This article is rated Start-class on Wikipedia's
content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||||||
|
Actually the number in the title is incorrect, it should have been several trillion. There's actually some info on that in the foreword in later editions. Would anyone happen to have that on hand?
Kim Bruning 21:47, 23 Dec 2004 (UTC)
86.133.44.50 ( talk) 18:30, 22 October 2018 (UTC)
Infinite # of names, but can be summed up by "OM"
~mctwists
Hmm. 9e91/9 is about 12.7. This is not totally unreasonable: the Rotokas alphabet only has twelve letters. Now, about the three-letters-in-sequence restriction... a quick Monte-Carlo calculation suggests that with nine-character names, and a 13-character alphabet, only about 1.9% of possibilities are blocked by this rule.
Incidentally, how many "electromatic typewriters" did they need to do the job? A 110 baud Teletype machine, with 2 stop bits and one start bit, will output at 10 cps (the noise still rings in my ears), so that would take a minimum of 9e10 chars/10cps = 9e9 seconds = 104166 days on a single teletype machine (and don't even get me started on the carriage returns and idle stuffing). To do it in "a thousand days", they'd need approximately 100 teletypes: still, that's not beyond the bounds of reason, considering the capabilities of 1967-era mainframes like the IBM 360/67, and assuming a custom terminal concentrator setup.
A modern implementation of this, with output written to disk, would likely be filesystem-limited, so at (let's say) around 10Mbytes/s sustained disk-write throughput, the job would take 9000 seconds, which is 2.5 hours, or 9600 times faster than the "thousand days" of the story.
However, let's assume the monks want a hard copy. A fast laser printer can do 37 ppm, and assuming a clearly legible type size which would give 166 x 80 chars on a sheet, we can print 37 * 166 * 80 / 60 = 8189 chars per second. So, one of these would take 9e10 / 8186 = 127 days to complete the run. A bank of 100 could do the work in 1.27 days. Indeed, with a PostScript program, there probably would be no need for central computer control: each printer could be sent a PostScript program to do its part of the job, and then allowed to get on with it by itself.
How big would the paper output be?
Assuming 9e10 chars, and 166 x 80 character pages, there would be 6,777,108 output pages. If these were bound in 1000 page volumes, there would be 6.7 thousand volumes, which would take up roughly 5cm x 6777 = 339 meters of shelf space at 5cm/book, = 452 x 75cm bookshelves, = 91 5-shelf bookcases.
-- Karada 13:38, 22 August 2005 (UTC)
Really fast laser printers (used for bank statements etc) can do more than 1000 impressions per minute, or 2E7 pages/month. One of these would take ten days to the job. Jds13 ( talk) 17:28, 4 May 2009 (UTC)
The link added by Machuka is a violation of Clarkes copyright. I'm not yet deleting it, but I believe posting links to illegal online copys of copyrighted material is against Wikipedia policies. If you are reading this Machuka, don't do things like this again because you will never manage to get way with it. Loom91 17:57, 10 August 2005 (UTC)
I am removing the link in accordance with Wikipedia policy; we cannot link to an outside site that is violating copyright. Too bad. It was nice to have it just right there. Please still assume good faith. Lebroyl 21:27, 3 January 2007 (UTC)
The quote is slightly off--missing the word "always".
In the story, the visiting lama states "...What would have taken us fifteen thousand years it will be able to do in a thousand days." However, the lama hires the engineers for three months. George and Clark leave the monastery three months and a week after their arrival and the stars start going out as the project had been finished. a thousand days =/= ninety days (3 months) Had it said 100 (one hundred) days, it would have worked, as three months and a week comes out to around 100 days. however the story explicitly says "a thousand" Has anyone else noticed this? The math here is surely inconsistent... MAKNZ493 ( talk) 20:13, 6 September 2010 (UTC)
Nahin's point is incorrect. Nothing would prevent an omnipotent deity from destroying the photons emitted from a star along with the star itself. Even if the deity were constrained to acting at the speed of light, the wave of destruction could actually be starting near the Earth and working outward, so that the stars that seemed to disappear might actually still have centuries left to go. Mahousu ( talk) 14:53, 13 June 2018 (UTC)
Well, to be fair, the relevant bit in the article--as it currently stands--specifies an "ominiscent" God, not an omnipotent God. Why exactly this peculiar Nahin person should want his God omniscient but not omnipotent is another question. TheScotch ( talk) 07:44, 25 June 2022 (UTC)
I read this story just once, many decades ago when I was twelve years old or so. It appears to be out of print now (or at least the story collection with the same name appears to be out of print). I'm hazy about the premise. Are we actually given enough information in order to calculate the outcome with simple combinatorial arithmetic as certain editors geeking out above seem to think? Are we told how many characters the monks's alphabet contains? Are we to assume that any permutation of these characters not exceeding nine in length spells some name of God? Or are there other constraints? I ask this not just out of curiosity; I think the article should tell us. TheScotch ( talk) 07:38, 25 June 2022 (UTC)
This article is rated Start-class on Wikipedia's
content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||||||
|
Actually the number in the title is incorrect, it should have been several trillion. There's actually some info on that in the foreword in later editions. Would anyone happen to have that on hand?
