You make some good points, if one assumes that DLL's are used as shared libraries - same code used by different programs. That's the intent, and it happens of course.
But on the other hand, it also often happens that some software package uses a number of DLLs that were written specifically for that package. That wouldn't be a problem if they would be installed in the application's directory - but often they are installed in Common Files or System or System32 (a lot of Windows developers seem to think that's the easiest or only way to make sure the DLLs can be located at program start).
Which is one of the reasons the system partition of a Windows system keeps growing and growing, and you can't do anything about it except format c: and reinstall.
I would expect winpcap to be very difficult to install remotely. And it requires a reboot.
It's not that difficult. Just put an installer program into some email attachment. Don't even bother to reboot, Windows machines reboot often enough without any help from the outside. Even if you don't need a reboot because of some new installed package or a BSOD, most Windows users turn their PC off when they're done using it.
Do you think you can make it much faster using hand-crafted assembly code? Admittedly, I used C instead of C++, but that doesn't make any difference for anything as small as this.
...it was placed south of Greece, not in middle of Atlantic
Neither is this... It's off the Egyptian coast. With just a bit of imagination, one could describte it as south of Greece. Especially since the Greeks where quite 'Greece-centric'.
I'm not sure if this is right, but it might even be the case that if you are born in Germany, but your father is not a citizen, you don't automatically get citizenship?
I'm not sure about it, but I think in France you get automatically French citizenship when you are born in France. Don't know about Germany, though.
Somewhat offtopic, but, does anybody realize that in order to have free software, we must have proprietary software? Somebody somewhere has to pay these programmers, or they'll starve to death.
Whether the law is right depends on whether you think it should be legal to "steal" information. Personally, I think it should not be. Economics relies on competition, and copyright and other intellectual properties laws are monopolies by definition. If CDDB wants to protect the data, have them establish direct account, with limits, and poisoned data, and non-disclosure agreements.
Call me stupid, but I don't really understand this. The information in the CDDB database is not some kind of secret or anything, it is public information. Walk into any record store and you'll find the same information.
Granted, the technology used to correlate the CD with it's track list might be proprietary (but then why Gracenote sue FreeDB for patent infringement?), but I can not understand how Gracenote can have any right whatsoever to the information itself.
Arresting small-time copyright violators isn't just a Belgian thing.
I feel the need to clarify this a bit...
There is no such thing as arresting small-time copyright violators here in Belgium.
(1) Indeed there *was* a threat from the IFPI (www.ifpi.org: "IFPI is the organisation representing the international recording industry")
(2) But that threat was only to a small number (100) of users, who had received a warning before
(3) Everyone and his brother says IFPI is not able to identify these users unless they violated the privacy legislation
(4) And, most importantly: Marc Verwilghen, our minister of justice, has declared that the prosecution of small-time piracy has the very lowest priority. This places it, I think, just above cannabis use.
Exactly what do they mean with 500.000 years computing time? 500.000 years computing time on a 8086 is something entirely different than the same amount of computing time on an Athlon 1 GHz, so I don't really know what their statement is supposed to mean.
No, true realtime OSes must guarantee latencies. Not necessarily low ones. Of course, systems with low latencies are preferred over systems with high latencies, but that doesn't make them any more real time.
the trouble with flash is that it's a closed format
I don't know the full story on wether Flash is open or closed, but I do know that the file format is available for everybody interested. Everyone can make applications that export Flash, and everyone can make Flash players. Even Adobe, Macromedia's most important concurrent, has a Flash editor. I don't know if we need much more openness(sp?) than that.
It was never intended for hyper-precise description of the layout of the material.
No it wasn't. Unfortunately, 99% of web designers don't know that. I have a big suspicion that most of them don't even know what HTML even is. If these guys want to be taken seriously by developers, then they need to start holding html code reviews, logging bugs and defects against their design, and instituting the bare minimums of software engineering.
