• It has been a while since my last post, but as always, there are reasons (or excuses?) behind that. 1) The weather got better and I prioritized doing some chores outside over spending evenings looking at a PC screen struggling to write some meaningful sentences. 2) I had a three-week-long vacation (and it has been a while since I had a vacation that long). And, that time, I was pretty busy working outside in the garden, visiting Turkey (long story short - first time flying by plane with kids, my phone stopped working on 2nd day there, therefore full attention went to the combination of sun, sand, and water, achieving high cocktail test coverage, and my Kindle, so I re-read “Start with Why” and probably found some answers there - best trip ever!), and helping our relatives to move house.
    Talking about vacation, there is a recent attempt to gaslight millennials by accusing them of “quiet vacationing”. IMHO, it is yet another flag (similar to or even a part of WFH debates) that some companies do not have a clue of how to deal with productivity topics (constructively) and lean towards a command-and-control/micromanagement style (destructively). Well done.

  • All that OpenAI ethicists drama happened this March (and I like “Doomers have lost the AI fight” article summarizing the key events and key takeaways briefly). IMHO, key sentence from the article is this one - “Altman has always talked about responsibility and caution, but under his direction, OpenAI continues to floor the pedal.” In other words, as often happens, there is a gap between messages being communicated and actual actions being taken, and “actions speak louder than words”.

  • Continuing on AI and LLMs, but more from testing perspective - I accidentally found this interesting open access scientific paper - “Testing theory of mind in large language models and humans”. It is not a short read, but considering the topic - how to test for mind? - it is pretty concise, descriptive and clear. It was interesting to read through the structure of Theory of Mind (hinting, false belief, faux pas, strange stories, irony). Typically, when I read such experiments that were did with humans or human mind, I think of how I would have done under such tests myself (especially, when I already recognized myself struggling with catching some hints or irony and that leading into awkward situations, but oh well…) And looks like ChatGPT is excelling humans in all but one area of Theory of Mind (phew, I am not the only one lacking). However, there is still hope - Michael Bolton and James Bach keep updating the list of known LLM issues, a.k.a. Large Language Model Syndromes.

  • Deming. This is a name you should know if you take testing and quality topics seriously. It is a shame that after spending close to 10 years in IT I found Deming’s work only recently. I noticed articles like this appearing more and more frequently in the newsletters I follow, but I probably did not understand the significance until the recent Quality Bits podcast episode with John Willis. Long story short, many “modern” software development ideas are not so modern as they can be traced back to the work of William Edwards Deming from the 1950s-1980s. Therefore, soon you will find me reading “Deming’s Journey to Profound Knowledge” (as well as “Thinking in Systems” as System Thinking was also something I heard a lot yet did not have a chance to dig deeper).

  • Finally, a less formal recommendation - Netflix documentary about Ashley Madison hack. I wanted to write “ethics and morals aside”, but I simply cannot as there are numerous stories that make your eyebrows go up. As always, you can argue that storytelling can be more condensed and they can focus more on other things, but this time the balance between business and customers stories felt right delivering a nice bigger picture. The picture of sh*t hitting the fan when the digital product does not deliver its main promise - keeping a secret.