some are prefix, some are infix, some are postfix and some are more two of those. That gets me from time to time, not the that operators can be overloaded. It's not like the code would compile if the overloads were not there...
I usually skip over all these OS bashing comments, but your comment "with NFS it is possible to assume root on your own box" caught my eye. After reading it a couple of times I read it like this.
assume root on own box, which has access to other's file
mount NFS
access NFS as root.
Which won't get you far on most NFS, as the option root_squash the default. So I guess I am not catching your drift after all. Care to elaborate?
Not suitable for desktop use? How come me and my wife use it every day, then? On the desktop? And my wife is a scientist, not computer geek:) I admit, I my be a computer geek. But back when I used SuSE and knew nothing about Linux, I had less trouble installing Linux than windows. OK, so linux didn't recognize my sound card, big deal, wrote to the support email and got a working response. When I tried installing windows on the same machine, it couldn't recognize my hard drives(!). Yes, I had a driver disk, but it took me some work to get the drivers loaded. In the end, I gave the windows away. Not worth the bother, I thought.
Remember that you have probably gotten used to windows. I use windows (2000) at work, and have enough trouble that I could say; Windows 2000 is just not ready for the desktop. Couldn't they at least make copy & paste work properly?;-)
To all non-plussed people here: It's refering to a novel by Terry Pratchett, the one with Holy Wood. The title eludes me right now. It's quite good, actually, featuring several talking animal and humans.
The author spends a lot of time trying to discredit the author on such terms as impartialness and experience. While such can lead credence to a strong case, it bodes when mentioned as the very first points. Also note the beginning of the article: "Many misled CS student...".
The author has no statistical or published backings for his claim
Most of the arguments are flawed, in my opionion. Yes, the corpus was trained on SpamAssassin, but the other filters' mistakes were, as far as I recall, examined for errors individually. Thus, any mistakes would be spotted or credit each filter equally.
I also always find it suspect when someone claims: "Yes, the program did not perform, but with a different configuration it might/in the latest version it might". While it could be true, such claims needs backing.
He claims that X's email was atypical, even for geeks. I would like to state here that I have 3 email accounts, of which none lie near his "typical" spam quotient (60%): 2 with >90% spams and 1 with <1% spam.
That said, he does raise a few valid points, such as the timeline:
If filters expunge old data based on time, this would not work in the test. That gives SpamAssisins' static rules an egde
Configurations should really have been published. I see no reason why not.
While I find the line-metric amazingly stupid, I would point out the that the most important reason to use lines over bytes is that the line count is independent of whitespace. Especially important for the indent-8 guys:)
Oh yeah, the perl -le '=1; =2; print " + = ", + ' program doesn't compile. And why must it be specified on the command line? The best I can do is:$a=1;$b=2;print "$a+$b=".$a+$b."\n";
This exploit has been
reported used to take down several "lame free-shell providers" servers (this is illegal in most parts of the world and strongly discouraged)
I don't know if this is a legend, but I have read that, according to the formulas used by aerospace engineers, a bumblebee can't fly. It's useless to extrapolate empiric rules, the only thing that matters is the practical result.
It is true that if a bumblebee used fixed wings, and if we ignored turbulence effects (safe if the object we consider is more than 1m or so long) a bumblebee wouldn't fly. That is hardly shocking news.
The example is usually used to demostrate that you cannot extrapolate indefinitely and expect results to hold true. One of my favorites: It would be impossible, studying fauna in a glass of water, to realize the existence of whales.
while you are right that Microsoft can't be blamed for not providing a patch in time, you could (rightly) ask why those ports were open in an operating system meant for non-technical users? For most users, you could probably block all incoming requests, and everything they use would keep working.
You will notice that others (such as RedHat) receives some flames over this as well.
Looking over the security advisories from the list you posted, I can see that the last relevant for me was in March. And that is only because I happen to host my own webpages, and thus uses apache. Remember that that those lists include martel-near-all software that runs in Linux. I remember scorched earth 3d in that list... not exactly a common program to use:)
However, I understand you reaction. This board isn't exactly Microsoft-friendly. I myself loathe to use, maintain and administer windows, but as long as I am not actively hindered in avoiding this, I see no reason to hate all-things-Microsoft. Well, except when I am forced to use it (work) and it crashes on me. Again. I hate blue-screen-of-reboot, as it has become:-( Can't even do anything about it, the machine is locked down tight:-(:-(
The problem is, 9.2.2 is the most recent version by portage's knowledge, and it keeps trying to downgrade it (which means I can't just blindly "emerge world", or portage will downgrade bind).
