Linux-Advocacy Digest #650, Volume #33 Mon, 16 Apr 01 20:13:06 EDT
Contents:
Re: OT: Treason (was Re: Communism) (Roberto Alsina)
Re: Pete Goodwin is in good company (Pete Goodwin)
Re: there's always a bigger fool (Chris Ahlstrom)
Re: So much for modules in Linux! ("Gary Hallock")
Re: there's always a bigger fool (Roberto Alsina)
Re: Pete Goodwin is in good company (Pete Goodwin)
Re: To Eric FunkenBush (Donovan Rebbechi)
Re: To Eric FunkenBush (Chronos Tachyon)
Re: Pete Goodwin is in good company (Pete Goodwin)
Re: Pete Goodwin is in good company (Pete Goodwin)
Re: To Eric FunkenBush (Chronos Tachyon)
Re: OT: Treason (was Re: Communism) (Chad Everett)
Re: To Eric FunkenBush (Donovan Rebbechi)
Re: To Eric FunkenBush (Roberto Alsina)
Re: t. max devlin: kook (Donovan Rebbechi)
Re: To Eric FunkenBush (Chronos Tachyon)
Re: Communism, Communist propagandists in the US...still..to this day. (Gunner �)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Roberto Alsina)
Crossposted-To: alt.destroy.microsoft,us.military.army,soc.singles
Subject: Re: OT: Treason (was Re: Communism)
Date: 16 Apr 2001 22:46:11 GMT
Reply-To: [EMAIL PROTECTED]
Chad Everett <[EMAIL PROTECTED]> wrote:
>On 16 Apr 2001 22:12:44 GMT, Roberto Alsina <[EMAIL PROTECTED]> wrote:
>>
>>
>>I wouldn't know, my knowledge of bible dates is sketchy. If you say it
>>happened in 1300BC, then so be it, and consider my 3000BC date incorrect.
>
>OK. thanks.
>
>>However, are you saying that hundreds of years of direct contact to god made
>>the israelits MORE RUTHLESS?
>
>Thousands of years of contact with God didn't do the Israelites much good.
>Their fault, not God's.
So, say, 1700 years earlier, would they have behaved in pretty much the
same way?
>>>Second:
>>>
>>>I specifically challenged your false assertion of "In 3000BC, the
>>>standard included massive rape of the women of the defeated". I
>>>never challenged that there was a whole lot of killing going on.
>>
>>Mind you, if you replace "massive rape of the women" with "massive
>>slaughter of babies" in the original post it makes absolutely no
>>difference. This is all just a digression.
>
>If you had made the same statement about a "standard" of massive killing
>of babies in 3000BC, I would have called you on that too.
You surely like to nitpick.
>> [ - snip -]
>>
>>>There was certainly no "standard of massive rape of the women of the
>>>defeated". In fact, the exact opposite was the standard because this was
>>>specifically prohibited by their laws, and God certainly never told them
>>>to.
>>
>>Care to cite where that is? Do you put rape as morally worse than killing
>>babies?
>>
>I think they are both equally reprehensible.
I don't.
>>>>> Show us an example in 3000BC where
>>>>>>>>>>>massive rape of the defeated women occurred. Since you throw
>>>>>>>>>>>out such blatantly false statements as fact, everything you
>>>>>>>>>>>say is questionable.
>>>>
>>>>Always happy to serve.
>>>>
>>>
>>>Thank you for proving my point
>>
>>What precisely was your point?
>>
>>> and conceding so happily.
>>
>>What do you believe I am conceding? That the 3000BC date was wrong?
>
>That there was no standard in 3000BC of massive rape of women of the defeated.
I am not conceding that yet. Let's work a little more in that.
[below things were edited for clarity]
>>Hell, that's such a collateral issue it makes no difference whatsoever
>>to the real argument: does god allow war?
> Yes
Where does he allow war? I see him ORDERING war. Not the same thing.
>> does war go against the commandments?
> No.
Are you implying that any act occuring in a war is not against
the commandments?
>> Is the commandment about any kind of killing or a
>> specific kind?
>A specific kind.
What kind? It doesn't seem to forbid killing babies after the war
is over. It doesn't seem to forbid human sacrifices. It doesn't
seem to forbid slaughter of defenseless prisoners.
What kind of killing IS forbidden?
