This was a thing I made pretty much solely for myself outta frustration for how lazily Metacritic handle their aggregation of this data after going to the effort of collecting all the data.
I scraped the data direct from their pages and then made efforts to sanitize it by writing scripts to pull extra data from other APIs. The original goal was to try an do some data analytics to try and identify some trends on the basis of genres and whatnot over the years (especially comparing how films done in some of the earlier lists compared to the 2010s decade lists) but it moreso just became a tool for me to find some recent films to watch when I've seemingly exhausted every remotely obvious recommendation from recent years.
The UI is a bit of a mess in part because I was playing around with some stuff and in part because it was my first Svelte project so I decided to chuck all the CSS into one file so I could focus on Svelte itself. Doing either one of these would've fine but doing both together was extremely dumb. There's probably some easy fixes I could add in via prompt messages to help guide a new user but I figured it'd be more useful to share it now when people are more interested in these kinds of lists and looking for stuff to check out over Christmas than waiting until after.
Svelte seems really really cool. I don't think this project is anything like a good example of a Svelte project but I will say for a scrappy thrown together thing that got bloated fast, the speed of Svelte has helped cover up an awful lot of cracks.
On the more negative side, for the first while I had a lot of trouble with the life cycle and how updates are triggered. It wouldn't surprise me at all if there's still some nested stuff that just doesn't update.
Please read the home page for my attempts to explain in greater detail if you're interested. And if you have any ideas of things I could do with the data I'd love an excuse to try digging into it.
In and of itself it looks great. Fantastically clear. Works on smaller screens. Easy to navigate.
Ratings are tough to represent but as your work has its own style it's giving it at least the air of an alternative (that's a compliment) which is a good thing.
Thanks! I'm probably the only person I still know on an original iPhone SE (320px) so I tend to make things with a borderline fixation on squeezing them onto smaller screens.
Think this would def benefit with some labelling that just becomes hidden on mobile view or something though. Easy wins as opposed to trying to explain what it's doing beyond "here's some films people who are supposed to know about films liked a lot but using a different metric than average score or whatever, and here are some tools to sift through it"
You make a good point but always consider the trade off (as you have done) between all the info you now know is out there and what to incorporate. That's always the trick, what you leave out must enhance. But you knew that!
Don't hold your breath, as I'm tied up with other things, but I'm working on a user movie scoring technique now. Testing on immediate friends for now.
5 comments
[ 2.8 ms ] story [ 24.4 ms ] threadThis was a thing I made pretty much solely for myself outta frustration for how lazily Metacritic handle their aggregation of this data after going to the effort of collecting all the data.
I scraped the data direct from their pages and then made efforts to sanitize it by writing scripts to pull extra data from other APIs. The original goal was to try an do some data analytics to try and identify some trends on the basis of genres and whatnot over the years (especially comparing how films done in some of the earlier lists compared to the 2010s decade lists) but it moreso just became a tool for me to find some recent films to watch when I've seemingly exhausted every remotely obvious recommendation from recent years.
The UI is a bit of a mess in part because I was playing around with some stuff and in part because it was my first Svelte project so I decided to chuck all the CSS into one file so I could focus on Svelte itself. Doing either one of these would've fine but doing both together was extremely dumb. There's probably some easy fixes I could add in via prompt messages to help guide a new user but I figured it'd be more useful to share it now when people are more interested in these kinds of lists and looking for stuff to check out over Christmas than waiting until after.
Svelte seems really really cool. I don't think this project is anything like a good example of a Svelte project but I will say for a scrappy thrown together thing that got bloated fast, the speed of Svelte has helped cover up an awful lot of cracks. On the more negative side, for the first while I had a lot of trouble with the life cycle and how updates are triggered. It wouldn't surprise me at all if there's still some nested stuff that just doesn't update.
Please read the home page for my attempts to explain in greater detail if you're interested. And if you have any ideas of things I could do with the data I'd love an excuse to try digging into it.
Ratings are tough to represent but as your work has its own style it's giving it at least the air of an alternative (that's a compliment) which is a good thing.
Think this would def benefit with some labelling that just becomes hidden on mobile view or something though. Easy wins as opposed to trying to explain what it's doing beyond "here's some films people who are supposed to know about films liked a lot but using a different metric than average score or whatever, and here are some tools to sift through it"
You make a good point but always consider the trade off (as you have done) between all the info you now know is out there and what to incorporate. That's always the trick, what you leave out must enhance. But you knew that!
Don't hold your breath, as I'm tied up with other things, but I'm working on a user movie scoring technique now. Testing on immediate friends for now.
A note though: City Hall is probably a reference to the Frederick Wiseman film and not the Al Pacino film from 1996.