I realize you're making a joke. However, there are people that want huge amounts of computing power occasionally, and it isn't cheap.
Right now, the guys down the hall are distributing complex physics simulations to about one hundred computers across three or four distributed sites. We're planning on scaling it up soon. These physicists will take all of the computing power we can throw at the problem. They don't necessarily need the power to be in-house--running remotely is fine for them.
Recently we helped an economics graduate student use about twenty four years of CPU time. Economics grad students aren't generally wealthy, and that kind of CPU time doesn't come cheaply.
And then there are the biologists doing protein analysis, and the astronomers analyzing images of remote regions of the universe, and...
IBM has a reasonable goal. If you want to get a lot of computing done and you don't want to invest in a full-time staff and a machine room and tens or hundreds of PC, it might be pretty reasonable to buy some computer time from them.
This isn't as simple as timesharing. It's also not for your average user that wants to browse the web, write email, and chart the expenditures in their checking account.
-alain
Ealier online smileys are known: MacKenzie, Apr79
on
The First Smiley :-)
·
· Score: 2, Interesting
It is well known that online smilies go farther back than the example provided here. The famous MSGGROUP (a very early mailing list, begun in June of 1975) had an earlier example of the smiley.
On 12-April-1979, Keving MacKenzie wrote:
Date: 12 APR 1979 1736-PST From: MACKENZIE at USC-ECL Subject: MSGGROUP#1015 METHICS and the Fast Draw(cont'd) To: ~drxal-hda at OFFICE-1 cc: msggroup at MIT-MC, malasky at PARC-MAXC
In regard to your message a few days ago concerning the loss of meaning in this medium:
I am new here, and thus hesitate to comment, but I too have suffered from the lack of tone, gestures, facial expressions etc. May I suggest the beginning of a solution? Perhaps we could extend the set of punctuation we use, i.e:
If I wish to indicate that a particular sentence is meant with tongue-in-cheek, I would write it so:
"Of course you know I agree with all the current administration's policies -)."
The "-)" indicates tongue-in-cheek.
This idea is not mine, but stolen from a Reader's Digest article I read long ago on a completly different subject. I'm sure there are many other, better ways to improve our punctuation.
Any comments?
Kevin
The MSGGROUP archives used to be easily browsable, I think. I found this mail message a couple of years ago. Today when I looked for it, I only found the compressed archives. You can find them online at: http://ftp.std.com/obi/Networking/archives/msggrou p/
The message in question is the file named "msggroup.1001-1100-z".
I'm not the first person to note this. If you search for msggroup with Google, you'll find other people that have noted it. Even the Economist notes this earlier occurence.
-alain
Apple has proposed Zeroconf has an internet standard. It's still only in draft form: http://www.zeroconf.org/
If you're pushing for an Internet standard, it certainly helps to have an open source implementation of the standard. Check out the RFC describing the Internet Standards Process. You'll note that a proposed standard: A Proposed Standard specification is generally stable, has resolved known design choices, is believed to be well-understood, has received significant community review, and appears to enjoy enough community interest to be considered valuable.
This doesn't state that you have to release open source, but it helps the case when you want to can get other people to look at your code and see what you've done.
Also note that a draft standard requires: A specification from which at least two independent and interoperable implementations from different code bases have been developed, and for which sufficient successful operational experience has been obtained, may be elevated to the "Draft Standard" level. Yes, they are independent implementations, but it can help other people develop theirs if they can see your source code. When my implementation doesn't interoperate with Apple's, I can actually figure out why, and whose fault it is.
So yes, releasing the source code is good if Apple wants this to become a standard.
You might be interested in GriPhyn, which is on of the American data grid projects that has been funded. It is a collaboration between physicists and computer scientists, with similar goals to the EU Data Grid.
Buried on the web site is the original proposal they made, and it gives you some idea of the amount of data we're working with.
Some approximate statistics from the paper:
SDSS gets data at 8MB/s, 10TB/year.
