Another way to phrase this is that guarantees can impose constraints that are overly limiting.

Having said that, the parallel with ML seems to be bit of a stretch. What exact high level guarantee is given up by being probabilistic?

How is the battery life? Rechargeable sure is nice, but the older models lasted forever on 4 AAAs (at least my TI-83). That's one aspect that would justify the low processing power for today's standards for portable computing devices.

This seems to be sloppy wording, with the intent of "we only offer the bounty under these terms". Maybe my interpretation is too charitable.

I do not speak for MSFT, but last time I spoke with MSRC indeed they would be happy to receive your vulnerability report even if you did not wish to participate in any particular bug bounty program.

I wonder if "if you contact us... you automatically agree" stands in court. That's just ridiculous.

Reader, it does not.

I wouldn't. This is very specialized to the type of the elements.

Some languages, such as C++, allow for specialisation via templates and compile time evabulation (constexpr). It would be possible to detect when the size of the data type matches one of the integer types, is a POD, is comparable via memcmp, etc to use SIMD optimised algorithms.

It is looking like C++ 26 will get compile time reflection, which would make things like this even more feasible.

1. You still need std::launder in that case.

2. It doesn't initialize the object that is implicitly created, even if the storage has initialized chars.

The paper that introduced the implicit lifetime mechanism suggests that std::launder is not required: https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2020/p05....

And they are legally required to license the play to do that, if it's still in copyright.

Only to perform it, not learn it.

And LLMs perform when you prompt them.

I feel uneasy about this after the xz story.

Great, the UX feature I probably hate the most in Jira, now on Github.

This was exactly my thought. It breaks every bit of intuition I have using a browser, and makes pages run even slower.

Every bit of intuition you have using a browser, really? You click a link, the current page changes, you click back, it goes away. You cmd/ctrl click it opens in a new window, you right click and select "open in new tab/window" and it opens in a new tab / window.

Now, when you click a link in GitHub, the current page doesnt change. I want to look at the linked issue on its own page. That doesn’t occur anymore.

The page i wanted to go to pops up in a small overlay on the right hand side. The body text and content that I wanted to view is in a new, weird location, with the old page still behind it in the normal spot. It’s very unintuitive.

Thankfully either the behavior has reverted or I’m no longer in the A/B test. I can’t get the popup to happen anymore for me. (edit, nvm, behavior varies depending on repo or something? it acts completely differently on different pages, sometimes links are normal and sometimes they open in a popup. extremely annoying)

Also it breaks copying links. If I want to link to an issue I copy the URL. But now there's two different issues open at the same time, which one am I linking to? Original? Popup? Both?

Right, not saying it's not annoying, but "every intuition about using a browser" is a bit over the top. A link can open a dialog, has been happening for decades.

And GitLab, too!

Recently I tried out Brompton Bike Hire in London for a week. Can recommend the bike, and the price is reasonable. The bikes are hired from automatic storage lockers, which makes sense as a concept. The app is atrocious though, and I had a lot of trouble returning the bike at the end of the week.