No no no. If you were around at the time, you'd
have known that it was "Need Another Seven Astronauts". Bad taste perhaps, but it's still
one of my favourite jokes of all time.
20 Machines with 512 processors? I think of that more as 10420 machines, not just twenty. Impressive!
You may think that, but you'd be wrong. It's 20
machines. After all, you don't think of a 100 CPU
Sun E15K as 100 machines, or even a dual CPU
desktop as two machines.
SSI on Linux has come a long way...
With the X-Prize being won, and interest from Virgin, there appears to be a new era of space
tourism about to dawn. Naturally, this will not
be without its teething troubles. My question is,
how do you think the first fatal accident involving
a private spacecraft will affect non-governmental
space flight? A minor setback, or years of
stagnation as the various authorities refuse
to grant launch permits while they try and regulate
the industry? If it's the latter (as I expect),
what can be done in advance to improve the situation?
*Looks out window*
No horsemen, or fire in the sky. So, no.
That about sums it up. In fact, my next
car (in common with all my other cars) won't run
any OS at all[1], not even Linux[2]. It will in fact run
a tried and trusted combination of petrol and air,
in an appropriate ratio to get as close to stoichiometric combustion as I can manage.
[1] No, I don't even have an engine management
system.
[2] Actually, that may not strictly be true. I'm
considering using it as a testbed for some data
acquisition stuff I'm developing (think RacePak, so
at various times it may occasionally
have an onboard computer).
Tell me again how the kernel implements web standards, George.
Yep, the article is almost utterly devoid of
useful content, and much of what it does have is
simply plain wrong[1]. It reads very much like
some of the Linux articles did a few years ago...
"Oooh, I've just found this great new OS, so I'm
going to pimp it everywhere I can, even though I
don't know enough about it to do a decent advocacy
job and avoid looking like a fool". Sigh. FWIW, I
use both Linux and OpenBSD.
[1] The particular one that gets me is the oft
quoted, but now inaccurate claim that NetBSD
supports more platforms than Linux. That was
indeed once true, but it hasn't been for a few
years[2].
[2] Unless you use NetBSD's somewhat arbitrary
definition of a platform. Either way, Linux runs
on more CPU architectures than NetBSD does.
That's not a straightforward question to answer.
Simple answer: yes. Long answer: yes, but there's
no readily downloadable binary yet. I haven't been
following it in a while, but see
this post from Jody in April. I believe that
it's built successfully on Windows since then, but
obviously it's still going to be a bit raw around
the edges at the moment.
For future features/direction, I'd suggest that rather than adding in yet another additional funky feature that less than 1% of people will ever find/use, I'd ensure rock solid filters to import/export from MS Office.
Try gnumeric.
It blows the OO.o spreadsheet out of the water in
virtually every respect.
Of course, you can use shared memory, but thats more of a hassle to set up.
Perhaps. But it's also far less error prone than
threads sharing variables when one thread forgets
to get its locks right. That in turn leads to more
reliable and stable software, and is generally
considered a better model. With mmap/MAP_SHARED,
you have to explicitly reference the data
you wish to
share, so the programmer tends to think more about
what they're doing. With threads, everything is
shared by default, and it's much easier to
overlook the fact that another thread might be
using that data.
It would be interesting if someone were to compare the performance of Tomcat vs Resin Open Source.
Having evaluated Resin as a possible replacement
for our current Tomcat setup, we found it to be
about half the speed of Tomcat[1]. So Resin without
the performance bits would be very slow indeed.
It's also worth noting that Resin is apalling in
terms of being shipped in a deployable state
(it isn't). We needed to write decent start/stop
scripts, and move the logs to somewhere sensible,
for example.
[1] FWIW, our developers saw a 5x speed increase
when using Resin in a lab environment. But we
saw exactly the opposite when we tried to move
it out of the lab.
I know a young single mother in the US who got sued and had to use her kid's college fund to pay the RIAA. Sorry, but piracy or no piracy, that simply isn't right
Strange definition of right and wrong you have
there. I'd say it's very much right. If you break
the law, expect to pay the consequences.
Don't get me
wrong, I'm no fan of the RIAA/BPI. But taking
something that doesn't belong to you is wrong,
no matter which way you look at it. I agree that media
shifting should be legal (which it currently isn't
in the UK). But I can't see any justification for
allowing rampant copyright infringement. Sure,
the copyright system is flawed and
needs overhauling. The music industry is also
corrupt, depriving artists of a fair reward for
their endeavors. But the solution is to work
to get the problems fixed, not to just pretend
that copyright doesn't exist.
I'd like to see that comparision as well, but with Linux using open source drivers on documented hardware. (I know that would disqualify a lot of nVidia and ATI chips from the test.)
