an internet of action

I am fed up with the cliche about the internet of things - the cool thing with the net
so far is that it is an internet of ideas - if we want the next exciting stage, it is not an internet of things (light switches) it is an internet of action -

The All Action Internet (AAI) - this will be an internet where you can make stuff happen (your heart beats faster as you click on an icon and your shoes change colour)

You slide on a scroll bar at the side of the car dash board and the car in front moves further away

you tab through various house designs on a fashion site, and the walls and furniture in your house morph

thats what I'm talking about

oh, and i dislike people getting credit for terms they didn't make up - dave clark was talking about the
internet of things before yahoo was even dreamed of.

public broadcasting on the internet

the bbc has long been an innovator and I am very happy that they've revived Dr Who so fabulously

but the plan to allow downloads of programmes using iplayer to only one OS platform is, for me just plain unacceptable, and I want a portion of my license fee back for 2 reasons
1. I have 3 Macs at home - none will run windows unless I spend money on a VM
or alternatively, I have to buy a windows machine AND a license
2. if some of my public broadcast service license fee has been used to subsidizew the development of a piece of software that only benefits one main vendor, this is a gross misuse of public funds.

I am not a religously anti-microsoft person at all - I believe they are a typical giant company, and they (especially in their research) are trying to improve their products and services as fast as possible, but I think the BBC could have developed an open platform using more open DRM technology -given a large part of their content is already captured and re-disted using bit torrent, they are fighting a losing battle and they are the wrong people to fight it - leave that to commercial channels who have clout and
reasons to avoid losing advertising revenue.

There are a lot of people who would help the beeb develop and distribute an open tool that time limited programmes, and compiled and ran on all major platforms - the issue of CODECs has long been solved for multiplatform audio/video - the DRM module in such software could easily be made portable (it is not media or network or OS dependant or can be made independant if you must insist on media specific watermarking techniques too)

sorry, but this is the wrong way for the british broadcasting corporation to evolve and surely they know it? see their report on
eu action in this space for other critique

n.b. zattoo ( for example, do the OS/Codec independant iptv stuff


meanwhile, other mejeeya newz:

scientolgists to play anti-nazi conspirators is
just too much for germans

It'd be ok for me if it was a comedy (e.g. like THe Producers, with tom cruise singing
springtime for hitler, only not)

internet mileage metering - how to measure?

how would you measure how "far" you have gone in the Internet?

Obviously one could just log URLs visited (most browsers do), but some urls are nearer to each other than others (small world etc).

At the IP level, one can run traceroutes and see how many IP networks (prefixes) and how many ASs (ISPs) one has visited or traversed (again, some ISPs are next to each other - some nets are "within" other nets, so it makes for a fun challenge - the AS level graph is well known to have an (approximately) power law node degree).

Why would you want to do this? Because that's what we do in the Internet: mostly harmless, mostly useless stuff:-)

So another metric might be (sort of like googlewhacking) - how many places do you go that very few others do (whether IP addrs, ports, or URLs)? That might then act as a "weirdness" or "geekiness" metric. On the other hand, it might just be sad:)
Like only going to the field of Lost Vagueness in Glastonbury and never standing neck-deep in mud by the Pyramid Stage grooving to the Arctics...

total cost of ownership & grids

so networks went from circuit to packet mode partly because of the stat muxing gain.
is there a stat muxing gain for CPUs hosting virtual services? looking at eScience, I doubt it very much - most the apps want all the cycles they can get and then some so why dont those greedy physicistas lay off and buy their own computers for the LHC then, eh?

and why dont they just use botnets if its so useful - irc is so much more efficient than SOAP as a control plane and even with dyndns the anonimity seems good enough to stop them being closed down very often...

or xenoservers (oh stop it:)

new approach to email

I'm trying an experiment just now - all my e-mail is deleted - if you want to send me a message you need to put it on a web log (sorry, blog) or wikipedia, and I will get a google alert on a (private, whilelist only) mail address and will get back to you. This both rate limits how often new people can send to me, and scales my mail to google's search/alert system, which is probably better than the university's - of course, if everyone did it it might be very interesting!

three things one hopes to learn
a. how fast google scan/alert stuff runs
(it hasn't found michael dales blog of my email autoanswer yet, but its early days
b. how many people care
c. if anyone thinks of an attack...

the angels have the phone box

I'm not usually a sucker for 1 liners but this one yesterday on a certain british tv show takes the biscuit - I want the T-shirt - i want the transcript

and no, if you don't know what I'm talking about, too bad:-)
if you do, blink:-(

p.s. quantum neutral aliens who don't exist when you observe them,
how cool is that, eh?

possibly the best scifi transcript (with the coversation between a live person and easter eggs on 40 year old dvds) in decades.

magic and not a dalek in sight

so alas half the folks selling (very nice) T shirt design's have sold out:(

meanwhile, when will the Dr get an honarary chair in some decent university (Cambridge, obviously since he's a techie) and then become
Professor Who!

mobilising networks visually

just came across the Colorado visualisation tools for mobile devices - very nice stuff - if only i was teaching soemthign now I could use it!!!

a use for spam - verifying availability (e.g. for torrent)

so I can prove how available (and how much bandwidth i have, modulo asymmetry) by how much spam I get - i dont have to see the spam, but I can prove that i got it (e.g. summary hashes) and then use this to verify I am a high avaialbility, high capacity site -i could bootstrap my trust in (say) bittorrent from this...

how about this?

would there be a perverse incentive for people to spam themselves to get more bittorrent tokens? no - they would do better to just actually use the acpacity fore bittorrent

would spammers have an incentive to boost artificially, some sites over others? not really (as far as i can see) - a "popular" person would stay (at least pro rate) popular...