I'm in search of war-stories!
I'm now the de-facto anthology/collection editor of /The "Not Invented Here" Chronicles,/ in ACM Queue. It’s about all the times when something brilliant fell out of use because someone (or their employer) was unwilling to look and see if it was already done. It “wasn’t invented here.” So I’m looking for stories about stuff where a technique was rejected, or just fell out of use due to carelessness. The stupidest case I know of is extended versioning for controlled change. It was invented in Multics(!), reinvented in Solaris, and re-re-invented in glibc. It’s described in Paul Stachour’s CACM article[1], as a superset of the kind of ordinary versioning /everyone/ does. I start the series with “The Hardest Way is the Easiest”[2], a story about getting full-path test coverage without infinite labor. So what have you run into that didn’t get enough respect, and could be usefully reincarnated? As with Paul, I’ll play editor or co-author, if you’re shy at submitting articles to the ACM. Heck, brain-dump on me on the phone and I’ll be your secretary (:-)) --dave [1 https://queue.acm.org/doi/pdf/10.1145/1626135.1640399 ] [2 A draft is at codeberg, https://codeberg.org/davecb/NIH/src/branch/main/Exhaustion/Exhaustion.pdf ] -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest davecb@spamcop.net | -- Mark Twain
David Collier-Brown via Talk said on Tue, 6 Oct 2026 13:14:56 -0400
I'm now the de-facto anthology/collection editor of /The "Not Invented Here" Chronicles,/ in ACM Queue.
It’s about all the times when something brilliant fell out of use because someone (or their employer) was unwilling to look and see if it was already done. It “wasn’t invented here.”
So I’m looking for stories about stuff where a technique was rejected, or just fell out of use due to carelessness.
The most glaring case I know of is the decades-long snubbing of daemontools, just because it was invented by that difficult to deal with guy djb. Daemontools, invented in 2001, could be added to svsinit to address every sysvinit problem. I started using it in the mid-00's. Several people made daemontools workalikes that were complete init systems (s6, runit, perp, etc), but the powers that be decided that the solution to sysvinit's shortcomings (and upstarts) was to create from scratch an overcomplexificated mess called systemd. We had everything we needed in 2001, but it was invented by djb, so it was ignored. SteveT Steve Litt http://444domains.com
From: Steve Litt via Talk <talk@lists.gtalug.org>
The most glaring case I know of is the decades-long snubbing of daemontools, just because it was invented by that difficult to deal with guy djb.
DJB-created programs were difficult to redistribute. His license said that you could only distribute unmodified code. That makes it impossible for most distros to deal with. If something isn't in a distro, it has to be really valuable for me to install it (and install it again after each relevant distro upgrade). I'm willing to do this for a very few programs (one, currently). DJB's stuff hasn't been worth the investment for me. I don't even look at it. His cryptographic work is important and does not suffer from the same problem. His stubbornness does show up there too, sometimes for the good.
If you look in the bible, their are many war stories. Tom On Thursday, October 22nd, 2026 at 10:04 AM, D. Hugh Redelmeier via Talk <talk@lists.gtalug.org> wrote:
From: Steve Litt via Talk <talk@lists.gtalug.org>
The most glaring case I know of is the decades-long snubbing of daemontools, just because it was invented by that difficult to deal with guy djb.
DJB-created programs were difficult to redistribute. His license said that you could only distribute unmodified code. That makes it impossible for most distros to deal with.
If something isn't in a distro, it has to be really valuable for me to install it (and install it again after each relevant distro upgrade). I'm willing to do this for a very few programs (one, currently). DJB's stuff hasn't been worth the investment for me. I don't even look at it.
His cryptographic work is important and does not suffer from the same problem. His stubbornness does show up there too, sometimes for the good. ------------------------------------ Description: GTALUG Talk Unsubscribe via Talk-unsubscribe@lists.gtalug.org Start a new thread: talk@lists.gtalug.org This message archived at https://lists.gtalug.org/archives/list/talk@lists.gtalug.org/message/4JQ5BPP...
participants (4)
-
D. Hugh Redelmeier -
David Collier-Brown -
Steve Litt -
tom.fran