Somewhat related, Netheads vs. Bellheads https://www.wired.com/1996/10/atm-3/

I briefly worked at a place fully dominated by Bellheads early in my career. They were bemoaning IP and holding on to ATM stuff as late as 2001. I'm told that the last of their tribe (after I left) fought off VoIP as a passing fad well into the 2000s.

Hard to believe this was nearly a quarter century ago. T he references are getting dated:

> "How do you scare a Bellhead?" he begins. "First, show them something like RealAudio or IPhone. Then tell them that right now performance is bandwidth-limited, but that additional infrastructure is being deployed." > … > One result is undergrads who, for $29.95 a month, clog up the Internet with CU-SeeMe sessions.

The apparently out of place reference to the iPhone in this article from 1996 is actually referring to the IPhone, one of the first internet appliances.

The article is dated 1996. How does it have an IPhone reference?

Wrong iPhone.

https://en.wikipedia.org/wiki/Linksys_iPhone

Thanks. That's closer, but the time to release is still 2 years. Is the article referencing pre-release rumors? Or was it amended later?

Probably it was about this product: https://www.zdnet.com/article/hey-look-heres-an-iphone-ad-fr...

I think I still have one somewhere. I believe it was Cisco branded though.

Its crazy to think that the iPhone only came out in the great recession in 2007. Seems like we had it forever.

We not being me, brandishing my Psion gold card as evidence of my first adopter enthusiasm...

ATM is still actively in use in broadband networks all over the place. Its a real pain when you're on the ISP side.

It might very well be a pain for the admins, but for low latency applications, ATM has clear advantages to the user.

That explains its massive success I suppose?

Only with ADSL right?

It’s gone in newer DSL variants and not in DOCSIS? Or am I wrong?

ATM. The horrors of working at a Navy installation that was trying to get ATM to the desktop (!) to work.

Fun times.

No, it won because it could be carried over all other protocols. It had a simple design and real world use cases too, which helped. A classic example of "worse is better".

Wouldn't real world use cases be a technical advantage?

Thanks. I haven't thought of X.25 in 25 years.

Heh I'm currently working on a toy X.25 stack implementation as a side project

github link, please? :)

Is IPv6 winning yet??

Depends what you're looking at, on mobile it absolutely is. Not sure what the situation around the world is but in the US pretty much all LTE connectivity is via v6. T-Mobile at least doesn't even dual stack, if you need to go somewhere that's v4 only they do NAT64

It's hit 35%, and has broken 40% in many countries: https://www.google.com/intl/en/ipv6/statistics.html

IPv4 will probably never go away because it's a better choice for private networks.