>You were trying to make your false assertion a supporting point in your
>argument about the above.
Parse error.
>>Heck, let's start another digression: do you believe this massacre was a
>>holy endeavour?
>
>I don't know.
How can you not know what you believe?
You may not know if it was, but that is not what I am asking.
Are you a worshipper of a brutal deity that requires the mass
slaughter of babies under penalty of plague?
--
Roberto Alsina
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: Pete Goodwin is in good company
Reply-To: [EMAIL PROTECTED]
Date: Mon, 16 Apr 2001 22:39:21 GMT
Terry Porter wrote:
>> Well, stop using ISA and switch to PCI as it works?
> Why?
Why not? Why use old cards? That's what other advocates have already said
to me.
> I have heaps of ISA cards and I use ISA prototyping cards for various
> projects. Linux is excellent for getting off the 'Upgrade Bandwaggon'.
I have two network cards. They're both cheap and they're both PCI. Why use
old technology when I can use new?
> Hahahah, I believe you again, but you dont *have* to use KDE do you ?
>
> What does KDE offer you that other proven Linux WM's and apps don't ?
KDE gives me desktop close to what Windows offers in some respects.
>> THEN WHY IS IT NOT MARKED AS UNDER DEVELOPMENT???
> I have to admit,I dont use it myself, but I'm sure the doc would state
> that it is?
> Do you really believe KDE is finished ???
> Or are you saying that KDE *is finished*, but unreliable ?
What does KDE say about itself:
"KDE is a mature desktop suite providing a solid basis to an ever growing
number of applications for Unix workstations. KDE has developed a high
quality development framework for Unix, which allows for the rapid and
efficient creation of applications."
A _mature_ desktop? Note there's no mention of it being beta!
>> And what do you think I use my PC for?
> Tell us ?
Programming.
>> Suddenly the so called choices on Linux start to drop. GNOME is the
>> alternative. What happens if I don't like GNOME? Tough, is that it?
> Not at all, there are other window managers that are reliable and easy to
> use. Blackbox is one, why not try it ?
Why not fvwm2? Why not just a menu? Can't you guess?
> Is it a problem for you to hear that Linux is NOT Windows ?
No. But it is nowhere near a replacement for Windows!
> It never was, it never will be (I hope).
But it _should_ be a replacement for Windows, except it isn't.
--
Pete
Running on SuSE 7.1, Linux 2.4, KDE 2.1
Kylix: the way to go!
------------------------------
From: Chris Ahlstrom <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: there's always a bigger fool
Date: Mon, 16 Apr 2001 22:57:11 GMT
"Aaron R. Kulkis" wrote:
>
> Roberto Alsina wrote:
> >
> > On Mon, 16 Apr 2001 15:48:28 -0400, Aaron R. Kulkis <[EMAIL PROTECTED]> wrote:
> > >Matthew Gardiner wrote:
> > >>
> > >> Just as a follow up for Pete to read, isn't libc one of the most
> > >> important libraries on the OS? hence, its not just one of those minor
> > >> ones, its a major one, and requires no reboot. Now thats what I call
> > >> convenient.
> > >
> > >Yes. without libc, NOTHING works.
> >
> > Well, here again Aaron shows how little he knows about the inner
> > workings of a unix system.
> >
> > Had he said "nothing works" I would have let it pass. But no,
> > he had to say "NOTHING works".
> >
> > Bullshit, Aaron. Read this and weep (if you understand it):
> >
> > [ralsina@pc03 ralsina]$ ldd /sbin/* |grep "not a"|wc -l
> > 16
> >
> > Ok, you probably didn't. Here it is in detail: no dynamic
> > binary will work. But all static ones will.
>
> Who uses staticly linked apps these days?
That's one bug up my ass these days. Who the hell cares if you link
statically? It's faster and safer, and the extra space is generally
meaningless with today's big-ass PC hard-drives.
To HELL with DLL Hell.
Chris
--
"Where do you want to hang today?"
------------------------------
From: "Gary Hallock" <[EMAIL PROTECTED]>
Subject: Re: So much for modules in Linux!
Date: Mon, 16 Apr 2001 18:57:40 +0000
In article <VJwC6.4655$[EMAIL PROTECTED]>, "Pete Goodwin"
<[EMAIL PROTECTED]> wrote:
> I moved DHCP to after the network loaded. It still loaded DHCP before
> loading the network modules.