If you don't mind going for the ~x86 packages for bind, you could use
That sounds seriusly broken to me. My linuxboxen never swap before the entire RAM is filled (with cache or otherwise). Even so, the swappiness stays quite low. Perhaps you show turn the
A lot of people believe that pot is addictive random google link. And from going to "school" with a few who smoked pot, I'd say they were visibly impaired for about a week. Totally unscientific, I know.
Still, I think that it must be the choice of every individual, but children should not have access until they are old enough to understand the consequences.
hhh... having just read the actual topic, this is rather off topic. Oh well.
.. but this is insane by any standard. Only the most extreme economic offenses should be punishable by imprisonment. Fines and compensation can do for the rest.
There is little math in /playing/ poker
on
Geeks and Poker?
·
· Score: 1
There is little math in playing poker. You just have to learn some values, and then build an intuition.
There is, of course, some calculation to be done to arrive at these values, and even some theory you could learn besides. But you don't want to do THAT during the game:-)
Perhaps it is just a matter habits... I never learned to use tables or document.all, and I have never missed either. Nor do I find the layout to be radically different between browsers, though I do wish Konqueror would start supporting min-height --- but then I could just implement it, right?
And now, totally offtopic, I must state that for creating tables, (X)HTML absolutely rocks! It is the only system I have ever found that I get the results I want. Not Word, Not KOffice, not LaTeX --- but HTML.
The cache on the disk. And while we do not REALLY know, the described behaviour is a common and proven strategy. Remember that the disk does not have a lot of information to go by; it basically just sees request to read individual sectors. More or less.
The main problem with fragmentation is cache-faults. The disk drives assume that you will be reading the following sector;: when you don't, you'll have to wait for the sector you requested to be brought in from disk. This applies even in the face of the tricks you mention.
Two problems:
1, I can't really fork Java I think Sun is killing it.
2. That test-suite should then be publicly availble, so Joe Hacker can test his YAJ-compiler-etc against the suite.
But who cares? It's not like there is a lack of languages out there, and Java lacks some pretty vital stuff anyhow (e.g fork() --- and forking the JVM is not really a solution).
Slashdot Mirror
It's like how IBM happened: We started out small and grew :)
It's called evolution. Self-replication can start in several ways; and once you got that life just happens.
Remember that commmon sense and intuition are poor tools when trying to understand stuff outside the sphere of human, daily interaction.
I've never understood why people found operator overloading to be hidden or anything. When you see
you should think , unless a and b a native/primitive/whats-it-called-in-python types. Easy, I find.What makes operator overloading hard (sometimes) is the operator precedence. E.g., in C++, the following does not do what you would expect:
This is, of course, because somebody thought it a good idea to take the bit shift operator and use it for stream operations...
If Perl, Python and C++ can be criticized re operators, it must be the (exceedingly) long hierachi. I believe C++ is the shallowest with only 11 (!) levels. Then for extra fun, some are prefix, some are infix, some are postfix and some are more two of those. That gets me from time to time, not the that operators can be overloaded. It's not like the code would compile if the overloads were not there...
Here you go: Python!
/me unable to resist the temptation
whatever, I'm sure we'll soon be able to install Linux on it in any case :-D
I usually skip over all these OS bashing comments, but your comment "with NFS it is possible to assume root on your own box" caught my eye. After reading it a couple of times I read it like this.
Which won't get you far on most NFS, as the option root_squash the default. So I guess I am not catching your drift after all. Care to elaborate?
Not suitable for desktop use? How come me and my wife use it every day, then? On the desktop? And my wife is a scientist, not computer geek :) I admit, I my be a computer geek. But back when I used SuSE and knew nothing about Linux, I had less trouble installing Linux than windows. OK, so linux didn't recognize my sound card, big deal, wrote to the support email and got a working response. When I tried installing windows on the same machine, it couldn't recognize my hard drives(!). Yes, I had a driver disk, but it took me some work to get the drivers loaded. In the end, I gave the windows away. Not worth the bother, I thought.
