The year 2020 has been weird for everyone, but one good thing that has come of it is that it's been easier to distance (ha!) oneself from others for a bit and do some quiet contemplation. It's something that was long overdue for me, and it is starting to pay real dividends. It's been over eight months since the last post on this blog, and now that I'm feeling more collected and energized, it's time to put some new life into it. There's about half a dozen half written blog posts that I want to finish and publish at some point, but I thought I'd make this post about the recent developments in the open source projects I'm involved in as an easy starter.

Work on pulldown_cmark has settled a lot since last year, with only two point releases in 2020. All known parsing bugs have been addressed and the latest 0.8 release saw the introduction of smart punctuation, automatically generating fancy opening and closing “quotes” and em-dashes from simple ASCII. The project is in a pretty good place right now, with the only missing feature being a math mode. Maybe 2021 will be the year where we'll finally do a 1.0 release.

Today I've also published a new release for sortedvec after more than a year of inactivity. I was prompted to make some changes after some issues were filed, which was a pleasant surprise as I didn't know the library was actually being used. A funny thing that happens that when you look at work you've done such a long time ago is that you forget what problem you were trying to solve. The project readme did state what the library does, but not why. I'm too embarrassed to share how long it took for me to figure it out. I've added a paragraph to the readme so others won't have to wonder as long.

One of my special interests have been video coding as of late, especially new codecs like AV1 and VVC. I've been working on a blog post on the current state of visual codecs and where they may be heading in the future. It'll probably be the next post on this blog. One of the central components of many modern compression methods is arithmetic coding, and I've been implementing my own encoder and decoder from first principles (in Rust of course). There are still some issues to sort out regarding numerical stability, but once they are addressed I will publish the project on GitHub.

There's plenty of stuff in the works, and I hope that translates to more activity on this blog as well. The best way to keep up is to subscribe to the RSS feed, if you haven't already. Hope to see you soon!