I happen to know a few web designers, and discussions I have had with them have led me to this conclusion: it's not so much that they don't know what HTML is supposed to be; that's not even relevant.
Reality is that there are lots of people who don't want to use content-based markup -- they want full control over the layout. They know (from experience, if nothing else) HTML doesn't offer that (and, insofar it does, it does so completely different and the different platforms). Therefore the web designers I know try to use HTML only when they really need to, and use Flash for everything else. It gives them complete control over the layout, it handles interactivity, it is uniformly supported on a whole range of platforms.
Re:Programmers Make Computers Slower Year by Year
on
Netscape 6 Vs. 4.7x
·
· Score: 1
When I suggested that we use, depending on Java's bitwise limitations, that we either get rid of one boolean, or use C/C++ style bit fields, he essentially said that while that would make the code faster and more compact, it would never be necessary.
You're sure bit fields would make it faster? I agree of course it would take less memory than two booleans, but I think it would be slower. Booleans can be retrieved from memory in one step, while a bit field can only be interpreted by XOR'ing and AND'ing against a mask, which takes extra time.
Openmail IS ported to Linux - we've been running it for over half a year. We are a small company with litte financial resources. We have one Linux server, and when the time came to install a mail system, management wanted to use the calendaring features of Exchange, but didn't want to buy the hardware and software it takes to run Exchange on NT.
It works, but administering it is quite complex (since nobody at the company knows enough about it or has the time to learn it, we still only use it for internal mail - still using an external server with POP and SMTP for external mail).
It's quite annoying to have to install Openmail client drivers on every client PC, it's impossible to change username and/or password on a client without hacking the registry, and it seriously slows down some of the client machines.
We're using dynamic DNS updates with BIND v8.2.1 and the ISC DHCP Server V3.0b2pl6, with no problems.
The relevant lines from/etc/dhcpd.conf:
ddns-domainname "ourdomain";
ddns-rev-domainname "in-addr.arpa";
In/etc/named.conf:
in the relevant zones, include the line
allow-update { list of hosts that are allowed to update the dns };
From http://www.docguide.com/dgc.nsf/html/English-Dicti onnary.htm:
Main Entry: sarcastic
Pronunciation: sär-'kas-tik
Function: adjective
Date: 1695
1 : having the character of sarcasm
2 : given to the use of sarcasm : CAUSTIC
- sarcastically/-ti-k(&-)lE/ adverb
synonyms SARCASTIC, SATIRIC, IRONIC, SARDONIC mean marked by bitterness and a power or will to cut or sting. SARCASTIC implies an intentional inflicting of pain by deriding, taunting, or ridiculing . SATIRIC implies that the intent of the ridiculing is censure and reprobation . IRONIC implies an attempt to be amusing or provocative by saying usually the opposite of what is meant . SARDONIC implies scorn, mockery, or derision that is manifested by either verbal or facial expression .
According to that dictionary, there doesn't seem to be that much difference between sarcastic and ironic. Still, it seems to me that ironic is more relevant in this case.
Not declaring a return type for main() is a member of the first group, since exit codes can be awfully useful in a Unix environment... But a typeless main should default to int.
In C, not declaring a return type for a function is the same (but more clear) as declaring it as an int; that's not the problem. The problem is that you effectively have to return something, which doesn't happen here, otherwise it's not legal C. But it's quite possible gcc will just issue a warning and then happily continues to compile. I don't remember if gcc issues a warning or an error, I haven't tried to forget the return lately.
Slashdot Mirror
Not to mention the unilateral cancellation of the Kyoto and ABM treaties, showing off that typical American arrogance towards the rest of the world.
Sorry if I insulted any of you by that statement, but that's the image you get by doing all those things.
