3 ms·

I think the author should look at other countries before declaring the Finnish system an outlier. The only "major" difference I've managed to find with the Dut

I think the author should look at other countries before declaring the Finnish system an outlier. The only "major" difference I've managed to find with the Dutch system is a slightly bigger emphasis in individual candidates due to "unsorted" party listings (although the article isn't quite clear about it, if the Finnish parties can sort their own list on the ballot, the systems are pretty much the same). Other than that, it's mostly minor cultural differences. These kind of "election engines" are all over the place in Dutch elections, I doubt whether that is any different in any multi-party democracy. Edit: Just looked it up for fun: these things pre-date OkCupid by almost a decade (more if you count the offline versions).

jsnell 11y ago
I did cover all of these points to some extent in the post, but could of course be wrong. - I'm aware that similar sites exist in other countries. From talking to people in many such countries my understanding is that they are not as influential (see numbers in the post on the reach and influence; e.g half the voters using one of these sites, and a sixth of those users just taking the top recommendation). - There is no ordering at all to the list, it's formed completely by the voters. (Well, there's an ordering in that everyone candidate is randomly assigned a 2-3 digit id number). This is to me a fundamental difference, since there is no obvious fallback option for "I just want to vote for this party", and that case becomes entirely about name recognition. It should be possible to check whether this really does make a difference, by checking what proportion of celebrities got "accidentally" elected under the different systems. - Yes, the first Finnish version was from the mid-90s, and the story of the site I built predates OkCupid too. But when you're explaining X in terms of Y, there's no assertion of causality (or even of Y existing before X). What matters is whether Y is a cultural referent that's likely to be shared by the readers.

jsnell 11y ago

I did cover all of these points to some extent in the post, but could of course be wrong. - I'm aware that similar sites exist in other countries. From talking to people in many such countries my understanding is that they are not as influential (see numbers in the post on the reach and influence; e.g half the voters using one of these sites, and a sixth of those users just taking the top recommendation). - There is no ordering at all to the list, it's formed completely by the voters. (Well, there's an ordering in that everyone candidate is randomly assigned a 2-3 digit id number). This is to me a fundamental difference, since there is no obvious fallback option for "I just want to vote for this party", and that case becomes entirely about name recognition. It should be possible to check whether this really does make a difference, by checking what proportion of celebrities got "accidentally" elected under the different systems. - Yes, the first Finnish version was from the mid-90s, and the story of the site I built predates OkCupid too. But when you're explaining X in terms of Y, there's no assertion of causality (or even of Y existing before X). What matters is whether Y is a cultural referent that's likely to be shared by the readers.