> ALGOL decided to stick close to mathematical tradition

And now we have all 3 in Python (=, ==, and :=) which makes me sad.

Splitting = and := in python is very intentional though, It makes writing bugs like this impossible:

    if is_logged_in=True:
        allow_access()

I've got... opinions on the way python separates statements and expressions, but there's some real benefits to it too.

Splitting = and := is intentional, but not for the reason you stated. We could have used := for all assignments from the beginning:

    is_logged_in := True

or

    if is_logged_in := True:

I agree that this would require blurring the statement/expression distinction. You can still do that in a weird way, by disguising your assignment as an expression. This is valid:

    (is_logged_in := True)

The reason it was done that way because := was an afterthought, and making it the assignment operator would have introduced a breaking incompatibility. That lead to having 3 different symbols for 2 use cases (assignment and comparison).

except if := was the default then accidentally typing ":=" instead of "==" would be the easy bug to make.

:= was bolted on after of course, and I think that was a mistake, but no one asked me, nor do I expect anyone to ;P

> Wikipedia is cheap compared to creating and training models.

DeepSeek said it spent $5.6M [1] on training V3, which doesn't sound too much for a near-SOTA model.

An open source entity can come up with a hybrid business model, such as requiring a small fee from those who want to host the model as a business for the first n months following the release of a new model, but making it fully free for individuals.

[1] https://arxiv.org/pdf/2412.19437

There are very good open weight models (such as DeepSeek v4 Flash) that can run on consumer level hardware.

Note that we are talking about 95% of everyone's use cases, not your specific use cases (which could require better models all the time).

Just curious, I have no idea how eSIM reselling business works: How can you offer 2-3x cheaper prices? Is it because others have very high margins, or because you have a secret optimisation method?

Great question. Cheaper prices come from a mix of focus on customer value primarily. That includes making sure that the eSIM is high quality (low latency, data routing in the region) and also focusing on getting the best eSIM plan to the customer - transparency, better FUP etc. So there's an element of optimisation involved here.

The U.S. government spends approximately 4.1 cents [1] to produce each $1 bill. It would probably be more expensive to counterfeit it because of the volume, but I doubt it would be more than $1.

[1] https://www.federalreserve.gov/faqs/currency_12771.htm

> He was also made to pay a fine of $1.

I see what they did there.

You can also to to jail for selling chewing gum. And you are legally required to flush a public toilet after using it.

Neither one is really enforced

I don't fully disagree, but the bullying will not stop if they see that they can beat you up easily. It might even get worse.

I can think of only 3 times where i or someone else confronted a bully and made it stop. In 2 of those cases the bully was stronger but such stuff can always carry risks that the bully might not like or scenarios they can't take.

In the first the bully eventually got hit with a school desk (they were fairly light but hard) pretty bad by the victim that finally crashed out and the bully actually looked like a wimpering fool in front of a ton of people. As far as I know he didn't try to get back at the victim.

In the second it was I that flipped out and had some luck. I didn't seriously hurt him but he realized the blind intent in the moment was there. He just seemed shocked and no longer bothered

The 3rd guy had some Moroccan machismo thing going and kept picking on people he couldn't beat and it always happened fairly conventionally without suprise.

How about the amount of times when the bully didn't stop? That's the useful metric here. When there was resistance, how effective was it?

Judging from your description you didn't lay out any examples of where fighting back failed.

Because I can't really think of examples. Mind you that might be biased because people do often have an intuition for when they'd either get their ass handed to them or it would otherwise backfire so I assume there'd be some examples otherwise if they didn't take that into account.

Also these examples were mostly from incidents where the bullying got physical anyway. Much of it isn't and making it physical even as the victim tends to have drawbacks.

To be fair they said "I can think of only 3 times [...] and made it stop", implying that there were comparatively more cases where it didn't stop.

Yes I think the implication didn’t fully materialize in his head. If he stop to think about it he doesn’t actually have any other cases.

Which sort of explains why there’s no subsequent response.

> it was too hard at such a massive scale... of 858 TB

There are probably many S3 buckets in existence that are bigger than that.

Not saying that they should've used S3, but it's definitely possible configure multi-regional backup (and a government can afford it).

My home theater setup has more storage than that.

It doesn't change the fact that the workflow gp explains is a duct taped construct.

It's hyperscalable and highly available today, until the API changes.