
So this just spawned out of a random conversation with a co-worker. ---- NMI (I now have you attention.) ---- Anyone else got some good hardcore comp-sci hummor? -- Scott Sullivan

"Parity is for farmers." -- Seymour Cray That's a famous quote that is hard to figure out. It was from some time in the 1960's. This was after his CDC 6600 was released (without parity) and before his CDC 7600 (which had parity). - Cray is the most famous designer of supercomputers. - "parity" has two meanings in this quote - the main meaning is: extra memory bits to detect some memory or bus failures (ones with an odd number of bit-flips) - the mainstream belief was that parity checking was really useful to detect when things have gone wrong. In that era, lots of things went wrong. - my interpretation is that Cray just didn't want to waste circuitry, ferrite cores, and speed on parity bits. Why speed? Because it would surely affect fan-out or fan-in of some circuits on the critical path of his processor, which would in turn affect how fast they could be driven. Other designers were more humble: they liked to know if the results were corrupt. Cray advocated running test programs once in a while to see if the processor was still working. This would actually reduce the computation power available less than parity checking. - I'm pretty sure that every current serious supercomputer uses not just parity but ECC. This may not be the case for GPU computing. - "parity" was also a term used to describe certain features of US farm subsidies of that era. This (possibly fringe) paper <http://c457332.r32.cf2.rackcdn.com/pdf/the-freeman/0604Folsom.pdf> says: Two concepts in the AAA are fascinating. First is the idea that because farmers overproduce some crops the government ought to pay them not to grow on part of their land. Second is the idea of "parity," that farmers ought to be protected from falling prices by fixing them so that they were comparable to the purchasing power of their crop in the excellent years 1909-14. Note that AAA was the US Agricultural Adjustment Act of 1933 or 1938 <https://en.wikipedia.org/wiki/Agricultural_Adjustment_Act> This seems to have been part of Roosevelt's New Deal. So: does this famous quote win the obscurity competition? "Farmers buy computers too" -- Seymour Cray, after CDC 7600 was introduced. If you want to know more about Cray, here's a slide deck from Gordon Bell (a great computer architect, practical and academic). <http://research.microsoft.com/en-us/um/people/gbell/craytalk/> It's from a talk given shortly after Cray's death. Funny point: the abstract calls Cray "the ultimate "tall, thin man"*." but the notes on slide three call him the penultimate one. I wonder who the ultimate one is if the notes are correct.

On Mon, Aug 31, 2015 at 06:21:53PM -0400, D. Hugh Redelmeier wrote:
- I'm pretty sure that every current serious supercomputer uses not just parity but ECC. This may not be the case for GPU computing.
Modern nvidia quadro's have ECC. It can be disabled if you value a bit more performance over ensuring memory consistency. I don't believe any other video cards have ECC. -- Len Sorensen

| From: D. Hugh Redelmeier <hugh@mimosa.com> | Funny point: the abstract calls Cray "the ultimate "tall, thin | man"*." but the notes on slide three call him the penultimate one. I | wonder who the ultimate one is if the notes are correct. I emailed Gordon Bell about this and he says that he was mistaken about the meaning of "penultimate" when he wrote that. I used "penultimate" in a draft of my masters thesis but my supervisor suggested replacing it with "second last" because it was more likely to be understood and in no way inferior. He was right.

On Mon, Aug 31, 2015 at 2:41 PM, Scott Sullivan <scott@ss.org> wrote: <snip>
Anyone else got some good hardcore comp-sci hummor?
NACK, NACK. Who's there? ATM. ATM who? NACK, NACK. "NACK" is a "negative acknowledgement", a very old communications protocol for "I don't understand". NACK usually forces a resynchronization or renegotiation between end points. ATM is "Asynchronous Transfer Mode"; a couple of decades ago it was a relatively high-speed packet technology that lost out to other technologies. - via <http://stackoverflow.com/a/1590009/773209> ... The TCP joke is a personal favourite: "Hi, I'd like to hear a TCP joke." "Hello, would you like to hear a TCP joke?" "Yes, I'd like to hear a TCP joke." "OK, I'll tell you a TCP joke." "Ok, I will hear a TCP joke." "Are you ready to hear a TCP joke?" "Yes, I am ready to hear a TCP joke." "Ok, I am about to send the TCP joke. It will last 10 seconds, it has two characters, it does not have a setting, it ends with a punchline." "Ok, I am ready to get your TCP joke that will last 10 seconds, has two characters, does not have an explicit setting, and ends with a punchline." "I'm sorry, your connection has timed out. Hello, would you like to hear a TCP joke?" ... There's a few good one-liners over at <http://attrition.org/misc/ee/protolol.txt>

On 31 August 2015 at 23:55, Scott Elcomb <psema4@gmail.com> wrote:
On Mon, Aug 31, 2015 at 2:41 PM, Scott Sullivan <scott@ss.org> wrote: <snip>
Anyone else got some good hardcore comp-sci hummor?
NACK, NACK.
Who's there?
ATM.
ATM who?
NACK, NACK.
"NACK" is a "negative acknowledgement", a very old communications protocol for "I don't understand". NACK usually forces a resynchronization or renegotiation between end points. ATM is "Asynchronous Transfer Mode"; a couple of decades ago it was a relatively high-speed packet technology that lost out to other technologies. - via <http://stackoverflow.com/a/1590009/773209>
...
The TCP joke is a personal favourite:
"Hi, I'd like to hear a TCP joke."
"Hello, would you like to hear a TCP joke?"
"Yes, I'd like to hear a TCP joke."
"OK, I'll tell you a TCP joke."
"Ok, I will hear a TCP joke."
"Are you ready to hear a TCP joke?"
"Yes, I am ready to hear a TCP joke."
"Ok, I am about to send the TCP joke. It will last 10 seconds, it has two characters, it does not have a setting, it ends with a punchline."
"Ok, I am ready to get your TCP joke that will last 10 seconds, has two characters, does not have an explicit setting, and ends with a punchline."
"I'm sorry, your connection has timed out. Hello, would you like to hear a TCP joke?"
With thanks to Scott Sullivan who told me this one, I think it's a great follow-up to the TCP joke: "I'd tell you a UDP joke but you might not get it." Which also says a lot about the difference between the two protocols ... -- Giles http://www.gilesorr.com/ gilesorr@gmail.com

That's cute. /gary On 15-09-01 01:42 PM, Bob Bosiljevac wrote:
"I'd tell you a UDP joke but you might not get it."
Which also says a lot about the difference between the two protocols ...
Wouldn't that be more like:
"I'm telling you a UDP joke but you might not get it".
BB
--- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
participants (7)
-
Bob Bosiljevac
-
D. Hugh Redelmeier
-
Gary
-
Giles Orr
-
Lennart Sorensen
-
Scott Elcomb
-
Scott Sullivan