This is classic technological disruption. I can ask ChatGPT the most mundane questions, while it is very high effort to ask questions on Stack Overflow. Nobody really wants to RTFM. ChatGPT hallucinates, but many Stack Overflow results are of questionable quality as well. One major benefit I see with Stack Overflow is that people can score and discuss the answers and can specify alternatives whereas ChatGPT does not have a way of moderating results.
> I think we are at the point where Patreon & podcasts can keep the best researchers self-funded...
What I believe you are saying is that the "popular" researchers will get ad-spend to fund their "research" that won't be peer reviewed. Why even bother publishing research, if no one reads anymore? It would just devolve into a popularity contest and following trends. Those trends will just be co-opted by monied interests.
The esoterism is due to the fact that there is a body of research that you need to know to understand the new research. Just because you can't understand the topic in a short sound bite does not mean it is not worth researching. Not all of the research is intended to be consumed by a lay public either.
Many podcasts and Patreon exclusives are behind paywalls and there is no expectation of peer-review.
In regards to calling this a "college model", not all research is done at college there is also thinktanks (institutions) and industry research which are funded by governments as well.
I think governments should be accountable for making sure the research is rigorous, has a social benefit, and is publicly available.
You are quite correct that Substack will favor the popular, not the best. But universities will favor what is popular too. Just popular with the different audience who controls university budgets instead of the general public. And how can governments do any better? The way they are held accountable is an election, or in other words a popularity contest.
I don't see why my tax dollars need to fund Proust studies, nor Elvis and Hip Hop researchers for that matter. It's all for elites to feel like they are doing something useful, "research", that no one would ever voluntarily fund otherwise. Or if they would, they should find a way to get paid voluntarily.
I know this sounds judgmental, but this reminds me of the idiom “touch grass”. Children should be outdoors observing real life and not be consuming AI slop. You are not overthinking this, this will most likely be bad for children and everyone in the long run.
This is the classic innovator's dilemma. Any interesting new innovation from Intel gets snuffed out due to the fact it takes time to get to market. The success of x86 was blinding. Intel was impatient, so it killed products in their infancy (which leads to customer mistrust).
Intel did a great job coming up with great ideas before everyone else, then killing them off once someone decided they weren't going to immediately make 60% gross margins. I'm constantly amazed at how quickly the company managed to throw away good ideas because it could only make 40% gross margin, or was going to take another few years to develop a market.
You could argue Microsoft had the same disease. Ultimately, it's not as terminal because turning around a software-sized cruise ship is alot easier than turning around a hardware one.
Google can be refocused in a 4 years. Intel, at minimum, seems to need a decade.
Microsoft has a worse case of the same problem but greater monopoly force. Entirely dependent on the captive markets but rapidly snatching up everything they can to enforce it on other markets.
That said Google could be doing worse. They at least have something around phones and AI. They're trying to attack some new markets. Google was structured so shareholders have no power but share price is still something that is going to drive decisions and hasn't been that amazing recently.
Intel has actually had many quarters of declining revenue, Google has been far more consistent in the last 15 years. Intel has had over 20 quarters where revenue declined since 2010 and they are spread out across 2010-2024, while Google only lost revenue in two quarters.
I would argue that Intel's poor performance is quite detached from Google's consistently exceptional performance. Heck, they aren't even in the same business (primarily hardware vs. primarily media (advertising) and software).
Google is:
- A member of the smartphone app store duopoly
- A member of the cloud compute big 3
- Nearly a pure monopoly in Internet search in Western countries
- Has a higher office productivity software marketshare than Microsoft Office
- Internet cable TV provider with the highest subscriber count
I'd like to understand how it "hasn't been that amazing recently."
YTD S&P 500 has a 26.9% return and GOOG has 25.7%.
Intel has been around since 1968 and Google since 1998. As you say they're also in a different business. But Intel has executed pretty well up to maybe 2007 so over about 50 years. vs. Google's 26 years of existence.
Most of your bullet points aren't anything new, this is Google riding on work that's been done a while back. The question is what will be the new bullet points and whether it will be able to stave competition eating into its existing bullet points. The pressure to get the stock price higher is going to take its toll as the bean counters try to artificially inflate it.
