HN user

hansword

782 karma

made some money with bitcoin a decade ago, but I am not the 'invest' type, I am more the 'spend and share' type, so it's all gone. I am very happy to have no connection to the crypto scene anymore. :)

I believe the platforms are the scourge of the web.

Posts1
Comments169
View on HN
[dead] 4 years ago

Some friends in the US are really non-ironically worried about the country coming apart within the next few years.

You are missing the central point of the Walkie-Talkie and Disney-Concert-Hall controversies. These are Frank-Gehry(-style) buildings, and in those roughly half of all surfaces is a concave and half is a convex mirror. THIS concentration to one point is what turns the summer sun into death rays.

Reflected sun, by definition, has a little less energy than direct sun. So, given the mirror walls are straight, as the designs show, you get (at maximum) 200% of normal desert sun energy, which isn't comfortable, but not a Gehry-style death ray.

You are right. And I do understand why he does it. I can see his side.

But I sit on the other side. I hate it and will tell everyone how much I hate it.

--- Edit:

Or, if you prefer, in market terms: Because I am not alone, there is a market for non-clickbaity titles, non-exaggerated thumbs, and there are channels that cater to this market.

I want to very much second excluding Veritasium. Whenever he talked about something I was knowledgeable in, he made mistakes. This combined with branding himself as 'the element of truth' and the increasing over-clickbaity titles and thumbs made me stop watching him.

It's an entertainment channel, and if you want light science entertainment, and expect to forget everything within a few hours, it's a good channel. If you want a science/learning/education channel, it isn't.

Think about how the hosts.txt file was originally used before the DNS. It's closer to that (with a lot of extra sophistication and 40+ year learning)

There is another consideration that people on YouTube ignore, but Apple doesn't.

You don't want the underside of the computer to reach substantially over 40deg for a considerable amount of time. This is because some people put their laptop on their lap and if it is too hot, it starts slowly cooking the leg. This has happened before, and thus nowadays, all new computers make sure the underside doesn't exceed 40deg for longer periods.

So I wouldn't be surprised if the people on YouTube traded performance for on-lap-safety. Apple can't, for obvious reasons.

StretchText 4 years ago

The way this is described it reminds me of the <details> and <summary> tags in HTML, most often used for spoilers.

But I am not sure if I understood it correctly. This might be more for different level of comprehension (eli5?).

Why is this titled "Advice for New Developers"? The actual blog post is called "The best investment".

A proper title for HN, imo, would be "A personal fitness trainer is the best investment".

I wouldn't have clicked on that one and saved myself from reading an article that is irrelevant to me.

I would say it helps to first think what you want to get out of it.

If your task is "show me something that breaks through our hyperspeed media", then I guess some obscure museum is a better place than an ML model.

If your task is "find the best variation on theme X" or "quick draft visualization", they are often very useful. I am sure there will be many further tasks to which current and future models will be well suited. They are not magic picture machines. At least not yet.

I had the same problem, was a size cutoff caused by me setting the scaling to 200% or something. Once I reset scaling, date is back.

Hope it is that easy for you, and it isn't a bigger problem.

You seem to have completely missed that NNs run on physical hardware.

A contemporary CPU and/or GPU is 'a bunch of connections that are pre-encoded' - just not in DNA, but in silicon.

Is there any chance

There is a chance.

To get a better answer, you have to specify what exactly you mean by 'sampled' and by 'without context local to the code in question'.

Neuro-scientists and cognitive scientists and developmental psychologists and linguists spent the last century, give or take, looking at this question.

The most basic, most broken down, most simplified answer is what is called "poverty of stimulus".

A NN has an abundance of stimuli, orders of magnitude greater than any biological being could ever possibly take in. Finding enough correlations (enough for productivity) in such abundance is not actually that surprising. What is surprising is kids being competent at speech after less than 2000 days of stimuli, often really low-quality stimuli.