In late November I put together a homebrew RSS reader to address some of my challenges in trying to keep up with thousands of RSS feeds. I knew at the time that I wasn’t going to immediately get a sense of how well it worked because we were heading into the December holidays. I’d have to wait until at least mid-January.

Well, now it’s mid-January and the verdict is in: this RSS feed reader works really well. Really really well. As in, I’m going to have to get more strict about what ends up in the newsletter.

I don’t normally add an article to ResearchBuzz as soon as I find it. Each ResearchBuzz newsletter has several sections and, if I added each item in the order I came across it, I’d have several source duplicates and I might overload a single section (and the only section I will allow for overload is New Resources.) Instead, I queue the resources into a Google Sheet via IFTTT and Raindrop.io. Each article has its own row, including a color-coded cell to let me know what section the article belongs in. This enables me to get a sense of my queued articles at a glance and alerts me if I have too many or too few articles for a certain section.

It also lets me know if I’m going to have enough articles in my queue to fill the 22 slots in the two daily newsletters. Unfortunately, this new RSS feed reader lets me go through my content so fast that my queue is already ridiculous — 73 items. I turned the queue on its side and zoomed out so everything’s visible at once — it’s pretty but I’m going to have to do some weeding!

73 items is over three days’ worth of content and that is too much. My task now is to decide what the largest acceptable queue size is or what the oldest acceptable news date is (I try not to keep news older than a week.) Once I figure that out, the last thing on my daily ResearchBuzz newsletter is to take the spillover items and index them in ResearchBuzz Firehose. It shouldn’t take long; it’s going through my RSS feeds / alerts and reviewing the articles that’s time-consuming.