Hacker News new | past | comments | ask | show | jobs | submit login

> Bernard Fabrot spent the last three and a half years computing the solution to a puzzle

TacticalCoder last HN comment is from Dec 5, 2015. Checks out.

Kudos to you, well done!




I actually thought this might get posted to HN, so I went through a lot of pain to find back a way to get my HN login back (basically had no access to the email with which I opened an account here: I had to dive deep down in some Google Suite tied to a domain name to find a way into the user account and eventually reset my HN password). I'm glad I did in time and didn't miss this thread!


Could have used the world's best 2fa. "I can prove my identity because I have the answer to LCS35".

Pretty sure @dang would have accepted that!


Well, you sure seems like the kind of guy who will go through a lot of pain to get quick results.

Was this a side project? did it require much attention after the initial setup?


> Was this a side project? did it require much attention after the initial setup?

Totally a side project.

It hardly required any attention: just making sure the computation was running and making sure to backup the intermediate results once in a while (I had backups everywhere of the intermediate results: remote servers, USB keys, DVD, other computers at home, email, etc.).

I only needed to do a few tiny steps once I found the secret message (which contains a number which allows to compute one of the two prime numbers).


Did you knew for certain that your program will solve the puzzle at some time? Did you have an estimation of the date?


> Did you knew for certain that your program will solve the puzzle at some time?

The description of the problem itself explains a way to verify that intermediate results are correct.

> Did you have an estimation of the date?

Yes and that one is easy: I wrote a little piece of code (in Clojure, whatever, it's unrelated to finding the solution) which would check at which intermediate result the computation was and then compute an "ETA".

Fun story: I'd regularly send my brother something like this...

{ :time-left-in-years 1.239341704718417 :eta "1 year 2 month(s) 26 day(s)" :percentage-done 0.6284244211583109 }

(that's an actual estimate I found in my old emails)

: )


> TacticalCoder last HN comment is from Dec 5, 2015. Checks out.

I didn't realize... Your comment is actually extremely funny: I love it : )

Yeah, it's more or less when I started computing that. This is hilarious.


I just want to be a part of this thread. Congrats :)




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: