Keep it beautiful
Everything here is my opinion. I do not speak for your employer.
April 2011
May 2011

2011-04-02 »

What you can't say

Normally I don't write a new post just about updates to a previous post, but I have added quite a lot of clarifications and notes to I hope IPv6 never catches on, in response to copious feedback I've received through multiple channels. Perhaps you will enjoy them.

One very interesting trend is that the comments about the article on news.ycombinator were almost uniformly negative - especially the highly upvoted comments - while I simultaneously saw tons and tons of positive comments on Twitter (it was my most popular article ever) and numerous people wrote me polite messages with agreement and/or minor suggestions and clarifications. Only one person emailed me personally to say I was an idiot (although it was my first, at least from people I don't know).

The trend is curious, because normally news.yc has more balanced debate. Either I'm utterly completely wrong and missed every point somehow (as a highly upvoted point-by-point rebuttal seems to claim)... or I seriously pinched a nerve among a certain type of people.

All this reminds me of Paul Graham's old article, What You Can't Say. Perhaps some people's world views are threatened by the idea of IPv6 being pointless and undesirable.

And all that, in turn, reminded me of my old article series about XML, sadomasochism, and Postel's Law. I was shocked at the time that some people actually think Postel's Law is wrong, but now I understand. Some people believe the world must be purified; hacky workarounds are bad; they must be removed. XML parsers must not accept errors. Internet Protocol must not accept anything less than one address per device. Lisp is the one truly pure language. And so on.

Who knows, maybe those people will turn out to be right in the end. But I'm with Postel on this one. Parsers that refuse to parse, Internet Protocol versions that don't work with 95% of servers on the Internet, and programming languages that are still niche 50+ years later... sometimes you just have to relax and let it flow.

Update 2011/04/02: Another big example of a "less good" technology failing to catch on for similar reasons as IPv6: x86 vs. non-x86 architectures. Everyone knows x86 is a hack... but we're stuck with it. (ARM is making an impact in power-constrained devices, though, just like IPv6 is making an impact in severely IPv4-constrained subnets. Who will win? I don't know, I just know that IPv4 and x86 are less work for me, the programmer, right now.)

Thinking about the problem in that way - why "worse" (hackier) technologies tend to stick around while the purified replacements don't - reminded me of Worse is Better by Richard Gabriel. In retrospect, when I divided people above into "purists" and "Postel's Law believers", I guess I was just restating Gabriel's much better-written point about "worse-is-better" vs. "the right thing." If you haven't read the article, read it; it's great. And see if you're firmly on one side or the other, and if you think the other side is clearly just crazy.

If you think that, then that, in turn, brings us back to "What You Can't Say." The truth is, you can say it, but people will jump down your throat for saying it. And not everybody will; only the very large group of people in the camp you're not in.

That's why we have religious wars. Figurative ones, anyway. I suspect real religious wars are actually about something else.

I'm CEO at Tailscale, where we make network problems disappear.

Why would you follow me on twitter? Use RSS.

apenwarr on gmail.com