>
Then you didn't do what you think you did. Did you make sure that you
removed any starting of DHCP that had previously been done before
starting the network - are you trying to load DHCP twice?
Gary
------------------------------
From: [EMAIL PROTECTED] (Roberto Alsina)
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: there's always a bigger fool
Date: 16 Apr 2001 22:59:52 GMT
Reply-To: [EMAIL PROTECTED]
On Mon, 16 Apr 2001 22:57:11 GMT, Chris Ahlstrom <[EMAIL PROTECTED]> wrote:
>"Aaron R. Kulkis" wrote:
>>
>> Roberto Alsina wrote:
>> >
>> > On Mon, 16 Apr 2001 15:48:28 -0400, Aaron R. Kulkis <[EMAIL PROTECTED]> wrote:
>> > >Matthew Gardiner wrote:
>> > >>
>> > >> Just as a follow up for Pete to read, isn't libc one of the most
>> > >> important libraries on the OS? hence, its not just one of those minor
>> > >> ones, its a major one, and requires no reboot. Now thats what I call
>> > >> convenient.
>> > >
>> > >Yes. without libc, NOTHING works.
>> >
>> > Well, here again Aaron shows how little he knows about the inner
>> > workings of a unix system.
>> >
>> > Had he said "nothing works" I would have let it pass. But no,
>> > he had to say "NOTHING works".
>> >
>> > Bullshit, Aaron. Read this and weep (if you understand it):
>> >
>> > [ralsina@pc03 ralsina]$ ldd /sbin/* |grep "not a"|wc -l
>> > 16
>> >
>> > Ok, you probably didn't. Here it is in detail: no dynamic
>> > binary will work. But all static ones will.
>>
>> Who uses staticly linked apps these days?
>
>That's one bug up my ass these days. Who the hell cares if you link
>statically? It's faster and safer, and the extra space is generally
>meaningless with today's big-ass PC hard-drives.
Actually, the problem is not disk space, but memory usage
and cache efficiency.
--
Roberto Alsina (that gueses a desktop linux would need some
300MB of RAM if everything was static)
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: Pete Goodwin is in good company
Reply-To: [EMAIL PROTECTED]
Date: Mon, 16 Apr 2001 22:42:06 GMT
Matthew Gardiner wrote:
> Haven't had a crash yet. Maybe I'm just lucky, however, I have a pretty
> generic computer. 384MB RAM, 500MB Swap, 60gig hdd, TNT 2 Graphics Card,
> Sound Blaster Live!CD-R and CD-ROM, Zip100 USB, etc, etc. Just a quick
> query, have you upgraded your kernel? I have upgraded mine from 2.4.0
> (which was provided in the CD-ROM), to the new one from the
> update/kernel directory on the suse site, which is 2.4.2-4GB. The
> stability is very good. I don't know whats wrong with your computer,
> but I have not experienced any of those problems. If the problem
> persists after updating the kernel, conduct a memory test, which is a
> selection in the boot menu, just to make sure that a dodgy memory module
> isn't at fault.
Mine is a generic machine: 128MByte RAM, 500MByte swap, 10 + 14 gig hdd's,
Voodoo 5500, various CDROM's, DVD's, ZIP 100 SCSI.
Why would upgrading my kernel fix my problems? They look to me as though
they're in KDE itself, not Linux.
--
Pete
Running on SuSE 7.1, Linux 2.4, KDE 2.1
Kylix: the way to go!
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Subject: Re: To Eric FunkenBush
Date: 16 Apr 2001 23:06:26 GMT
On Mon, 16 Apr 2001 14:13:43 GMT, Chronos Tachyon wrote:
> On Sun 15 Apr 2001 10:45, Donn Miller wrote:
> Haven't you noticed that C++ programs take five times as long to compile,
> and the result is considerably larger? Compiling Hello, World! with gcc
> yields a stripped executable size of 2984 bytes; compiling the exact same
> program by giving the exact same options to g++ produces a stripped
Are you sure it's *exactly* the same program ? (you're not using
<iostream> in the C++ version ?)
