It's a very real problem with all projects
for which FSF owns the copyright:
In order to submit a patch you have you get
a copyright assignment physically signed by
the holy hand of RMS. This can take literally
many months. I just gave up on GCJ myself.
"Your patch doesn't have to be 'accepted' to be
used by you" -- true enough, but why bother?
If I was interested in my own personal benefit,
I would have used a closed source product in the
first place.
No, usoft is the biggest drain on the closed-
source world, costing the economy literally
billions of lost man-hours as a result of
malignance and incompetence, while the biggest
analogous impediment to human progress in the
free world is the FSF.
.
Agreed. I never put two IDE HDs on the same
(non-RAID) controller when I care about
performance. But I'd rather have 2 IDEs on
two onboard controllers than 2 SCSI drives on
one controller.
Isn't this an obsolete belief? With ata/133
and udma-4, the only advantage i know of for
scsi comes from tagged command queueing. If
you have a bofunk scsi controller, you don't
even get that.
I know that SCSI used to be better. I just
don't know any reason to believe that it still
is.
It's a cute ploy,
to bless VC++ with the C++ logo by
waving Lippmann's name over it, but it's quite
clear in the text of the article that MS has no
intention of implementing C++ or of being
interoperable with conformant implementations of
C++. VC++ is in fact being crushed in the vice
of embrace-and-extend. They don't WANT you to be
able to write portable code, and it's getting
harder every day, not easier.
I'm repeating myself on purpose because
it's true and important, and i don't care
about my karma if the truth is at stake.
Let's see now, 2.1GB/sec per CPU (PC2100 DDR),
it would take less than 68 Athlons to provide the
bandwidth to memory that you describe. To get
600 MB/sec between system boards would require
3 bonded gig-e nics per system. For 34 dual cpu
(COTS) boxes, that's less than 100 ports on the switch. Yeah, you can do with with COTS. Now if
you want *cross-sectional* bandwidth, it takes a lot more switches, but the PC's aren't the problem.
There are a number of real-time GC systems,
all of them incremental, some of them exact.
Embedded Java is big business, and in many
cases uses real-time GC.
Everybody who writes at the code project site
is whoring for the beast. Lippmann is now
whoring for the beast. It's a cute ploy,
to bless VC++ with the C++ logo by
waving Lippmann's name over it, but it's quite
clear in the text of the article that MS has no
intention of implementing C++ or of being
interoperable with conformant implementations of
C++. VC++ is in fact being crushed in the vice
of embrace-and-extend. They don't WANT you to be
able to write portable code, and it's getting
harder every day, not easier.
The powers that be are obviously idiots,
so the question you should be asking instead
is something like "what is the best recruiter
board"? I like dice.com. Whenever it becomes
obvious that management has chosen the downward
spiral, I just hop over to an enterprise that
might have a better future. Since my experience
and qualifications have compounded, I generally
get a better compensation package each time.
I'd really really really like to buy a Mac
of some sort, but they always come with these
one-button mice.... Completely unusable.
Did they fix it this time?
Versions of gcj before gcc 3.0.1 were essentally
unusable. It's pretty solid now. The hotspot
compiler is better still, but it doesn't let
you deliver native code, or transparently intercall
between Java and C++ classes. libgcj is pretty
much JDK 1.2 compatible now. Yes, there's no AWT
or Swing.
Re:Haldeman doesn't get it -- minor spoiler
on
The Forever War
·
· Score: 1
Economic factors *are* a misunderstanding -- a
misunderstanding of what has true value. In order
to understand something, you must stand under it.
If you try to comprehend it, you are enveloping it,
and asserting your superiority. As a result, you
can never understand, much to your own detriment.
Or at least, it is only a speedbump. I know no
Chinese to speak of, but I chat up the college
girls from Guangzhou to Haerbin on a regular basis
using OICQ and Babelfish. Eventually they all
start to learn English, which takes the bloom off
the rose for me, and I dump them.
Funny thing is, that Chinese-speaking majority
is going to be spending most of its free time
learning English, so they can get out of China.
.. will always be with us. Let's just accept
them as such, and be sure to inform their bosses
that they are such, since otherwise they will
harm many other people. One day they may change.
Slashdot Mirror
It's a very real problem with all projects
for which FSF owns the copyright:
In order to submit a patch you have you get
a copyright assignment physically signed by
the holy hand of RMS. This can take literally
many months. I just gave up on GCJ myself.
"Your patch doesn't have to be 'accepted' to be
used by you" -- true enough, but why bother?
If I was interested in my own personal benefit,
I would have used a closed source product in the
first place.
