Re: bin/60521: dump(8): mystery of Hanoi

The following reply was made to PR bin/60521; it has been noted by GNATS.

From: Christos Zoulas <[email protected]>
To: [email protected]
Cc: [email protected], [email protected],
 [email protected]
Subject: Re: bin/60521: dump(8): mystery of Hanoi
Date: Sat, 1 Aug 2026 17:17:02 +0300

 Sorry, I thought the answer was clear: it trades more tape scheduling comple=
 xity for better data retention, tape use, and anegdotically, going through f=
 ewer tapes to recover a file, which is what I recall were the reasons we wer=
 e doing it in the early 80's with 6250 tapes. In the end, we were always wor=
 ried about making mistakes or having bad backups, so we ended up making more=
  full backups than the algorithm required. Many variables have changed now, t=
 he volume of data has skyrocketed, deduplication has become very important, t=
 apes are not the dominant backup media, so perhaps noting the rationale for u=
 sing the algorithm is just there for historical preservation.
 
 christos
 
 > On Aug 1, 2026, at 3:55=E2=80=AFPM, Taylor R Campbell via gnats <gnats-adm=
 [email protected]> wrote:
 >=20
 > =EF=BB=BFThe following reply was made to PR bin/60521; it has been noted b=
 y GNATS.
 >=20
 > From: Taylor R Campbell <[email protected]>
 > To: Christos Zoulas <[email protected]>
 > Cc: [email protected], [email protected]
 > Subject: Re: bin/60521: dump(8): mystery of Hanoi
 > Date: Sat, 1 Aug 2026 12:51:51 +0000
 >=20
 >> Date: Sat, 1 Aug 2026 14:40:53 +0300
 >> From: Christos Zoulas <[email protected]>
 >>=20
 >> Here is what chatgpt has to say:
 >=20
 > Thanks, but this just repeats the same information I noted above about
 > what _other_ references say `tower of Hanoi' means, and doesn't answer
 > any of the questions I posed.  I already spent a lot of time filtering
 > through LLM slop to get here; if I wanted more I could have gotten it
 > myself.
 >=20