2011-10-25

IPv6 has to be made sexy.

Switching to IPv6 is like taking out the trash: everyone hopes someone else is going to do it so they don't have to.

I've known a lot of people who have enthusiastically set out to IPv6ify the world only to realize that their mission is met by as much enthusiasm as experienced by those who turn up at your doorstep with pamphlets to guide your soul away from some inevitable doom.

If I were an economist (a proper one) I would probably ask: where are the incentives?

Because there are no incentives. In fact, there are only disincentives. It is hard enough to teach people how to set up a network using IPv4. If you start talking about IPv6 their eyes just glaze over.

For IPv6 to become a mainstream occurrence companies like Cisco and Intel need to hire people that understand the need to make products that are significantly easier to configure. The products have to be so easy to configure that they are sexy. And this has to go beyond just slapping some My First Pony GUIs on top. The sexiness has to be deeper. To appeal to morons and tinkerers alike.

Ever tried to configure a router using its command line interface?  Exactly.  They are rubbish almost no matter what the price range.  It seems no matter how many times I try, I can't get the broadband router in my home to behave sensibly.  It has been some years since I had to set up a Cisco router, and while those are made by somewhat brighter people, you are still going to need to study the manual to figure the thing out.

When networking gear becomes more sensible, and is designed by people who have an ambition to simplify their use adequately, IPv6 can succeed. Until then it will remain the domain of people with pony tails, suspect body odors and an undying belief that the rest of the world even gives a shit about the diminishing supply of public IP addresses.

2011-10-11

Android as a platform for launching a new language?

I'm not sure if I would say that there is a lot of focus on designing languages at Google, but with the recent announcement of Dart, I couldn't help but think that Google actually has what it takes to make a language successful quickly. If they develop a new language for the largest smart-phone platform on the planet, they can reach a lot of programmers in a very short time.

(Just look at how quickly the world learned Objective-C, which was all but an oddity, to make apps for iOS).

I think Oracle are mismanaging Java and that they have proved to be extremely short-sighted. Trying to extort Google for cash rather than seeing Android as an opportunity for becoming relevant on mobile platforms is a very weak response and one that tells me that Oracle does not have the long term in mind. And if you make an investment in a language you want a responsible "owner" that has the long term in mind.

Oracle is not that responsible owner, and as long as the attitudes of Larry Ellison persist, there will be a dark cloud hanging over Java.

I'm not sure if either Go or Dart are suitable for writing applications for a mobile platform. But it if Google should decide to create a replacement for Java on Android I think it could have a profound impact on the industry. They can reach a lot of developers very fast.