While the “Open Data Definition” argument has been won – on the web, for free, for anyone, for ever; reality has a habit of tossing up edge cases, that don’t come up in theory.
All have valid concerns, fears and desires. Paul, on the database state aspects, Lou on the impact on offenders, and Will on communities.
There is also the Rehabilitation of Offenders Act, which means that Government can not name people when those convictions have been spent (generally after 5 years). Government tends to be slightly careful about not breaking the law.
If we consider those as free and open data, then there clearly seems to be problems with the full list of names and data being published and copied widely.
However, that’s not, quite, what Will is asking for. Will would like to know very specific – “who committed crime AB1234?”, which he can get by walking down to the court and asking for it. Will is trying to get round the problem of having to go to the court (when it’s open and they can find the piece of paper), and just ask a computer. That’s something that should be worked towards.
I don’t believe that either Lou or Paul would argue that anyone should be prevented from publishing the name of the person convicted of a crime. It is purely a question of how they get that name.
The passage of time helps. Most blog posts have degraded after a few years; how many of your blog posts from 5 years ago are still findable?. Looking at Will’s site, they’re there, but take a good amount of search. Most blogs aren’t that well maintained (and even some KXE posts seem to have disappeared in the recent move). There is no takedown requirement after a number of years, but Government may not publish that information. It helps, but doesn’t solve the problem.
The unstated assumption of both Paul and Will could be described as, does the action of finding out need to be cognitively free?
Could there be some relatively trivial barrier to access, and an agreement on that access that it wont just be pasted, effectively alone, into a field in a database. If you want to know who committed a crime, or a number of crimes, in your area, then you can find the names. But that doesn’t mean the whole thing is on the web for everyone. It doesn’t stop KXE writing what they want, but it does mean that the Government stays legal.
And it doesn’t need to be a simple captcha; if good enough geocoded data is available (OSM is, OS should be), you could instead of asking for a word, you could ask for the name of the street next to the one that the crime took place on. Or the name of the church on the corner. Something that’s infeasible to automate; but easy to do for individual information, and which fundamentally doesn’t scale. I’m sure this idea for “local knowledge captchas” has other applications as well.
But for now, progress is good, and the progress is towards lots of detail freely available. The more rough and nuanced areas can come next when we see what’s working, and how, and what isn’t. As Lou points out, these decisions can have significant impact on lives, and it’s impossible to get this genie back in the bottle. Being right slowly is better than being wrong quickly.