Hydroelectric generation varies a lot annually due to naturally varying precipitation. In the 1990s, hydroelectric generation reached a maximum in 1997 with 356 TWh (this was also the all-time peak) and a minimum in 1992 with 253 TWh:
In 2025, hydroelectric power generation was 247 terawatt hours. Utility scale solar generation was 295 terawatt hours. Estimated small scale (e.g. rooftop) solar generation was 93 terawatt hours. Combined, that would bring solar generation to 383 terawatt hours for the year.
Solar generation is now higher than hydroelectric generation ever was in the US, but it's also true that hydroelectric generation is lower in recent years. The Ars Technica article focused on the utility scale solar generation. Counting that portion alone, solar generation has yet to surpass hydro's 1997 record of 356 TWh.
Reminds me of the story of Henry Ford, then an elderly man proudly showing off the first two robots on the assembly line to the representative of the automotive union.
To which the representative asked how many cars the robots will buy.
I've for decades have opted to use pvc conduit for underground installations. This way I can pull/replace wire. Reading this made me realize that putting fiber inside conduit in inaccessible areas would be a safe way to go.
Years ago some electronics sensors were replaced and started having problems. I came to realize it was a voltage drop with the new increased load. I was able to pull in new larger wire by using the old wire as the puller.
I often blacklist sites that cover content with unremovable ads or has unrelenting ads. They need a clear button that acknowledges I've seen it and to stop annoying me.
I've started using SQLite years ago for its use of a single file to store data. Along the way, I found more reasons to like it. Today, I discovered a new reason when I read this article.
I also noticed many software companies have been switching to SQLite for storing data. Audacity has switched from using hundreds of tiny audio files in multiple folders for a single recording to a single file using SQLite. I suspect it is the reliability that comes from sticking to C
Hmm
Could Jensen's statement be motivated that by forcing companies to build AI infrastructure would ensure more of his Nvidia AI chips to be sold? I keep seeing him say things that makes him appear bias towards more sales. Like dismissing any possibility that an AI bubble is growing. To admit the possibility may hurt sales.
I guess Micro$oft has so much invested into their AI that they're trying to insure a return on that investment by pushing it onto everyone.
I read this morning that their AI i to provide free heating for 6,000 nearby homes with its heat waste. So to use M$ AI is to do 'your good deed' and subsidize someone's heating.
In defense of Firefox, there is a reason for the iOS Firefox browser being so bad. Apple mandates all other browsers to use the WebKit rendering engine instead of their own. Firefox isn't able to use the Gecko engine. I guess Apple is afraid that others will show up the Safari browser.
Protecting your anonymity isn't a sign that you're hiding something because you have done something wrong. It is because you're cautious. Even the most benign information can be use against you. European Jews in the early 1900s gave census takers the info yhat they were Jews. Decades later that info was collated with punch card machines by the Nazis to build family trees to hunt them down. This is an example of how "innocent" info can be weaponized in ways you would never dream of.
This is also happening right now with ICE combing through the various federal databases that they previously didn't have access to.
But the worst thing is that the sheer amounts of compute available to even fairly small organizations, never mind governments, means that it's so easy to data mine stuff like that. And now there's AI to do the same with unstructured data. And meanwhile our corporate overlords are happy to profit from it: https://news.ycombinator.com/item?id=46349281
reply