Lessons for AI prompting and retrieval

I started writing this early last week but Real Life Stuff happened and now you're getting the first-draft late this week. Warning, unedited thoughts ahead! New Logic for Programmers release! v0.9 is out! This is a big release, with a new cover design, several rewritten chapters, online code samples and much more. See the full release notes at the changelog page, and get the book here! Write the cleverest code you possibly can There are millions of articles online about how programmers should not write "clever" code, and instead write simple, maintainable code that everybody understands. Sometimes the example of "clever" code looks like this (src): # Python p=n=1 exec("p*=n*n;n+=1;"*~-int(input())) print(p%n) This is code-golfing, the sport of writing the most concise code possible. Obviously you shouldn't run this in production for the same reason you shouldn't eat dinner off a Rembrandt. Other times the example looks like this: def is_prime(x): if x == 1: return True return all([x%n != 0 for n in range(2, x)] This is "clever" because it uses a single list comprehension, as opposed to a "simple" for loop. Yes, "list comprehensions are too clever" is something I've read in one of these articles. I've also talked to people who think that datatypes besides lists and hashmaps are too clever to use, that most optimizations are too clever to bother with, and even that functions and classes are too clever and code should be a linear script.1. Clever code is anything using features or domain concepts we don't understand. Something that seems unbearably clever to me might be utterly mundane for you, and vice versa. How do we make something utterly mundane? By using it and working at the boundaries of our skills. Almost everything I'm "good at" comes from banging my head against it more than is healthy. That suggests a really good reason to write clever code: it's an excellent form of purposeful practice. Writing clever code forces us to code outside of our comfort zone, developing our skills as software engineers. Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you [will get excellent debugging practice at exactly the right level required to push your skills as a software engineer] — Brian Kernighan, probably There are other benefits, too, but first let's kill the elephant in the room:2 Don't commit clever code I am proposing writing clever code as a means of practice. Being at work is a job with coworkers who will not appreciate if your code is too clever. Similarly, don't use too many innovative technologies. Don't put anything in production you are uncomfortable with. We can still responsibly write clever code at work, though: Solve a problem in both a simple and a clever way, and then only commit the simple way. This works well for small scale problems where trying the "clever way" only takes a few minutes. Write our personal tools cleverly. I'm a big believer of the idea that most programmers would benefit from writing more scripts and support code customized to their particular work environment. This is a great place to practice new techniques, languages, etc. If clever code is absolutely the best way to solve a problem, then commit it with extensive documentation explaining how it works and why it's preferable to simpler solutions. Bonus: this potentially helps the whole team upskill. Writing clever code... ...teaches simple solutions Usually, code that's called too clever composes several powerful features together — the "not a single list comprehension or function" people are the exception. Josh Comeau's "don't write clever code" article gives this example of "too clever": const extractDataFromResponse = (response) => { const [Component, props] = response; const resultsEntries = Object.entries({ Component, props }); const assignIfValueTruthy = (o, [k, v]) => (v ? { ...o, [k]: v } : o ); return resultsEntries.reduce(assignIfValueTruthy, {}); } What makes this "clever"? I count eight language features composed together: entries, argument unpacking, implicit objects, splats, ternaries, higher-order functions, and reductions. Would code that used only one or two of these features still be "clever"? I don't think so. These features exist for a reason, and oftentimes they make code simpler than not using them. We can, of course, learn these features one at a time. Writing the clever version (but not committing it) gives us practice with all eight at once and also with how they compose together. That knowledge comes in handy when we want to apply a single one of the ideas. I've recently had to do a bit of pandas for a project. Whenever I have to do a new analysis, I try to write it as a single chain of transformations, and then as a more balanced set of updates. ...helps us master concepts Even if the composite parts of a "clever" solution aren't by themselves useful, it still makes us better at the overall language, and that's inherently valuable. A few years ago I wrote Crimes with Python's Pattern Matching. It involves writing horrible code like this: from abc import ABC class NotIterable(ABC): @classmethod def __subclasshook__(cls, C): return not hasattr(C, "__iter__") def f(x): match x: case NotIterable(): print(f"{x} is not iterable") case _: print(f"{x} is iterable") if __name__ == "__main__": f(10) f("string") f([1, 2, 3]) This composes Python match statements, which are broadly useful, and abstract base classes, which are incredibly niche. But even if I never use ABCs in real production code, it helped me understand Python's match semantics and Method Resolution Order better. ...prepares us for necessity Sometimes the clever way is the only way. Maybe we need something faster than the simplest solution. Maybe we are working with constrained tools or frameworks that demand cleverness. Peter Norvig argued that design patterns compensate for missing language features. I'd argue that cleverness is another means of compensating: if our tools don't have an easy way to do something, we need to find a clever way. You see this a