Google's Amazing New Information Tool, Notebook-LM
In the two or three years “AI” has been in our active lives, we’ve gone from thinking it would be the end of the human race to regarding it as yet another way to cheapen politics and the internet. Could there be a happy medium? Google has a new tool out that illustrates a use case halfway between those extremes and begins to hint at the possibilities AI could finally deliver. The product is called NotebookLM, and it’s sort of like a project manager, collaborator, and research assistant rolled up into one. Anyone who has large amounts of data they need to organize and analyze will find it useful. One of the creators Google tapped was writer Steven Johnson, so it also has a literary flavor, but its application is going to be far broader. Researchers working with qualitative data will love it, as will historians. Archivists? Oh my. Lawyers tired of thumbing through old cases, doctors looking for a diagnosis; NotebookLM could be an off-the-shelf solution for bespoke databases and libraries, searchable in minutes instead of weeks. (Breweries may find it of limited use—though even there I see the potential for it to help refine brewery processes and recipes.)
One of the more amazing features is a “podcast” the program will generate where two natural-sounding people discuss a topic. I’m not totally sure how useful that’s going to be for most folks, but as Patrick and I predicted on our final pod, it does suggest that at least in terms of fluid delivery and organic banter, AI is now doing a better job than 75% of the humans out there. I know this subject is a bit far afield from the regular beer talk on this site, but hey, you get what you pay for. Read on if you want to glimpse the future—and especially if you’d like to hear a 9:29 podcast review of The Beer Bible generated in about a minute. I have been impressed with NotebookLM and I think many folks out there might like taking it for a spin (for the moment, it’s free).
The innovation with NotebookLM is that you define your dataset. You’re not asking the program to scrape the internet for potentially bogus “facts.” Johnson, the writer, entered all his notes from 30 years of writing. You could dump a legal library in there. I loaded all my books and columns from All About Beer and Craft Beer & Brewing—maybe half a million words. It seems like many AI products perform a clever trick without providing any particularly useful work. With NotebookLM, you think about what you want before you load it with information.
As a writer, I was curious if the program could actually “understand” my writing and offer me advice. I started by asked it to analyze all the material and describe any explicit and implicit themes it found in my work. I think they exist, and I know how I’d answer this question, but before I trusted NotebooLM with other tasks, I wanted to see how much insight into the work it had. (I know, it doesn’t think or have insight, but those anthropomorphisms are useful in describing the way it feels to the user.) So, how’d it do?
Examining the provided excerpts and drawing upon our previous analysis of the author's writing style, several themes and subtle points emerge in Alworth's work:
Reverence for Tradition and History: A dominant theme in Alworth's writing is a deep respect for the historical and cultural context of beer and brewing traditions. He consistently emphasizes:
Evolution of Styles: Alworth frequently traces the development of beer styles across different countries and time periods, highlighting how factors like ingredient availability, technological advancements, and changing consumer tastes have shaped brewing practices. For instance, he discusses how the invention of pale malt in the 19th century revolutionized brewing and led to the creation of styles like Pilsner.
Similarly, he explores how the American interpretation of English brewing traditions ultimately resulted in the emergence of a distinct American brewing identity, characterized by styles like the IPA.
Importance of Place: Alworth underscores the significance of regionality in shaping beer styles. He argues that factors like local water sources, indigenous ingredients, and traditional brewing methods contribute to the unique character of beers from different places. This is evident in his discussion of the difficulty in replicating the flavor profile of Altbier outside of Düsseldorf, Germany, and in his assertion that to truly understand a beer style, one must understand the context in which it originated.
Appreciation for Craft and Process: Beyond simply describing beer styles, Alworth delves into the intricacies of brewing processes and techniques. He demonstrates a clear appreciation for:
The Art of Brewing: Alworth presents brewing as more than a mere industrial process; he frames it as a craft that requires skill, knowledge, and a deep understanding of ingredients and their interactions.