I’m not sure what less than one percent difference from the S&P 500 is supposed to prove good or bad. The S&P 500 is heavily weighted toward Alphabet, it literally is a big part of the S&P 500 itself.
And, you know, Exxon Mobil hasn’t really made a new product in quite some time but they’re doing just fine.
Still, I’d say that Google has had plenty of new products and bets that it has been making, along with great recent performance in many ways.
They’re finally hitting their stride with consumer hardware, with the FitBit acquisition clearly boosting their wearables business and selling a lot more pixel watches than any previous smartwatch effort.
Similarly, Pixel phones themselves have jumped from around 4% marketshare to 8-12% marketshare in the US just this year. Success in that arena is very new to Google, so I would call that a long term investment that is paying off.
Google is investing in custom Tensor processors which is also a recent product launch for them, starting in 2020.
Stadia was a new product that obviously failed in the end but did represent a decent technical achievement and major bet on a new product.
YouTube TV isn’t that old, either, it launched 2017, and being the top subscribed digital cable product is a big achievement.
The elephant in the room for newest major product launch is Bard AI and Gemini. It’s fair to say AI is less than proven and that these may be copycat products, but it’s also true that it’s going to be an area where your big competitors are basically Microsoft, Apple, and Google with little room for anyone else.
And let’s not forget that large companies can boost their value proposition and mitigate competitive threats through acquisitions. Even IBM’s lost decade was overcome, and a lot of it was through acquisitions and business unit spinoffs and closures that helped modernize their offerings (companies like Red Hat and Hashicorp).
So what you're telling me is that I should be listening to you, a random person on the Internet, and expect that one of the most valuable companies in the world with decades of consistent revenue growth, a monopoly in internet search, duopoly in smartphone operating systems/app stores, one of the big 3 social media companies, one of the big 3 cloud compute companies, one of the big 3 AI companies, and the owner of the digital cable TV provider with the highest subscriber count is just going to magically not be successful in the future?
You might as well be telling me that you think that Saudi Aramco is going to go under because they need to fix their "culture."
Culture is not important for their revenue, most of the company is not. The ads business is their cash cow which finances the rest of the company. They can have horrible culture and mismanagement everywhere and revenue will still grow - if they don‘t fuck up their teams working on ads.
> The management want return to office and that's it.
I believe this is the truth. I worked for a startup 20 years ago and the founder told us all to come in early in the morning and look as busy as possible, because key investors were touring the office. He told us that they wanted to see where their money was going and we need to play our part.
Normally we worked nights and weekends at home and in the office, but the management class actually want to see us sweat IRL.
Another part of this is the management class has a herd mentality. If they see another dominant manager with some modicum of success do something, they will mindlessly copy it.
Is your argument that organizations should not have a work place anti-discrimination policy and enforcement, because proponents of such policies are all bad people and we should just trust in inherent human kindness? Are you advocating that people do not speak up about discrimination? I am not sure what you are advocating?
spicymaki's comment is just pointing out how some people who talk big about equality can be pretty crappy themselves, which kinda makes it hard to take them seriously.
We definitely need rules against discrimination at work and ways to deal with it when it happens. People should absolutely speak up if they see or experience discrimination.
But the thing is, real kindness and respect should be at the heart of all this. We need both the rules and people actually being decent to each other. It's not one or the other, it's both; hence the false dichotomy.
I'm not sure how much clearer I can be about it: The more loudly someone complains about racism, the more likely they are racist. Likewise the other -ists and -isms.
This article was very educational. I did not know that Martin Luther King Sr. was such a prominent person, and the connection of the King family with Berlin. Thank you for posting it.
They would if they could. Optical physics is one factor that prevents the lenses from being flush with the back. If you want to support advanced optical features the lenses need depth.
Thats the thing, I don’t care about the optical features they rolled out. I have an interchangeable lens camera that is the tool for that job, 10 years old and still works great.
I already own a copy! I was just wondering whether it was against HN guidelines to post the entirety of a copyrighted work. Perhaps I should have been more direct.
reply