251–260 of 892 posts
A polling tool for quickly getting opinions on logo designs, product ideas, etc. I thought other people might like it too, but 5 months after publishing it I'm still making about 95% of the polls.
The 5,000 app users have now basically become my personal soundboard for ideas, which I'm more than happy to pay for.
Simple visualization of the relative size of different planets and moons I made when my daughter was asking if the moon is bigger than the earth - http://callumprentice.github.io/apps/celestial_bodies/index....
WoofJS is a simple JS canvas library and IDE I built for my students so they can learn JavaScript.
I made http://serializer.io 18 months ago and I still use it multiple times a day to follow tech news. It's how I found this thread.
looks somehow nice. gonna get on my list
In the same vain I put together a little archive for storing bookmarks under revision control:
https://github.com/skx/bookmarks.public/
Finally I put together a small archive of tools which seems to be quite popular for reasons I don't fully understand:
The projects I am the most proud of: http://rationalfiction.io - a collection of amazing science fiction stories. http://lumiverse.io - discovery platform for educational videos. http://digitalverse.io/rigs/ - several rigs that I have made, for practicing 3D animation in SideFX Houdini. Single scripts: https://github.com/raymestalez/rssdigest - sends me a daily email digest of my rss feeds. https://github.com/raymest…
I'm quite big on QA, but it's always been a problem for us, due to lack of tools. We have tons (millions) of time series being churned into proprietary files (neither of which can readily change). We've always had issues analysing these, be it analytically or visually. Two years ago, I wrote a parser in Python, which feeds the data into a browser interface. There, one can select values in a few () dropdowns - which denote dimensions, compare multiple files across these dimensions, and further manipulate these subsets of data. But the core are simple line charts from these data slices.
The whole thing is under 500 SLOC, it's blazing fast and it lets users cut through our data in no time. It has helped streamline our verification workflows, catch bugs, and allowed our clients to better understand the large amounts of data we send their way.