jump to navigation

Broadband problems May 20, 2007

Posted by mark in : General, Networks, Personal, Telecoms , add a comment

Finally got home DSL service again, after more than 3 weeks, now that my ISP’s fixed the fault (or, rather, taken more than 2 weeks to ask BT to fix it). D’you think that’s a long time? I think it’s a long time. Nobody’s apologised, though.

Anyway, I’ve a few articles and links that I hope to put up, later this week.

Links 2007-05-03 May 3, 2007

Posted by mark in : Culture, General, Geo, Networks, Web , add a comment

Links 2007-04-22 April 22, 2007

Posted by mark in : Geo, Networks, Telecoms, VoIP, Wireless , add a comment

Twitter April 1, 2007

Posted by mark in : Application, Networks, Telecoms, Ubicomp, Web, Wireless , add a comment

I’m still undecided about Twitter [blog][forums], the simple moblogging site to update friends on your current activity (or “public IM” etc, if you prefer). Its flexibility offers a ‘one size fits all’ platform for a lot of peoples’ requirements, but will the value of a larger ‘crowd’ outweigh personal compromises in other areas? It feels more like a clever hack, in the right place at the right time, than a preferred long-term solution.

In Sep?-06 I read a launch advert on Odeo, another Obvious Corp property, but didn’t ‘get’ how it differed from other group-SMS applications. Still, the buzz gradually grew, and since SXSW (Mar-07) articles are everywhere.

My page hasn’t been used since Jan-07, as friends remain disinterested after I ‘explain’ it. And I’m tired of the hype. Twitter has interesting possibilities. I don’t agree that it’s pointless or (simply) intrusive. I’ll probably find a reason to use it. But right now I’ll capture these few thoughts and links, and move on.

So why is it Twitter that captured the buzz, and became successful social software? Is it the open access – SMS, webpage and IM (though apparently not Trillian)? Is it the simple, open API for 3rd party developers? Is it the persuasive personalities of early adopters? (Is it just a niche success, talked up by a lot of US tech bloggers?) Probably all of those, at the moment.

Unlike other broadcast-SMS apps, Twitter is lightweight and already supported, so doesn’t need a (proprietary) mobile Java app. Also, PersonX on Twitter doesn’t broadcast directly to a group : she sends to a PersonX node on the Twitter server. The significance is (1) A node could represent an event or service (for breaking news), instead of just a person, which adds functionality; (2) in the UK (etc, where sender pays), a group only pays for one SMS per broadcast (…It’s less of an issue in receiver-pays USA, where monthly cellular plans have huge SMS bundles); (3) losing a mobile doesn’t lose a person’s groups; and (4) all of the data is captured by the Twitter servers.

PersonX is also able to send a message directly to PersonY=webservice node and receive a direct reply, which adds ‘mobile’ to an existing web service, and makes Twitter more of a platform than (just) a social software app …at least until mobile web browsing is a preferred alternative.

My first (and still clearest) thought was that Twitter acts as a ‘presence’ indicator like IM does, but SMS doesn’t, and gives the security of knowing what your group/ tribe is doing by continuous partial attention …and Liz Lawley [Many2Many] seems to agree. The convention of short (up to 140 characters), occasional, messages allows mobile users to participate more equally, because it doesn’t try to replace IM conversations, where mobile (SMS) users are disadvantaged by connectivity, typing speed, expense, etc. [Around the time that Twitter launched, there was discussion about whether mobile IM app's would succeed SMS, and a planned telco IM platform (to compete with Y!M, MSN, AOL, etc).]

Displaying public conversations on a webpage attracts people (see: Jakob Nielsen’s 90-9-1 rule), even if they don’t post, or belong to groups; and it keeps a history (public or private). The website is the heart of the application. But posting is shown to be easy, because it’s short and doesn’t NEED to say anything clever, and so the initial hurdle is low (but the reason to continue is usually your group). Group-forming is easy and Reed’s Law says it has the potential to add huge value to a network (so should Twitter encourage it more, by adding an option to post to sub-groups of your contacts?); and a reluctance to allow it is one of many reasons given for Friendster surrendering its lead.

Lots of users, groups and posts means a lot of metadata, and the open API allows 3rd parties to repackage it …to add value, to offer different data visualisation tools, or simply to display interesting usage patterns.

Twitter [Jason Kottke]
“Newbie’s guide to Twitter” [Webware]
A Twitter primer [theory.isthereason]
“Dodgeball vs. Twitter” [TechCrunch]
Twitter commands [Twitter.com]
“Six ways to improve Twitter” [Webware]
“Twittering your home” [O'Reilly]
Twitter Fan Wiki …for widgets, etc
twittermap
twittervision [via O'Reilly and Google Maps Mania]
Twitterholic.com
TwitterBuzz
Twitterbar extension [Flock]
Twapper [via Lifehacker]

Update: Origin of the Twitter idea [via Jason Kottke]

The Information Factories (Cloudware) November 6, 2006

Posted by mark in : Design, General, Networks , add a comment

George Gilder’s article in Wired magazine describes how the cloud computing model (ie. servers ’somewhere on the Internet’) moves processing away from the desktop to massive server farms (Google, etc). It mentions historic quotes “The network is the computer” (Sun) and “global computing market of five mainframes” (IBM) and some interesting numbers.

In every era, the winning companies are those that waste what is abundant – as signalled by precipitously declining prices – in order to save what is scarce. Google has been profligate with the surfeits of data storage and backbone bandwidth. Conversely, it has been parsimonious with that most precious of resources, users’ patience.

Unfortunately that strategy means massive energy cost/waste (cooling, replication), and few locations will be able to provide enough (cheap) electricity as the need scales up. Fortunately “semiconductor and optical technologies are on the verge of a new leap forward” and the current economic advantages of a ‘centralised’ solution will decrease.