I'm a non-native English speaker so I grew up with subtitles, and I still use them today, even though I use the English subs now and not in my native language.

I don't agree with the premise that it has become harder to hear dialogue. In some movies yes, and in some of them that is deliberate. But most of them time I can perfectly hear what they say, but I find that subtitles help me to remember certain details, such as names. It also helps when the talking is going really fast in intense scenes.

Weirdly enough I have experienced audio issues at cinemas where the voices are difficult to hear, but everything else is fine. Watching those same movies at home and it's all good. Probably just an issue at my local cinema though.

And the size difference appears to be only 0.2kB? 6.1kB from the zipped version and 6.3kB for the unzipped one

I see advanced.css uncompressed size is 25K and compressed (by zopfli) - 5.4K

The current (and very recently updated) OWASP recommendation[1] is a minimum of 600 000 iterations

[1] https://cheatsheetseries.owasp.org/cheatsheets/Password_Stor...

I read that as a "total of 600 000" iterations, so 300 000 locally and 300 000 on the server. Am I wrong?

According to the OP article, the server side iterations are ineffective for adding security in bitwarden, so you need 600,000 on the client. This would not be the case if the design was correct.

(I'm not a security expert, so I'm going by the article)

If this was a frequent event then yes that would be an issue. But Python 3 released in 2008, more than 14 years ago now, and since then there have been no sign of such a new big change. While the core developers briefly talked about making Python 3.10 into 4.0 that was scrapped, and currently the plan is just to continue with Python 3.x "forever".

I should maybe test out pyenv again. I tried it several years ago but I ran into some limitations. Or at least I thought I did, I might just have used it wrong. My main usecase, back then at least, was to have just one global-like environment for every Python version I was working with. And sometimes and additional one for debugging/testing some different packagecombinations. Back then it seemed like, to me, that pyenv only worked on a per-project basis, but I might just have followed incomplete guides and not looked into it deep enough.

Lol, the first song I searched for game 0 results. (Searched for "More With You").

Oh and a bug: If I first search for something and get a result, then enter a new search and scroll down before this new search has completed, it starts searching for page 2-7 of this new search term. Even better if you've already triggered the search for page 2-7 on that first term. If you then scroll down after updating the search, it is now searching for page 8-11 of that new search term :)

Thanks, I registered the issue on GitHub.

It's 5 years from End of Sale, and not just 5 years from the initial sale date. It sucks for those that bought the router close to End of Sale, of course.

Don't know for how long these routers have been around, but for one of them I found a overview document dated 2011.

If I buy a washing machine and it breaks after 4 years, it doesn't matter that the model is 10 years old at that point. My rights are from time of purchase.

Should be same for devices. If not, they should at least carry a "best before date" like food, so you know at time of purchase how long you can expect to get updates.

> Apple supports iPhones for 7 years

That's 7 years from launch, not 7 years from end of sale.

Not defending Cisco here, but looking up the RV110W I can find documents for it dating back to 2011, so that's 11 years of support.

For the RV130 I can find an administrators guide that was revised in Aug 2014, so at least 8 years there. probably more given that it's a revision.

It is irrelevant.

Using your argument, you could stop all support right after the devices stopped being sold.

So it was being sold for 15 years and the software has been in support for 15 years and it is old so we stop selling it and supporting it at the same time.

Now hope you are not the person who bought it on the last day...

A fresh load of the page (according to Chrome) is 21.5kB transmitted, 59.1kB uncompressed. That's when loading with nothing cached. The HTML is just 7.9kB, but then the CSS is 2.2kB, JS is 2.3kB, and the favicon is 7.9kB which is a bit funny (but it's of course irrelevant for the actual page).

HN could put the CSS and JS into the generated HTML file and still stay under 14kB for the initial load, which then would give you almost everything needed to render the page except a few gifs.

Favivon isn't blocking the initial request so it really shouldn't matter.

> HN could put the CSS and JS into the generated HTML file and still stay under 14kB

At the expense of caching those resources that stay static. With http/2 the benefit of merging into one resource should be negligible anyways.

> With http/2 the benefit of merging into one resource should be negligible anyways.

That's not true, by including them in the HTML you save an extra round trip for requesting them. That's what HTTP2 Push was supposed to solve, but it's being deprecated & removed.

That's a fair point.

In marketing and on cables they've chosen to use the terms USB4 20Gbps and USB4 40Gbps, so at least that's explicit. There's also officials ways to mark cables as being 100W or 240W capable.