Worth Reading: The IPv6 Agnostic Blog
Ole Troan, an excellent networking engineer working on IPv6 for decades, has decided to comment on the color of the IPv6 kettle, starting with:
- Is the transition to IPv6 inevitable? (hint: Betteridge’s law of headlines)
- The mistakes and missed opportunities in the design of IPv6 - episode 1 (aka Second System Effect1)
I’m pretty sure Ole won’t stop there, so stay tuned.
Behind the Scenes
Ole initially published his articles on Medium, so I sent him a polite email asking whether he could move them to a platform with somewhat less blatantly aggressive in-your-face monetization attempts. He quickly responded and asked for ideas, and one of my suggestions was the Hugo/GitHub/CloudFlare Pages pipeline I’m using.
A day later, I received another email from him, asking a rhetorical question: “How can we make self-hosting of anything easier?” His Hugo-built website was up and running on a Linode S3 bucket.
If you want to do something similar:
- Read the Linode Hugo tutorial
- Explore Hugo hosting and deployment options
- Cloudflare Pages might be an interesting hosting option that I’m using for all my projects. They handle domain registration, TLS termination, website deployment, and global hosting for you. Another free alternative is GitHub Pages.
However, never forget that “perfect is the enemy of published.” It took a friend of mine forever to set up his Hugo web site because (in his own words):
I tried to run it locally and applied some templates, and it went… well, until I felt the urge to “improve it”. Perfection is the mother of all disgraces. It all went down from there.
He got his site up and running in the meantime, but probably spent way too much time on irrelevant details (trust me, I’ve been there).
-
I love that Wikipedia authors used IPv6 deployment as an example of second-system effect. ↩︎
Refreshing to see more real life IPv6 takes like these!
And Ole's blog just needs an RSS feed now ;)
This is an interesting read especially as someone who is firmly on the IPv6 train. Calling it a failure is interesting. Maybe in some western countries? I'm not sure calling it a failure works globally, see India. It also doesn't really address the scalability issues with IPv4. Yes we could keep working on the protocol and extending it to make it scale better, but there are so few addresses and all of them have already been assigned. I just don't see how there is any practical or viable way to keep pushing IPv4 forward.
It's really nice to hear from someone being happy with IPv6 ;) and I totally agree with you that we have no alternative.
However, how do you call a protocol that solves an unquestionably dire problem (total IPv4 address exhaustion) and has less than 50% adoption rate (less than 50% of all Internet users can access Google over IPv6) more than two decades after we had production implementations?
I mean it's definitely not a success but I also don't think it's doing THAT poorly. In 2010 (30 years from the introduction of the IPv4 spec) the total internet(IPv4) adoption in the developed world was just 67%. Yes 50% 30 years after the introduction of the IPv6 spec is lagging behind(and ironically the developing world is ahead) but it's also not a big enough gap to be called a failure when you compare it to how long the Internet itself took to gain mass adoption. It's moving slower than IPv4 did but that's to be expected. IPv4(the Internet) was a revolutionary technology and IPv6 is something only the network professionals care about. Maybe my logic here is flawed but from this lense it just doesn't seem like a failure, just par for the course really. I do wish it would hurry up though lol.