> executable size of 11660 bytes, nearly four times the size. Even with the
elflord@ruffbruff ~ $ cat test.cc
#include <cstdio>
using namespace std;
int main()
{
printf ("hello world\n");
return 0;
}
elflord@ruffbruff ~ $ g++ test.cc
elflord@ruffbruff ~ $ strip ./a.out
elflord@ruffbruff ~ $ ls -l ./a.out
-rwxrwxr-x 1 elflord elflord 3364 Apr 16 19:05 ./a.out
elflord@ruffbruff ~ $
> is O(n log n) when compiling code, while C++ is O(n^2). KDE would be a
> prime example, although any massively large C++ project will do. On my
> poor widdle 300MHz box, almost every source tarball in KDE took between one
> and two hours to compile. Is KDE really *that* big?
Yes, KDE really is "that big". Try counting the lines of code some time.
Last I counted, kdebase and kdelibs both had about 800 C++ source files,
and each of those are probably around 1000 lines each.
But you have a point, C++ does take a long time to compile. Moreover, it's
trickier to get dependencies right in C++.
Of course, it's not clear that C would be any faster (or smaller) to
compile if you wanted to do OO programming.
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
From: Chronos Tachyon <[EMAIL PROTECTED]>
Subject: Re: To Eric FunkenBush
Date: Mon, 16 Apr 2001 23:06:26 GMT
On Mon 16 Apr 2001 05:37, Erik Funkenbusch wrote:
> "Chronos Tachyon" <[EMAIL PROTECTED]> wrote in
> message
>> Haven't you noticed that C++ programs take five times as long to compile,
>> and the result is considerably larger?
>
> C programs compiled as C++ are larger because C++ programs have features
> that C doesn't have, such as global object instantiation and exception
> handling. The startup code must do more for a C++ program. Of course,
> the linker could be smarter about this and it could link in the standard C
> startup code when the expanded features aren't in the code.
>
> This is an implementation detail though and not something defined by the
> language. It could be fixed if it were that big of a deal, but apparently
> it's not.
>
Wrong, I'm afraid. It *is* defined by the language. Surely you're not
implying that it's possible to write a C++ compiler (targeted to a much
larger and more [mis-]feature rich language) that can compile code as
efficently as a C compiler, are you?
> C++ programs on average take longer because they include more stuff,
> causing the compiler to do more work. This is aggravated by people
> tending to #include stuff more often in C++ when they don't really need
> to. This is more of a programmer discipline issue. Dependancies should
> be as minimal as possible, but many people simply don't do this.
>
This conveniently skirts the fact that, in C, an #include is usually
harmless both to compile time and binary size.
> Again, it's really not technically the compilers fault, but rather
> programmer fault. The same C program compiled as C++ should include the
> exact same files as the same program compiled as C. This may not be the
> case with g++, but it certainly is with compilers like MetroWerks or VC++.
>
It's certainly the case with g++. The problem is language bloat,
especially exception handling.
>> prime example, although any massively large C++ project will do. On my
>> poor widdle 300MHz box, almost every source tarball in KDE took between
>> one and two hours to compile. Is KDE really *that* big?
>
> Yes, it's that big. Remember, it includes all of QT as well.
>
No, it doesn't. Qt is already compiled as a shared library. You don't
recompile MSVCRT40.DLL every time you write a Windoze app, do you?
>> On a side note, I find most of C99 rather disappointing, for exactly the
>> same reasons.
>
> You don't have to use all the new features, and there are ways to reduce
> the hits that you percieve. You just have to know what you're doing.
>
Specifically, I refer to the K5 story about the changes that C99 brings:
<http://www.kuro5hin.org/?op=displaystory&sid=2001/2/23/194544/139>
A, D, K, L, M: Useful.
B, C, I, J, O: Syntactic sugar.
E, F, G, H, N: Pure language bloat.
Although only G, H, K, and N seem to have the potential to bloat code size
(with K and N mostly confined to the runtime library), most of these
features will make C99 a notably more complex language to parse, which will
slow down compile times and make my life as a programmer suck more.
--
Chronos Tachyon
Guardian of Eristic Paraphernalia
Gatekeeper of the Region of Thud
[Reply instructions: My real domain is "echo <address> | cut -d. -f6,7"]
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: Pete Goodwin is in good company
Reply-To: [EMAIL PROTECTED]
Date: Mon, 16 Apr 2001 22:48:18 GMT
Tom Wilson wrote:
> Have any of your systems ever screamed obscenities in Latin or
> bazooka-barfed pea soup all over the place?