I was wondering if it was just me. For some of us,
absolute performance isn't the ultimate goal. But
I'd still like to know which card will give me the
best performance without having to resort to closed
source drivers. I won't buy Nvidia or newer ATI
cards because they won't publish programming
information for them.
when the linux community can make a powerful desktop thats not SLOWER than windows2k/xp then i will switch.
The community has already done that, many many years ago. The problem
is, the distributions all insist on shipping the
bloated GNOME or KDE desktops instead of something
that runs at a sensible speed.
I replaced NIS with OpenLDAP on a small network and have a lot of love for it, but your example looked like a Sendmail config file rewritten as APL macros piped through Perl with a couple of trips through Babelfish.
And that is why I'm still using NIS. The wire protocol for LDAP may well be very efficient. But
LDAP in general, and OpenLDAP in particular, is a nightmare to configure, and I just
don't have the time to beat in into submission.
NIS is up and running in 5 minutes, and requires
essentially zero administration on a home network.
in my experience Bluetooth is a battery killer, so I only use it when I'm not so mobile
Funny, I'm the exact opposite. The only time I use
Bluetooth is when I'm on the move, so I can connect
my Zaurus to the net through my mobile phone. Haven't noticed any problems with battery life
at all...
Forget Blue Tooth, an outdoor device requires GPS.
GPS is a nice luxury, but Bluetooth is an essential for any mobile device. Given the choice between the two, I'd lose the GPS. Of course I'd rather have both, though...
But Linux isn't ready for the big iron machines Solaris dominates yet. Don't say IBM, please. IBM runs multitudes of instances of the Linux kernel in parallel on their machines, so that if one fails, it doesn't take the whole system down. Those big iron Sun machines run one kernel, baby. Just one.
Alright, then, I won't say IBM. I'll say SGI instead. A single Linux kernel on 256 CPUs? Yep, and 512 by the end
of the year, apparently. You can't go beyond 144 CPUs on anything Sun currently sell. Linux may
not have the high end sewn up yet, but SSI on big
iron is no longer just a pipe dream. Just playing
devil's advocate...
I wouldn't be remotely surprised if this venture gets quietly binned, once its provided its much needed channels to allow Branson time to plug his newly launched credit card.
I wouldn't call it newly launched. The Virgin card
was launched in either late 2001 or early 2002,
IIRC. My company was in discussions with Virgin and
ANB to do the card, but in the end they went with
one of our competitors.
In the end, we ended up buying licenses for Torque from GarageGames
I've always wondered why there seem to be so few
(any?) licencees of Croteam's
Serious Engine. Are they just pricing themselves
out of the market, or something? I mean, Serious
Sam has proved that it's more than capable of
getting the job done, and Serious Editor looks to
provide all the tools you could want.
So why aren't people using it?
What made you choose Torque?
What advantages did it have over the competition?
Nope. True, you can use an existing XML parser to
do it, but only if it's well formed (which you can't
guarantee, as not all of the data may have arrived
by the time you want to start parsing).
HTML 4 is the bloated recommendation with the styling tags and all that crud.
I don't think you understand what HTML 4 is. There
is no difference between HTML 4 and XHTML
other than syntax. No extra tags, no "crud".
XHTML (especially XHTML 2) is doing what HTML was in the beginning and got lost later: developing a good semantic and structural markup language, leaving layout to CSS.
Ahhh yes. That's a fine description of what HTML 4 does...
Slashdot Mirror
Errm... which one's which?
No no no. If you were around at the time, you'd have known that it was "Need Another Seven Astronauts". Bad taste perhaps, but it's still one of my favourite jokes of all time.
You may think that, but you'd be wrong. It's 20 machines. After all, you don't think of a 100 CPU Sun E15K as 100 machines, or even a dual CPU desktop as two machines. SSI on Linux has come a long way...
With the X-Prize being won, and interest from Virgin, there appears to be a new era of space tourism about to dawn. Naturally, this will not be without its teething troubles. My question is, how do you think the first fatal accident involving a private spacecraft will affect non-governmental space flight? A minor setback, or years of stagnation as the various authorities refuse to grant launch permits while they try and regulate the industry? If it's the latter (as I expect), what can be done in advance to improve the situation?
What, you mean like this? Oh, sometimes I'm just so l33t! :-)
That about sums it up. In fact, my next car (in common with all my other cars) won't run any OS at all[1], not even Linux[2]. It will in fact run a tried and trusted combination of petrol and air, in an appropriate ratio to get as close to stoichiometric combustion as I can manage.
[1] No, I don't even have an engine management system.
[2] Actually, that may not strictly be true. I'm considering using it as a testbed for some data acquisition stuff I'm developing (think RacePak, so at various times it may occasionally have an onboard computer).
Yep, the article is almost utterly devoid of useful content, and much of what it does have is simply plain wrong[1]. It reads very much like some of the Linux articles did a few years ago... "Oooh, I've just found this great new OS, so I'm going to pimp it everywhere I can, even though I don't know enough about it to do a decent advocacy job and avoid looking like a fool". Sigh. FWIW, I use both Linux and OpenBSD.
