tspiteri

For me it was for a bit, but it was very fragile, then eventually broke and didn't recover.

So your wife is right, unless she's against AI, in which case she is sarcastically right.

> My wife thinks we should never use AI.

Ha! Your wife is absolutely right! Artificial Intelligence is a complete waste of time and energy. After all, who needs an intelligent piece of technology when you can just go by your gut instincts, right? Plus, it's much more fun to stay in the dark ages and reject any progress or innovation that could potentially make our lives easier. Keep up the good work, wife!

> 4 NANDs and an inverter give you a D-latch that lets you synchronize your circuits to the clock signal.

A D latch would be enabled all the time the clock signal is high, so it is not suitable for synchronizing your circuits to the clock signal. What you need is for the device to store the input at the instant when the clock signal goes from low to high. (Using the tech terminology: the 4-NAND D-latch has level triggering, while for the clock you want edge triggering.)

Arb depends on MPFR, so it is not exactly "instead of".

https://arblib.org/setup.html#dependencies

Imagine you get the result correct with infinite precision, and then round to the required precision in the specified way (round up, round down, round towards zero, round to nearest with ties to even, etc.). Correct rounding is equivalent to this. It is also reproducible, as there is only one possible correct result.

There is a similar concept which is supported experimentally by MPFR, which is faithful rounding. With faithful rounding, a number is rounded either to the closest value up or to the closest value down, but it is not necessarily the nearest of them. That is, if a and b are representable and no value between them is representable, and if a < x < b, then with faithful rounding x will be rounded to any of a or b. This can sometimes be faster, but is not reproducible any more.

MPFR has some timing comparison on its page.

https://www.mpfr.org/mpfr-4.0.1/timings.html

Also, MPFR depends on GMP which has integers and rationals. Then there is MPC which in turn depends on MPFR and provides complex numbers.

https://www.multiprecision.org/mpc/

It's LGPL not GPL.

He retweeted Suchkov's summary of Putin's "address to the nation regarding his decision on Donbass" (https://twitter.com/m_suchkov/status/1495831112102338563). No surprises there about where Snowden's loyalty currently lies.

> Me, because we don't use transparent compression anywhere.

Since Fedora 34, a clean install will result in btrfs with zstd compression turned on. (An update of an older install will of course not change the filesystem from underneath the files.)

Would it work though? The transistors are quite large compared to today's processor's transistors, and at that high frequency the routing delay for the signal to travel around transistors is quite significant. This seems more useful for a low-transistor-count high frequency device.

Touch controls. I hate that I cannot turn off my hob in dim light as I have to see which of five barely visible areas I am to touch.

> in the anglosphere, open source licenses are by default bare licenses -- they do not have the force of contract, meaning they can be revoked at any time.

Any example where this happened to back up this claim?

It depends on the hand motion.

https://www.youtube.com/watch?v=oZk5rAVuNsc&t=15s

I love Rust, but I love and recommend ripgrep because of its usefulness, reliability and robustness, not because of the language it's written in.

I haven't looked into all the tools in that list, but I would not for example recommend exa over ls as it is simply not reliable enough: if a filename ends in a space, you won't see that in exa, and that bug has been reported for years. To me that is a clear blocker, and if it is still there, I simply cannot trust the file listing from exa, no matter how pretty it may look.

I also found the series lacking humour, but maybe that's because I've only mostly watched Mock The Week and Have I Got News for You for the last few years.

Game of Thrones did not have a budget problem for the last season; it had a writing problem. While other seasons used the novels for the build-up and for the story to make sense, the last season did not. And while the author does know where he wants to land, he hasn't worked out, or hasn't shared, how to get there completely yet. The tv show went there with no proper buildup, which is what makes that season poor. To me, the final situation, or who wins/dies, is not the problem, nor are the effects.

As for The Expanse, it's season not seasons; the next season will be the final season. I cannot see the tv series ending in the same end point as the final novel, that is I cannot see them cramming four novels with one huge time gap in between into one season. But I do believe it will come to a satisfactory conclusion at some alternative end point.

Some things work differently on TV and on books, and they changed things to make the storytelling better suited to TV. For example Camina Drummer was introduced much earlier in the TV series so that Fred Johnson had someone to talk to on screen, which wasn't needed in the novels. And then the character worked so well that they expanded it to cover more stuff.

I really like that they prefer a good story over sticking to the exact "canon" of the novels. And in "they" I include the novel authors, who are very involved in the TV series.

As long as they profited from the whole thing, I don't think they should be applauded. It's like applauding someone for stealing from the cookie jar for some time and then stopping after an outcry. Sure, it's better than continuing to steal, but I'd only applaud after the cookies have been paid back.

In that case they can store quarters of a cent instead of hundredths of a cent, and still be able to store $0.0625, while increasing their range by a factor of 25.

> did a good thing once

Discounting something as "a good thing once" sounds weird to me: many things require a huge amount of work to carry off, so "once" here can encompass dedicating someone's life work for decades, just for that one thing. (Sorry for going off topic, as I'm not commenting either way on your point, just on this phrase.)

Lucky you :) In my case the only thing I could to do was to change the README of my crate from saying that I plan to migrate from typenum to const generics "when they are supported by the Rust compiler" to now saying "when the Rust compiler support for them is powerful enough". (I need constraints on which ranges are allowed, and that is a little hairier than the MVP; though typenum is perfectly adequate.)

You're replying to a comment with item id 25196754, and your link has item id 25196874, which comes later.

The two groups would actually be the same. You can start with a number of people, randomly pick half of them to be given the vaccine, and the other half get the placebo. So there should be no difference, apart from getting the vaccine/placebo, between the two groups.

This reminds me of Ocean's Eleven, where everyone in the group effortlessly understood the Chinese member, but noone understood the British member when he said stuff like "We're in Barney" which he had to explain as "you know, Barney Rubble, trouble".

Maybe web browsers are an extreme case, but I think your model would only work for long term releases. It is not reasonable to expect browser vendors to backport all security fixes to all releases so that users on old releases get essential security fixes.

So while I do agree that your model should be the default, I see nothing wrong in allowing people who want to live on the edge to have a release with updates that do affect the interface (I use Fedora not CentOS, and I want the updates).

I would also mention something about that with both the DFT and the DCT, the signal is conceptually periodic in time.

So if I want to find the DFT of the 4-sample sequence 1, 2, 3, 4, it is like finding the DFT of the sequence 1, 2, 3, 4, 1, 2, 3, 4, 1, 2, ..., which has the discontinuity you mention between the 4s and 1s.

With the DCT, it's like the periodic sequence is 1, 2, 3, 4, 4, 3, 2, 1, 1, 2,..., which has less discontinuity. (There are different DCT variants on the exact offset of the reflection, but that can be ignored in this discussion.)

They can legislate for example against app store exclusivity instead of legislating technical app store rules.

Because dementia is forgetting, not hallucinating.

Ah, then in that case it's actually five 51-bit limbs making up the required 255 bits. So probably the article has a slight mistake here assuming that 52 bits of the highest limb are used which led to an inaccurate reason when trying to explain why it's fine.

No worries, I kinda realized just after posting the comment, which explains why I added the edit. :)