LIGO will get data at 10MB/s, 250TB/year.
CMS will get data at 100MB/s, 5 Petabytes per year.
Work has already been done with simulated data for CMS, and a demo of virtual data (may be pre-calculated, or calculated on demand) for CMS was shown at the Supercomputing 2001 conference last week. They used Condor clusters from a few different sites. I'm not sure which sites made it into the final demo, but it may have included U. Florida, Argonne, and U. Wisconin.
Although I haven't used it recently, I used to regularly own and love the Metrowerks IDE, called Codewarrior. It is a very nice IDE that support Jave, C, and C++. The debugger is quite good.
I used it on the Mac for C and C++, and I haven't used it in two major revisions. So possibly my experiences won't reflect yours. But I do think it's worth checking out.
Codewarrior runs on a bunch of platforms, including Mac, Windows, Solaris, and Linux. I only have experience with the Mac and Windows version, personally.
Slashdot Mirror
I realize you're making a joke. However, there are people that want huge amounts of computing power occasionally, and it isn't cheap. Right now, the guys down the hall are distributing complex physics simulations to about one hundred computers across three or four distributed sites. We're planning on scaling it up soon. These physicists will take all of the computing power we can throw at the problem. They don't necessarily need the power to be in-house--running remotely is fine for them. Recently we helped an economics graduate student use about twenty four years of CPU time. Economics grad students aren't generally wealthy, and that kind of CPU time doesn't come cheaply. And then there are the biologists doing protein analysis, and the astronomers analyzing images of remote regions of the universe, and... IBM has a reasonable goal. If you want to get a lot of computing done and you don't want to invest in a full-time staff and a machine room and tens or hundreds of PC, it might be pretty reasonable to buy some computer time from them. This isn't as simple as timesharing. It's also not for your average user that wants to browse the web, write email, and chart the expenditures in their checking account. -alain
Apple has proposed Zeroconf has an internet standard. It's still only in draft form: http://www.zeroconf.org/
If you're pushing for an Internet standard, it certainly helps to have an open source implementation of the standard. Check out the RFC describing the Internet Standards Process. You'll note that a proposed standard: A Proposed Standard specification is generally stable, has resolved known design choices, is believed to be well-understood, has received significant community review, and appears to enjoy enough community interest to be considered valuable.
This doesn't state that you have to release open source, but it helps the case when you want to can get other people to look at your code and see what you've done.
Also note that a draft standard requires: A specification from which at least two independent and interoperable implementations from different code bases have been developed, and for which sufficient successful operational experience has been obtained, may be elevated to the "Draft Standard" level. Yes, they are independent implementations, but it can help other people develop theirs if they can see your source code. When my implementation doesn't interoperate with Apple's, I can actually figure out why, and whose fault it is.
So yes, releasing the source code is good if Apple wants this to become a standard.
-alain
You can read about it at: www.griphyn.org
Buried on the web site is the original proposal they made, and it gives you some idea of the amount of data we're working with.
Some approximate statistics from the paper:
SDSS gets data at 8MB/s, 10TB/year.
LIGO will get data at 10MB/s, 250TB/year.
CMS will get data at 100MB/s, 5 Petabytes per year.
Work has already been done with simulated data for CMS, and a demo of virtual data (may be pre-calculated, or calculated on demand) for CMS was shown at the Supercomputing 2001 conference last week. They used Condor clusters from a few different sites. I'm not sure which sites made it into the final demo, but it may have included U. Florida, Argonne, and U. Wisconin.
Although I haven't used it recently, I used to regularly own and love the Metrowerks IDE, called Codewarrior. It is a very nice IDE that support Jave, C, and C++. The debugger is quite good. I used it on the Mac for C and C++, and I haven't used it in two major revisions. So possibly my experiences won't reflect yours. But I do think it's worth checking out. Codewarrior runs on a bunch of platforms, including Mac, Windows, Solaris, and Linux. I only have experience with the Mac and Windows version, personally.