Yes. Try a 100 user machine. Each user running their own Java environment with swing apps etc. How many cpu's/memory will you need? (rethorical)
Two things here. Firstly, to answer your question: in my experience you could get by with a four-CPU box of say, 600MHz processors, and maybe four gigabytes of RAM if you have nice fast hard drives. Now that's not a very high-specced machine in the world of multi-user machines. In fact, most people would go for the more expensive option of having a two-CPU fileserver then giving each user their own workstation. This is, as well as costly, a taller order for admins. I'd prefer the single-machine approach using SunRays or similar stateless clients, as you need much less hardware and only one OS/JVM between the users.
Secondly, a rhetorical question is one that presents an idea of viewpoint solely in its asking. Your question is merely one that you would prefer people not to answer, so it contains no rhetoric at all.
You're probably able to answer a question of mine then - if MKLinux uses a Mach microkernel, and Linux is a different kernel [also with a different design, being megalithic], how is MKLinux actually Linux? Is it Linux in the way that, say, Darwin is BSD? By providing an emulation layer?
Or possibly for someone to tell him how to cache certain requests, how to ensure that all code is compiled to native machine language, that the FreeBSD version of the runtime is x% faster, etc. etc. etc. The answer "no" neither confirms nor denies the possibility of speeding up.NET code.
The real question is:
Will it handle a lot of clients at once on very little hardware?
The answer is: No.
You missed out the bit of your post where you explain why you don't believe that.NET can scale. And the answer because it's Microsoft and therefore slow and evil isn't going to be considered admissable evidence, neither is it aer not haev teh unix as this is a question of scalability not operating system architecture.
Try again, only this time please expand on your answer, explaining why it is that you don't think the.NET platform is scalable. I'd be interested to hear your answer too; I've only used.NET so far on a test box, hosting a local user (and the beta version of the software I was using ate system resources, so I stopped) and wonder whether MS are expecting all small to medium enterprises to want to shell out lots of money on hardware as well as software.
This many-worlds theory was fashionable in the 1980s among philosophers who couldn't get their heads around this whole quantum mechanics thing, but don't forget that the Copenhagen interpretation provides answers that are identical but without having to invoke additional universes. As another poster in this thread said, Occam's razor prefers the simpler explanation.
None of these other universes would be anything we could interact with, yet they determine (by the particular combination of laws that they don't contain) the fundamental laws of our universe.
That's just an exercise in quasi-philosophical tautology. The two situations, (i)our Universe is unique and obeys the Physics we observe, (ii)our Universe is one of many, and obeys the Physics we observe, are completely identical within the realm of experiment. To argue whether one or the other is the true situation is like trying to argue whether or not the Universe is a computer sim; every piece of 'evidence' for one case is identically 'evidence' for the other.
The Universe means the system that completely encloses all of our interactions and experiences. If we find something, it cannot not be within our Universe, as we could not interact with it. Therefore even if other Universes 'exist', they cannot exist within our experience.
Of course, I forgot to mention the 13.6 electron Volts of free energy one can get by putting one's Electron near to a BBC Model B (obscure, but funny enough to mention;-)
This is the home of positron, the synchronization manager for the Neuros Audio Computer. This software is primarily intended to support Linux, but will probably be portable to other POSIX systems such as OS X and the BSDs. Currently there are USB protocol issues with OS X, so please don't try it there yet.
The above was taken from a link in the above story. Now, the problem I have is that I own three of these, Acorn Electrons. Can I simultaneously use an Electron and Positron, or am I going to end up annihilating something?
Thanks to the help of people here as well as others online, I now have my hardware working fine under Linux (would like to point out that I hadn't not tried before, I just didn't know the tricks). Still a big fan of Solaris and my Linux admin-ness is getting there reasonably quickly. I was cheated a little by starting out with Linux (on the PC at any rate) with SuSE then Red Hat, which let you install a system without having to administer it:-)
If I'd dropped myself in at the deeper end (say, with Gentoo or Sourceror) then I might have found learning how to administer the system a lot easier.
One thing I can think of, is that if you're using ALSA, you shouldn't be building the kernel modules (or compiling support into the kernel) for your sound cards.
