In light of HealthCare.gov struggles, here are the UK's design principles. | Hacker News Reader

In light of HealthCare.gov struggles, here are the UK's design principles.

gov.uk

20 pointsquickpost9 comments

Gov.uk is, amazingly and surprisingly, well run, and I think this advice is solid. But broadly speaking, it's not like the British government has done much better with large computer projects.

http://en.wikipedia.org/wiki/NHS_Connecting_for_Health

Many UK government big IT projects have been disastrous.

http://www.computerweekly.com/news/1280096822/Government-IT-...

> But in central government there is no such thing as an IT disaster: project names are usually changed or the failed scheme continues indefinitely. A project to deliver unified systems for magistrates has continued for about 17 years, complete with name changes.

http://www.zdnet.com/the-top-10-it-disasters-of-all-time-303...

> Edward Leigh, chairman of the Public Accounts Committee, was outraged when the National Audit Office subsequently picked through the wreckage: "Ignoring ample warnings, the DWP, the CSA and IT contractor EDS introduced a large, complex IT system at the same time as restructuring the agency. The new system was brought in and, as night follows day, stumbled and now has enormous operational difficulties."

> It was the summer of 1999, and half a million British citizens were less than happy to discover that their new passports couldn't be issued on time because the Passport Agency had brought in a new Siemens computer system without sufficiently testing it and training staff first.

http://www.channel4.com/news/articles/science_technology/mod...

> Over £4bn is being spent on a huge new defence computer system designed to help Britain's troops operate more effectively in Iraq, Afghanistan and across the globe.

> But a joint investigation by Channel 4 News and Computer Weekly magazine has uncovered internal emails from civil servants describing parts of the system as "an unmitigated disaster" and "not fit for purpose"

The notion that British healthcare IT is a model for success is somewhat disputable: http://www.bbc.co.uk/news/uk-politics-24130684

Yes, I know that the gov.uk guidelines are very good, but they relate primarily to customer-facing websites and user-interface design, and don't really cover the gnarly, painful back-end integration jobs. To the extent that healthcare.gov is an expensive failure, it's due to the complexity of the back-end systems and not due to, say, poor UX (although it might have poor UX too).

Government Digital Service (GDS) covers this and more, see their service manual https://www.gov.uk/service-manual

>We should share what we’re doing whenever we can. With colleagues, with users, with the world. Share code, share designs, share ideas, share intentions, share failures.

…and also share data about the interactions between UK citizens and their government with a foreign corporation.

Who on earth thinks it is appropriate for Google Analytics to be informed about my interactions with my government? Or for Google Analytics to be informed when I anonymously report a crime[1]?

[1] http://webmasters.stackexchange.com/questions/47069/is-it-et...

Government should only do what only government can do. If someone else is doing it — link to it. If we can provide resources (like APIs) that will help other people build things — do that. We should concentrate on the irreducible core.

That's good advice for all of government, not just digital services. For way too many government services, the irreducible core is simply the wealth transfer to make sure the better off are taxed to pay for services to be available to the poor, not actually providing the service.

These seem to me like they'd be good principles regardless of who your client is.

Gets on hobby-horse

For the jobcentre site (as was), 'Start with needs' means IDS needs to be able to monitor these shysters for not getting a job.

Gets back off

Is this in response to the problems with the website or the system as a whole?

Reply on news.ycombinator.com