He highlights the role of intuition and experience in making subtle adjustments throughout the brewing process to achieve the desired flavor profile. This is evident in his discussion of Italian brewing, which he characterizes as being guided by a deep-rooted appreciation for balance and sensory experience, drawing parallels to the meticulous approach of Italian cuisine.
The "Invisible Hand" of Tradition: Alworth often refers to an "invisible hand" that guides brewing traditions. This concept encompasses the unspoken knowledge, shared practices, and accumulated wisdom that have been passed down through generations of brewers within a particular region or brewing culture. This notion emphasizes that replicating a beer style involves more than simply following a recipe; it requires an understanding of the subtle nuances and unspoken conventions that have shaped its development over time.
Beyond the explicit themes, Alworth subtly conveys several underlying points:
The Value of Experimentation: Alworth celebrates brewers who are willing to push boundaries and experiment with new ingredients, techniques, and flavor combinations. He sees innovation as essential to the continued evolution of beer and encourages both brewers and consumers to embrace new possibilities. This is apparent in his discussion of breweries like Dieu Du Ciel and their unconventional approach to brewing styles like Witbier. He lauds their willingness to deviate from established norms and create unique and flavorful beers.
The Importance of Community: Throughout the sources, Alworth highlights the collaborative spirit within the brewing community. He emphasizes the role of shared knowledge, mentorship, and collaboration in advancing the craft of brewing. This is evident in his account of the relationship between English and American brewers, tracing how the exchange of ideas and techniques has influenced the evolution of brewing traditions on both sides of the Atlantic.
The podcast”NotebookLM created based the text of The Beer Bible.
This is actually … remarkable. I don’t know what I expected, especially given the varied source material, but I think this captures the essence of what I’ve tried to communicate in my writing. I do have a point of view, and increasingly an agenda, and it starts with the culture of beer—or in the language of NotebookLM, its “history and tradition.” That is, in my books and pieces like the Style School column and “Making of a Classic” series, I’ve tried to illustrate how beers and styles of beer come to be through this amazing tangle of causes—and how they are “sticky” in that they continue on once they’re invented. They constitute a historical throughline, a tradition. And that’s the first thing it mentions.
I love that it captured both my love of tradition and the cræft of beer—but, equally, experimentation and evolution. They would appear to be contradictory impulses, and a human might have tried to reconcile them. Maybe that’s an advantage of AI—it doesn’t think, so it doesn’t worry about that stuff. I also love that it picked up on my love of the way beer has always been a function of community, something else I do highly prioritize but wouldn’t expect to see.
Once it had produced the above analysis—it took perhaps five minutes to “think”—I asked it some other stuff as well. I’ll spare you all those readouts, though the photo at the top shows that it can effectively critique writing style. This was equally as impressive to me. I’ve worked with a number of editors over the years, and this program made professional-level observations. I primed the question by calling out my parentheticals (they are many and possibly unnecessary), and it acknowledged their heavy use. It went beyond parroting this truth back to me, however, and actually provided more depth, giving categories to my use and more or less exonerating me. Unexpected! It was harder on my use of “But” to start sentences—something my book editors also dislike—and it discovered a tic I didn’t know I had by too often starting sentences with “This is….” (Ironically, NotebookLM used the “this is” sentence formulation several times in its response to me, whereas I have not used it once in this post, but I’m not bitter about this tendentious hypocrisy. Not at all.)
I am not the kind of writer who typically uses these kinds of programs. I have crude and inefficient ways of collecting and organizing my notes and research, and as a writer I am more improvisational than systematic. I’m not a luddite, but I’ve yet to find a program that actually streamlines my process given the mode of my work. Each person has a process. I don’t know that this program will make me work more efficiently, either, which is fine. It might act as a collaborator and editor, however, and that would be valuable indeed. I have already created a couple other “notebooks” in the program to see what help it might offer. Go give it a look.