Antonio G. Di Benedetto, reporting for The Verge:

Google is announcing a new line of laptops coming in the fall called Googlebooks. Details are sparse for now, as the tease is just a small part of various Android announcements during Google’s Android Show. But we do know this is a major new initiative in the laptop space for Google, seemingly designed to succeed Chromebooks with something more capable: a platform running a long-rumored new operating system based on a fusion of Android and ChromeOS…

So, what do we actually know about Googlebooks and their operating system that’s not Aluminium but also not not Aluminium? For starters, Googlebooks are built on the Android technology stack. They’ll run Chrome for web browsing and also run Android apps. They’ll even be able to directly access files from your Android phone and run your apps right off of it so you don’t have to temporarily move your attention across devices. And they’re going to have Gemini Intelligence baked into just about everything — right down to the cursor.

Googlebooks will have a Magic Pointer feature that offers contextual suggestions whenever you shake your cursor and point it at something on the screen. Google’s examples include setting up a meeting by pointing at a date in an email or selecting images of furniture and a living space to visualize them together. Beyond your mouse pointer, Googlebooks will also feature the custom AI-created widgets that Google is also debuting today for Android phones and Wear OS smartwatches.

“Googlebook” is indeed an asinine name, but I was half-expecting it to be called “Geminibook.” Nevertheless, the Googlebook platform is built with “Gemini Intelligence,” a suite of Gemini features that aim to do agentic work in apps people have installed on their devices. If this sounds familiar, it is. Apple announced the same feature, powered by App Intents, almost two years ago. It is yet to ship, of course, and Google has a track record — through Project Mariner and Gemini Agent — of shipping features similar to Gemini Intelligence. From Allison Johnson:

Task automation is apparently one of those “best of Gemini” features. It’s already on some recent Pixel and Samsung Galaxy phones, and it enables Gemini to use certain apps on your behalf. It’s been limited to a handful of rideshare and food delivery apps until now. That’s changing “soon,” says Google, when task automation will open up to a wider range of apps.

It will also add multimodality; previously, Gemini could only use voice or text prompts to inform its actions. Now you can throw a screengrab or a photo into the mix, which kind of seems like something you should have been able to do from the start. You’ll be able to give Gemini a screenshot of a grocery list in your notes app and it will add those items to your cart. You know, provided you have an Android phone that supports Gemini Intelligence.

The task automation feature is intriguing because, while Google probably wants to use the Model Context Protocol and application programming interfaces to control apps long-term, it has, in the short term, settled on a computer-use agent trained to tap around apps. Think of Codex’s computer use or the browser-use agent in ChatGPT Atlas, OpenAI’s web browser. This is also more or less how Project Mariner worked, mostly because the MCP servers tech firms are publishing — if they make them at all — are limited in scope. That’s why ChatGPT’s “Apps” feature has failed to reach velocity — it just isn’t useful. (I have said many times that computer-use agents and MCP-powered apps are highly inefficient, and I stand by this. APIs, or now command-line interfaces, will be the future of this kind of work. App Intents are just APIs.)

There is utility to an agent that can work in other apps, but it depends on the implementation. I don’t see the usefulness of an agent that can order DoorDash or Starbucks. Such implementations are useless because people usually prefer to choose what they want to eat. (I realize this might be surprising to the Silicon Valley types.) I also don’t think people are in any hurry to give an agent their credit card information so it can spend thousands of dollars on their behalf booking flights or hotels. As Brian Chesky, the chief executive of Airbnb, said on social media, chatbot-style agents are the wrong user interface for these choices. There is little wrong with an intake form to make granular choices.

Back to the Googlebook, a “platform” for Google’s next-generation, unnamed desktop operating system based on Android. The Chromebook platform is bespoke, built on Google Chrome, which has limited the variety of apps that can run on ChromeOS. The Googlebook is not a hardware laptop model — it is an operating system and a list of specifications for third-party hardware manufacturers. It will nominally succeed the Chromebook, but in practice, Chromebooks are mainly used by schools that use them because they’re based on Chrome. Years of Chromebook usage in elementary and middle schools across the United States have made this generation practically tech-illiterate. They don’t know what a file system is, they don’t know how applications work, and they abstain from using the keyboard and trackpad. It’s far too late to transition them to a “real” desktop operating system — it’s an operational nightmare for the Chromebook’s most popular demographic.

So that makes the Googlebook a competitor to the MacBook Neo, which is regrettably preposterous. The MacBook Neo’s singular strength is that it’s a great Mac. It runs full-fledged macOS with all the programs people love, and attention to detail and beauty. It’s Windows, but better. And if people aren’t interested in an alternate operating system, they’ll purchase a cheap Windows laptop with no learning curve and all of the programs they’re used to. The market for the Googlebook is unclear. I don’t think anyone loves Gemini enough to switch their entire computing life to a blown-up version of Android just to shake their cursor and have Gemini summarize whatever is on screen. I can see why Google wants to make a desktop OS — I support the endeavor! — but that ship has, unfortunately, sailed long ago.

I said earlier on Tuesday that the Android Show this year was exciting, and I maintain my enthusiasm. There’s a lot of new stuff here, especially Gemini Intelligence. But, much like all of Google’s products, I’m afraid they won’t have much staying power.