Books, Generative AI
Meet the Author: The Complete Obsolete Guide to Generative AI
The complete interview with David Clinton, the author of The Complete Obsolete Guide to Generative AI, published by Manning Publications
Today's special guest is David Clinton, the author of The Complete Obsolete Guide to Generative AI, published by Manning Publications.
Before we get into the details of the interview, I would like to share some background with you. After preparing the questions, I sent a LinkedIn message to David, who kindly said he would answer the questions the following day. And so it was. No sooner said than done. After less than 24 hours, each of my eight questions had an answer. And what a response!
David sent me a message on LinkedIn letting me know that he had answered the questions. I rushed to read the answers. I read them all in one go. I won’t tell you the immense joy I felt. David Clinton had answered my questions about the book! A wonderful thing is to be able to speak directly with the author of a book!
The thing that surprised me the most was the style of the answers! By reading them, you will realize Clinton’s unmistakable style, which can be seen in all the pages of his book. It gave me the impression of reading another chapter of his book…in fact, we could ask Manning (the book’s publishing house) to add this interview as an appendix to the book because it is truly something exceptional! Like the whole book, after all!
But I won’t keep you on tenterhooks any longer. Here is the full interview. And a big thank you to David Clinton for his availability and speed in response!
1 Dear David, please introduce yourself briefly by highlighting your skills and interests.
I guess I spend most of my time creating books and course content that teach skills like Linux and AWS administration, IT security, and containers. But data analytics has been playing an increasingly dominant role in whatever else I’m doing. And these days, you can’t really separate data from generative AI. Well, I guess you could separate the two if you wanted, but you’d be giving up your competitive edge, so what’s the point?
2 What motivated you to write a book about an “obsolete guide” to generative AI?
The “obsolete” gag was a result of conversations I’d been having with Manning over many months. They kept pushing me to write a book about AI and I kept telling them there was no point: things change so fast in the industry that whatever I could possibly write would be obsolete long before the book hit the stores. Proof in point: there was a moment a month or so back when we weren’t sure the OpenAI company was even going to survive. How would we explain that one?
Manning refused to give up and, eventually, I settled on writing the book, but lodging a tongue-in-cheek formal protest on the front cover. But it’s really our way of communicating to readers that, indeed, a lot of what they’ll see will have changed, but it’ll all be handled in a lighthearted way. And, by the way, even the stuff that has gone stale is still useful, because it teaches the core principles.
3 What challenges did you encounter while exploring and writing about this complex topic, and how did you address them?
The biggest problem really was how fast things were — and still are — moving. Perhaps 50% of the code examples from the book can now be performed through the ChatGPT Plus UI (if you pay for a subscription). And at least 25% of the code itself from my first draft no longer runs and had to be updated. But, since a lot of the real value of gen AI is in how effectively you can incorporate it into your code infrastructure, I’d say that everything in the book is absolutely still valuable. By the way, “still valuable” is a crazy thing to say about a book considering that it isn’t even available on Amazon yet!
4 Compared to when you wrote the book, do you think anything has changed in the generative AI landscape?
Well, of course, everything has changed. But I’d say that there are two things that really stand out:
- The subscription-only ChatGPT Plus has set an insanely-high standard for genAI. Not only is it multi-modal — meaning you can feed it your documents, spreadsheets, and images — but its effectiveness and accuracy are way better than anything else I’ve seen. I just gave it a large, complex spreadsheet with around 40 columns and 200,000 rows of data and, in seconds, I got credible analytics out of it. I checked it out myself using normal Python analytics code, but GPT nailed it.
- Google is VERY good at making product announcements that show amazing AI applications. Unfortunately, they’re not quite so good at delivering products that are actually available.
5 What takeaways do you hope readers will gain after finishing your book?
The biggest thing readers can learn is that they’ll only get really good at using these tools by diving in, making mistakes, and working through code that doesn’t work. All the books, courses, and documentation in the world won’t accomplish all that much unless they’re followed up with real projects. But I guess that’s the main takeaway of just about everything I’ve taught over the years.
6 Reading the book, there is a certain optimism towards generative AI. In your opinion, is an apocalyptic scenario in which Generative AI takes over humans totally to be ruled out?
Knowing how things might work out is way beyond my pay scale. I can see what worries people, but I have no idea how likely those scenarios are. But one thing I can say with confidence: it’s far too late to do anything about it. Even if governments understood AI (which they don’t) and even if they were capable of coherent and focused policy implementation (which, by and large, they’re not), there’s no way they could ever regulate AI. It’s out in the wild now. You can train a new LLM on a smartphone.
At this point, it’s like the subtitle from the Dr. Strangelove movie from the 60s: How I learned to Stop Worrying and Love the Bomb” (which I borrowed for one of the chapter titles in my book). We might as well just enjoy the ride, even if we don’t know where it will end.
7 Could you suggest other resources (books, courses, articles, etc.) the readers can consult after reading your book?
That’ll depend on what you’re after. This is a bit of a big-picture, introduction book. But Manning really has been going full steam into the gen AI market and they’ve got some excellent new books that go deep on all kinds of topics, including building your own LLM, software testing, and coding. Check those out.
8 As an author, what was the most intriguing or surprising discovery you made while researching and writing “The Complete Obsolete Guide to Generative AI”?
Well I was forced to dig much deeper into the OpenAI API than I had previously. Even though the ecosystem is still a bit of a mess from a “best-practices” perspective, there’s still so much you can do and so many ways you can integrate AI into your own work environment. And imagination: there really is no limit to the range of what you can do with gen AI!
Conclusions
I believe that speaking directly with the author of a book is the greatest aspiration of every reader! And I had the honor of speaking with David Clinton, the author of the book The Complete Obsolete Guide to Generative AI.
A big thank you to David and to you who had the patience to get to the bottom of this article!
Until next time!
In case you missed my review of David Clinton’s book, here is the link:
