I think the AI community is sleeping hard on proper symbolic recursion. The computer has gigabytes of very accurate "context" available if you start stacking frames. Any strategy that happens inside token space will never scale the same way.
Depth first, slow turtle recursion is likely the best way to reason through the hardest problems. It's also much more efficient compared to things that look more like breadth first search (gas town).
> The second approach is to just write to the terminal like any CLI program, appending content to the scrollback buffer
This is how I prototyped all of mine. Console.Write[Line].
I am currently polishing up one of the prototypes with WinForms (.NET10) & WebView2. Building something that looks like a WhatsApp conversation in basic winforms is a lot of work. This takes about 60 seconds in a web view.
I am not too concerned about cross platform because a vast majority of my users will be on windows when they'd want to use this tool.
If you use WPF you can have the Mica backdrop underneath your WebView2 content and set the WebView2 to have transparent background color, which looks nice and a little more native, fyi. Though if you're doing more than just showing the WebView maybe isn't a choice to switch.
I like the idea of using a transparent background in the webview. That would compose really well.
The primary motivation for winforms was getting easy access to OS-native multiline input controls, clipboard, audio, image handling, etc. I could have just put kestrel in the console app and served it as a pure web app, but this is a bit more clunky from a UX perspective (separate browser window, permissions, etc.).
> The wafer isolation method in Tesla TeraFab involves containing individual wafers within protective enclosures throughout the fabrication process to shield them from airborne particles and contaminants.
This seems kind of redundant. I would just use flexbox for something like this. Grid is already an extremely rare item for me. I only ever use it when I need to control the overall layout for an app that has to work on a wide range of viewports. I'd never use grid just because it can do clever brickwork.
You very much cannot use flexbox for this. The whole point of these gridlanes is that not only can you have elements that automatically move across lanes when resizing the container, you can also have elements spanning multiple gridlanes and also fix the positions of elements in the grid, something wholly impossible in pure CSS flexbox. They link to the article[1] that even describes all the use cases this covers right below the first image.
Grid covers a lot of very subtle use cases that have historically required hacks like a list of select options where some can have icons on the left and some don't. You just need a subgrid that will automatically position every element in the select correctly to align them, regardless of whether there is an icon or not within the element in all select items. Previously you'd have to add a fixed width padding to the left and check if all the select items had icons. It also correctly scales the width and height of a row of items like cards where you want to ensure the alignment of headers, content, image etc depending on if that stuff is in there or not. You can have text missing and the card will still take up that size because your subgrid has defined it so. All of this needed JS, complex CSS hacks and so on. These aren't obscure features these are commonly used layouts that required a lot of time and effort to make it look nice.
I don't actually see what's stopping European firms from figuring this out in terms of hardware infrastructure.
Buildings full of computers aren't that difficult a problem to solve compared to things like semiconductor manufacturing or energy.
Perhaps the issue is more on the software and architecture side. Getting sucked into weird cloud products that don't translate clean to other premises is perhaps the more difficult aspect of this for larger firms. I've made a very strong point to only use EC2, Route53, S3 and Azure AD. Moving between environments is a lot easier when you stick with the VM as the unit of deployment. Getting out of something like a MSSQL hyper scale instance is simply not possible without switching to a different SQL provider or accepting new operational risks.
From a purely engineering perspective I think it becomes difficult to argue with the gas turbine once you get into the gigawatt class of data center. The amount of land required for this much solar is not to be understated. In many practical scenarios the solar array would need to be located a distance away from the actual data center. This implies transmission infrastructure which is often the hardest part of any electrical engineering project. You can put a gigawatt of N+1 generation on a 50 acre site with gas. It's dispatchable 24/7/365 and you can store energy for pennies on the dollar at incredible scale.
Having both forms of generation available at the same time is the best solution. Once you put a data center on the grid you can mix the fuel however you want upstream. This should be the ultimate goal and I believe it is for all current AI projects. I am not aware of any data center builds that intend to operate on parking lot generators indefinitely.
For inference you don’t need gpus to be clustered together as much (generally training has lots of synchronisation steps so you can be bottlenecked on that instead of ‘real’ work) as they can handle separate tasks in parallel. But maybe other economies of scale still make you want to put them together (and therefore on average further from the power).
I guess there was a bit of thought about transmission with the reference to high voltages. Another interesting thing: batteries allow you to reduce the needed capacity for transmission lines – if you have batteries near generation and then transmit power at a lower maximum, same average rate than if you only have batteries near use, you can more efficiently use the available transmission.
I guess the main reason for gas to be a problem is if you can’t get new generation (eg lack of turbines).
Sadly, I agree until we get SMRs (I think we are few years off). Obviously it would be more ideal to use grid+solar with curtailment but not super realistic.
For context, at the moment they hope to have them operating some time in the 2030s. That’s a best case, just like the cost estimates (which operating practically and safely may be more than what people are forecasting)
Not operating today like it sounds from the comment.
IBM is the undisputed king of backward compatibility. There is code running on mainframes right now that is going on 50 years old. Microsoft is a close #2 with windows.
I'd probably consider using IBM if it wasn't so goddamn weird and expensive. I suppose all that backward compatibility does have its downsides. Windows feels a bit weird in some places too, but at the same time it didn't start out life as a typewriter.
reply