The biggest shame of my life is that I had one of these and somehow managed to lose it.
utopcell 3 hours ago [-]
If that's your biggest shame then you must have lived a good life.
hbcdbff 2 hours ago [-]
Or perhaps a life not lived
gradus_ad 41 minutes ago [-]
This comment may trigger life changing reflection in hundreds or thousands of souls...
kens 2 hours ago [-]
I lost one of mine but Knuth kindly sent me a replacement.
ape4 4 hours ago [-]
The error was saying that "infinitely many alphabets can be generated" but there is a finite number of parameters which have a finite range of values. If Knuth had just written "a humongous number of alphabets" he would be correct.
JK-Swizzle 43 minutes ago [-]
If I am reading this right, you can generate 268 435 456 ^ 62 alphabets.
I would say he could have even gotten away with “functionally infinite number of alphabets”.
rawling 12 minutes ago [-]
Some of the parameters are bools. So somewhere between 2^62 and your figure...
WillAdams 2 hours ago [-]
It was a bit more than 25 years ago for me, check #790 for $2.88 DT 33, (143).
Really need to find another error or point of improvement so as to get the digital account --- I've been reading through TAoCP Vol. 4 Fascicle 7 on _Constraint Satisfaction_, but it's been slow going (though I did find out that it is intentional that he not list beginning of chapter quotes in the index).
assumed_throwaw 4 hours ago [-]
I'm surprised we haven't heard of anyone running an AI on Donald Knuth's books to generate the largest collection of checks in existence...
WillAdams 2 hours ago [-]
He is quite methodical and meticulous --- I doubt that there is much in the training data which isn't already accounted for in his research for his writings.
I only managed to get one check (for an error and a point of improvement) for _very_ minor points in _Digital Typography_ which was facilitated by my having been fascinated by typography since reading Oscar Ogg's _The 26 Letters_ and his cousin Warren Chappell's _The Living Alphabet_ when I was 8 or so.
xyst 20 minutes ago [-]
The next headline will read: "AI pilled vibecoders bankrupt a once prominent computer scientist, the figure will surprise you!"
(Hmm maybe I should also write a blogpost about this)
jll29 3 hours ago [-]
I also got one for catching DEK out for claiming "Alan Turing broke the Enigma".
That claim goes to the Polish cryptoanalyst Marian Rejewski instead, of course (https://en.wikipedia.org/wiki/Marian_Rejewski); what Alan Turing and Tommy Flowers did was automate the breaking, which was important because it had to be done again every day (due to the many combinations of configurations of the Enigma).
utopcell 3 hours ago [-]
I bet the mailing cost was more than 0x$2.00 for Knuth.
largbae 4 hours ago [-]
What a brilliant teaching technique, think of how many people studied his work for bounties that they would never cash.
dhosek 3 hours ago [-]
I’m on the mailing list where errors related to TeX are submitted and the level of scrutiny behind submissions is truly mind-boggling.
ryandrake 2 hours ago [-]
Honestly, I wish all software were subject to that much scrutiny.
jacquesm 4 hours ago [-]
Interesting how no other writers in the field seem to have adopted Knuths' strategy to get their books proofread to the point of being bullet proof.
chompychop 2 hours ago [-]
No other writers have the stature of Knuth. The motivation to proof-read here is not the monetary reward after all.
mococa 4 hours ago [-]
I know a guy (ricbit) who have 4 or more
jdw64 4 hours ago [-]
But what's the most Reward checks anyone has ever held?
> "These account balances reflect only the rewards for errors reported since the beginning of 2006. Many reward checks were issued during 1968--2005, too numerous to mention here."
There's 20 years on that chart vs 37 not on it.
Also that's "account balance" rather than "the most Reward checks" which I take to mean "the most [number of] reward checks" -- there's probably a correlation, yes, but [0] shows a variety of ways where check values are non-constant.
There is only one person that can rule that out, but since this is their statement on the subject I would take it as authoritative unless there is a better source.
You probably have a substantial subset of the people on that list on HN, so it might be possible to recover a chunk of the 'paper era'.
blltprfmnk 3 hours ago [-]
I received two checks in the very early 2000s and my name isn’t on this list.
For the record I cashed both and have zero regrets. At the time it definitely felt like the right thing to do as I am not into collecting tokens.
zimpenfish 2 hours ago [-]
> I would take it as authoritative unless there is a better source.
I'm not doubting the authority on the balances; it's the correlation to the number of checks I'm querying - high balance doesn't necessarily mean high number of checks, especially since the checks can have different values.
dhosek 3 hours ago [-]
Although that omits the paper check era.
My guess for the top slot off the top of my head was actually #4.
jdw64 4 hours ago [-]
thanks!
nice_byte 1 hours ago [-]
i was expecting an article about farming these with claude.
I would say he could have even gotten away with “functionally infinite number of alphabets”.
Really need to find another error or point of improvement so as to get the digital account --- I've been reading through TAoCP Vol. 4 Fascicle 7 on _Constraint Satisfaction_, but it's been slow going (though I did find out that it is intentional that he not list beginning of chapter quotes in the index).
I only managed to get one check (for an error and a point of improvement) for _very_ minor points in _Digital Typography_ which was facilitated by my having been fascinated by typography since reading Oscar Ogg's _The 26 Letters_ and his cousin Warren Chappell's _The Living Alphabet_ when I was 8 or so.
(Hmm maybe I should also write a blogpost about this)
That claim goes to the Polish cryptoanalyst Marian Rejewski instead, of course (https://en.wikipedia.org/wiki/Marian_Rejewski); what Alan Turing and Tommy Flowers did was automate the breaking, which was important because it had to be done again every day (due to the many combinations of configurations of the Enigma).
https://www-cs-faculty.stanford.edu/~knuth/boss.html
> "These account balances reflect only the rewards for errors reported since the beginning of 2006. Many reward checks were issued during 1968--2005, too numerous to mention here."
There's 20 years on that chart vs 37 not on it.
Also that's "account balance" rather than "the most Reward checks" which I take to mean "the most [number of] reward checks" -- there's probably a correlation, yes, but [0] shows a variety of ways where check values are non-constant.
[0] https://en.wikipedia.org/wiki/Knuth_reward_check#Reward_amou...
You probably have a substantial subset of the people on that list on HN, so it might be possible to recover a chunk of the 'paper era'.
For the record I cashed both and have zero regrets. At the time it definitely felt like the right thing to do as I am not into collecting tokens.
I'm not doubting the authority on the balances; it's the correlation to the number of checks I'm querying - high balance doesn't necessarily mean high number of checks, especially since the checks can have different values.
My guess for the top slot off the top of my head was actually #4.
thank god it wasn't that lol
One can infer that the post's URI is like a "directory" of the subresources for the post, including those images; the URL of
(n.b. the trailing slash) adjusts the resolution of the image's relative-ref now to: And the images load. Yes, trailing slashes in URLs are significant in the case of resolving a relative ref like that.If the author is here, you might consider redirecting the non-slash to the slash.
> But why will you find me there with the entry “0x$1.20”, i.e. decimal $2.88?
I know Knuth has his own notation here, (it is clearly noted at the bottom of the BOSS page) but 0x1.20 should really be decimal 1.125.