No, why have yours? You must have strange peripherals attached - is the pea
soup dispenser an essential item?
--
Pete
Running on SuSE 7.1, Linux 2.4, KDE 2.1
Kylix: the way to go!
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: Pete Goodwin is in good company
Reply-To: [EMAIL PROTECTED]
Date: Mon, 16 Apr 2001 22:48:17 GMT
Tom Wilson wrote:
> Have any of your systems ever screamed obscenities in Latin or
> bazooka-barfed pea soup all over the place?
No, why have yours? You have very strange peripherals attached to your
machine in that case. Is the pea soup dispenser an essential item?
--
Pete
Running on SuSE 7.1, Linux 2.4, KDE 2.1
Kylix: the way to go!
------------------------------
From: Chronos Tachyon <[EMAIL PROTECTED]>
Subject: Re: To Eric FunkenBush
Date: Mon, 16 Apr 2001 23:13:19 GMT
On Mon 16 Apr 2001 06:06, Donovan Rebbechi wrote:
> On Mon, 16 Apr 2001 14:13:43 GMT, Chronos Tachyon wrote:
>> On Sun 15 Apr 2001 10:45, Donn Miller wrote:
>
>> Haven't you noticed that C++ programs take five times as long to compile,
>> and the result is considerably larger? Compiling Hello, World! with gcc
>> yields a stripped executable size of 2984 bytes; compiling the exact same
>> program by giving the exact same options to g++ produces a stripped
>
> Are you sure it's *exactly* the same program ? (you're not using
> <iostream> in the C++ version ?)
>
Yep, it's exactly the same program. I didn't even change the extension.
>> executable size of 11660 bytes, nearly four times the size. Even with
>> the
>
> elflord@ruffbruff ~ $ cat test.cc
> #include <cstdio>
> using namespace std;
>
> int main()
> {
> printf ("hello world\n");
>
> return 0;
> }
>
> elflord@ruffbruff ~ $ g++ test.cc
> elflord@ruffbruff ~ $ strip ./a.out
> elflord@ruffbruff ~ $ ls -l ./a.out
> -rwxrwxr-x 1 elflord elflord 3364 Apr 16 19:05 ./a.out
> elflord@ruffbruff ~ $
>
Odd. My GCC is 2.95.2, what's yours? I used -O2, BTW...
>> is O(n log n) when compiling code, while C++ is O(n^2). KDE would be a
>> prime example, although any massively large C++ project will do. On my
>> poor widdle 300MHz box, almost every source tarball in KDE took between
>> one and two hours to compile. Is KDE really *that* big?
>
> Yes, KDE really is "that big". Try counting the lines of code some time.
> Last I counted, kdebase and kdelibs both had about 800 C++ source files,
> and each of those are probably around 1000 lines each.
>
Yep, I'm fully aware of how large KDE is in terms of lines-of-code. My
impression was that it should have been a 5-hour overnight compile, not a
two-day affair. It certainly took longer to compile than Gnome did.
> But you have a point, C++ does take a long time to compile. Moreover, it's
> trickier to get dependencies right in C++.
>
> Of course, it's not clear that C would be any faster (or smaller) to
> compile if you wanted to do OO programming.
>
I'm not arguing that. I'm arguing that for those who want a nice, tidy
language shouldn't have to put up with the bloat of turning C into (C++)--.
--
Chronos Tachyon
Guardian of Eristic Paraphernalia
Gatekeeper of the Region of Thud
[Reply instructions: My real domain is "echo <address> | cut -d. -f6,7"]
------------------------------
From: [EMAIL PROTECTED] (Chad Everett)
Crossposted-To: alt.destroy.microsoft,us.military.army,soc.singles
Subject: Re: OT: Treason (was Re: Communism)
Reply-To: [EMAIL PROTECTED]
Date: 16 Apr 2001 17:59:19 -0500
On 16 Apr 2001 22:17:54 GMT, Roberto Alsina <[EMAIL PROTECTED]> wrote:
>Chad Everett <[EMAIL PROTECTED]> wrote:
>
>>>>>>, all of you.28 This is my blood of the covenant,
>>>>>>which is poured out for many for the forgiveness of sins." He, of course,
>>>>>>was referring to the cup ("drink from it"), as in when he later asks God:
>>>>>>"if it be your will, take this cup from me".
