Great discussion guys (greetings Rob!). I hear your argument Rob and understand where you’re coming from. But I don’t think time can be the measure here or in other media — time is not in itself an evenly distributed resource. So don’t think time spent (consuming X) can work as a measure for articles, posts, videos, music, films, etc.
But I do think the mechanics of clap (!) can be tweaked to mitigate some of the follow on social FX and dynamics you fear (and which I’m sure Ev and team have hashed out ad nauseam):
- Taper off the clap rate on the button itself so that it’s non-linear and so that the slope discourages users from clapping too long
- Recognize that it’s not only number of claps but time spent clapping that shapes a user’s clapping actions and activity — and build that into metrics and analytics if possible. The first clap, for example, is worth far more than the second.
- What’s a second of clapping? Two? Three? Six? See if there are numbers of claps that look like common units of time spent paying attention. We know from live performances what’s a short round of applause; a good round; a wild and ecstatic round; and so on. Are there similar and natural “periods” of clapping here engendered by the button itself, and if so, can these be used in analytics to smooth out some of the curves?
- Don’t compare all clap totals with each other, but when appropriate, distinguish amongst categories of genre and author so that the rating systems are appropriately scaled to communities of authors/readers. In this way use segmentation to mitigate the impacts of an undifferentiated scale (as soon as you display the number of claps on a socially-consumed content item, you create a meta value (valuation of the product, represented as a number), and that meta value system will be (consciously or not) applied by users to everything.