We see the last-minute tickets dripping in. Curious to see if any extra pull the trigger after ElixirConf in Chicago as well. I do have a bit of FOMO for that one but I am not hopping over that pond unless someone pays me at this moment. I heard the Nerves workshop went really well.

I saw the last-minute ticket pushes from ElixirConf and maybe I haven't underlined that we try to have a reasonable ticket price. But for the price of an ElixirConf virtual ticket you get a 3 day physical ticket at Goatmire.

There is no virtual ticket. Because the point is to meet. We'll just upload the videos as soon as we have them ready.

join us

Signal, gain and noise

As someone with a bit of notoriety in my small pond people are under the impression that I'm drowning in communication, feedback and response. This is never the case at this tiny scale. A million subscriber youtuber is typically drowning. I am not.

If you've ever worked with radios, cameras or sound you know about signal, noise and gain (or camera ISO I suppose). When the signal we are trying to get is weak we need to crank up the gain to even hear it. And it becomes hard to differentiate between signal and background noise. When we get frequent and relatively loud signal you get a much better idea of what the noise floor is.

Anyone running a podcast has very little feedback to work from. The medium doesn't contain a low-friction feedback mechanism. Blogs are quite similar. If you have a comment section you can get a bit more but the odds are still not in your favor.

I just presented Goatmire to a class of students who are studying to work on events. I think I did an okay job, not my best, not my worst. I got exactly one piece of response. One blessed young woman asked a question and we had a small follow-up exchange. They are a new class and all of them quite nervous and unsure about the whole thing. So that plays into it. But it has been very similar when I talk to the programming classes as a guest lecturer. 1-2 people interact.

After a lecture like that you have two options for how to interpret the data. You have a few spikes in the form of the questions you received. You have the level of engagement, attention, facial expressions of the group .. to whatever extent you manage to ingest that. That's a weak signal. If I get some laughs, some smiles, some nods it helps to calibrate against the noise floor. If I just cranked the gain and tried to sus out how I did against their carefully neutral faces and patient listening I'd feel like I bombed. It isn't signal. And with a 5 min spiel I won't necessarily succeed in unlocking a lively exchange.

Most of the time the signal for when you do well and poorly are a few clear responses, if anything at all. For publishing things, videos, writing, etc, I tend to lean on peer review. I check if the response is the expected one when I bounce it against a known topology (a friend). I have a sense of how they'd typically respond. Usually quite positive, so I mostly care about if the important bits made an impression. If I want a sober critique I bounce it on some germans. Even when biased by friendship they offer analysis and suggestions, I think it may be their love language. If I want to know about clarity I have some good people who frequently misunderstand me that I can try it on.

Overindexing on the popular response I think is often a mistake. Trust your sensors if you know they work decently well. Calibrate against known-good people and compensate for their typical bias. And be prepared for the silence, the lack of signal. If I relied on the response I get for these things there are many groups I would never bother trying to reach. And the stuff I do would be lesser for it. There is no easy answer, you just put yourself out in front of people repeatedly if you believe it serves a purpose. You might be able to measure the value of things repeated over years. Impact is rarely immediate.

Thank you for reading. I appreciate it.