>>>>>
>>>>>So, you say that the important thing was not drinking, but drinking
>>>>>from a specific cup?
>>>>>
>>>>
>>>>No. The important thing was eating and drinking in remembrance of the sacrifice
>that
>>>>Christ was making of his body and blood. What you eat and drink made/makes little
>>>>difference to the "sacrament" that he introduced.
>>>
>>>Well, he seems to be specific about the eating and drinking being of bread and
>>>wine. Where does he say that it doesn't matter if you do it with broccoli and
>>>orange juice?
>>>
>>>If you gonna take it seriously, you can't just ignore stuff. He gave them wine
>>>and said "this is my blood". Why do you believe any other fluid would do?
>>>
>>
>>Follow your own advice. Matthew doesn't quote him as saying "this is my blood"
>>he quotes Jesus as saying "this is My blood of the new covenant". Not the same
>>thing, now is it?
>
>What's the difference? BTW: I saw "my blood of the covenant" not "of the new
>covenant".
>
"new covenant" is in all translations I am aware of.
Christ was establishing a NEW covenant between God and Man. He said "this is My
blood of the new covenant, which is shed for many for the remission of sin."
He was referring directly to Jeremiah 31:31 and to Exodus 24:8. His disciples
would have been very familiar with both of these passages and understood
immediately the signigicance and symbolism of what he was saying.
>>>In fact, why do catholics believe any current day wine may do? He said THAT
>>>wine was his blood.
>>
>>No he didn't. Matthew quotes him as saying "This is My blood of the new covenant".
>
>Ok, whatever. "In fact, why do catholics believe any current day wine may do? He
> said THAT wine was his blood of the new covenant". happy?
>
Roman Catholics believe in "transubstantiation". I do not. I would have trouble
explaining WHY someone believes something I do not. Maybe because the Pope tells
them to?
>I don't see how that makes any difference whatsoever to the question.
>
>Care to answer?
>
Sure. You originally made ridiculous statements claiming that there was some
sort of command from God to drink wine on Sundays. You made this ridiculous
statement in an attempt to support some sort of strange argument about God, and
murder, and killing, and war.
I simply stated that God never made any such command, which is true of course.
You have since made several rather silly attempts at using poor and incomplete
biblical quotes to question the fact that your original statement about drinking
wine on Sundays was indeed completely absurd and incorrect.
You seem to have a knack for changing the subject and playing merri-go-round with
questions when someone calls you on completely false statements you make in your
arguments.
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Subject: Re: To Eric FunkenBush
Date: 16 Apr 2001 23:19:57 GMT
On Sun, 15 Apr 2001 00:34:20 -0700, GreyCloud wrote:
> Donovan Rebbechi wrote:
> I'll end this thread from my viewpoint: if you try to use the examples
> in a well known book using VC++ and find it can't compile it and then
> try another compiler on UNIX and it works just fine... I'll take the
> UNIX compiler anyday.
You'll run into problems on any compiler. For example, try using
stringstream in a default g++ installation. The other killer misfeature
is that it accepts broken code (namely, the std library is not in namespace
std)
I wouldn't consider lack of support for partial specialisations to be
substantially worse than g++'s misfeatures (though I agree that it's
a useful feature)
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
From: [EMAIL PROTECTED] (Roberto Alsina)
Subject: Re: To Eric FunkenBush
Date: 16 Apr 2001 23:21:25 GMT
Reply-To: [EMAIL PROTECTED]
On 16 Apr 2001 23:06:26 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>On Mon, 16 Apr 2001 14:13:43 GMT, Chronos Tachyon wrote:
>> On Sun 15 Apr 2001 10:45, Donn Miller wrote:
>
>> Haven't you noticed that C++ programs take five times as long to compile,
>> and the result is considerably larger? Compiling Hello, World! with gcc
>> yields a stripped executable size of 2984 bytes; compiling the exact same
>> program by giving the exact same options to g++ produces a stripped
>
>Are you sure it's *exactly* the same program ? (you're not using
><iostream> in the C++ version ?)