That was exactly what was wrong - thanks very much. rmmodded the soundcard kernel modules, and everything started working again. Just shows that Linux can be easy to set up, as long as I get off my arse and find out how:-)
It didn't have a precompiled image for my kernel, and was building from source. However, it always failed the CC sanity check, even with IGNORE_CC_MISMATCH set. Eventually, I managed to remember which CC I used to build my kernel, rolled back to that temporarily and everything worked. So that's the NVidia bit sorted out:-)
As for alsa, I'm stumped. I have two soundcards, an intel i810 on-board and an emu10k1 (i.e. SB Live! 1024). I do 'apt-get alsa-source' then 'make-kpkg' to create the modules.deb, which I install. However, alsaconf complains that it can't find a soundcard, and if I try to manually set up a driver anyway it still (obviously) doesn't work. I've just deleted all the alsa stuff and am about to start again from scratch, but any advice you could give would be warmly welcome.
I for one only use Linux because the hardware support for Solaris/x86 isn't up to the job, and I can't afford a Mac (yet; don't want to have to sell my PC). I also use Windows, because there are moments when I need to use my sound card and so far setting up alsa with kernel 2.4.20 hasn't proven too fruitful. Come to that, neither has installing the genuine NVidia drivers for my graphics card, so GLX has been out of the question too.
But for day-to-day computing I'd always much rather have a Unix-style system (hope the Open Group don't sue me for using a hyphen there), so even when I do go into Windows I have Cygwin/XFree/Window Maker there, should I start feeling pangs of nostalgia.
I'm not sure I understand your logic. The reason that 'Libre' can't be used is because Americans wouldn't understand it, is what I think you're saying. But RMS requires largenumber of pages to explain the word 'Free' as in freedom, which is a word used in America. So I don't think it would make many odds:-)
Libre Software would be appropriate, but it uses foreign words, which is confusing.
To Linus Torvalds, the words 'Free' and 'Software' are foreign. Actually, they're foreign to most of the people who work at SuSE, or Mandrake (they're French, so 'Libre Software' would be quite nice), many of the GNUstep community, the people at Red Flag Linux, and so forth. And in a time when many of the Free Software projects are striving for ubiquitous i18n support, is using the odd alien (from the Old French) word such am bad thing?
Hint for considering which words people will find foreign: the language spoken by most people in the world is Mandarin. English is (I think) fourth, with French (from whence, the word 'libre') being second.
Did it ever occur to you to (i)get your facts right, and (ii)stop being a tosswit before you started posting? It looks like I received mod points for being right, and you have received none for being wrong.
The truth gets one the respect from ones peers that one rightfullys deserves, should one choose to emit the truth. You really should consider giving that a go.
FWIW, Hotmail ran on BSD for a number of years, before Microsoft bought it out. They then sent a huge crack team of MCSEs (if such a thing exists:-) in to switch everything over to Windows, and they did everything apart from the advertising servers. It was run like this for a couple of years, then some Linux fanboi said "look! Microsoft use Unix!" and they changed the ad servers too. I've had my Hotmail account for around six years, and have been receiving stupid volumes of spam for about three years. Even when Microsoft took over, it was a useful service for a few years.
Of course, we all know Microsoft don't use UNIX at all, do they? In fact, they neverdid.
In the words of the time honoured Spectrum games, Congraturation! You sucsess!
You're exactly right. The eStar project is a robotic telescope network, as well as having various databases of previous observations. The telescopes and the databases can be queried in the same way, with the databases returning rows and the telescopes zooming off to look at the object in question. All of this is controlled in Perl.
This takes some of the glamour out of astronomy in favour of more rapid results. I still remember the images of Hubble sat at the lower end of a very very big telescope, gripping a notebook in one hand and a pipe between his teeth as he readied himself for a night-long observing session. Still, if Larry Wall can find us some aliens then more power to him, I say.
I'm a Physics UG at the University of Oxford, and have done the odd part-time job writing code for some of the groups (I worked for CMP last year and will start in the Teaching Course next month). All of the code that I have written was released as OSS, because other research groups can then modify it and use it in their experiments, without affecting the originality of our work.
The systems I worked on in CMP were almost entirely Red Hat, and some of the code I had to work with was being ported to C/Linux from Delphi/Windows it was stupid to ask people to pay for Delphi licenses just to tweak a few lines of code in a detector controller. Some of the older computers were running Win95 with X11 and SSH clients and were just used as terminals for the beefier Linux boxen.