You make some good points, if one assumes that DLL's are used as shared libraries - same code used by different programs. That's the intent, and it happens of course. But on the other hand, it also often happens that some software package uses a number of DLLs that were written specifically for that package. That wouldn't be a problem if they would be installed in the application's directory - but often they are installed in Common Files or System or System32 (a lot of Windows developers seem to think that's the easiest or only way to make sure the DLLs can be located at program start). Which is one of the reasons the system partition of a Windows system keeps growing and growing, and you can't do anything about it except format c: and reinstall.
It's not that difficult. Just put an installer program into some email attachment. Don't even bother to reboot, Windows machines reboot often enough without any help from the outside. Even if you don't need a reboot because of some new installed package or a BSOD, most Windows users turn their PC off when they're done using it.
First off, please neglect my previous post... hit the wrong button by accident.
.rodata
.LC0
.LC16
.L36
> C++ is not a very efficient language
Well, this little program:
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>
const double a = 1.1;
int main()
{
double d = a;
unsigned long i;
for (i = 0; i < ULONG_MAX; i++)
{
d += a;
}
printf("%lf\n", d);
return 0;
}
Compiled into this:
.file "repadd.c"
.version "01.01"
gcc2_compiled.:
.globl a
.section
.align 8
.type a,@object
.size a,8
a:
.long 0x9999999a,0x3ff19999
.LC1:
.string "%lf\n"
.align 8
.LC0:
.long 0x9999999a,0x3ff19999
.align 8
.LC16:
.long 0x99999999,0x40319999
.text
.align 16
.globl main
.type main,@function
main:
pushl %ebp
movl %esp, %ebp
pushl %eax
fldl
fldl
pushl %eax
movl $15, %eax
.p2align 4,,7
.L36:
fadd %st(1), %st
addl $30, %eax
cmpl $-2, %eax
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
fadd %st(1), %st
jbe
fstp %st(1)
subl $12, %esp
fstpl (%esp)
pushl $.LC1
call printf
xorl %eax, %eax
movl %ebp, %esp
popl %ebp
ret
.Lfe1:
.size main,.Lfe1-main
.ident "GCC: (GNU) 2.96 20000731 (Linux-Mandrake 8.0 2.96-0.48mdk)"
Do you think you can make it much faster using hand-crafted assembly code? Admittedly, I used C instead of C++, but that doesn't make any difference for anything as small as this.
/* some test code */
if (i == 0)
Neither is this... It's off the Egyptian coast. With just a bit of imagination, one could describte it as south of Greece. Especially since the Greeks where quite 'Greece-centric'.
Mod this up!!
I'm not sure about it, but I think in France you get automatically French citizenship when you are born in France. Don't know about Germany, though.
I suggest you read Eric S. Raymond's "The magic cauldron" for a thoughtfull essay on that subject: http://tuxedo.org/~esr/writings/magic-cauldron/
Call me stupid, but I don't really understand this. The information in the CDDB database is not some kind of secret or anything, it is public information. Walk into any record store and you'll find the same information.
Granted, the technology used to correlate the CD with it's track list might be proprietary (but then why Gracenote sue FreeDB for patent infringement?), but I can not understand how Gracenote can have any right whatsoever to the information itself.
If FreeDB is infringing on a CDDB patent, shouldn't CDDB be suing FreeDB instead of Roxio??
I feel the need to clarify this a bit...
There is no such thing as arresting small-time copyright violators here in Belgium.
(1) Indeed there *was* a threat from the IFPI (www.ifpi.org: "IFPI is the organisation representing the international recording industry")
(2) But that threat was only to a small number (100) of users, who had received a warning before
(3) Everyone and his brother says IFPI is not able to identify these users unless they violated the privacy legislation
(4) And, most importantly: Marc Verwilghen, our minister of justice, has declared that the prosecution of small-time piracy has the very lowest priority. This places it, I think, just above cannabis use.
So the whole thing is just some FUD from IFPI.
Or if you run multiple apps concurrently, like compiling the kernel while applying some heavy filters in the Gimp, or even when using make -j.
