Tuesday, May 17, 2005
100 Welsh Heroes (the end or something close to it)
The letter is very detailed and I do not have much time so I will set out their main conclusions now and deal with the specific allegation later on. The letter states that in short the Wales Audit Office staff found:
- Culturenet Cymru following proper procurment procedures, procured website design, development and production services and related promotional services from external contractors;
- the main contractor delivered at just under the budgeted price;
- Culturenet Cymru designed the poll to have certain voting rules, but some voters went to considerable lengths (with both on-line and paper ballots) to circumvent these rules to promote their favourite 'heroes';
- to attempt to eliminate illegitimate votes identified through screening figures in the database underpinning the poll, Culturenet Cymru made regular and systematic adjustments to voting;
- Culturenet Cymru did not formally document decisions taken by the project team (the Director, Project Manager and the IT Manager) on precisely what weekly amendments to make to voting figures;
- Culturenet Cymru did not operate formal controls over whether the adjustments actually made to the database by the IT Manager were those agreed by the project team as a whole.
Culturenet Cymru officials have acknowledged there were shortcomings in documentation and control. In mitigation they have said that the poll was an opinion survey not an election, and that it attracted a far greater number of votes than had been expected, which put significant pressure on them in terms of screening and processing votes.
Despite these shortcomings my staff found no reliable evidence to suggest that Culturenet Cymru systematically manipulated the voting figures to secure a particular result, or that external pressure was brought to bear on the company to do so. However, given the control environment in which the project team worked together largely on trust, it would have been quite possible, for example, for the former IT Manager himself to have manipulated the voting figures had he wished to do so.
The last paragraph is a daft and maladroit attempt to shrug off responsibility. I wrote the program that added or removed votes after being asked to find a way of fixing the figures. I was asked each week to adjust the votes by an amount determined by other senior staff members - I didn't have any particular interest in who won and who lost at all.
Yes it was a poll not an election but it cost £150,000.
My job of course is to scrutinise Government expenditure and to act on behalf of the public in seeking value for money. I was forced to involve the Wales Audit Office because the Minister refused to conduct his own independent inquiry. As such this is very much a constructive use of my time.
No database adjustments were made because of paper votes - paper votes would be ignored if the marketing department, whose responsibility it was to enter them onto the database, decided not to count them.
The other thing that I suppose the Audit Office will gloss over is, why were made-up votes added to the poll? That wasn't a case of eliminating 'illegitimate' votes.
Oh, and by the way, I did the third and final version of the site, not an external contractor. The external contractor did the first two sites, using Microsoft .NET and IIS and I used PHP, which is what the site is currently delivered in - I don't think the external contractor we used develops in PHP.
The £150,000 tag doesn't include internal staff costs.
It's easy to create a fixed poll that doesn't accurately reflect votes cast and it's easy to do it very cheaply. Question is, should the Assembly fund it, or fund the body that managed to spend so much on a poll that everyone now acknowledges was fiddled.
Here's a thought for Alan Pugh - forget about money for Culturenet and instead commission articles for wikipedia, translated into Welsh on wicipedia (cy.wikipedia.org).
Culturenet paid a media consultancy in Cardiff to write many of the brief biographies on the 100 Welsh Heroes site. £150,000 might easily have bought more than 10 times that content and made it all available for addition, amendment, and reuse through Wikipedia.
No imagination shown, partly because very few people connected to the project understand online issues (or offline issues, either, memorably forgetting to credit the WAG on the accompanying book or reserving an ISBN for it, making it unsellable in bookshops).
Don't know how they became involved in the first place.