Dr Mark Searle – on I am not a number – I am a tag ..

February 8, 2006 By

number.JPG

Like Judy in the previous blog, I have a lot of respect for Dr Mark Searle‘s views. He does an excellent job of explaining the I am not a number – I am a tag concept - perhaps better than I did!

Mark says ..(again at forumoxford )

I can’t speak for Tony of course and his idea is fairly vague but please let me explain what I think he means.

At present, when we set up a communication session we use various ‘addresses’. I use the word advisedly as in reality most addresses are tags in both legacy POTS telecoms and IP-based services. The exception is E164 which developed out of mechanical switch technologies. For example, an MSISDN is a tag as it is not used for addressing but for the identification of an operator and user account. A tempory routing number is used in this case to address a mobile user.

Now things are complicated in IP because this has opened up the possiblities for communication, Email, IM, SMS VOIP etc. The number of ‘addresses’ has increased enormously. This is where ENUM and NAPTR help by allowing a single ‘address’/tag to allow applications to identify the most appropriate ‘address’ for communications e.g sip+2EU, mailto:+E2U etc.

Now where I think there is confusion is in the next stage. I think tony is refering to tags as arbitary pointers to an individual like what you throw up when you do an ordinary google search on ‘Tony Fish’. I think what he is saying is why should I have to remember even the one address that ENUM needs (1.1.2.3.4.5.6.arpa.E164) . I can search simply by requesting ‘Tony Fish’. So far so good. However what google can’t do is the next stage. The search engine works out from my own information that I must mean a particular Tony Fish and that his contact details are open to you because you are a business collegue and have been authenticated as such. Google can’t make a search based on information Tony entered about his status one minute ago. Google can’t tell me the Tony Fish I found on the search is really the person I want to talk to and not a Nigerian Engineer who wants my bank account details to that he can deposit $25 millon. Oh, and by the way the search also throws up that ‘Tony Fish’ is a bridge player on a thursday so if my mobile is switched off contact him there.

I think we had this system once before you know. It worked very well. It was called a switchborad operator. Strowger has a lot to answer for.

Regards Mark

Filed Under: Uncategorized

cognitive aggregation of tags ..

February 8, 2006 By

number.JPG

Over at forumoxford, Judy Breck whose work I admire expanded on out I am not a number – I am a tag idea

From a purely content angle, there is a fundamental difference between a tag and a number. A tag carries cognitive meaning and a number carries little if any cognitive meaning. The tag “New York” conveys the meaning of the city to most people. The number “212,” assigned as a phone area code to New York, becomes a bit cognitive as it is learned by those who use it, but most numbers do not even convey that much meaning.

With apologies to linguists for over simplifying, the cognitive aggregation function of tags is the use of something that has been happening on the Internet from its beginning – if informally. An early example was the “Hittite Homepage” (http://www.asor.org/HITTITE/HittiteHP.html) where virtual cognitive material about the ancient Anatolians has been aggregated by its experts since at least the mid-1990s.

It is observable that the cognitive aggregation that makes the Internet so valuable for learning (and for commerce, too) – that this aggregation follows the laws of small world network theory. (My 2004 book “Connectivity” describes that cognitive networking). Informal tags like “Hittite” are the underlying mechanism, seems to me. It would have been pretty hard to aggregate cognitively if the Hittites were designated only as something like #889345.

I realize that at a techical level every address is no more than a number, But the cognitive networking occurs at a virtual level that emerges from the technical level. At the virtual cognitve level numbers are a nuisance. Humans are going to tag webpages regardless of what transmission does. Would it be simpler just to use the tags humans choose for their cognitive appropriateness?

Thanks Judy. We seem to be developing this idea nicely with help from the community

Filed Under: MOBILE WEB 2.0

Is enum the answer to the I am not a number problem ..

February 5, 2006 By

enum.JPG

I got two comments about enum to my article on I am not a number – I am a tag

Is enum the solution to the I am a tag question?

My take is – enum is a much lower level of abstraction – and hence subject to the slow uptake/control issues/standards problems. The beauty iof tags is – they are at a much more ‘user’ level of interaction. Seek thoughts?

Steve Kennedy says

ENUM goes someway towards this, people think of it as a “electronic directory” but it isn’t.

ENUM allows a user (or the owner of an electronic number) to set their own rules on how other can contact them.

It’s based on the DNS system (and if successfull will mean DNS needs a complete overhaul to cope with the added demand).

It uses what are called NAPTR (pronounced napter – i.e like napster without the s) records and these are routing/preference records for how a particular number should be routed and to where.

ENUM was specified by the ITU-T and there’s an RFC for it. It uses the e164.arpa DNS zone (E.164 was the ITU-T group that specified it).

It only supports numeric numbers which are stored in reverse direction, so take a random telephone number +44-20-7123-4567, this would be stored as: -

7.6.5.4.3.2.1.7.0.2.4.4.e164.arpa

In DNS terms this is easy to parse, and fits into a hierarchical storage system (as does DNS in general).

When someone does a query for this it would start at e164.arpa which is managed by RIPE, they then delegate 4.4 to the UK (there is only ONE registry per country which is in UK terms will be allocated by the DTI).

The registry then holds the info for 4.4

(In the UK how the infrastructure will be run has not yet been agreed/organised, there have been some limited trials. Austria has a working ENUM infrastructure and is probably the most advanced country to date. It has been decided there will only be one registry and many registrars. A registrar can not be a registry, there are other parts beyond scope here).

So for an invidual number there’s a NAPTR entry which has URI’s telling the system how to reach that number and preferences. So it might say between 9 and 5 try the SIP address SIP:steve@gbnet.net, at other times use telephone number tel:0777777777. Any type of service may be specified in the URI.

What this means is that users (or number owners) control how calls should be routed to their numbers, which potentially is disasterous for telecoms providers as they lose control (i.e. dont pass traffic over the PSTN, use IP instead), but it does go someway to the end user controlling how people contact them,Steve Kennedy

Paul Golding says ..

I think Tony’s presentation must have been aimed at old-school telco managers who don’t understand IP. Ultimately, to “find” someone in the IP-connected world, you need their IP address. This is how to “locate” someone and therefore connect. You can tag an IP address any way you like and stick those tags in a search index. If you want to use a telephone number, you can (ENUM is only one method – a “standardised” one).

However, you still need to locate the IP address somewhere on the network. In the “Next Gen” operator world (e.g. IMS, TISPAN etc.), they still intend to control that process and have fully embraced tags already, which is what SIP addresses are essentially all about.

Filed Under: MOBILE WEB 2.0 Tagged With:

Carnival of the mobilists

February 4, 2006 By

carnival_mobilists.JPG

The carnival of mobilists is at the Mobile Enterprise Blog this week. Topics covered include Web 2.0, mobile behaviour and the latest apps. As usual, it makes great reading.

Filed Under: carnival
Newer Posts »