Remember that you have probably gotten used to windows. I use windows (2000) at work, and have enough trouble that I could say; Windows 2000 is just not ready for the desktop. Couldn't they at least make copy & paste work properly? ;-)
Indeed. And from my 300Mhz ditto.
It just bodes. In general. Just as Gaspodes says.
To all non-plussed people here: It's refering to a novel by Terry Pratchett, the one with Holy Wood. The title eludes me right now. It's quite good, actually, featuring several talking animal and humans.
There are several warning signs in this article.
That said, he does raise a few valid points, such as the timeline:
While I find the line-metric amazingly stupid, I would point out the that the most important reason to use lines over bytes is that the line count is independent of whitespace. Especially important for the indent-8 guys :)
Oh yeah, the perl -le '=1; =2; print " + = ", + ' program doesn't compile. And why must it be specified on the command line? The best I can do is:$a=1;$b=2;print "$a+$b=".$a+$b."\n";
It is true that if a bumblebee used fixed wings, and if we ignored turbulence effects (safe if the object we consider is more than 1m or so long) a bumblebee wouldn't fly. That is hardly shocking news.
The example is usually used to demostrate that you cannot extrapolate indefinitely and expect results to hold true. One of my favorites: It would be impossible, studying fauna in a glass of water, to realize the existence of whales.
while you are right that Microsoft can't be blamed for not providing a patch in time, you could (rightly) ask why those ports were open in an operating system meant for non-technical users? For most users, you could probably block all incoming requests, and everything they use would keep working.
You will notice that others (such as RedHat) receives some flames over this as well.
Looking over the security advisories from the list you posted, I can see that the last relevant for me was in March. And that is only because I happen to host my own webpages, and thus uses apache. Remember that that those lists include martel-near-all software that runs in Linux. I remember scorched earth 3d in that list... not exactly a common program to use :)
However, I understand you reaction. This board isn't exactly Microsoft-friendly. I myself loathe to use, maintain and administer windows, but as long as I am not actively hindered in avoiding this, I see no reason to hate all-things-Microsoft. Well, except when I am forced to use it (work) and it crashes on me. Again. I hate blue-screen-of-reboot, as it has become :-( Can't even do anything about it, the machine is locked down tight :-( :-(
I'm probably too tired to see the joke, but in the 2.6.x series, you can set it to any value between 0 and 100...?!
Still, I think that it must be the choice of every individual, but children should not have access until they are old enough to understand the consequences.
hhh... having just read the actual topic, this is rather off topic. Oh well.
you can buy a bit of space in a publication for your own message. This is what we in europe call ad-ver-ti-cing ;-
.. but this is insane by any standard. Only the most extreme economic offenses should be punishable by imprisonment. Fines and compensation can do for the rest.
There is little math in playing poker. You just have to learn some values, and then build an intuition. There is, of course, some calculation to be done to arrive at these values, and even some theory you could learn besides. But you don't want to do THAT during the game :-)
And now, totally offtopic, I must state that for creating tables, (X)HTML absolutely rocks! It is the only system I have ever found that I get the results I want. Not Word, Not KOffice, not LaTeX --- but HTML.
The cache on the disk. And while we do not REALLY know, the described behaviour is a common and proven strategy. Remember that the disk does not have a lot of information to go by; it basically just sees request to read individual sectors. More or less.
Reiser has a kernel option "Stats in /proc/fs/reiserfs". I do not know what information is output, though.
The main problem with fragmentation is cache-faults. The disk drives assume that you will be reading the following sector;: when you don't, you'll have to wait for the sector you requested to be brought in from disk. This applies even in the face of the tricks you mention.
Two problems: 1, I can't really fork Java I think Sun is killing it. 2. That test-suite should then be publicly availble, so Joe Hacker can test his YAJ-compiler-etc against the suite. But who cares? It's not like there is a lack of languages out there, and Java lacks some pretty vital stuff anyhow (e.g fork() --- and forking the JVM is not really a solution).