Conversation

Notices

1. Embed this notice
   Bradley M. Kühn (bkuhn@fedi.copyleft.org)'s status on Thursday, 30-Apr-2026 16:56:15 JST Bradley M. Kühn

   (1/2) ≈ every 6 months, I have trouble sleeping specifically because I think about how, circa 1996, I attended some sysadmin (probably SANS) event in Baltimore.

   I went to a tutorial on #IPv6. I got there early as we knew it'd be standing room only & it was.

   One of the first slides the instructor put up was a graph of how quickly #IPv4 numbers were being assigned. He said with complete confidence: “We all need to learn IPv6 as IPv4 will surely be deprecated by the end of the century”. …

   • Embed this notice
     LisPi (lispi314@udongein.xyz)'s status on Thursday, 30-Apr-2026 16:56:11 JST LisPi

     in reply to

     • Craig Brozefsky
     @craigbro @bkuhn I'd hardly call it a dire prediction.
     
     IPv6 lagging behind is in considerable part due to perverse incentives for ISPs to maximally rent-seek IPv4 addresses and "business" features.
     
     IPv6 obsoletes a lot of their rent-seeking schemes, which they dislike.
     
     The SANS instructor wrongly assumed companies cared more about doing things properly than screwing over customers for money.
     Rich Felker repeated this.
   • Embed this notice
     Craig Brozefsky (craigbro@infosec.exchange)'s status on Thursday, 30-Apr-2026 16:56:12 JST Craig Brozefsky

     in reply to

     @bkuhn I think this is also an artifact of getting older, you collect a raft of dire predictions that don’t occur because complex systems are often remarkably resilient when they have actors embedded in them with massive resources.

     Rich Felker repeated this.
   • Embed this notice
     Bradley M. Kühn (bkuhn@fedi.copyleft.org)'s status on Thursday, 30-Apr-2026 16:56:13 JST Bradley M. Kühn

     in reply to

     (2/2)…Why insomnia?: After 30 years, this became my archetype of how ridiculously bad we technologists are at prognostication.

     IPv6 remains spotty. IPv4 addrs oughta be worthless, but ARDC became a wonderful grantmaker b/c Bezos paid $108mil for 25% of a Class A. 😲

     Truth is stranger than prediction.

     Most who did prognosticate accurately probably got lucky.

     This terrifies me b/c the primary job of an activist policy maker is to anticipate future events & have policies ready to solve them.😬

   • Embed this notice
     Bradley M. Kühn (bkuhn@fedi.copyleft.org)'s status on Friday, 01-May-2026 01:21:40 JST Bradley M. Kühn

     in reply to

     • LisPi
     • Craig Brozefsky

     @lispi314

     That was part of my point. At that time in the mid-1990s, sysadmins generally thought that capitalism needed those who actually understood how the Internet worked.

     As you say, it turns out we had no power at all against both the rents you mention and how excessive use of NAT turned the Internet into a primarily broadcast medium rather than an egalitarian one we all arrogantly assumed couldn't be coopted.

     Cc: @craigbro