>
>> executable size of 11660 bytes, nearly four times the size. Even with the
>
>elflord@ruffbruff ~ $ cat test.cc
>#include <cstdio>
>using namespace std;
>
> int main()
> {
> printf ("hello world\n");
>
> return 0;
> }
>
>elflord@ruffbruff ~ $ g++ test.cc
>elflord@ruffbruff ~ $ strip ./a.out
>elflord@ruffbruff ~ $ ls -l ./a.out
>-rwxrwxr-x 1 elflord elflord 3364 Apr 16 19:05 ./a.out
>elflord@ruffbruff ~ $
>
>> is O(n log n) when compiling code, while C++ is O(n^2). KDE would be a
>> prime example, although any massively large C++ project will do. On my
>> poor widdle 300MHz box, almost every source tarball in KDE took between one
>> and two hours to compile. Is KDE really *that* big?
>
>Yes, KDE really is "that big". Try counting the lines of code some time.
>Last I counted, kdebase and kdelibs both had about 800 C++ source files,
>and each of those are probably around 1000 lines each.
This is a very rough measurement:
wc says header files have in kdelibs have 140571 lines, and source
files 366379 lines, for a total of about half a million.
In kdebase: 70540 and 279871 for a total of about 350000.
>But you have a point, C++ does take a long time to compile. Moreover, it's
>trickier to get dependencies right in C++.
>
>Of course, it's not clear that C would be any faster (or smaller) to
>compile if you wanted to do OO programming.
Lots of the time it takes to compile C++ is caused by the header files.
You see, in C, a header file contained prototypes, type definitions,
macros, and perhaps small inline functions.
On C++, they also contain templates.
So, including a C++ header often has a very non-trivial effect on the compile
time.
Anyway: I've heard rumours that precompiled headers are coming to g++,
and that they improve compilation speed almost by an order of magnitude.
--
Roberto Alsina (compiling)
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Subject: Re: t. max devlin: kook
Date: 16 Apr 2001 23:23:23 GMT
On Sat, 14 Apr 2001 22:57:24 -0700, Tim Hanson wrote:
> Donovan Rebbechi wrote:
> I'm just starting to play around with it on a spare computer, Reiser but
> not RAID. So far so good; Everything seems to map just fine. 2.4
> kernel.
Cool. The RAID is in hardware, we're going to use a 3Ware IDE RAID card,
so it shouldn't be that much of a problem.
I think I'll just take the plunge and use Reiser, the outside chance of
having NFS acting up for a few months is not as dire as putting critical
data on ext2. (Apart from anything else, we should be encouraging users
to connect to the RAID servers and run their apps directly from there)
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
From: Chronos Tachyon <[EMAIL PROTECTED]>
Subject: Re: To Eric FunkenBush
Date: Mon, 16 Apr 2001 23:26:57 GMT
On Mon 16 Apr 2001 06:21, Roberto Alsina wrote:
[Snip]
>
> Anyway: I've heard rumours that precompiled headers are coming to g++,
> and that they improve compilation speed almost by an order of magnitude.
>
Woohoo, if the PCH's follow some sort of binary standard and people who
write C++ code get into the habit of including them in the source tarball.
--
Chronos Tachyon
Guardian of Eristic Paraphernalia
Gatekeeper of the Region of Thud
[Reply instructions: My real domain is "echo <address> | cut -d. -f6,7"]
------------------------------
From: Gunner � <[EMAIL PROTECTED]>
Crossposted-To:
misc.survivalism,alt.fan.rush-limbaugh,soc.singles,alt.society.liberalism,talk.politics.guns
Subject: Re: Communism, Communist propagandists in the US...still..to this day.
Date: Mon, 16 Apr 2001 16:29:06 -0700
On 16 Apr 2001 21:00:34 GMT, [EMAIL PROTECTED] (Roberto Alsina) wrote:
>>
>>Sliverdick forgets that THE ENTIRE FUCKING PLANET WAS IN A DEPRESSION
>>(INCLUDING HIS BELOVED SOVIET UNION!!!)
>
>Depression? Maybe you meant recession?
>
>--
>Roberto Alsina
Not in 1925-34
Gunner
"So it was that four hours later, carrying two hundred cigarettes, completely drunk
and with a
half-naked, unmarried Filipino lady, I emerged, behind the writing desk in the
Headmaster's
study-simultaneously breaking a hundred and twenty-seven school rules. The Chaplain,
now
seventy-four and impatient to get his Archbishopric, had finished the tunnel just a
hundred yards
too early."
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.advocacy.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
