Hacker Newsnew | past | comments | ask | show | jobs | submit | layout's commentslogin

> Now he's making less because they're directing less

No. There are less visits to his site from Google because they’re directly displaying information from his website on their page.

That is not an accurate statement.


> But I think it's more complicated than tech companies taking away work through greed

When technology companies choose to take others information without compensation, what do you call it?

> Some small publisher likely felt the same a few years ago

Your analogy has no relationship to this situation. Unless you’re alleging small publishers, Work was stolen by websites and copied.


The website in question is not the originator of any information. It is a collection and a summarization of bureaucratic rules and laws. The man behind the website does not own the information he has re-published there.

Where did I say that the website was the originator? What does that have to do with this discussion?

Clearly, the collection and summarization of bureaucratic rules and laws is valuable, if Google is taking what he created and summarizing it?


>take others information

>Work was stolen

You have to own something before you can say that other people stole it. If Google or anybody are copying parts from the website verbatim, yes that is breaking copyright. But taking info from a source or several and summarizing isn't stealing. Then the person behind All About Berlin also stole his information from government regulations, laws, etc.

It would be very easy for him (and everybody like him) to make an honest living from his website traffic, and if I remember right I've talked with him about it here on HN the last time a similar article was on the front page. Here's how you do it:

Keep offering all this information for free to your website visitors. But offer to take care of the problems they're looking to solve for a fee. There are tons and tons of people who appreciate such services. And having the information available on your website proves that you're reliable and know what you're talking about. It's a win-win-win situation for everybody.

Most people aren't like hackers looking to solve everything by themselves and spend unlimited time to save an infinitesimal amount of money. A lot are very happy to pay an expert to make the problem go away.


> You have to own something before you can say that other people stole it

We own our labour. In this case, the AAB’s author spent many hours discovering and organising that information. If the AI could have done this construction from primary sources then have at it. But if, as others have said, the AI’s information goes stale because the actual originators are disincentivised, what then?


When it comes to information, we do not own our labour. At most we can own copyright, trademarks, and original (patented) ideas. Working hard to make a user friendly collection of information from other sources is neither of these.

If you put information out on the public web and make it available for free, then you can't turn around and say that "actually I own that information". It's too late.

Even before AIs, you always had the risk that somebody would just steal the content of your website and make a copy or a business from it. Word it a bit differently for copyright's sake. This is not new.

I'm in the same boat as OP, but what I've noticed is that the more information you put out for free and the more detailed you are, the more customers you get. I thought that AI would have taken my job by now, and maybe (probably?) it will in the near future. But so far it hasn't happened.

What won't survive is making a living on website ad revenue. That ship has sailed.


> If you put information out on the public web and make it available for free, then you can't turn around and say that "actually I own that information"

That all hinges on what we mean by the word “own”. Unlike a material product, information can exist in multiple places at once. Copyright is how we define “ownership rights” to information. The copyright holder owns the rights to distribution and the terms under which it is published. Copyright is automatic the moment something is created and does not disappear just because it’s been made freely available for human consumption. Licensing terms can still limit or constrain downstream use and re-distribution, subject to “fair use” provisions.

> Even before AIs, you always had the risk that somebody would just steal the content of your website and make a copy or a business from it

I’m glad you acknowledge it as theft. Irrespective of the ease with which it can occur, that is exactly what it is. If AI companies want to protect their IP rights it would behoove them to have made at least some a-priority attempt at licensing or some attempt at restitution.

Those that live by the sword will also die[0] by the sword. I fully expect that leaks and/or exfiltration of model weights from frontier labs will occur. I for one, look forward to it.

[0] or at the very least suffer a few deep cuts.


I have no problem acknowledging copyright, and if an AI is breaking copyright then that is grounds for legal action. But you can't copyright information, facts. Especially facts that you yourself have taken from somewhere else, such as in the case of All About Berlin.

> I’m glad you acknowledge it as theft.

I should have phrased it better. But yeah, somebody can take the crucial information which you have worked hard to find out and republish it. There are thousands and thousands of newspapers, and all of them are quoting news which somebody else found out originally. Most of the times without giving any credit to the source.


> expectations were tepid to begin with

Tesla projected selling 250k of these a year or $15-20b in revenue per year (20%+ of vehicle revenue for Tesla). This is a big number and would’ve added significant total vehicle sales. Instead, they’ve sold less than 70,000 total trucks over its production history. That’s why it’s a rounding error.

This is an absolute disaster and failure by any rational analysis. Its failure helped puncture “Tesla sales will continue to increase” and stock price justification storyline.


Let's not throw the 250k /yr number around like it was an official company projection. It was a top-end pie in the sky number looking at pickup sales in the US. Even at it's peak, the designed and installed line could only output a maximum of 125,000 units annually. Wild for you to claim "rational analysis", when there are actual facts showing that it didn't perform as well as expected but you decided to hand pick numbers like you're Fred Lambert.

The CEO actually gave a range of 250k - 500k. I get it now. We’re not supposed to take the statements of the CEO of the company seriously, got it.

Makes sense. The CEO of the company statements are only pie in the sky claims and not based in reality in any way shape or form.


> It didn't fail

Tesla forecasted selling 250,000 trucks per year by 2025

Actual sales

2024 - 38,965

2025 - 20,237

2026 to date - 11,586

This includes trucks that have been “sold” to other Musk companies.

This product was a complete failure, making the Ford Edsel look like a roaring success


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: