Ask HN: How would you represent a downvoted comment in text-to-speech? | Hacker News Reader

Ask HN: How would you represent a downvoted comment in text-to-speech?

14 pointsmwcampbell17 comments

This is meta, but might still be interesting to think about.

On screen, a comment with significant downvotes has a different color. If it has enough downvotes, the contrast is such that the comment is harder to read; presumably that's the point. But when reading HN with text-to-speech (e.g. a screen reader for the blind), this information is lost. Sure, HN could be modified to add the word "downvoted" or the like such that only a screen reader can see it. But that doesn't have quite the same effect as changing the color. So, if you could solve this problem by modifying HN and/or the screen reader, how would you do it? Maybe add a background noise to the reading of the comment, and make the noise louder based on the number of downvotes?

View on HN

I imagine the text being said while hearing a number of booo's in the background. The number of voices boo'ing would increase as the number of downvotes increases. I think that even if a downvoted comment may not be popular, it should still be heard so distorting the comment in some way, as mentioned by a previous poster, would take away that commenter's voice being heard.

Nice. Having booo's in the background is such a natural way to represent a downvoted comment.

Ha ha! I totally dig that! Would be a reason for me to start using text-to-speech software. :-)

My first thought would be an ever-increasing accent, which becomes unintelligable when the comment is so downvoted that it can't be seen. Dunno how practical that would be to implement however.

Might be possible with text-to-speech engines based on pure synthesis (like eSpeak) rather than concatenation of actual voice recordings (every mainstream text-to-speech engine).

Downvoted comments should sound increasingly like Gilbert Gottfried.

I can't think of a way to adjust speech synthesis parameters to effectively produce that kind of voice. And for the mainstream text-to-speech engines (based on actual voice recordings), forget it, unless we get a Gilbert Gottfried voice (and people actually have it installed on their computers and mobile devices).

A good downvote sounds like when your parents/friends tell you to behave yourself and you somehow know that they are right.

A bad downvote sounds like when your parents/friends are asking you to behave yourself and they are not right. Bias will always exist. In HN they exist too.

Why not just parse the color and preface the comment with "Downvoted comment". The same can be done for flagged and dead comments without parsing the color.

This is the correct solution.

I think this is one of those things where you want to talk to the end user, a sight impaired person with a screen reader, and see what they want.

> talk to the end user ... and see what they want.

A maxim that when applied en mass could save trillions.

add some artifact in the background? maybe something unpleasant that gets louder with the more downvotes?

Or possibly the more 'shunned' the comment is, the more it sounds like it's in an empty space? reverb delay increases, the more isolated the comment is?

I think the reverb idea is the best one so far.

Maybe lower the volume of that comment? :) A really downvoted comment would be almost mute...

Except that whispering has a tendency to increase attention. Maybe speed up the reading?

Fart noises.

Reply on news.ycombinator.com