Quotes

Part of the concept of ‘Disruption’ is that important new technologies tend to be bad at the things that matter to the previous generation of technology, but they do something else important instead. Asking if an LLM can do very specific and precise information retrieval might be like asking if an Apple II can match the uptime of a mainframe, or asking if you can build Photoshop inside Netscape. No, they can’t really do that, but that’s not the point and doesn’t mean they’re useless. They do something else, and that ‘something else’ matters more and pulls in all of the investment, innovation and company creation. Maybe, 20 years later, they can do the old thing too - maybe you can run a bank on PCs and build graphics software in a browser, eventually - but that’s not what matters at the beginning. They unlock something else. What is that ‘something else’ for generative AI, though? How do you think conceptually about places where that error rate is a feature, not a bug?

It is my judgment in these things that when you see something that is technically sweet, you go ahead and do it and you argue about what to do about it only after you have had your technical success. That is the way it was with the atomic bomb. I do not think anybody opposed making it; there were some debates about what to do with it after it was made. I cannot very well imagine if we had known in late 1949 what we got to know by early 1951 that the tone of our report would have been the same.

On two occasions I have been asked, — "Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out ?" In one case a member of the Upper, and in the other a member of the Lower, House put this question. I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.

The landlady hurried into the backyard, put the mousetrap on the ground (it was an old-fashioned trap, a cage with a trapdoor) and called to her daughter to fetch the cat. The mouse in the trap seemed to understand the gist of these proceedings; he raced frantically in his cage, threw himself violently against the bars, now on this side and then on the other, and in the last moment he succeeded in squeezing himself through and disappeared in the neighbor's field. There must have been on that side one slightly wider opening between the bars of the mousetrap...I silently congratulated the mouse. He solved a great problem, and gave a great example.

Questions for developers: - “What’s the one area you’re afraid to touch?” - “When’s the last time you deployed on a Friday?” - “What broke in production in the last 90 days that wasn’t caught by tests?” Questions for the CTO/EM: - “What feature has been blocked for over a year?” - “Do you have real-time error visibility right now?” - “What was the last feature that took significantly longer than estimated?” Questions for business stakeholders: - “Are there features that got quietly turned off and never came back?” - “Are there things you’ve stopped promising customers?”

It's okay to complain and vent, I just ask you be able to back it up. Saying, "Python packaging sucks", but then admit you actually haven't used it in so long you don't remember why it sucked isn't fair. Things do improve, so it's better to say "it did suck" and acknowledge you might be out-of-date.

I plan to introduce hard Rust dependencies and Rust code into APT, no earlier than May 2026. This extends at first to the Rust compiler and standard library, and the Sequoia ecosystem. In particular, our code to parse .deb, .ar, .tar, and the HTTP signature verification code would strongly benefit from memory safe languages and a stronger approach to unit testing. If you maintain a port without a working Rust toolchain, please ensure it has one within the next 6 months, or sunset the port.

15. Everything should be built top-down, except the first time.…\&30. In programming, everything we do is a special case of something more general—and often we know it too quickly.…\&31. Simplicity does not precede complexity, but follows it.…\&58. Fools ignore complexity. Pragmatists suffer it. Some can avoid it. Geniuses remove it.…\&65. Make no mistake about it: Computers process numbers—not symbols. We measure our understanding (and control) by the extent to which we can arithmetize an activity.…\&56. Software is under a constant tension. Being symbolic it is arbitrarily perfectible; but also it is arbitrarily changeable.

Web apps are typically continuously delivered, not rolled back, and you don't have to support multiple versions of the software running in the wild. This is not the class of software that I had in mind when I wrote the blog post 10 years ago. If your team is doing continuous delivery of software, I would suggest to adopt a much simpler workflow (like GitHub flow) instead of trying to shoehorn git-flow into your team.

Loveliest of trees, the cherry now Is hung with bloom along the bough, And stands about the woodland ride Wearing white for Eastertide. Now, of my threescore years and ten, Twenty will not come again, And take from seventy springs a score, It only leaves me fifty more. And since to look at things in bloom Fifty springs are little room, About the woodlands I will go To see the cherry hung with snow.

At first, I struggled to understand why anyone would want to write this way. My dialogue with ChatGPT was frustratingly meandering, as though I were excavating an essay instead of crafting one. But, when I thought about the psychological experience of writing, I began to see the value of the tool. ChatGPT was not generating professional prose all at once, but it was providing starting points: interesting research ideas to explore; mediocre paragraphs that might, with sufficient editing, become usable. For all its inefficiencies, this indirect approach did feel easier than staring at a blank page; “talking” to the chatbot about the article was more fun than toiling in quiet isolation. In the long run, I wasn’t saving time: I still needed to look up facts and write sentences in my own voice. But my exchanges seemed to reduce the maximum mental effort demanded of me.

I’m not bowled over much these days. But Guardian Open Platform is a chasmic leap into the future. It is a work of simplistic beauty that I’m sure will have a dramatic impact in the news market. The Guardian is already a market leader in the online space but Open Platform is revolutionary. It makes all of their major competitors look timid.

