?

Log in

No account? Create an account
A point on large projects and ID - The Ex-Communicator

> Recent Entries
> Archive
> Friends
> Profile

April 15th, 2010


Previous Entry Share Next Entry
10:01 am - A point on large projects and ID
I think that the drive to establish a transferable secure ID for British citizens is tied to the drive to break up large government programmes into smaller chunks which can be bid for by SMEs (businesses smaller than the massive ones like Capita who currently dominate government IT contracts).

Currently a Brit like me has a unique identity on each of a number of large IT systems - DVLC, NI, passport control, customs and revenue etc. These major systems can neither speak to one other, nor be subdivided, because the limits of my secure identity on the system, and the processing of my information within the system, must share the same boundaries. You can't split National Insurance into sub-programmes because I don't have a transferable secure identity external to and independent of the NI system, so any processing activity which utilises my NI-identity has to occur within the single mega-programme within which alone that identity is meaningful.

This means that only three or four massive organisations like Capita can bid for the big government IT projects, and it makes them lazy, complacent and expensive IMHO. The Conservatives (for example) have said that 'we will open up government contracts to SMEs and open source by breaking up large ICT projects into smaller components'. But they have also said they will work against establishing a single transferable identity system.

I think these aims, both of which are good in isolation, are incompatible. Criticism of ID has mostly been taken forward by the most well educated and intelligent of pro-Tory voters, so I have been very disappointed that the big advantage of secure transferable ID - that it allows fragmentation of monolithic IT projects as demanded by the same people - has not been addressed. But point me to it if I'm wrong, I'd be really interested if this circle has been squared. (ETA - I'm not being sarcastic, if there are solutions I don't understand then they should be explained to people like me who are uninformed)

(7 comments | Leave a comment)

Comments:


[User Picture]
From:andrewducker
Date:April 15th, 2010 02:21 pm (UTC)
(Link)
Absolutely. Some people want a separate identity in each database, so that things can't be tied all together. Personally I don't mind having a single ID, but I know that I'm odd.
[User Picture]
From:communicator
Date:April 15th, 2010 03:05 pm (UTC)
(Link)
What seems odd to me is a consensus that should both break up the big databases and not have transferable ID. I think it's one or the other - big closed systems or a distributed multi-focus system with transferable identity, you can't get rid of both as far as I can see
[User Picture]
From:andrewducker
Date:April 15th, 2010 03:14 pm (UTC)
(Link)
You can do both - but it's horribly inefficient, because each one has to keep its own identity information, it's out of synch with all of the other ones, and when you move home you have to tell 45 different companies.
[User Picture]
From:matgb
Date:April 15th, 2010 11:53 pm (UTC)
(Link)
These are actually good points, some of which I'd not considered. My principle objection to the NIR scheme was the planned one massive database with all the info in one "unhackable" place, combined with the massive threatened fines if you forgot to update, etc.

However, the objective of breaking the projects up into much smaller units is a good one.

But wouldn't a single unique identifier like NI number, alongside public data like name, address and listed phone number, be transferable but the rest be private and non-transferable?

I'm not a data tech, at all, been there, would rather not.

Definitely completely opposed to the scheme as proposed initially, but they've kept moving the goalposts.
[User Picture]
From:andrewducker
Date:April 16th, 2010 08:21 am (UTC)
(Link)
Well, you could hold these identity pieces centrally - like name, age, date of birth, and that way they'd be easily updatable without replications errors.

Julie's Dr recently passed her address on to the hospital for an appointment. The address her Dr had was 2f1 46 Some Street. The hospital's address system couldn't cope with addresses in that format, so the address became flat 2, 46 Some Street. Which was the wrong address, and meant that the appointment information went to the wrong people. As a computer person this kind of design drives me _insane_, as does the stupidity that means that when we move we'll have to tell both places that we've done so.

I'm against massive fines for not being on the system. I think they stem from stupidity on the part of the government - because I think that if they are given the choice, 95% of people will _want_ to be on the system, and the other 5% will just have to live with their lives being more complex than they might have been.
[User Picture]
From:sugoll
Date:April 15th, 2010 08:29 pm (UTC)
(Link)
Personally, I don't want all those systems tied together. We live in a digital age, and the only thing that keeps us all safe is our incompetently-designed and incompatible systems. Get them all working together, and your identity can be torched like tinder. And I'm not even talking about malicious identity theft - simple mistakes propagate everywhere, and it can be very hard to prove there's even anything to undo, because people trust what's on the screen in front of them.

The US has had decades of people implementing systems based on social security number, and it often raises its head in comp.risks (or did, back when I used to read it).
[User Picture]
From:communicator
Date:April 15th, 2010 08:54 pm (UTC)
(Link)
The consequence is that we must remain beholden to two or three massive private companies, who make a total mess of handling our data, and exclude smaller companies, innovation, and open source

> Go to Top
LiveJournal.com