There are actually some really interesting problems to solve there:
It's now widely understood that online reviews can have a large impact on the success of a small business.
- What rating do you leave if you have a disappointing service from a really kind proprietor (like if the best humans make you the worst food)?
- Are we entering a world where there will be ramifications for the reviews you give? Will a restaurant be less likely to seat you if you left a middling review? As more places require you to identify with a phone number before you can be seated, will you receive worse service if you left a disappointing review or tip? It feels like reputation is about to flow in both directions.
- How do you avoid rating inflation when people who have bad experiences are reluctant to write about them?
And there are a bunch of little bugs in the current rating ecosystem:
- Culture impacts a rating. Americans are conditioned to start from 5 and deduct stars, which makes it harder to identify truly great places. Contrast this with Japan, where 3.5 stars is a really good rating, because Japanese people start from the median.
- If a place has thousands of reviews and a really high score, they're probably bribing people to rate them.
- How do you protect against spam? That includes reviews being bought from call centers, but also shitposts from people who don't like that something exists, or the way its staff behaves outside work.
- If people who eat fast food like a fast food place, it could have a better rating than an objectively better place that caters to more discerning clientele. How do you communicate that the people leaving reviews are/aren't representative of your tastes?
And as you alluded to, writing reviews (and HN comments) takes time that would often be better spent doing other things. What incentives do people have to take the time to leave a useful review? Can we find a way to make the process less burdensome?
