A few weeks ago I finished writing a tutorial for a web framework I've built. The tutorial teaches programmers how to create a barebones, toy version of Slack/Discord/etc. I called it "eelchat." Finishing that little project has been enlightening.
In particular, any excitement that I previously had for creating a new, non-toy discussion app has been sapped away. Making apps is hard! Getting an initial version out is easy—10 days is my average for the past few things I've launched—but refining it, getting it into a state where it's genuinely valuable to other people? Takes forever. Months! It's a huge drag, and it's precisely why I was so reticent to pivot from The Sample to Yakread instead of pivoting to getting a job or similar. (I started building Yakread in August, and it's taken me four months to get to a point where I feel like working on growth/marketing instead of solely improving the core product is a good idea.)
On top of that, any time I spend working on a discussion app would occur exclusively during my "open-source Fridays." So yeah, I have very little desire to spend all my Fridays for the next year working on eelchat. As such, I've been thinking more about the stuff I wrote in Discord could almost replace Twitter. Summary of that post:
Due to that last point, my conclusion then was that it wasn't worth spending much time on a Discord forum publisher bot; better to just create a new discussion app from scratch.
I have now changed my mind.
I still don't think it'd be worth, like, starting a movement or anything. Discord still isn't and won't be a Twitter replacement. But for my own use at least, I think the forum publisher bot could be really nice. I built a proof-of-concept of it already, back when I wrote the previous article. But the UI sucks, and it needs better RSS/email stuff.
So at some point, I'll probably put some more work into that. I think it could make Discord a really nice place for sane people who don't want to check 50 different text chat channels every day. And then maybe later someone will decide to turn eelchat into a real thing.
If you're thinking about making a little Discord server for your own newsletter or whatever, I think the following minimalist structure is a great starting point:
So if you just want to post a meme or say "how do you do," you can throw it in #random. If you have something that might possibly be worth coming back to in more than 48 hours, you can make a post for it in #stuff. A decent heuristic might be "does giving this message a title feel like a drag?" If yes, just put it in #random.
If you do that, you can also go ahead and add the forum publisher bot to your server! Only takes a click. And be sure to send me an invite.
Normally I'd put this section last, but since I wrote about Discord: here's an exclusive, holofoil-rare invite link for the tools-for-online-speech Discord server. Also feel free to email me.
I did several things last week! The biggest one is that Yakread's timeline has been turned from a drab list of blue, flavorless links into a parade of never-ending written excitement:
(I had to scroll down to find one that had an image—newsletter emails rarely/never include image metadata, at least not in the place I'm looking for it, and I haven't yet put in any effort to deal with it.)
Besides that:
I also started working on an "Open in Yakread" feature. I only managed to finish the UI before the weekend hit:
(Still not sure how I feel about that "Open" button...)
If you paste a URL into that little box (which will be shown at the top of every page on Yakread), it'll fetch the contents and then show you various actions you can take, such as:
When this is finished, I'll also provide a little bookmarklet so you can have an "Open in Yakread" button on your browser toolbar. I'd even like to make some mobile apps. It might not be that hard: I just need a couple apps that add an item to your phone's share menu. When you share something to Yakread, the app takes the URL and then opens up Yakread in the regular web browser, passing the URL as a parameter.
However having done a decent amount of Android stuff in the past, I generally prefer to not touch mobile development with a 10 foot pole. (If you're a mobile dev and you think you could bang one of these out in under an hour or two, please let me know!) But whenever I do get around to this, I think it'd be nice to release the apps as open-source. Then anyone who wants to make a web app that supports opening URLs can copy the apps, change a couple settings, and use them for their own project.
I'm gonna try just doing links without excerpts or commentary. Less work!