The computer I'll be using in my next project is a little meatier, a four CPU Sparc box. There are a couple of these in the department, each with a few dozen SunRays attached. The code developed on here will still be OSS, and the box has a fair amount of the GNU userland sitting side-by-side with the Solaris rubbish.
Other than that, my knowledge is a bit sketchy and is inferred through what I hear or see around the department. There are a number of Windows workies around, because software like SPSS or Origin or Minitab only exists on Windows, and there either don't exist Unix or Linux alternatives, or they aren't yet mature enough to want to switch onto. Many of these workstations also have Cygwin or at least DJGPP.
But the main point to be answered is whether or not a research group should be OSS. My opinion is yes as far as in-house code is concerned, because this facilitates collaboration between groups using similar code, hence quickly smoothing bugs. OSS also neatly fits the philosophy of shared information many scientists have.
Should the computers all use Open Source operating systems? I just don't think this should be a requirement in most cases. Yes Linux and the BSDs are stable and mature enough to be used in a research environment, but then so are Slowlaris and Mac OS X, and you can develop and build your code on both of these systems too. The only situation OTTOMH in which having an OSS OS would be directly beneficial is when custom-built embedded devices are required, for instance in HEP detectors or beam controllers. In these circumstances the ability to modify the OS kernel would be useful, but if you're just developing your C code in a Unix environment to be used on bog-standard Unix or Linux machines, then a good set of manpages and a functional cc is really all that's needed.
Once you are used to multiple desktops (no, that measly 4 add-on powertoys desktops don't count.),
When I'm in Windows [.NET server rc2] I use Litestep, which has four nice proper virtual desktops. It also has a much smaller footprint than the usual Explorer.exe GUI. As you could probably guess I use Window Maker everywhere else (including XFree for Windows;-)
For some reason, British Telecom operators refer to the # key as "the square key". Well on my phone the two upright lines are not perpendicular to the two horizontal lines, so it's more of a rhombus than a square, but when I explain that to the ops, they hang up...
Slashdot Mirror
Two things here. Firstly, to answer your question: in my experience you could get by with a four-CPU box of say, 600MHz processors, and maybe four gigabytes of RAM if you have nice fast hard drives. Now that's not a very high-specced machine in the world of multi-user machines. In fact, most people would go for the more expensive option of having a two-CPU fileserver then giving each user their own workstation. This is, as well as costly, a taller order for admins. I'd prefer the single-machine approach using SunRays or similar stateless clients, as you need much less hardware and only one OS/JVM between the users.
Secondly, a rhetorical question is one that presents an idea of viewpoint solely in its asking. Your question is merely one that you would prefer people not to answer, so it contains no rhetoric at all.
I usually attempt to steal definitions from dictionary.com, but if this definion thing floats your boat, bully for you.
:-)
Also, I only use dictionary.reference.com when I don't have access to dictionary.oed.com
Mmmmm...GNU/Linux/Open/Mach...nice
Thanks
You're probably able to answer a question of mine then - if MKLinux uses a Mach microkernel, and Linux is a different kernel [also with a different design, being megalithic], how is MKLinux actually Linux? Is it Linux in the way that, say, Darwin is BSD? By providing an emulation layer?
The German Web server of the SCO Group GmbH is not at present attainable.
And even if you don't know German, I believe Google language tools is available in America these days.
Or possibly for someone to tell him how to cache certain requests, how to ensure that all code is compiled to native machine language, that the FreeBSD version of the runtime is x% faster, etc. etc. etc. The answer "no" neither confirms nor denies the possibility of speeding up .NET code.
You missed out the bit of your post where you explain why you don't believe that .NET can scale. And the answer because it's Microsoft and therefore slow and evil isn't going to be considered admissable evidence, neither is it aer not haev teh unix as this is a question of scalability not operating system architecture.
Try again, only this time please expand on your answer, explaining why it is that you don't think the .NET platform is scalable. I'd be interested to hear your answer too; I've only used .NET so far on a test box, hosting a local user (and the beta version of the software I was using ate system resources, so I stopped) and wonder whether MS are expecting all small to medium enterprises to want to shell out lots of money on hardware as well as software.
