Author Options:

Why the parsimonious ratings? Answered

Since the staff changed the ratings paradigm, I've seen only one instructable with a rating of 4 or above. And I've personally rated a few with 4.5.... Why are we suddenly so stingy with our ratings? (Is it like the "negative rating" days? down-rating because we can?) Or is the new system weighted, also? (with views, etc., factored, too?)


the rating system is too weighted, it would take hundreds and hundreds of 5* ratings to get to 4* all the ibles hover around 3.10-3.20

If you look at the top rated instructables...

Wall-E Robot has a rating of 4.60, and 79 votes. I assume not *all* of those are perfect 5-star votes.

So no, it definitely does not take "hundreds and hundreds" of 5-star ratings...

PS: some instructables do indeed have "hundreds and hundreds" of votes, and a rating very close to 4.0. The screen printing instructable is a good example: 352 votes, rating of 3.97.

But that's simply because all the old "plus" votes were translated to 4-star votes. So all the old favorites are stuck at 4.0, and will have a very hard time to ever rise above that. Definitely an issue that is open to improvement...

The new rating system is not yet complete. Right now, when you rate something, it is queued to be processed during our overnight Bayesian rating algorithm-run and does not update immediately. This sucks, and we're working to make the ratings update immediately.

Closer to complete. The ratings now update immediately.

Oh wow, I was just about to PM you about that. This Comment helps.

Hmm... should be fairly easy to do an immediate approximate update, then do the correct Bayesian update overnight. Chance are, nobody would notice the difference, and the users would get that instant gratification of seeing the rating respond to their votes. (Plus you're likely to get fewer "the ratings don't update correctly" bug reports...)

What did you do to the comment system? You can't go directly to the ible from your email, and it doesn't say which ible its from. I liked the old way better


10 years ago

This deserves a new topic, but...

I don't care for the new "images" layout of the Explore menu (the grid format.)

Why? The lead image is now 75% of the item preview--so the title is much smaller, and no description. Lets be honest--the photos almost never effectively illustrate what each project actually does.

And maybe only 1 out of 10 photos are intriguing enough (or well-executed) to capture my attention. One way or another, it's harder to glean useful info from the new format (usually I have to click for the old format--again, extra work.)

The new format works reasonably with featured instructables, but they the cream....

Also, I've moved this to the help category. Future things like this should be in the same category.

Right you are--"help" is where this belongs... The Bayesian rating schema might explain why there are no highly rated projects (yet.)

Ugh, lets go back! (please!)

hey, are your getting weird comment emails and then you follow the link and its different. I don't like it

Yes! Are you posting comment replys but they reply to the wrong thing!?

*sigh* The only way to fix this would be to reset all those IMO.... Also, when I (and other people from what I've noticed) rate something, that doesn't really count. For example, an instructable might say 4.00 and 5 ratings, yet on the hey Like it page, theres clearly over 20 people I rated something, yet it says no ratings......I'll try that....I'll rate this

I think the new system is screwed up. I gave an instructable a rating, but it didn't register.