lot in formal methods like TLA+. Need to check a hyperproperty? Cast your state space to a directed graph. Need to compose ten specifications together? Combine refinements with state machines. Most difficult problems have a "clever" solution. The real problem is that clever solutions have a skill floor. If normal use of the tool is at difficult 3 out of 10, then basic clever solutions are at 5 out of 10, and it's hard to jump those two steps in the moment you need the cleverness. But if you've practiced with writing overly clever code, you're used to working at a 7 out of 10 level in short bursts, and then you can "drop down" to 5/10. I don't know if that makes too much sense, but I see it happen a lot in practice. ...builds comradery On a few occasions, after getting a pull request merged, I pulled the reviewer over and said "check out this horrible way of doing the same thing". I find that as long as people know they're not going to be subjected to a clever solution in production, they enjoy seeing it! Next week's newsletter will probably also be late, after that we should be back to a regular schedule for the rest of the summer. Mostly grad students outside of CS who have to write scripts to do research. And in more than one data scientist. I think it's correlated with using Jupyter. ↩ If I don't put this at the beginning, I'll get a bajillion responses like "your team will hate you" ↩

Whether we like it or not, email is widely used to identify a person. Code sent to email is used as authentication and sometimes as authorisation for certain actions. I’m not comfortable with Google having such power over me, especially given the fact that they practically don’t have any support you can appeal to. If your Google account is blocked, that’s it. Maybe you know someone from Google and they can help you, but for most of us mortals that’s not an option.

In his book “The Order of Time” Carlo Rovelli notes how we often asks ourselves questions about the fundamental nature of reality such as “What is real?” and “What exists?” But those are bad questions he says. Why? the adjective “real” is ambiguous; it has a thousand meanings. The verb “to exist” has even more. To the question “Does a puppet whose nose grows when he lies exist?” it is possible to reply: “Of course he exists! It’s Pinocchio!”; or: “No, it doesn’t, he’s only part of a fantasy dreamed up by Collodi.” Both answers are correct, because they are using different meanings of the verb “to exist.” He notes how Pinocchio “exists” and is “real” in terms of a literary character, but not so far as any official Italian registry office is concerned. To ask oneself in general “what exists” or “what is real” means only to ask how you would like to use a verb and an adjective. It’s a grammatical question, not a question about nature. The point he goes on to make is that our language has to evolve and adapt with our knowledge. Our grammar developed from our limited experience, before we know what we know now and before we became aware of how imprecise it was in describing the richness of the natural world. Rovelli gives an example of this from a text of antiquity which uses confusing grammar to get at the idea of the Earth having a spherical shape: For those standing below, things above are below, while things below are above, and this is the case around the entire earth. On its face, that is a very confusing sentence full of contradictions. But the idea in there is profound: the Earth is round and direction is relative to the observer. Here’s Rovelli: How is it possible that “things above are below, while things below are above"? It makes no sense…But if we reread it bearing in mind the shape and the physics of the Earth, the phrase becomes clear: its author is saying that for those who live at the Antipodes (in Australia), the direction “upward” is the same as “downward” for those who are in Europe. He is saying, that is, that the direction “above” changes from one place to another on the Earth. He means that what is above with respect to Sydney is below with respect to us. The author of this text, written two thousand years ago, is struggling to adapt his language and his intuition to a new discovery: the fact that the Earth is a sphere, and that “up” and “down” have a meaning that changes between here and there. The terms do not have, as previously thought, a single and universal meaning. So language needs innovation as much as any technological or scientific achievement. Otherwise we find ourselves arguing over questions of deep import in a way that ultimately amounts to merely a question of grammar. Email · Mastodon · Bluesky

In mid-March we released a big bug fix update—elementary OS 8.0.1—and since then we’ve been hard at work on even more bug fixes and some new exciting features that I’m excited to share with you today! Read ahead to find out what we’ve released recently and what you can help us test in Early Access. Quick Settings Quick Settings has a new “Prevent Sleep” toggle Leo added a new “Prevent Sleep” toggle. This is useful when you’re giving a presentation or have a long-running background task where you want to temporarily avoid letting the computer go to sleep on its normal schedule. We also fixed a bug where the “Dark Mode” toggle would cancel the dark mode schedule when used. We now have proper schedule snoozing, so when you manually toggle Dark Mode on or off while using a timed or sunset-to-sunrise schedule, your schedule will resume on the next schedule change instead of being canceled completely. Vishal also fixed an issue that caused some apps to report being improperly closed on system shutdown or restart and on the lock screen we now show the “Suspend” button rather than the “Lock” button. System Settings Locale settings has a fresh layout thanks to Alain with its options aligned more cleanly and improved links to additional settings. Locale Settings has a more responsive design We’ve also added the phrase “about this device” as a search term for the System page and improved interface copy when