orra

To put numbers on this, you can buy a C rated fridge freezer in the UK which uses 176 kWh of energy per annum.

That's an average power consumption of 20W. But electric heaters are say 1200W or 2000W.

(These costs rocketed last year. Incorporation is now £50, and an electronic confirmation statement is £34.)

So, the older DeepSeek-V3 model weights are sadly not permissively licensed.

But the recent DeepSeek-R1-Zero and DeepSeek-R1 have MIT licensed weights.

Ahah! I worked on a project which used dotnet format in a commit hook. That was a frustrating experience, trying to rebase code.

(Unlike most formatters which are instant, dotnet format takes at least half a minute, because it performs a build just to format your code.)

Thanks for linking this. I wasn't surprised Microsoft made their AI auto completion proprietary (they did similar for C# in VSCode). But it grated me that semantic highlighting wasn't open source.

What about ruff + pyright? From a skim it seems like a lot of the missing features would be covered by a type checker?

(Pyright isn't written in Rust, but adequate performance is still a design goal)

We tend to breed specialised sheep for wool, and specialised sheep for meat. Same thing with different animals.

Besides, why the analogy? We’re talking app stores. It's not apparent why we need a joint supply analogy.

This almost got me too, but you missed GP’s point. See their next paragraph beginning “Or”. (It's about double standards for individual versus corporate copyright infringement.)

OP puts up a straw man that criminal harassment is any trivial slight in the eye of the alleged victim. However, for an instance of harassment to be criminal harassment, the perpetrator must [1(b)] know their behaviour is harassment. Morever, if perpetrator is behaving reasonably, then it's not harassment [3(c)].

Section 7 is also key to statutory interpretation. 7(b) gives persuasive examples (you had not seen these and so called section 1 tautological). Morever, 7(3) means that the criminal offence only applies to repeat offenders.

This comment is being deservedly downvoted. It's misinformation. Here's the statutory threshold for harassment: https://www.legislation.gov.uk/ukpga/1997/40/section/1

> Which sounds nice, but you can't really get something for nothing.

This is hyperbole. First, nobody's getting something for nothing. Second, the eligibility is limited. Besides, this is hardly the first time a price regulation has been applied to a failing market.

> I really don't want me ISP to track my income.

Then don't apply for the discount.

> You were using the TfL schematic map which is very much a form of art

The site was taken down by a trademark complaint, not a copyright complaint.

Nah, pip is still brittle here because it uses one package resolution context to install all your global tools. So if there is a dependency clash you are out of luck.

So that's why pipx was required, or now, UV.

Very nice work. But could you please explain why counting digits quickly (or even slowly) is useful for a JSON serializer? This is lower level than I'm used to working. Is this so you can alloc the right amount of memory, or something else?

A surprisingly detailed writeup, nice.

But what's the misleading anecdote from Raymond Chen? The claim piqued my interest but I didn't catch that part.

No, what you describe is not better. Quite the opposite.

First, USB PD is capable of much higher wattages than what SuperVOOC artificially caps USB PD to.

Second, proprietary charging standards are not compliant with the USB specification, so it didn't meet the regulation.

> A reason to not demand USB-PD, such law would prevent upgrades to later better version of that thing.

Can we apply some common sense please? You're right that not allowing revised standards would be silly. So, they simply update the law to reference newer versions of the standard. https://eur-lex.europa.eu/eli/reg_del/2023/1717/oj

This hopefully means the end to standards-violating nonsense like SuperVOOC? Originally SuperVOOC wasn't USB-PD compatible at all. Now, AIUI, SuperVOOC is partly USB-PD compatible, but only to lower wattages.

I linked N1256, which is essentially C99 plus the errata (Tchnical Corrigendum 1, 2, & 3).

Reading an indeterminate value is its own undefined behaviour. See the second bottom paragraph on page 492: https://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf

> This reminds me of the time Debian developers "knew better" with the openssh key handling...

This is a totally unfair comparison. The upstream OpenSSL code was broken: it was invoking undefined behaviour. Moreover, the Debian maintainer had asked upstream whether it was OK to remove the code, because it was causing Valgrind memory warnings.

I like the mediocrity of the example they headlined: the user asks for unit tests. Copilot writes only two test cases, so not exactly great coverage. Plus, the test cases uses python's unittest, which isn't as slick as pytest.

> If, when the GPL was first published, the world was cloud-hosted and SaaS-ful, the GPL would either have included some provision like the SSPL, or it would have had relatively little impact on the world.

Well, no on the first one. We know exactly what the GPL looks like to protect users against proprietary network hosted software: the GNU Affero General Public License.

Nice work, and well explained. Thanks.

Thanks for the correction and sorry for the mistake. I skimmed the changes but apparently not very well.

Whilst this is true, it looks like OpenWRT fixed the hash truncation but not the command injection.

I hope they're planning on fixing the command injection. As the blog post says, the created images are signed. Even without the signing, it's code execution from untrusted user input. And of course vulnerabilities can be strung together (just like in this hash collision case).

Thank you, I was wondering about the chemistry of this.

When they say to avoid grapefruit juice, it's not because of something innocuous like indigestion. But until recently that had been my assumption.

Sounds sensible. It's important to set boundaries, and enjoy time off.

For me Advent of Code is a slippery slope. The difficulty ramps up so at first it's easy, then it's rewardingly difficult. But then before I know it, it takes wayyyyy too much time. The danger is being emotionally invested by then.

I would love it if the first star of the day was required, but the second was a bonus. I love Advent of Code, but I don't have the time to get 50 stars.

Yeah, you'd definitely want baby jumbo frames if you live in the UK. ISPs here almost universally use PPP.