Casey Liss

And, you know, the reality is like I can try to do a lot of features around these file formats.

But again, because the world of DAI does not really support them, there's just not that much demand and there's not that much supply.

I will support them in the sense that I will download them and parse them and have a way to display them and maybe even display them by default.

I think that that would probably be the right move.

But I don't think I'm going to have them replace mine.

And honestly, I would be very surprised if most of my users ever saw a podcast that actually had them.

Overcast infrastructure uses tools that were proven and boring in 2013.

So what does that mean?

It means things like MySQL.

It means PHP.

It means Memcached.

redis and beanstalk d that's my server stack it's those it's those things you know nginx with the php fpm plugin like that's that's what i'm using here so what this queue is is beanstalk d which is very old very boring and it works fantastically it has almost no features it's

Almost no one that I've ever met uses it, but it's fine, and I've been running it forever.

I do use Redis' queues for feed crawling because they allow me to – the way Overcast is feed crawling, basically, I have these crawl servers that run the go process that I wrote a million years ago and then instantly forgot go.

No.

I learned Go just to write that one process.

Haven't touched it since and have since completely forgotten Go.

But I mean, I think today if I ever needed to add anything to it, I'd just have AI help me because I need to relearn this entire language.

But anyway, the way those work is they crawl the feeds.

They stuff the content of a changed feed into Redis, and then they add to a big Redis queue so that other processing PHP processes that parse those can go pick them up and crawl them.