In part 1 I discussed the origins of the ideas that have driven the development of SWIM.
In part 2 I talked about the history of the product from the early 00’s until 2006. Those were the most productive years, but after pivoting the idea I have worked on a couple proof-of-concepts since.
As part of my application to work at Collecta (which I did from 2009-2011, when it shut down), I developed what is really the second incarnation of SWIM. I called it Suggest-o-matic, and for a very quickly thrown together bit of HTML and Javascript, it worked, used the Google search API, and got me the job.
Then, at the end of 2016, I revisited Suggest-o-matic, re-adopting the SWIM name (and of course quickly whipping up a new logo), simply as something to play with. Really just a re-do of the original proof-of-concept, the Google search API was no longer free so it uses Bing for search results, and works moderately better the original. I also briefly played with Algolia search as a solution for storing/searching a personal corpus, but their pricing was prohibitive for such a small side-project. (It looks like they’ve adjusted their pricing, so now I’m tempted to try that road again.)
Aaand now Bing isn’t free either…