This many-worlds theory was fashionable in the 1980s among philosophers who couldn't get their heads around this whole quantum mechanics thing, but don't forget that the Copenhagen interpretation provides answers that are identical but without having to invoke additional universes. As another poster in this thread said, Occam's razor prefers the simpler explanation.
That's just an exercise in quasi-philosophical tautology. The two situations, (i)our Universe is unique and obeys the Physics we observe, (ii)our Universe is one of many, and obeys the Physics we observe, are completely identical within the realm of experiment. To argue whether one or the other is the true situation is like trying to argue whether or not the Universe is a computer sim; every piece of 'evidence' for one case is identically 'evidence' for the other.
The Universe means the system that completely encloses all of our interactions and experiences. If we find something, it cannot not be within our Universe, as we could not interact with it. Therefore even if other Universes 'exist', they cannot exist within our experience.
Of course, I forgot to mention the 13.6 electron Volts of free energy one can get by putting one's Electron near to a BBC Model B (obscure, but funny enough to mention ;-)
The above was taken from a link in the above story. Now, the problem I have is that I own three of these, Acorn Electrons. Can I simultaneously use an Electron and Positron, or am I going to end up annihilating something?
TIA LOL Me too!!!! etc.
Thanks to the help of people here as well as others online, I now have my hardware working fine under Linux (would like to point out that I hadn't not tried before, I just didn't know the tricks). Still a big fan of Solaris and my Linux admin-ness is getting there reasonably quickly. I was cheated a little by starting out with Linux (on the PC at any rate) with SuSE then Red Hat, which let you install a system without having to administer it :-)
If I'd dropped myself in at the deeper end (say, with Gentoo or Sourceror) then I might have found learning how to administer the system a lot easier.
Thanks to people for their help, Graham.
That was exactly what was wrong - thanks very much. rmmodded the soundcard kernel modules, and everything started working again. Just shows that Linux can be easy to set up, as long as I get off my arse and find out how :-)
Cheers, Graham.
It didn't have a precompiled image for my kernel, and was building from source. However, it always failed the CC sanity check, even with IGNORE_CC_MISMATCH set. Eventually, I managed to remember which CC I used to build my kernel, rolled back to that temporarily and everything worked. So that's the NVidia bit sorted out :-)
As for alsa, I'm stumped. I have two soundcards, an intel i810 on-board and an emu10k1 (i.e. SB Live! 1024). I do 'apt-get alsa-source' then 'make-kpkg' to create the modules .deb, which I install. However, alsaconf complains that it can't find a soundcard, and if I try to manually set up a driver anyway it still (obviously) doesn't work. I've just deleted all the alsa stuff and am about to start again from scratch, but any advice you could give would be warmly welcome.
Thanks very much :-)
I for one only use Linux because the hardware support for Solaris/x86 isn't up to the job, and I can't afford a Mac (yet; don't want to have to sell my PC). I also use Windows, because there are moments when I need to use my sound card and so far setting up alsa with kernel 2.4.20 hasn't proven too fruitful. Come to that, neither has installing the genuine NVidia drivers for my graphics card, so GLX has been out of the question too.
But for day-to-day computing I'd always much rather have a Unix-style system (hope the Open Group don't sue me for using a hyphen there), so even when I do go into Windows I have Cygwin/XFree/Window Maker there, should I start feeling pangs of nostalgia.
I'm not sure I understand your logic. The reason that 'Libre' can't be used is because Americans wouldn't understand it, is what I think you're saying. But RMS requires large number of pages to explain the word 'Free' as in freedom, which is a word used in America. So I don't think it would make many odds :-)
To Linus Torvalds, the words 'Free' and 'Software' are foreign. Actually, they're foreign to most of the people who work at SuSE, or Mandrake (they're French, so 'Libre Software' would be quite nice), many of the GNUstep community, the people at Red Flag Linux, and so forth. And in a time when many of the Free Software projects are striving for ubiquitous i18n support, is using the odd alien (from the Old French) word such am bad thing?
Hint for considering which words people will find foreign: the language spoken by most people in the world is Mandarin. English is (I think) fourth, with French (from whence, the word 'libre') being second.
