The site uses cookies that you may not want. Continued use means acceptance. For more information see our privacy policy.

What’s in a name? Version numbers as names

There’s a warm feeling to know that this humble blog is accessible to users over IPv6. For now I can only wait for the day that I can actually get to it over IPv6 myself.

What’s in a name? that which we call a rose
By any other name would smell as sweet;

— William Shakespeare, Romeo and Juliet

In practically every discussion I’ve seen of Firefox’s new accelerated release cycles the topic of the version numbers stands prominent.  People have repeatedly claimed that it’s some sort of half-assed ploy to raise the version number to improve marketing.  I find that quite misguided, as I see the great promise the new development plan has for making Firefox the best browser it can be.

But people get bent out of shape over versioning, for whatever reason.  Though in this case the focus has been on the marketing angle, Linux has had heated discussions before regarding their versioning scheme.  And TeX just keeps on approaching Pi as its versioning scheme.

It’s just a name, and the only real rule that’s commonly followed is that the version should always be an increasing value (mainly because it’s good for programmatic comparison purposes).

Shakespeare was dead on in the famous speech by Juliet regarding the disputes that names cause.

Hell, the Internet Protocol simply skipped version 5, as ECMAScript skipped version 4.

Oh, which reminds me:

  1. APNIC (the Asian/Pacific Network Information Centre) is almost out of IPv4 addresses, meaning that for practical purposes all new allocations will be of IPv6 addresses and it’s time for the world to move on.
  2. Thanks to the awesomeness that hosts this blog, it now has a AAAA record (ie, an IPv6 address).

There’s a warm feeling to know that this humble blog is accessible to users over IPv6.  For now I can only wait for the day that I can actually get to it over IPv6 myself.

Where in the world is IPv6?

How big is the address space? Undecillion means a billion billion billion billion. So, around 340 of those. That’s a lot.

There are approximately 62 million IPv4 addresses left to be allocated, and estimates put exhaustion somewhere around August, 2011.  That means time is running out for the bulk of the Internet to begin rolling out IPv6.

What does all of that mean?

IPv4 is the current majority protocol that the Internet runs on.  The device you use to read these words almost undoubtedly used it to get them.  It defines the underlying behavior for the little packets of data that are sent and received to make the Internet work.  Part of that behavior is a way to say who the data is from and who it is for.

IPv4 addresses fill that role by having four parts of eight bits each.  That means that each portion of the address is a number from 0 to 255.  A full address might look like  That representation is referred to as “dotted decimal” because the delimiters are dots (periods) and the numbers are in base 10 (decimal).

But there’s a catch: The maximum number of addresses in a 32-bit address space is 232, or just beyond four billion addresses. The actual number is smaller, as there are several blocks of addresses that are reserved for various purposes, which are not part of the general allocation pool.

And the number of addresses that aren’t allocated is quickly approaching zero.  Wikipedia: IPv4 address exhaustion: Predictions of exhaustion dates has a couple of graphics worth examining.  One shows the exhaustion curves, while the other shows the allocations.

To continue the growth of the Internet beyond IPv4 address exhaustion, more addresses will be needed.  That means moving to IPv6, with its 128-bit addresses.  With 128-bit addresses there are not just four times as many addresses, but 296 as many. In total, the IPv6 address space holds around 340 undecillion addresses.

Undecillion means a billion billion billion billion. So, around 340 of those.  Again, minus some of the space for reserved and private uses, but a small enough amount to not worry about that.

The problem is that the world needs IPv6, but it’s been slow to roll out and a long time coming.  Many of the major web properties will be holding a test day in June, but it will probably be once the switch is flipped that a bulk of the Internet actually starts moving in the right direction, and there will probably be a bit of breakage along the way as people find hard-coded addresses from decades past and other such bitrot.

Anyway, I’ll leave you with a fun song that was sung back in 2007 at the RIPE conference, The Day the Router Died:

Daily News #000

Melting Away Defects on Microchips via Melting Microchip Defects May Extend Moore’s Law on Slashdot.
Zebra Striping: Does it Really Help? via Do Zebra Stripes Actually Help? on Slashdot:

Many believe that zebra stripes aid the reader by guiding the eye along the row. However, despite being in use in both paper and electronic mediums for almost half a century, there is practically no evidence that it actually assists users in this way.

This is actually one I’ve been putting some thought into for the web. My hypothesis is that it can help if a few UI tweaks are made (most likely via javascript).

Up to 300 Megawatt Worth of Keepalive Messages to be Saved by IPv6? via IPv6 would save 300 Megawatts ? on Planet Debian. To summarize this one, the increasing use of NAT, especially for mobile devices, means keepalive messages must be sent periodically (the connection is poked to ensure it remains active). For mobile devices that means shorter battery life, but it also means more energy use overall.

Eh, got sidetracked by the primary results; next time I’ll try to cover more than three items. North Carolina went strong for Senator Obama while Indiana went weak for Senator Clinton. Is she done yet?

I don’t think she is. One of the Kosers put it pretty well saying that the delegates plus superdelegates garnered from this race will likely put Obama on the green for Oregon. He’ll just have to make that putt. That’s May 20, 2008. Two weeks.