That is still cheaper than the index when it came out, and it sounds like a general improvement in all areas. Flagship vr for less than the cost of the latest smartphone seems pretty reasonable given how low adoption is.
Yep I have this and it's total garbage. The scheduling options feel designed to troll, serious MacBook wheel vibes. No way to temporarily disable the schedule if you're going out of town. Either turn your HVAC off entirely, or delete the entire schedule and manually reenter the whole thing when you get back (who doesn't love life-wasting menu diving after a long trip?).
Yeah. It’s awful. And there’s no reason they couldn’t have given us one last firmware update that just at bare minimum allows us to use it with HomeKit or Alexa or whatever else.
Vaguely projecting forward the relationship between total cumulative funding and experimentally achieved q value, it's not unreasonable to think that this money could have brought us to commercially viable fusion power if spent judiciously over a longer period of time. It's pretty typical that we collectively spent the money on an orgy of waste instead.
This could be useful for undergrads getting their feet wet with LaTeX and the world of publishing. However, I'm very skeptical that the options are broad enough to conform to the exacting style guidelines typical of PhD theses. Usually people use a template from their school which already formats everything, autogenerates the table of contents, etc.
In fact it isn't clear what this system offers over a LaTeX document started from a template and hosted privately on GitHub. The CI pipeline can even be set up to compile the document for the people who want to be able to do everything in a web browser.
LaTeX was much easier with Overleaf for my PhD thesis. I still recommend that for friends starting a thesis or a book project. I even used it for recent book project with a friend.
As you noted, one needs a lot of fine tuning to meet publication rules & guidelines. Compared to a local LaTeX editor or Overleaf, this looks too generic to meet the needs I've had in the past. Sure, LaTeX can require a lot of tinkering, but PhD students ought be able to figure it out for themselves, whether through documentation, forums, or asking labmates.
I wouldn’t, solely because it’s still in version 0.X - for any long-term, important project (e.g., PhD dissertation) I’d recommend LaTeX due to A) it’s mature and B) many universities provide LaTeX templates.
This isn't exactly what you want, but you could get the same kind of breaker interlock used for traditional generators and use that for the solar inverter. The downside is that you can't blend grid and battery power at the same time, but this may not actually be a problem in a practically sized setup where solar is the primary.
It doesn't matter whether it's possible for a highly motivated and educated person to distinguish the lemons. This will usually be true. The important thing is the experience of real consumers participating in the market as it actually exists. A better framing would be "of all products in category X sold last month on amazon, what fraction were lemons?"
> It doesn't matter whether it's possible for a highly motivated and educated person to distinguish the lemons.
It does though, since the entire premise of the "lemon market" idea is that buyers can't differentiate between good and bad products. What matters is where the bar is to be a good differentiator, and how many people can cross it.
It also doesn't necessarily matter how many lemons are for sale, if the good ones are easy to find. Imagine that there are 1 million trashy sleep masks on Amazon and only 3 good ones, but the 3 good ones are right at the top, with superior ratings and reviews.
The better question is: what % of buyers are buying lemons?
I work with some undergrads and see this delegation increasing year over year. Unfortunately it's also happening at the expense of reading books, using library search tools to find proper sources, and information gathering in general.
"An LLM might be able to explain something to you, but it can never understand it for you."
Spreadsheets occupy a weird position in many organizations. They are "functional" programs, but most people who make them have no concept of how "functional programming" compares to other paradigms, nor do they need to.
Spreadsheets just work, perhaps so well that they have become victims of their own success. They are such effective tools that they are lumped in with other user facing office productivity software, and as a result they seem to be overlooked by more technical people with programming skills in other languages. I always thought the spreadsheet format could be expanded and improved to be able to handle more use cases and behave more like a normal program, but there has been shockingly little effort put into trying this out. At one point there was something that acted like a spreadsheet where any cell could be a python expression, but I think it's defunct now.
There will be no innovation from Microsoft on this front. We should really decouple the spreadsheet concept from excel the product.
it is already decoupled, lots of products have been been using .xlsx (format) as data interchange/template driven inputs for a long time. Libraries in almost all programming languages to create .xlsx with formulas/formatting etc have been around for a long time
Garbage software still makes lots of money. Sometimes more than good software, and certainly more than the best software. Just look at the windows vs Linux ecosystem for a simple task like screen recording with decent compression. Open source tools are superior, but can be harder to even find on windows due to the poorer signal/noise ratio from all the terrible for profit/enshittification software. Some graphics drivers have dashboards that do screen recording, but that is not a universal solution.
The most viable free options on windows seem to be sketchy cloud stuff designed to be inconvenient enough to upsell you. On Linux it's either built in already or trivial to install something that records locally and doesn't rug pull the user demanding money.
reply