No, usoft is the biggest drain on the closed-
source world, costing the economy literally
billions of lost man-hours as a result of
malignance and incompetence, while the biggest
analogous impediment to human progress in the
free world is the FSF.
.
Hehe. Can you say "halting problem"?
Agreed. I never put two IDE HDs on the same
(non-RAID) controller when I care about
performance. But I'd rather have 2 IDEs on
two onboard controllers than 2 SCSI drives on
one controller.
Because nobody with an ethical bone in their body
would buy RAMBUS.
Nor, for that matter would anyone who cared about
getting a fair value for their dollar.
Isn't this an obsolete belief? With ata/133
and udma-4, the only advantage i know of for
scsi comes from tagged command queueing. If
you have a bofunk scsi controller, you don't
even get that.
I know that SCSI used to be better. I just
don't know any reason to believe that it still
is.
.
it's just a rumor, unless you can provide
a link to the driver itself. if you can, i will
mirror it.
So where can I get this, since Yahoo took it down?
I want to mirror it.
It's a cute ploy,
to bless VC++ with the C++ logo by
waving Lippmann's name over it, but it's quite
clear in the text of the article that MS has no
intention of implementing C++ or of being
interoperable with conformant implementations of
C++. VC++ is in fact being crushed in the vice
of embrace-and-extend. They don't WANT you to be
able to write portable code, and it's getting
harder every day, not easier.
I'm repeating myself on purpose because
it's true and important, and i don't care
about my karma if the truth is at stake.
Let's see now, 2.1GB/sec per CPU (PC2100 DDR),
it would take less than 68 Athlons to provide the
bandwidth to memory that you describe. To get
600 MB/sec between system boards would require
3 bonded gig-e nics per system. For 34 dual cpu
(COTS) boxes, that's less than 100 ports on the switch. Yeah, you can do with with COTS. Now if
you want *cross-sectional* bandwidth, it takes a lot more switches, but the PC's aren't the problem.
There are a number of real-time GC systems,
all of them incremental, some of them exact.
Embedded Java is big business, and in many
cases uses real-time GC.
No, an exact garbage collector does not scan all
the data in memory. Yes, if you have a lot of
allocations, GC takes longer. So does free().
I CAN'T BELIEVE I'm not reaping Funny karma
from this one. Amazing. Y'all are a bunch o'
philistines!
Everybody who writes at the code project site
is whoring for the beast. Lippmann is now
whoring for the beast. It's a cute ploy,
to bless VC++ with the C++ logo by
waving Lippmann's name over it, but it's quite
clear in the text of the article that MS has no
intention of implementing C++ or of being
interoperable with conformant implementations of
C++. VC++ is in fact being crushed in the vice
of embrace-and-extend. They don't WANT you to be
able to write portable code, and it's getting
harder every day, not easier.
> What would Alan call his version of kernel?
I know what I would call it: Coccyx.
The powers that be are obviously idiots,
so the question you should be asking instead
is something like "what is the best recruiter
board"? I like dice.com. Whenever it becomes
obvious that management has chosen the downward
spiral, I just hop over to an enterprise that
might have a better future. Since my experience
and qualifications have compounded, I generally
get a better compensation package each time.
Can I get one will all three mouse buttons
intact? I'll dump this presario for an
ibook if i can upgrade from 2 buttons to the
full 3.
> My company is going to MS based web services and > .net too
Please do tell me the name of your company,
so that I can boycott it.
I'd really really really like to buy a Mac
of some sort, but they always come with these
one-button mice.... Completely unusable.
Did they fix it this time?
Versions of gcj before gcc 3.0.1 were essentally
unusable. It's pretty solid now. The hotspot
compiler is better still, but it doesn't let
you deliver native code, or transparently intercall
between Java and C++ classes. libgcj is pretty
much JDK 1.2 compatible now. Yes, there's no AWT
or Swing.
That's why Makefile-based Java projects should
use jikes instead of javac.
It included a link to goatse.cx, so it deserved to be modded off the face of the earth.
Happy Kwanza indeed.
Economic factors *are* a misunderstanding -- a
misunderstanding of what has true value. In order
to understand something, you must stand under it.
If you try to comprehend it, you are enveloping it,
and asserting your superiority. As a result, you
can never understand, much to your own detriment.
Or at least, it is only a speedbump. I know no
Chinese to speak of, but I chat up the college
girls from Guangzhou to Haerbin on a regular basis
using OICQ and Babelfish. Eventually they all
start to learn English, which takes the bloom off
the rose for me, and I dump them.
Funny thing is, that Chinese-speaking majority
is going to be spending most of its free time
learning English, so they can get out of China.
.. will always be with us. Let's just accept
them as such, and be sure to inform their bosses
that they are such, since otherwise they will
harm many other people. One day they may change.