I’m an early adopter. I was a wiz with Alta Vista (a predecessor to Google) back in the day. Actually, a full decade before that I was using one of the first networks that aggregated scientific papers called Dialog Information Services. It costs something like $100 an hour to connect and 25 cents per citation. I could spend a few hundred bucks in no time but I’d walk away with several relevant citations on my research topic, something that might take me days to do in a library.
This was the perfect ground to hone my searching skills because if I wasn’t very specific with my queries it would cost me a lot of money!
Fast forward to today. Even though many of the practices around searching and prompting are different, many of the principles are similar. They both involve set theory, logic, and the use of abstraction. In both cases, it is about clarifying, refining, and drawing out relevant information.
Like everything, prompting is a skill, and doing it well takes practice. We want to learn from others and also learn by doing. I also like to have references. For years I kept a dictionary on my desk. I like to look stuff up. But I found no references for Prompt Engineering.
OpenAI has excellent developer documentation but I can’t find any user documentation. I think the assumption is that we’ll just figure it out and for primary information access that’s true. However, ChatGPT is more than just a different kind of search tool. It represents a different order of information available to us that can inform and enrich us in ways we are just beginning to recognize.
There are dozens of different kinds of prompts, each of which is useful in different kinds of situations. So I have been setting out to document them, and I discovered many powerful ways of using ChatGPT in the process.
This is ongoing. Recently, I added fifty more pages of detailed examples to the chapter on Advanced Prompts in my book. I believe that Prompt Engineering for Everyone is the most complete resource for prompting ChatGPT at the moment. And it’s only getting better.
We’ve just scratched the surface of what ChatGPT can do for us. Knowing a variety of techniques for prompting can help you get more from ChatGPT and other LLMs. I’m excited to learn more and share what I discover.
Note: I will continue to update Prompt Engineering for Everyone. If you have the Kindle version you can get updates for the current edition as they come out. If you prefer paperback then I have no way to send you updates unless you sign up for the bonuses (instructions in the book) and then I will also send you free PDF updates for the current edition as they come out. Here’s a link for more information about the book – https://passprog.com/pefe/https://passprog.com/pefe/.