Exactly what do they mean with 500.000 years computing time? 500.000 years computing time on a 8086 is something entirely different than the same amount of computing time on an Athlon 1 GHz, so I don't really know what their statement is supposed to mean.
No, true realtime OSes must guarantee latencies. Not necessarily low ones. Of course, systems with low latencies are preferred over systems with high latencies, but that doesn't make them any more real time.
I don't know the full story on wether Flash is open or closed, but I do know that the file format is available for everybody interested. Everyone can make applications that export Flash, and everyone can make Flash players. Even Adobe, Macromedia's most important concurrent, has a Flash editor. I don't know if we need much more openness(sp?) than that.
No it wasn't. Unfortunately, 99% of web designers don't know that. I have a big suspicion that most of them don't even know what HTML even is. If these guys want to be taken seriously by developers, then they need to start holding html code reviews, logging bugs and defects against their design, and instituting the bare minimums of software engineering.
I happen to know a few web designers, and discussions I have had with them have led me to this conclusion: it's not so much that they don't know what HTML is supposed to be; that's not even relevant.
Reality is that there are lots of people who don't want to use content-based markup -- they want full control over the layout. They know (from experience, if nothing else) HTML doesn't offer that (and, insofar it does, it does so completely different and the different platforms). Therefore the web designers I know try to use HTML only when they really need to, and use Flash for everything else. It gives them complete control over the layout, it handles interactivity, it is uniformly supported on a whole range of platforms.
You're sure bit fields would make it faster? I agree of course it would take less memory than two booleans, but I think it would be slower. Booleans can be retrieved from memory in one step, while a bit field can only be interpreted by XOR'ing and AND'ing against a mask, which takes extra time.
It works, but administering it is quite complex (since nobody at the company knows enough about it or has the time to learn it, we still only use it for internal mail - still using an external server with POP and SMTP for external mail).
It's quite annoying to have to install Openmail client drivers on every client PC, it's impossible to change username and/or password on a client without hacking the registry, and it seriously slows down some of the client machines.
We're using dynamic DNS updates with BIND v8.2.1 and the ISC DHCP Server V3.0b2pl6, with no problems. The relevant lines from /etc/dhcpd.conf:
ddns-domainname "ourdomain";
ddns-rev-domainname "in-addr.arpa";
In /etc/named.conf:
in the relevant zones, include the line
allow-update { list of hosts that are allowed to update the dns };
Main Entry: sarcastic /-ti-k(&-)lE/ adverb
Pronunciation: sär-'kas-tik
Function: adjective
Date: 1695
1 : having the character of sarcasm
2 : given to the use of sarcasm : CAUSTIC - sarcastically
synonyms SARCASTIC, SATIRIC, IRONIC, SARDONIC mean marked by bitterness and a power or will to cut or sting. SARCASTIC implies an intentional inflicting of pain by deriding, taunting, or ridiculing . SATIRIC implies that the intent of the ridiculing is censure and reprobation . IRONIC implies an attempt to be amusing or provocative by saying usually the opposite of what is meant . SARDONIC implies scorn, mockery, or derision that is manifested by either verbal or facial expression .
According to that dictionary, there doesn't seem to be that much difference between sarcastic and ironic. Still, it seems to me that ironic is more relevant in this case.
Don't you think he was being ironic?
Nah... unfortunately (?) it's spelled zee-eend nowadays
In C, not declaring a return type for a function is the same (but more clear) as declaring it as an int; that's not the problem. The problem is that you effectively have to return something, which doesn't happen here, otherwise it's not legal C. But it's quite possible gcc will just issue a warning and then happily continues to compile. I don't remember if gcc issues a warning or an error, I haven't tried to forget the return lately.
Wel, als het niet meer gebruikt wordt is het archaïsch. En kan je veel beter "dal" gebruiken. Q.E.D.
Well, if it's not used anymore, it's archaic. En it's better to use "dal". Q.E.D.