The developer here. Moreso it's a vacation project. I started it after the Netflix redesign and put it down until two weeks ago when I was motivated to finish. I woke up and saw the Netflix Prize article on HN, and thought it was an interesting foil. It contains absolutely no elements of the recommendation system - relying on search speed as the main benefit.
Thanks, I wrote the backend in Django - but most of the heavy lifting is javascript. Django more or less is just populated by a cronjob and then caches the results of the Netflix odata API.
Thanks for the suggestion, I'm definitely considering some form in independent verification of the Netflix user rating system and was considering a series of icons that would display underneath the program info. The only concern is that I'd want to keep it visually not too busy - there's some visual design considerations that would need to be thought through.
I would like the ability to present the Netflix streaming titles with their associated imdb ratings... And filter accordingly. I find the high imdb ratings a better indication of what I actually enjoy than the Netflix rating.
Excellent point - I'd noticed that too, but hadn't gotten around to delaying the transition or alternatively placing it elsewhere. I just wrote it down on the todo list. Thanks!
I did stumble on to Instant Watcher midway through development. It's a great concept - I especially like the aggregation of review information. IW covers the full spectrum of search, recommendations and queue management. Qwikstant is much simpler by comparison, partially by virtue of not cutting off more than I could chew (also working fulltime), but also because I was scratching my own itch and search was the issue I was trying to solve.
Thanks! Yeah, I can see how it looks a little out of place. It's clickable (hides the filter sidebar), but the fact that wasn't obvious is good input..
This really makes searching so much better than what netflix offers. My only suggestion would be to keep the search options open since there are not that many and modifying the search requires opening and closing to verify settings.
That's a good point, and one I've struggled with a bit. I'd envisioned it being mobile friendly (android only so far) which places serious constraints on available screen real estate. On my tablet, the existing search filters run up against the limits. I'm not sure how far a one size fits all approach goes, and forking the display for desktop might be one way to solve that - but it has it's own set of issues.
20 comments
[ 2.6 ms ] story [ 53.7 ms ] threadEdit: grammar.
Probably the best use of the Netflix API I've seen - accounts for the search areas where Netflix lacks, without losing discoverability.
Your project is obviously much closer to the netflix design than IW is, but check it out, might give you some ideas!
I just hope I can remember the url next week though, not sure how to recall "stant" (or what it means).
http://blog.synopsi.tv/post/17773566495/histogram