It's a good job you're so sure of your self that you're able to post the URLs to the netcraft results. They look a little like this:
Passport Login running Windows/IIS
Hotmail Main Page running Windows/IIS
Advertising Server now running Windows/IIS, as I promised it would be a couple of posts back.
Image Server running Windows/IIS.
Did it ever occur to you to (i)get your facts right, and (ii)stop being a tosswit before you started posting? It looks like I received mod points for being right, and you have received none for being wrong.
The truth gets one the respect from ones peers that one rightfullys deserves, should one choose to emit the truth. You really should consider giving that a go.
FWIW, Hotmail ran on BSD for a number of years, before Microsoft bought it out. They then sent a huge crack team of MCSEs (if such a thing exists :-) in to switch everything over to Windows, and they did everything apart from the advertising servers. It was run like this for a couple of years, then some Linux fanboi said "look! Microsoft use Unix!" and they changed the ad servers too. I've had my Hotmail account for around six years, and have been receiving stupid volumes of spam for about three years. Even when Microsoft took over, it was a useful service for a few years.
Of course, we all know Microsoft don't use UNIX at all, do they? In fact, they never did.
In the words of the time honoured Spectrum games, Congraturation! You sucsess!
You're exactly right. The eStar project is a robotic telescope network, as well as having various databases of previous observations. The telescopes and the databases can be queried in the same way, with the databases returning rows and the telescopes zooming off to look at the object in question. All of this is controlled in Perl.
This takes some of the glamour out of astronomy in favour of more rapid results. I still remember the images of Hubble sat at the lower end of a very very big telescope, gripping a notebook in one hand and a pipe between his teeth as he readied himself for a night-long observing session. Still, if Larry Wall can find us some aliens then more power to him, I say.
I'm a Physics UG at the University of Oxford, and have done the odd part-time job writing code for some of the groups (I worked for CMP last year and will start in the Teaching Course next month). All of the code that I have written was released as OSS, because other research groups can then modify it and use it in their experiments, without affecting the originality of our work.
The systems I worked on in CMP were almost entirely Red Hat, and some of the code I had to work with was being ported to C/Linux from Delphi/Windows it was stupid to ask people to pay for Delphi licenses just to tweak a few lines of code in a detector controller. Some of the older computers were running Win95 with X11 and SSH clients and were just used as terminals for the beefier Linux boxen.
The computer I'll be using in my next project is a little meatier, a four CPU Sparc box. There are a couple of these in the department, each with a few dozen SunRays attached. The code developed on here will still be OSS, and the box has a fair amount of the GNU userland sitting side-by-side with the Solaris rubbish.
Other than that, my knowledge is a bit sketchy and is inferred through what I hear or see around the department. There are a number of Windows workies around, because software like SPSS or Origin or Minitab only exists on Windows, and there either don't exist Unix or Linux alternatives, or they aren't yet mature enough to want to switch onto. Many of these workstations also have Cygwin or at least DJGPP.
But the main point to be answered is whether or not a research group should be OSS. My opinion is yes as far as in-house code is concerned, because this facilitates collaboration between groups using similar code, hence quickly smoothing bugs. OSS also neatly fits the philosophy of shared information many scientists have.
Should the computers all use Open Source operating systems? I just don't think this should be a requirement in most cases. Yes Linux and the BSDs are stable and mature enough to be used in a research environment, but then so are Slowlaris and Mac OS X, and you can develop and build your code on both of these systems too. The only situation OTTOMH in which having an OSS OS would be directly beneficial is when custom-built embedded devices are required, for instance in HEP detectors or beam controllers. In these circumstances the ability to modify the OS kernel would be useful, but if you're just developing your C code in a Unix environment to be used on bog-standard Unix or Linux machines, then a good set of manpages and a functional cc is really all that's needed.
Waves AmEx These aren't the droids you're looking for...
Obiwan was a bribe merchant!
When I'm in Windows [.NET server rc2] I use Litestep, which has four nice proper virtual desktops. It also has a much smaller footprint than the usual Explorer.exe GUI. As you could probably guess I use Window Maker everywhere else (including XFree for Windows ;-)
For some reason, British Telecom operators refer to the # key as "the square key". Well on my phone the two upright lines are not perpendicular to the two horizontal lines, so it's more of a rhombus than a square, but when I explain that to the ops, they hang up...