a restart is required to finish installing updates based on your feedback. Plus, Stanisław improved stylus detection in Wacom settings preventing a crash when no stylus is found. AppCenter We now show a small label next to the download button for apps which contain in-app purchases. This is especially useful for easily identifying free-to-play games or alt stores like Steam or Heroic Games Launcher. AppCenter now shows when apps have in-app purchases Plus, we now reload app icons on-the-fly as their data is processed, thanks to Italo. That means you’ll no longer get occasionally stuck with an AppCenter which shows missing images for app’s who have taken a bit longer than usual to load. Get These Updates As always, pop open System Settings → System on elementary OS 8 and hit “Update All” to get these updates plus your regular security, bug fix, and translation updates. Or set up automatic updates and get a notification when updates are ready to install! Early Access Our development focus recently has been on some of the bigger features that will likely land for either elementary OS 8.1 or 9. We’ve got a new app, big changes to the design of our desktop itself, a whole lot of under-the-hood cleanup, and the return of some key system services thanks to a new open source project. Monitor We’re now shipping a System Monitor app by default By popular demand—and thanks to the hard work of Stanisław—we have a new system monitor app called “Monitor” shipping in Early Access. Monitor provides usage information for your processor, GPU, memory, storage, network, and currently running processes. You can optionally see system information in the panel with Monitor You can also optionally get a ton of glanceable information shown in the panel. There’s currently a lot of work happening to port Monitor to GTK4 and improve its functionality under the Secure Session, so make sure to report any issues you find! Multitasking The Dock is getting a workspace switcher Probably the biggest change to the Pantheon shell since its early inception, the Dock is getting a new workspace switcher! The workspace switcher works in a familiar way to the one you may have seen in the Multitasking View: Your currently open workspaces are represented as tiles with the icons of apps running on them; You can select a workspace to switch to it; You can drag-and-drop workspaces to rearrange them; And you can use the “+” button to create a new blank workspace. One new trick however is that selecting the workspace you’re already on will launch Multitasking View. The new workspace switcher makes it so much more accessible to multitask with just the mouse and get an overview of your workflows without having to first enter the Multitasking View. We’re really excited to hear what people think about it! You can close apps from Multitasking View by swiping up Another very satisfying feature for folks using touch input, you can now swipe up windows in the Multitasking View to close them. This is a really familiar gesture for those of us with Android and iOS devices and feels really natural for managing a big stack of windows without having to aim for a small “x” button. GTK4 Porting We’ve recently landed the port of Tasks to GTK4. So far that comes with a few fixes to tighten up its design, with much more possible in the future. Please make sure to help us test it thoroughly for any regressions! Tasks has a slightly tightened up design We’re also making great progress on porting the panel to GTK4. So far we have branches in review for Nightlight, Bluetooth, Datetime, and Network indicators. Power, Keyboard, and Quick Settings indicators all have in-progress branches. That leaves just Applications, Sound, and Notifications. So far these ports don’t come with major feature changes, but they do involve lots of cleaning up and modernizing of these code bases and in some cases fixing bugs! When the port is finished, we should see immediate performance gains and we’ll have a much better foundation for future releases. You can follow along with our progress porting everything to GTK4 in this GitHub Project. And More When you take a screenshot using keyboard shortcuts or by secondary-clicking an app’s window handle, we now send a notification letting you know that it was succesful and where to find the resulting image. Plus there’s a handy button that opens Files with your screenshot pre-selected. We’re also testing beaconDB as a replacement for Mozilla Location Services (MLS). If you’re not aware, we relied on MLS in previous versions of elementary OS to provide location information for devices that don’t have a GPS radio. Unfortunately Mozilla discontinued the service last June and we’ve been left without a replacement until now. Without these services, not only did maps and weather apps cease to function, but system features like automatic timezone detection and features that rely on sunset and sunrise times no longer work properly. beaconDB offers a drop-in replacement for MLS that uses Wireless networks, bluetooth devices, and cell towers to provide location data when requested. All of its data is crowd-sourced and opt-in and several distributions are now defaulting to using it as their location services data provider. I’ve set up a small sponsorship from elementary on Liberapay to support the project. If you can help support beaconDB either by sponsoring or providing stumbler data, I’d highly encourage you to do so! Sponsors At the moment we’re at 23% of our monthly funding goal and 336 Sponsors on GitHub! Shoutouts to everyone helping us reach our goals here. Your monthly sponsorship funds development and makes sure we have the resources we need to give you the best version of elementary OS we can! Monthly release candidate builds and daily Early Access builds are available to GitHub Sponsors from any tier! Beware that Early Access builds are not considered stable and you will encounter fresh issues when you run them. We’d really appreciate reporting any problems you encounter with the Feedback app or directly on GitHub.