I then start a fresh conversation, make it analyze the design document and code, and for larger changes, generate a high-level implementation document which includes concrete phases or steps. I review this plan and iterate if necessary.

Then for each phase I make it generate a detailed plan for that phase and save it along side the other documents. Once the phase is over, I make it write a summary of what was done, decisions made and reasons for it. And typically a good point to compact the model's context.

These documents gives additional context for when I make another model do code review, and help illuminate drift or gaps from the main design document.

However in most cases the incarceration is used as punishment, with the length is related to the seriousness of the crime rather than the likelihood for repeating offenses.

Here in Norway we explicitly separate this, where most sentences are punishment, but some are explicitly for protecting society. In the latter case there is technically no end, just a minimum time and after that periodic reviews to determine if the person still poses a sufficient threat.

It's technically classified as a punishment due to legal reasons, like ensuring human rights and due process are respected.

I frequently make this error when I talk. My brain thinks of different ways to phrase what I want to say, but when I speak it starts with one and finishes with another. The result is almost always wrong in the way the title is, ie some variant of a double negation.

Sometimes it happens when I type, though I try to read it multiple times so often catch it.

I've made a few internal web browser-based tools recently, and I let Codex and Claude handle the entire frontend. Don't really care about the details there.

Tool is not exposed to the internet and they had limited scope. I handled the important details on the backend myself, and I reviewed the integration tests carefully.

These tools would just simply not been made without Claude and Codex.

YMMV as they say.

Then there's important code, like business logic, security related stuff and such. For that I'll happily use AI to brainstorm ideas and avenues, do code reviews where I'm prepared for it to be wrong and help with library suggestions etc, but where I want to write the code myself so I understand what's going on.

We'll see how it changes, been a wild ride so far.

Collected and stored sediment samples were found to have surviving T. namibiensis cells after over two years. The cells had no access to any added sulfide or nitrate during this time. In the surviving cells, there was a notable size decrease. To survive without growing the cells depended on the nutrient stores of the central vacuoles.

[1]: https://www.microsoft.com/en-us/trust-center/security/secure...

[2]: https://cybermagazine.com/articles/how-microsoft-is-securing...

I saw Metallica once, many moons ago, and it was at this big venue which of course was Live Nation/TM. It sucked ass. Sound was terrible, had to watch screens to see what was going on, beer was ridiculously priced and yet somehow long queues.

I decided then I wouldn't go to those venues anymore. If a band I like plays there, whatever, not worth it.

Meanwhile I've had many, many concert experiences that were 100x better than the Metallica concert for a fraction of the price of the Metallica ticket at small, local venues.

My buddy recently invited me to another such big-ass venue with some popular band, and it just cemented by view. So not worth it.

As I understand it, it would instead be

  k(x, x') = Cov[f(x), f(x')] = Cov[r_1 + x * r_2, r_1 + x' * r_2]

> What is 'x' here?

So as I understand it, a Gaussian Process is defined in terms of a set of random variables which are indexed, typically by either time (t), or space (x). So in the concrete example, x here would be the amount of cheese inserted into the magical machine. In general the "index" can be a vector. Say if the magical machine instead required inserting both cheese and milk to produce some amount of gold, the index x would be two-dimensional, to represents the various amounts of cheese and milk you inserted.

> It does not seem the 'f' here is intended to be the specific 'f' introduced at the beginning of the article.

Right, it's general, and it's kinda confusing to use f when everything else seems to use X_t or similar. Here f is actually a random variable index by x, so one example could be

  f(x) = r_1 + x * r_2

> The plots now have y and x, and x1 and x2. How are these related?

The left plot shows three realizations of y = f(x), ie for three different choices (samples) of the random variables that goes into f(x). The right-hand plot shows the output of the kernel function for two indices x and x'. In the first example, the kernel function was the dot product between the two inputs, but given the indices are 1-dimensional that reduces to just k(x, x') = x * x'.

Back to the example, you can feed the machine various amounts of cheese and record the various amounts of gold you get back. The amount of cheese are the indices which you use with the kernel function you picked, which you run through the Gaussian Process regression math, and you get a new function which takes an index (amount of cheese) and returns a normal distribution that predicts the amount of gold for that index (amount of cheese).

The process spits out the mean and the variance of the normal distribution, so you can look at the variance to determine how certain you can be about the prediction which will be centered around the mean.

As I understand it, the point of the left plot is that you can use it to get an idea for which kernel function to use for your measured data. And as mentioned you can easily make new kernel functions by adding (OR-like) and multiplying (AND-like) other kernel functions.

Also the author made a mistake, he mentioned kernel functions are parameterless, but he meant non-parametric. The kernel functions he shows like the periodic kernel has hyperparameters l and p for example.

At least that's my current understanding.

I recall my dad had a Tanzanian visit via work. My dad took him out for dinner, and as soon as they got a table the waiter came with glasses and a jug of water, as is customary here.

The guy got really upset, neither of them had asked for water and as such he thought it might go to waste.

My dad had to explain that in Norway water is not a scarce resource, at least not to the point of having to save a jug of drinking water.

The vector is always defined in a vector field which has a given dimension, and usually the dimension isn't that interesting. Typically it's either the same between the vectors you consider, or the vectors have one of a few fixed number of dimensions. Meanwhile the length of vectors is an interesting quantity.

For sets, since the values can be anything, nothing or everything in between, you can't really define many interesting functions or operations that work on the elements of sets in general. Meanwhile, the number of elements in a set is an interesting quantity.

Anyway, just my take, though I never did take much math.

[WHAM! CRUNCH!]

"Look, they nearly missed!"

"Yes, but not quite."

- George Carlin

For a long time I felt the construction of shows. The way they try to artificially raise tension and intrigue, and that most of them had such poor writing, be it predictable dialog or just characters doing illogical things for the plot to happen.

So now I watch mostly informative or edutainment on YouTube. Such a wealth of very varied content. And so easy to skip the overly fabricated stuff.

And to be honest, they shouldn't need to be. Browsers shouldn't be app hosts unless the user wants them to be.