Hacker News new | ask | show | jobs
by ssl-3 167 days ago
If the NAT function is running on a box that I can walk over and kick, then it is absolutely under my control. :)

CGNAT is a different discussion entirely. Neither the presence nor absence of upstream CGNAT changes my thoughts on locally-administrated NAT for my own LAN in IPv6 land.
1 comments
Dylan16807 167 days ago
When people complain about NAT they're mostly worried about NAT they don't control. CGNAT, or trying to deliver something that works on normal consumer computers without an expert user.
link
ssl-3 166 days ago
That's one perspective.

From my own perspective: I've been hearing people complain about local one-to-many NAT for a very long time, starting 30 or so years ago when fairly-regular people started introducing internet connections to their small networks.

These days, I hear about IPv6 being awesome mostly because it can used to eliminate the need for one-to-many NAT at the local border.

And that sounds great, in concept, except: This elimination introduces new issues that people didn't experience in their previous world of local NAT.

---

CGNAT is its own thing that was broadly introduced relatively recently. It can be similar in operation, but is generally very dissimilar in terms of scale and our ability to control its operation as end-users.

And people know it's different. We even use a different term to disambiguate it from other, more-local, types of NAT that are popularly implemented at the border between their LAN and the Internet: We call one of these things "NAT," and the other of these things "CGNAT".

---

And to be very clear: If I've ever meant to write about CGNAT, then I'd have done so -- and it would be obvious.

I'm very reluctant to defend a position that I have not presented, as entertaining such strawman arguments brings me to feel the opposite of satisfaction.

I'm richly disinterested in such discourse.

link
Dylan16807 166 days ago
Well I didn't misinterpret on purpose. It's not obvious in the other direction because there are many types of NAT and that is one of them. And yes lots of people do call it "NAT".

And it means you left a very important argument in favor of IPv6 unmentioned.

> This elimination introduces new issues that people didn't experience in their previous world of local NAT.

I didn't see you list any downsides of removing NAT in your earlier post, just mock the upsides. But maybe I misinterpreted part of the sarcasm.

link