USC hides reviews for other distro series instead of tagging

[Selene ToyKeeper]

A vendor brought a strange USC behavior to my attention: USC hides reviews for all distros except the one the user is currently running. The reason they don't like this is that it causes their sales to plummet every time Ubuntu has a new release, and then it takes a while for positive reviews to build up again.

I understand that we want to give customers the most relevant data for their purchases, but it seems counter-productive to hide data which may or may not be relevant instead of just tagging it to indicate that it may not apply. In many cases, the quality of a product is unaffected by the platform it's running on, and even more often, a history of good (or bad) reviews is helpful for building confidence that it will continue to be good (or bad) even on a brand new distro.

Instead of hiding the reviews for other platforms, we should probably attach a note of some sort indicating the version of Ubuntu the review was about, the version of the product, and possibly even some data about the hardware the reviewer used. These things could be used as search/filter patterns if the user explicitly searches, but otherwise I think we should show the full review history. I find this approach useful on my Android devices, at least... I can see whether the product is good in general, and then if I want to get more specific, I can search for reviews matching my particular device.

<https://wiki.ubuntu.com/SoftwareCenter/RatingsAndReviews#reviews-reading>:
"Next should be the five most relevant reviews for the item ... 'Most relevant' in this sense means, of all the reviews you have not flagged:
"1. any review that you have submitted previously (regardless of version);
"2. any reviews for the latest version that is available for your Ubuntu release (sorted by helpfulness or recency, depending on the current sort);
"3. any reviews of the next most recent version available for your Ubuntu release, and so on."

[David Pitkin (dpitkin)]

the client and web catalog must preserve users abilities to see previous distro and app version reviews. These are very important for developers and users.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Matthew Paul Thomas (mpt)]

In describing which reviews to show, up till now I hadn't mentioned Ubuntu versions at all. So I'm disappointed that reviews from other Ubuntu versions are being hidden altogether.

However, I had suggested that "The average rating for a software item, as shown to someone running a particular Ubuntu release, should be the mean of the last 50 ratings for versions of the software that are less than or equal to the latest version available for that Ubuntu release."

Imagine JazzWriter 1.0 was published for Ubuntu 12.04 and for 12.10. Later, JazzWriter 2.0 was published for Ubuntu 12.10 only. If you're running Ubuntu 12.04, the average rating should be based only on reviews of 1.0 -- it shouldn't be affected by reviews of a version you can't install. But it doesn't matter whether those reviews of 1.0 are from people using Ubuntu 12.04 or people using Ubuntu 12.10.

The same applies to which reviews you are shown. In the same example, you should see reviews for version 1.0 -- and only 1.0 -- regardless of whether the people reviewing it were using 12.04 or 12.10. I've now clarified this in the spec. <https://wiki.ubuntu.com/SoftwareCenter/RatingsAndReviews?action=diff&rev2=88&rev1=87>