Kim Bruning 21:47, 23 Dec 2004 (UTC)
86.133.44.50 ( talk) 18:30, 22 October 2018 (UTC)
Infinite # of names, but can be summed up by "OM"
~mctwists
Hmm. 9e91/9 is about 12.7. This is not totally unreasonable: the Rotokas alphabet only has twelve letters. Now, about the three-letters-in-sequence restriction... a quick Monte-Carlo calculation suggests that with nine-character names, and a 13-character alphabet, only about 1.9% of possibilities are blocked by this rule.
Incidentally, how many "electromatic typewriters" did they need to do the job? A 110 baud Teletype machine, with 2 stop bits and one start bit, will output at 10 cps (the noise still rings in my ears), so that would take a minimum of 9e10 chars/10cps = 9e9 seconds = 104166 days on a single teletype machine (and don't even get me started on the carriage returns and idle stuffing). To do it in "a thousand days", they'd need approximately 100 teletypes: still, that's not beyond the bounds of reason, considering the capabilities of 1967-era mainframes like the IBM 360/67, and assuming a custom terminal concentrator setup.
A modern implementation of this, with output written to disk, would likely be filesystem-limited, so at (let's say) around 10Mbytes/s sustained disk-write throughput, the job would take 9000 seconds, which is 2.5 hours, or 9600 times faster than the "thousand days" of the story.
However, let's assume the monks want a hard copy. A fast laser printer can do 37 ppm, and assuming a clearly legible type size which would give 166 x 80 chars on a sheet, we can print 37 * 166 * 80 / 60 = 8189 chars per second. So, one of these would take 9e10 / 8186 = 127 days to complete the run. A bank of 100 could do the work in 1.27 days. Indeed, with a PostScript program, there probably would be no need for central computer control: each printer could be sent a PostScript program to do its part of the job, and then allowed to get on with it by itself.
How big would the paper output be?
Assuming 9e10 chars, and 166 x 80 character pages, there would be 6,777,108 output pages. If these were bound in 1000 page volumes, there would be 6.7 thousand volumes, which would take up roughly 5cm x 6777 = 339 meters of shelf space at 5cm/book, = 452 x 75cm bookshelves, = 91 5-shelf bookcases.
-- Karada 13:38, 22 August 2005 (UTC)
Really fast laser printers (used for bank statements etc) can do more than 1000 impressions per minute, or 2E7 pages/month. One of these would take ten days to the job. Jds13 ( talk) 17:28, 4 May 2009 (UTC)
The link added by Machuka is a violation of Clarkes copyright. I'm not yet deleting it, but I believe posting links to illegal online copys of copyrighted material is against Wikipedia policies. If you are reading this Machuka, don't do things like this again because you will never manage to get way with it. Loom91 17:57, 10 August 2005 (UTC)
I am removing the link in accordance with Wikipedia policy; we cannot link to an outside site that is violating copyright. Too bad. It was nice to have it just right there. Please still assume good faith. Lebroyl 21:27, 3 January 2007 (UTC)
The quote is slightly off--missing the word "always".
In the story, the visiting lama states "...What would have taken us fifteen thousand years it will be able to do in a thousand days." However, the lama hires the engineers for three months. George and Clark leave the monastery three months and a week after their arrival and the stars start going out as the project had been finished. a thousand days =/= ninety days (3 months) Had it said 100 (one hundred) days, it would have worked, as three months and a week comes out to around 100 days. however the story explicitly says "a thousand" Has anyone else noticed this? The math here is surely inconsistent... MAKNZ493 ( talk) 20:13, 6 September 2010 (UTC)
Nahin's point is incorrect. Nothing would prevent an omnipotent deity from destroying the photons emitted from a star along with the star itself. Even if the deity were constrained to acting at the speed of light, the wave of destruction could actually be starting near the Earth and working outward, so that the stars that seemed to disappear might actually still have centuries left to go. Mahousu ( talk) 14:53, 13 June 2018 (UTC)
Well, to be fair, the relevant bit in the article--as it currently stands--specifies an "ominiscent" God, not an omnipotent God. Why exactly this peculiar Nahin person should want his God omniscient but not omnipotent is another question. TheScotch ( talk) 07:44, 25 June 2022 (UTC)
I read this story just once, many decades ago when I was twelve years old or so. It appears to be out of print now (or at least the story collection with the same name appears to be out of print). I'm hazy about the premise. Are we actually given enough information in order to calculate the outcome with simple combinatorial arithmetic as certain editors geeking out above seem to think? Are we told how many characters the monks's alphabet contains? Are we to assume that any permutation of these characters not exceeding nine in length spells some name of God? Or are there other constraints? I ask this not just out of curiosity; I think the article should tell us. TheScotch ( talk) 07:38, 25 June 2022 (UTC)