Tweets from the Vault is a service that will take a random[1] historical item from your RSS feeds and tweet a link to it.

When I started building the service, my goals were simple:

  • Solve my own problem
  • Get to a minimum viable product as quickly as possible

In this series of posts I’m going to look at each of those points in a little more detail.

Solve my own problem

There’s a bunch of content in my blog that is still very relevant today. Lots of stuff on agile; lots on software principles and some valuable odds and ends that were starting to be lost to the archives.

I have IFTTT tweeting content as soon as it hits my RSS feeds, which is great, and obviously that leads to people’s visiting those posts while that tweet appears in their timeline. Once that tweet falls off their timeline, though, all bets are off - nobody’s likely to see that tweet ever again.

I wanted a solution that would periodically fish out a historical but relevant article and tweet a link to it and there wasn’t a service that did that for me in a way that I liked. The closest I could find was an outdated Wordpress plugin (and I don’t use Wordpress). Well… I blog mostly about software so why wouldn’t I write one for myself? And, if I were to write one for myself, perhaps I could tweak it a bit and make it useful to other people. And thus, Tweets from the Vault was born.

Mimimum viable product

The minimum viable product for me was pretty simple:

  • Sign in using a Twitter account
  • Set up a small, recurring payment
  • Add and remove RSS feeds

  • On a schedule:

    • Pick a random article from that set of RSS feeds
    • Tweet it

And that’s pretty much it.

In the next post in this series, I’ll start looking at some of the technology choices for the app.

[1] For a given definition of “random”.