Sunday, April 23, 2006

apologize or else

apologize or else

So now it looks like Conencting for Health have put the ball back in the court of the academics and other signatories of the letter asking for a review of the
programme - viz
CFHfIT
say, ok so you want a review, please describe its terms of reference. Good call.

My take is that the review should be _by the people, of the people, for the people_ - i.e. it should be user centered. This means NHS employees and healthcare customers should define its contents, without fear (so some form of anonimity, which ought to be bread and butter to an NHS IT project should be assured for people who wish to be critical but are worried about their jobs- however, care should be taken that this doesn't lead to un-verifiiable whinges - this also, should not be beyong the wit of medical researchers to define:)

Thursday, April 20, 2006

apologize or else

apologize or else

Why would anyone want to review the NPfIT ?
FAQ
1/ its just a waste of time
2/ its going fine (see the websites)
3/ if it doesn't deliver, we don't pay
4/ any other reasons?

1. having an independant review of a large project doesn't typically increase project times - actually ,its a chance to improve documentation, and to get low cost objective fedback so you can re-factorize and change priorities. Aside from any accountability questions (and if the project fails, and we don't pay, we are left with a god almighty mess AND waste of NHS resources - time of training of medical and admin staff to use a failed system wouyld represent a massive sink of effort - the project must not be allowed to fail unless there's somethign crucially wrong with it, but it must also be bought in reasonably close to budget - this is not impossible, but I don't believe a self-monitoring programme has ever done that.

2. the website and other data on the programme are extensive, but so is the NHS - 14,000 siginficant sized sites with 1M employees - The story of how well its going down to grass roots is not captures through a few case studies and glossy PDFs or even some (laudible) Best Common Practice material. An _evaluation_ of the programme progress, and SWOT analyiss mid point should be public (the last annual report is not deep.

3. If the project misses part of its deliverables, the affect on not only the NGS but the suppliers is going to be fairly bad (think share price, think sunk cost, think committed opex as well as capex) -

The negative response to what is essentially an offer of help is suspicious.
The public will respond with "so what have they got to hide". Thats not helpful.

On the other hand, it might be good if an industrial conortium did an independant review (on the other hand, several of the candidate companies didn't bid for the NPfIT because thye assessed it as infeasible.)

Wednesday, April 19, 2006

apologize or else

apologize or else

So having spent some time to look through as much of the documentation on the NPfIT website as I could (most, exlcuding a few documents where links were broken) I Have to say that it is really impressive. I think there is ample evidence that the programme will deliver a lot of useful pieces (e.g. electronic presscriptins, picture arching and comms) and that is a fine thing - the pieces that look less likely to work for a while, to me, are the ones where policy is embedded in technology (.i.e. public facing systems), but they may yet work. The cradle to grave ambitions of the original NHS, translated into cyberspace, are clearly there, and the progress monitoring/reporting and visibility of a lot of the work is clearly also being carried out with the best intent. I wonder, just wonder, if this might not succeed? It would be very nice if the folks involved manage it (what would they do for an encore?).

A True History of the Internet

A True History of the Internet
Recent conversations about the
National Program for IT in the NHS or NPfIT, make it clear to me that the public understanding of academic computer science contributions to the world is sparse.

I've been an academic since 1981 (and even before that worked teaching remedial math in High Schools in Toronto, and spent 2 years as a systems/program advisor in North London Polytechnic. In all that time, I've observed that academics in CS like to build stuff. In fact, as part of both teaching and researching computing, one tries things out. Oftem the things one tries out are new (as thats more fun to teach and learn or because its research) - sometime later, you find lots of people using the thigns you built, often unaware of the origins of the stuff- the Net, the Web, lots of Unix stuff, sure, but what about more "commercial" systems? well, cisco, sun and google all started as stanford University EECS department internal projects. Microsoft's first TCP stack was done as a C port of a macro-11
stack that UCL worked on, and then handed via Spider Systems, Mentat and other companies to end up in Windows (since re-written, some of the re-write, e,.g the mobile IP code based on work by microsoft research with Lancaster University. There's a LOT of examples like this - it wouldn't surprise me if a significant fraction of software today didn't start out in a university lectuters spare time project. Some of the really "odd" theory things some folks in cambridge do like theorem proving are done using theorem proving programs. These programs turn out to be quite useful.

Some day, we'll do some cyber-archaeology and trace the family tree of software systems and discover whether my conjecture is even half way right. In the meantime, if you use ADSL to access the Internet, you have more than a 50% chance of using a modem with a chip in it designed by a colleague of mine here in the University. Oh, there's only about 6 million of you in the UK alone:)

Now, do we have the right to call into question whether the NPfIT programme should have due diligence (i.e. progress report/study, etc) done? Not only do we have the right, we have the duty AND the skills. One of the folks who signed the letter
asking for this to be done _invented_ software engineering.
Another, while he has written a quite good book, has also worked _in_ banks IT departments. Qualifications (real world or ivory tower)? well, it seems they wear them on their sleeves for all to judge, which is fine, so what is wrong with asking for another government group to do the same?

apologize or else

apologize or else

I wonder if it is Philip Gould who wrote some rather inaccurate
statements about my fellow academics on
Naughty Academics for daring to suggest the NFS National Program for IT might need to haev some transparency about its plans and progress, just like ANY well run project ought to

It has the same aggresive assertive style one associated with said person - see a response by a frend of mine,m which I thoroughly endorse - I don't know any academics in ivory towers any more, least of all in hothouises of technocracy such as Cambridge from where Ross ANderson hives, where most Computer Scientists have worked with industry and have to produce results to get funded, and have done more useful things for the world in their lunchbreak than any Government Department in a decade.

apologize or else

apologize or else

I wonder if it is Phillip Gould who wrote some rather inaccurate
statements about my fellow academics on


It has the same aggresive assertive (and factually wrong) style one associated with said person - see a response by a frend of mine,m which I thoroughly endorse - I don't know any academics in ivory towers any more, least of all in hothouises of technocracy such as Cambridge from where Ross ANderson hives, where most Computer Scientists have worked with industry and have to produce results to get funded, and have done more useful things for the world in their lunchbreak than any Government Department in a decade.