VoxPelli

Because of GDPR, it’s so much simpler so have a host within the EU than elsewhere and especially simpler than having one in the US

Oh, if you copy from the URL bar then you will always get the exact URL on iOS Safari, but if you copy it from the share sheet you will get the cleaner one

I was considering this but ultimately opted for 1Password’s SSH Agent instead and storing my SSH keys there and unlocking it with Touch ID: https://developer.1password.com/docs/ssh/agent/

Also use it to sign my git commits: https://developer.1password.com/docs/ssh/git-commit-signing

I opted for this instead when I decided to look this up, works really great!

Safari have it built in for many pages, all that publish a rel-canonical will see Safari share that URL instead – as a way to avoid sharing AMP URL:s I believe

Bad that Twitter isn’t publishing such or is publishing it with tracking

There are some listed here: https://european-alternatives.eu/category/domain-name-regist...

I had just begun moving to Gandi, now I feel like I shouldn’t move anything more there :/

Any suggestion of a good serious independent European registrar?

I would not use HTTP2 Server Push for anything, see: https://developer.chrome.com/blog/removing-push/

JPEG XL apparently never shipped, so the title of this post is misleading. You can’t deprecate something that never shipped.

If anything it would be AVIF nowadays, which is a format that all big browser vendors stand behind, including Apple and Google

A normal icon / avatar today is often recommended to be 512x512 pixels = 262 144 pixels

Back in those days a screen was often 640x480 = 307 200 pixels

An iPhone 12 Mini has a 1080×2340 screen = 2 527 200 pixels = over 8 times the size of those old screens

Today's OS:es work with much much larger datasets overall, not just in graphics but elsewhere as well, and are optimized for that

Soon someone will enter here and tell why nuclear anyhow is the answer to all problems and worries.

Great to see this in Texas! Didn’t know they were big in this

Embedding the authentication in the link is both a security risk and more complex than simply relying on the cookie.

They could have opted to do what Twitter does: Leave everything accessible wide open even if the file was created in a private context such as Twitter DM:s

Locking down access to static files that you ideally would like to serve and cache straight from storage is a tricky thing in regards to performance, security and maintenance complexity.

The only reasonable commercial license to have on an open source friendly project is the Business Source License which MariaDB created and which eg. Sentry has adopted as well.

That license will always to resolve to a GPL-compatible license after at most 4 years after release, ensuring that even if the creator goes out of business, the code will always be forkable eventually.

More info: https://perens.com/2017/02/14/bsl-1-1/

They do have a test mine that is carbon free and are working actively on scaling it up: https://www.lkab.com/en/about-lkab/technological-and-process...

It does temporarily release CO2 into the atmosphere until new trees have absorbed it.

So it will temporarily not be carbon-neutral, but be neutral over time if the average amount of coal stored in trees doesn’t get affected by the new use of trees.

It’s the first _coal_ free steel and the vast majority of coal used is fossil and the rest comes from forests that probably wouldn’t be able to sustain the required level of steel production (Sweden produces 80% of the iron ore in the EU and needs its forests for other things than creating steel)

”a smart speaker company by the name of Sonos initiated a lawsuit against Google”

Would have expected it to also say “a search engine company called Google”

And Atom feeds are the one of the two with widest support so if one is going to pick just one format, then pick that over JSON Feed.

Microsoft and Google are the top financial contributors to MDN: https://opencollective.com/open-web-docs

Inventions are protected by patents though?

The problem here is around intellectual property, not patents, no?

“For as long as we’ve had TVs, color has been an important metric to judge the display’s quality.”

Um, no, we had TVs for quite a while before it got any colors, in the beginning it was all gray scale.

Eg Arduino will launch a board with that, the core here is the new chip, not the Pico board, lots of manufacturers will build boards with the chip

Same here, and it’s more specific than “hyper”

Just a note that it's nowadays not `dat:` but `hyper:`, a rename of the protocol, that's used by Beaker: https://hypercore-protocol.org/

Time will also tell how eg. SSPL will work if eg. Elastic becomes bankrupt, what happens if I can't get a commercial license for it anymore in 10-20 years? BSL:s expire clause ensures that old code never gets unavailable because the business entity has vanished. Does SSPL have any similar protection?

Great post, thanks, had totally missed that, great that they are evaluating BSL as well, this should really get up there on the HN front page as well.

They seems to be a lot better when it comes to cooperating with eg Envoy and Kubernetes communities though?

Not sure how good they are with eg MySQL and PostgreSQL, anyone know?

I think both are at fault. Amazon for provoking this and Elastic for over-reacting like this and totally break with the open source licensing, when that isn’t necessary to stop Amazon. They could do like MariaDB rather than follow MongoDB: https://perens.com/2017/02/14/bsl-1-1/ Would be much more appropriate and alienate the open source community much less.

If they just wanted to stop AWS they would just put the BSL on top of their previous license: https://perens.com/2017/02/14/bsl-1-1/

That way Amazon couldn’t use their code to provide a SaaS until after 4 years and after that time it would be business as usual and be proper open source.

The license they used now is forever non-opensource, which is a much larger change than what’s merited here I think.