Your documentation is complete when someone can use your module without ever having to look at its code. This is very important. This makes it possible for you to separate your module's documented interface from its internal implementation (guts). This is good because it means that you are free to change the module's internals as long as the interface remains the same. Remember: the documentation, not the code, defines what a module does.

In fact, Microsoft goes so far as to promise that it cannot see the data collected by Windows Recall, that it can't train any of its AI models on your data, and that it definitely can't sell that data to advertisers. All of this is true, but that doesn't mean people believe Microsoft when it says these things. In fact, many have jumped to the conclusion that even if it's true today, it won't be true in the future.

What to try first? Run Claude Code in a repo (whether you know it well or not) and ask a question about how something works. You'll see how it looks through the files to find the answer. The next thing to try is a code change where you know exactly what you want but it's tedious to type. Describe it in detail and let Claude figure it out. If there is similar code that it should follow, tell it so. From there, you can build intuition about more complex changes that it might be good at. [...] As conversation length grows, each message gets more expensive while Claude gets dumber. That's a bad trade! [...] Run `/reset` (or just quit and restart) to start over from scratch. Tell Claude to summarize the conversation so far to give you something to paste into the next chat if you want to save some of the context.

Thus, if we do not lose ourselves in the dogmas of omnipotence and omniscience, the conflict between God and the Devil is a real conflict, and God is something less than absolutely omnipotent. He is actually engaged in a conflict with his creature, in which he may very well lose the game. And yet his creature is made by him according to his own free will, and would seem to derive all its possibility of action from God himself. Can God play a significant game with his own creature? Can any creator, even a limited one, play a significant game with his own creature?

It hangs on its / stem like a plum at the edge of a / darkening thicket.It’s swelling and / blushing and ripe and I reach out a / hand to pick itbut flesh moves / slow through time and evening / comes on fastand just when I / think my fingers might seize that / sweetness at lastthe gentlest of / breezes rises and the plum lets / go of  the stem.And now it’s my / fingers ripening and evening that’s / reaching for them.

One big thing that a lot of people love to do is create new role types. For any new thing a company wants to do, the tendency is to put up a new job description. I think a lot of people notice this and chafe at it when the role is for the new hotness. For example, every company wants to rub some AI on their stuff now, so they are putting up job descriptions for AI engineers. If you’re an engineer interested in AI sitting in such a company, you’re annoyed that they’re doing this (and potentially paying that person more than you) when you could easily rub some AI on some stuff.

When you're young, you look at television and think, There's a conspiracy. The networks have conspired to dumb us down. But when you get a little older, you realize that's not true. The networks are in business to give people exactly what they want. That's a far more depressing thought. Conspiracy is optimistic! You can shoot the bastards! We can have a revolution! But the networks are really in business to give people what they want. It's the truth.

But that was then, a long time gone. Now we have something different: we have "anything goes" without the spirit. "Transgression", "subversion", "deconstruction" are praise words bestowed as solemnly as "structure" and "order" once were, little gold stars awarded to rappers and television comics. Cakes on rakes are everywhere. A million cats cavort frantically for our attention. Even the fish has been co-opted (though what choice did he have?). "Enjoy!" cries the fish. "Consume! Everything will be fine when your mother gets home." But the fish is whistling into the wind. The mother has left, and she’s never coming back. It’s just us and that goddamn cat.

"Man has Nature whacked", said someone to a friend of mine not long ago. In their context the words had a certain tragic beauty, for the speaker was dying of tuberculosis. "No matter", he said, "I know I’m one of the casualties. Of course there are casualties on the winning as well as on the losing side. But that doesn’t alter the fact that it is winning."

Indexes are models: a B-Tree-Index can be seen as a model to map a key to the position of a record within a sorted array [...] Our initial results show, that by using neural nets we are able to outperform cache-optimized B-Trees by up to 70% in speed while saving an order-of-magnitude in memory over several real-world data sets.

Through me the way into the suffering city, Through me the way to the eternal pain, Through me the way that runs among the lost. Justice urged on my high artificer; My Maker was Divine authority, The highest Wisdom, and the primal Love. Before me nothing but eternal things Were made, and I endure eternally. Abandon every hope, who enter here.

H100s were prohibited by the chip ban, but not H800s. Everyone assumed that training leading edge models required more interchip memory bandwidth, but that is exactly what DeepSeek optimized both their model structure and infrastructure around. Again, just to emphasize this point, all of the decisions DeepSeek made in the design of this model only make sense if you are constrained to the H800; if DeepSeek had access to H100s, they probably would have used a larger training cluster with much fewer optimizations specifically focused on overcoming the lack of bandwidth.

"You see this goblet?" asks Achaan Chaa, the Thai meditation master. "For me this glass is already broken. I enjoy it; I drink out of it. It holds my water admirably, sometimes even reflecting the sun in beautiful patterns. If I should tap it, it has a lovely ring to it. But when I put this glass on the shelf and the wind knocks it over or my elbow brushes it off the table and it falls to the ground and shatters, I say, 'Of course.' When I understand that the glass is already broken, every moment with it is precious."