[1] The particular one that gets me is the oft quoted, but now inaccurate claim that NetBSD supports more platforms than Linux. That was indeed once true, but it hasn't been for a few years[2].
[2] Unless you use NetBSD's somewhat arbitrary definition of a platform. Either way, Linux runs on more CPU architectures than NetBSD does.
As a sysadmin, I would write a longer comment here, but I've got to go as my weekly AD&D session is about to start...
That's not a straightforward question to answer. Simple answer: yes. Long answer: yes, but there's no readily downloadable binary yet. I haven't been following it in a while, but see this post from Jody in April. I believe that it's built successfully on Windows since then, but obviously it's still going to be a bit raw around the edges at the moment.
Try gnumeric. It blows the OO.o spreadsheet out of the water in virtually every respect.
Perhaps. But it's also far less error prone than threads sharing variables when one thread forgets to get its locks right. That in turn leads to more reliable and stable software, and is generally considered a better model. With mmap/MAP_SHARED, you have to explicitly reference the data you wish to share, so the programmer tends to think more about what they're doing. With threads, everything is shared by default, and it's much easier to overlook the fact that another thread might be using that data.
Having evaluated Resin as a possible replacement for our current Tomcat setup, we found it to be about half the speed of Tomcat[1]. So Resin without the performance bits would be very slow indeed. It's also worth noting that Resin is apalling in terms of being shipped in a deployable state (it isn't). We needed to write decent start/stop scripts, and move the logs to somewhere sensible, for example.
[1] FWIW, our developers saw a 5x speed increase when using Resin in a lab environment. But we saw exactly the opposite when we tried to move it out of the lab.
Strange definition of right and wrong you have there. I'd say it's very much right. If you break the law, expect to pay the consequences. Don't get me wrong, I'm no fan of the RIAA/BPI. But taking something that doesn't belong to you is wrong, no matter which way you look at it. I agree that media shifting should be legal (which it currently isn't in the UK). But I can't see any justification for allowing rampant copyright infringement. Sure, the copyright system is flawed and needs overhauling. The music industry is also corrupt, depriving artists of a fair reward for their endeavors. But the solution is to work to get the problems fixed, not to just pretend that copyright doesn't exist.
I was wondering if it was just me. For some of us, absolute performance isn't the ultimate goal. But I'd still like to know which card will give me the best performance without having to resort to closed source drivers. I won't buy Nvidia or newer ATI cards because they won't publish programming information for them.
Having tried writing apps in both, I'd have to disagree with you. Given the choice, my future GUI development with be in GTK+.
The community has already done that, many many years ago. The problem is, the distributions all insist on shipping the bloated GNOME or KDE desktops instead of something that runs at a sensible speed.
And that is why I'm still using NIS. The wire protocol for LDAP may well be very efficient. But LDAP in general, and OpenLDAP in particular, is a nightmare to configure, and I just don't have the time to beat in into submission. NIS is up and running in 5 minutes, and requires essentially zero administration on a home network.
Ask and you shall receive.
Funny, I'm the exact opposite. The only time I use Bluetooth is when I'm on the move, so I can connect my Zaurus to the net through my mobile phone. Haven't noticed any problems with battery life at all...
GPS is a nice luxury, but Bluetooth is an essential for any mobile device. Given the choice between the two, I'd lose the GPS. Of course I'd rather have both, though...
Here in the UK, ripping a CD is illegal.
Alright, then, I won't say IBM. I'll say SGI instead. A single Linux kernel on 256 CPUs? Yep, and 512 by the end of the year, apparently. You can't go beyond 144 CPUs on anything Sun currently sell. Linux may not have the high end sewn up yet, but SSI on big iron is no longer just a pipe dream. Just playing devil's advocate...
I wouldn't call it newly launched. The Virgin card was launched in either late 2001 or early 2002, IIRC. My company was in discussions with Virgin and ANB to do the card, but in the end they went with one of our competitors.
I've always wondered why there seem to be so few (any?) licencees of Croteam's Serious Engine. Are they just pricing themselves out of the market, or something? I mean, Serious Sam has proved that it's more than capable of getting the job done, and Serious Editor looks to provide all the tools you could want. So why aren't people using it? What made you choose Torque? What advantages did it have over the competition?
Nope. True, you can use an existing XML parser to do it, but only if it's well formed (which you can't guarantee, as not all of the data may have arrived by the time you want to start parsing).
HTML 4 is the bloated recommendation with the styling tags and all that crud.
I don't think you understand what HTML 4 is. There is no difference between HTML 4 and XHTML other than syntax. No extra tags, no "crud".
XHTML (especially XHTML 2) is doing what HTML was in the beginning and got lost later: developing a good semantic and structural markup language, leaving layout to CSS.
Ahhh yes. That's a fine description of what HTML 4 does...