If there's one thing that I've learned from putting stuff on the Internet it's that I'm really not that great at predicting what people are going to find interesting to seek out and share. I hope that people reading things here do find what I'm writing about to be interesting or useful and that they'll hit the sharing buttons at the bottom of posts or otherwise share links with people who might be interested in this stuff. Now that it's been up a little over a month and the server logs have been rotated, I thought it might be interesting to look at what people have been looking at.

5. Learning to Roast Coffee at Home

The fifth most popular post for August was written at the end of July, but a lot of people read it in August. One of my goals with this blog is to reduce the volume of email that I get from random people asking questions so I'm using those messages as writing prompts. In this case, it was a home roaster who wanted me to teach him how to roast coffee. While I'm glad to take on consulting jobs to train roasters, I'm too expensive for that to make sense for a home roaster. That doesn't mean I can't offer some general advice on figuring out how to use a new roaster and provide some basic guided encouragement on experiments that will help a home roaster figure out what they like in terms of origin and roasting style.

4. The Pre-Roast

The fourth most popular post is another case of email as writing prompt and this one gave me a chance to reminisce about the history of Typica's roast profile translation feature. If you've looked at other frameworks for understanding roasting data and gotten confused by special casing related to batch size, you might find that understanding the concepts here leads to an approach to analyzing roasting data that is perhaps not as concise as others, but more elegant and more universal across different machines and roasting styles.

3. What I'm Doing at Roasters Guild Retreat

This is a post that I put together in advance of the 16th annual Roasters Guild Retreat, partly to explain that I'm wasn't going to be answering emails promptly while that was happening, but also to let people know about what sorts of things I planned to do there. The post-event recap was my least popular post, but also the one that was online for the shortest amount of time.

2. Modbus Simulation on Linux

This post was surprisingly popular. In it I describe how I used tty0tty, xrtx, and ModbusPal to test some code that I wrote to extend Typica's support for devices that communicate using the Modbus RTU protocol. This post shot up to the top for popularity immediately and stayed there for most of the month. There were a couple of follow up posts related to a pre-release build incorporating those code modifications that's currently available on the downloads page that came in at 7th and 9th.

1. Development Time Ratio Considered Harmful

While this post did end up as most read in August, it was rather slow in getting to the top, helped by a link getting shared on Home-Barista.com. Without that, this probably would have ended up closer to 4th. It's about how some people want to use this concept in ways that really don't work, why I'm not making changes to Typica to support an approach to quality assurance that's demonstrably misguided, and some better ideas that you can use now.

What's Next?

I expect September to be a busy month. I'm giving some talks for private groups and working on a new video series tentatively called "Evaluating Coffee with Purpose". I'll also be continuing work to update and expand Typica's documentation and get that uploaded to this site and continuing to work on the next release of Typica and some related software. I'm sure I'll find some things to write about here.