[listenallyall]

Despite what SO states, the karma system does matter. And it's flawed, as the vast majority of karma that will ever be awarded, has already been awarded. Very simple questions have thousands of upvotes, as do the answers, and all of the basic, common questions have been covered extensively. The only path for a new, energetic, enthusiastic user to grow his/her karma score (and associated credibility) is via a massive grind. The likelihood of a new question or answer gaining hundreds of upvotes, is very slim.

I don't claim to have all of the answers, but a decay function seems quite obvious to me. For one thing, technology evolves. The jquery way of doing something shouldn't have 10x the upvotes as the ES6 way. Old answers should lose a percentage of their upvotes over time, if they remain relevant they will gain new upvotes, but if not, they will get surpassed by newer, more relevant responses.

Oh, and the new fonts.

[busyant]

> I don't claim to have all of the answers, but a decay function seems quite obvious to me.

I don't have a super-informed opinion here, but a decay function sounds reasonable. I'd also suggest that the decay be uncoupled from the user's karma (i.e., the user keeps his/her karma, but the value of the votes decay over time to make way for changes in technology). I don't claim that this is a well thought out idea, but at first blush, it seems reasonable. Also, I just don't give a shit what my karma percentile is.

Finally, FWIW, I find SO ~85% useful and ~15% problematic. And I feel like sometimes we fixate on the 15% without acknowledging the genuine utility of the 85%.

[nailer]

The new fonts are just system fonts, which exist because they’re fast and good.

[listenallyall]

Segoe UI, which Windows users now see by default at SO, is not a good font for reading blocks of text. Actually, it's not just "not good," it's terrible.