I think I'm missing your point. I'm under the impression that electricity has never been easy or cheap to move very far, and it's not clear to me how solar-generated electricity is different in that respect.
You're mixing up two different things. Moving electricity with HVDC is easy and cheap... building the grid infrastructure is, well, not so easy or cheap.
The system cost depends on connection queues, transmission buildout, substations, transformers, planning permission, curtailment, etc.
Production of transformers, cables, and labour capacity are bottlenecks. Planning is a bottleneck.
On this very thread you already have people talking about "open weights" and similar nonsense. What is open about them? They're free to download, but that hardly qualifies as open. Where is the source? Where are the instructions to modify and build your own?
I'd never though I'd have to utter the expression "open as in beer".
The blatant attempt at manipulating vocabulary here is... quite blatant.
I'm a strong proponent of Open Source (TM) but I disagree with this take.
The weights are the useful artifact here. You can modify them, fine tune them and do what you want with them.
Unlike binary software there is nothing limiting that.
It is also useful to have access to the training recipes and to some extent the data. But I'm of the opinion that learning on something is not copyright infringement, so there are many circumstances where distributing the raw training data will not be possible.
For me this is like Open Office: it is open source, and largely inspired by and learned from Microsoft Office. But they don't need to distribute MS Office for Open Office to be Open Source.
In addition there are models that meet the criteria you appear to propose. The AllenAI models are a good example.
The analogy falls apart very quickly. Without the training data, your modifications amount to virtually nothing compared to what these "versions" are, and the idea that you can maintain and improve on these models without the continual support of the company that owns the training data AND harnesses AND in general build instructions is not very credible. This is why it's not rare that they "dump" old versions as freeware but at some point switch to not distributing them, and mostly get away with it. As this is really not open, and the threat of an effective fork is therefore non-existent, the pressure for any one who has released freeware models to "go SaaS" is too high.
While if "Open Office" switches to a more problematic license at some point, the existing source has all you need for an organization to support the project without regard to the original company (this has happened already!). If Qwen decides to stop distributing models for download, you're basically stuck, _even_ if you have unlimited resources, it's not clear how the released weights help you; your best bet is to start almost from scratch. This has also happened...
These models are not "Open" by any definition of the word. It is just freely redistributable. You can justify yourself in whatever way you want re a cowboy approach to copyright, but this doesn't change the fact that this is not open, and has almost none of the benefits of open, and therefore it is a huge abuse of the word "Open".
Ironically about the only thing that is copyrightable here is the sum of the training data (possibly) _AND_ the software used to build the model (most definitely). The model itself most likely isn't (databases are not copyrightable), which makes it even more pointless to abuse the word "open" for it. All the value is in the former two.
Jobs mischaracterized the innovation, and the author is technically correct (the best kind), but it's a shame that the piece appears to want to bury Holt's actual accomplishment. Holt's work was innovative in the same way that Woz's Disk II controller was. He didn’t invent the underlying technology, and he did create an elegant, product-defining implementation of a known (but difficult) technique.
Author here. Can you clarify what you consider Holt's innovative power supply accomplishment to be? I spent an excessive amount of time studying Holt's power supply to find out what made it special, but couldn't find anything particularly important.
Yeah it's nothing special. I find the whole Apple innovation thing quite tedious if I'm honest. I've seen some milspec stuff from around that era that purposely misused an LM723 as an SMPS. This was done without any fanfare at all even though it was quite interesting.
I wouldn't put Holt's power supply on the same pedestal as the Woz drive controller which I feel was actually innovative but perhaps what the angle they are looking at is just the apparent dearth of switched mode power supplies in the market for home computers, games etc at that time. If I think back to that time, Atari, Commodore, Radio Shack, TI etc - they all used heavy transformer bricks and linear regulators. I think the BBC microcomputer went to an Astec switch mode supply after a brick on launch and that was in 1982. Maybe it was this rise of Astec that you can somehow attribute to Apple and Rod Holt after a fashion.
I second your analysis, there's nothing special at all about it. By the standards of the time it was a unusual to see a switch-mode supply in a computer, but the supply itself isn't unique when compared to contemporary designs. Mix "unusual application of a known technology" with "Apple fanboyism" and you get "Holt revolutionized power supplies" or whatever the claim is.
https://charleswiltgen.github.io/Axiom/ – Suite of skills, agents, and tools that make general SOTA models actually good at building and/or auditing iOS/macOS apps. Built for myself initially, I FOSS'd it once I determined how generally helpful it was. It's helped me learn a lot about doing sophisicated things with LLMs in a token-efficient way.
https://charleswiltgen.github.io/TagLib-Wasm/ – Also built for myself initially, I FOSS'd it because there was nothing like Mutagen for TypeScript/JavaScript runtimes. (I don't dislike Python, but think it's a bit of a mess.) This was my first serious project to leverage LLMs for coding.
https://pwascore.com/ – Built because I wanted to quantify how bad Safari was at PWAs. Learned that, objectively, Safari is as bad as PWAs as Firefox (which is to say, not terrible, and not to blame for why PWAs continue to be mostly-irrelevant).
Apple presenters are coached on how to speak, how to stand/move, what to do with their hands, etc.
I can understand how it might seem culty, but it's in the service of clear communication to a global audience. Anyone who represents a company to important customers and/or the public goes through similar media training.
The comment is about how everyone in their videos does it. The over-use of it is the issue, like when you say a word too much and your brain stops understanding what it means.
In my experience, LLMs are great at reviewing changelogs for potential gaps from a user POV (and even creating draft changelogs wholesale, if you're backfilling) based on git history.
> So doubling isn't because it is doing above everyone's expectations, it is because Apple underestimated the demand.
Clearly it's doing above their expectations, and they had precise data in the form of their test selling the M1 Macbook Air at $599 (occasionally $499) since 2024. It's too bad you weren't at Apple so they could've avoided this mistake!
reply