ubuntu-phone team mailing list archive

[Rodney Dawes <rodney.dawes@xxxxxxxxxxxxx>]

On Wed, 2015-04-08 at 18:12 -0300, Sergio Schvezov wrote:
> On miércoles 8 de abril de 2015 17h'54:04 ART, Rodney Dawes wrote:
> > OK. I don't think we should add more ways of calculating things to solve
> > these problems though. I think there are a few distinct problems here to
> > solve, and sorting is only one of the issues. There's also the social
> > problem of people trying to use reviews as ransom for features or bug
> > fixes. That is not how reviews should be, and I think the solution to
> > that is mostly one of policy, and not technical implementation.
> >
> > For this I think we need a way for developers to report/flag
> > inappropriate reviews/ratings on their apps.
> 
> How should I review? I don't ransom reviews at all but I don't give 5 stars 
> to something that I feel has things missing and not stated in the 
> description (most apps have poor descriptions so I can think anything I 
> want of the intention of the app).

Well, a poor rating/review is certainly warranted if the app is quite
bad or malicious. But because it doesn't have extra features is no
reason to rate it poorly. Reviews should also not be used for feature
requests (nor demands). There have been other threads on this matter,
but not really any good solutions either. A 3 or 4 star rating for an
app that you like, is useful, and works fairly well, but has a small bug
where something doesn't work quite as expected or described, is probably
appropriate. Likewise, you probably shouldn't rate apps with 5 stars,
just because you're friends with the developer, and it's the only app
currently available that does what it does. But yes, it is very
subjective, and people are going to do all the wrong things with it,
because it is subjective. This is why I don't generally like ratings
systems at all. :)

> I review as I would review a restaurant; e.g.; a gourmet restaurant serving 
> regular beers (or wine depending on how you take gourmet) won't get 5 stars 
> but maybe 4 instead and comment mentioning that if better beers were 
> offered the place would be so much better. I do the same for apps.

But you wouldn't give it a rating of 1 because you wanted lobster, which
was out of season when you attended, and had to eat something else
instead. Or because the special you saw one day last week isn't
available any more. You'd give it a reasonable rating based on your
experience, and not on your demands.

> Scoring and staring systems are entirely subjective to say the least, but 
> Niklas' request of using ponderation for sorting makes perfect sense to me 
> regardless; I don't think we should focus too much on a lone example to 
> divert or expand the focus of discussion, it is a different request and 
> different thing to solve.

Ratings systems are indeed very subjective. As I stated already, I have
no problem with improving the sorting. I was simply trying to separate
the issues which Niklas presented in his original mail, into their
separate concerns.

A) Sorting is not great and it's hard to find apps due to ratings
B) Some people are apparently making ransom ratings

The former should be possible to solve pretty easily by just having the
server alter the algorithm used to calculate the average.

The latter is a much larger fix, and is a social one, relating to the
policies we have in place for how people should behave in the community.
We can't solve the social problem with code, but we can provide a decent
way for developers to easily report these sorts of reviews, when they
happen.

Attachment: signature.asc
Description: This is a digitally signed message part