When I started Preposter.us in 2013 I had a singular goal: to build a replacement for my favorite blogging software, Posterous . To this end I think the project was largely successful; within a few days I had an operational website, and within a few months the system was mature enough that I invited others to participate.
But after a year I grew weary of wrestling with the idiosyncrasies of parsing emails and, along with a mounting backlog of unimplemented requests I didn't have the time or gumption to push the project through to the next version. I moved-on to an off-the-shelf blog and Preposter.us went to sleep.
It's been half a decade since then, and having spent that time using several blogging tools I've found my writing productivity dropping-off to a trickle. The most recent attempt to streamline this was moving to Hugo , creating posts in Markdown and managing my content through Git. This seemed like a natural fit as I spent a lot of time using Markdown and Git , what I've found is that working with a git repository of this size is unwieldy, resulting in a lot of overhead that prevents me from writing unless I have a lot of time and energy to dedicate to the task.
Contrast this to writing with Preposter.us: I compose the post on anything that can accept text, and I can post it from anything with an email client. I can post any kind of media (text/photo/audio/video) with rich text formatting and the post goes live within a minute of having an Internet connection. All this with almost zero server maintenance, and the meagerest of server resources.
So why did I ever switch in the first place?
I think the primary reason is that I felt overwhelmed to make Preposter.us "more". Template customization, comments and a number of other expected features weighed heavily on me because implementing them threatened the simplicity and elegance of Preposter.us's email-based user interface. There were also bugs, and at the time I felt like I had to chose between writing and bug fighting, and eventually I felt like the writing, the purpose of the endeavor in the first place, was suffering. I made a decision that I wanted to spend my time writing English instead of Python, and used this to rationalize switching to an off-the-shelf solution. This seemed logical at the time but as the last few years have shown, the algebra wasn't quite so straightforward.
So now I'm revisiting Preposter.us. I'm already feeling a weight lifted off of my writing and the system is far from what I'd consider "production ready". But with only a couple hours I was able to get the code operational again and bring https://preposter.us back on-line.
Of course the old bugs are still there, and it's going to take a lot of work to shake them all out. But I've learned a lot about Python in the years since I last looked at this code, and this time I'm going to do more to get some help as well. I'm also reeling-back on features to allow more time for polish, and I'm on the hunt for 1000 true fans who are looking for the same thing in a blog that I am.
Preposter.us 2: press send, receive post
- Jason