zitterbewegung

Would think they would be doing it for their own hardware device for chatgpt not for developers.

I was studying Erdos problems by only taking ChatGPT 5.5 outputs and just asking it to keep on attempting to solve it by asking it to go further. I haven't started doing this with chatgpt 5.6 I have some partial results here https://chatgpt.com/g/g-p-69f03400f420819192418b18ca90ffee-d...

What was really interesting is that during the process it was able to find lemmas or theorems that might be related or relevant to be published.

While I was doing that I was also trying to use Aristotle to do the Lean formalization and I have a WIP system to do that at https://github.com/aconsapart/thesisus/

FWIW In the Apple ecosystem this usually gets filled in automatically based on the source.

I don't understand the code in the examples after simply reading it. That's a problem. You need to have comments to clarify or pick a better example of code.

You are in competition with Python and Javascript which has lots of code for training data. I wouldn't touch esoteric features unless it improves readability or is under the hood. The bigger problem with esoteric features is that its hard for humans in general to understand unless you cover it with enough syntactic sugar for people to write it down.

Xcode 27 which is in beta ships a much better codex integration. There are problems in Xcode that are frustrating . Also, git integrations are a nice touch that allow you to force push too. Would be interesting to find out how well this works in a team based setting.

I am working on a Jupyter notebook client for VisionOS. It allows for 3D data to be visualized on visionOS. Right now it supports point cloud data, USDZ models and Gaussian Splats. I am working on it to launch on the App Store. Sign up for more information at http://www.pulto.org

I love Claude Code and I don't use the rest of the models since I use ChatGPT for productivity work it. Fable is pretty great and the UI / UX is much better than codex

In America for my lifetime I have never been able to get fiber and it’s because America is too large and I live in an affluent suburb.

There is one simple thing you have to realize why Python is the optimal choice. You have so much training data. Python is the second most popular language on GitHub and is easy to read.

I agree with you on every point but it is interesting to see real world benchmarks like this. Showing the standard benchmarks that all LLMs use is not only boring but at this point likely gamed or even has issues (according to OpenAI) by every LLM.

The only thing I can possibly think of is that they could use it internally at possibly a lower cost and offer it to people who have a Tesla cheaply. Owning Cursor might help for integration or data collection.

Xbox around 2021 had around a 12% profit margin and the gaming industry as a whole was around 17-22% . In 2023 the target for the division was put to 30% . We see this new restructuring because the target was put this high. Microsoft really wanted Game Pass to be a steam competitor which is pretty much what everyone in the industry tries to do and fails. The push for Game Pass prices to be higher was to get the 30% margin and that didn't work out. They aren't operating at a loss they are operating at a goal and they failed the goal. From other child comments many studios they bought probably were below average. We can see this restructuring basically is that they failed the target, the old guard went out as the new guard came in.

https://www.bloomberg.com/news/articles/2025-10-23/microsoft...

This looks cool but this should be renamed without having Claude in the name.

Sort of interesting license not sure if anyone will do it long term.

The training data and the Apertus LLM may contain or generate information that directly or indirectly refers to an identifiable individual (Personal Data). You process Personal Data as independent controller in accordance with applicable data protection law. SNAI will regularly provide a file with hash values for download which you can apply as an output filter to your use of our Apertus LLM. The file reflects data protection deletion requests which have been addressed to SNAI as the developer of the Apertus LLM. It allows you to remove Personal Data contained in the model output. We strongly advise downloading and applying this output filter from SNAI every six months following the release of the model.

RIP Legend.

F.

Such a good composer for all of those games I can remember those songs still.

NASA always needs more competition to keep launch costs low and encouraging innovation and it seems like he hasn't been CEO for a long time. This is indicative of funding competition which is a good thing.

It is much better. Imagine if the whole Manhattan project could have been outsourced and costs you nothing. I expect in a short time that open source models will be almost or almost parity by 2030 and running on consumer devices.

I’m sorry but some of the comments are out of touch. Apple devices do not have any intent of supporting separate OSes . Asahi supports M1 to M2. I can see this as a PSA to not install a beta but I am confused who would install both Akashi and macOS 27 beta at the same time when you could run the beta in a VM for development ? Others have said that this has been a fix that will happen soon.

Much easier to get a moon mission due to politics .

Sort of expected this at the first attempt. Use their existing partnership for Google being the default search with Google and leverage Google's models.

How do I Google for this language……

I bet it will be something like a MacBook Pro but with a windows logo. I wouldn’t expect much different considering the other surfaces look practically the same including the keyboard . Not that that is a bad thing .

You can have a security solution but with large ecosystems like this it can’t be pushed to the ecosystem immediately and everyone will take longer to test and deploy.

Right now you could audit packages and make sure you don’t get the latest version

I use AI to generate diagrams but not having the time to just clean up the diagram is pretty bad.... at least take the time to do that.

Why are they saying to not distribute on YouTube they just want to give an indicator. Same with labeling if a video is an AD. I find some of the obvious AI content to be funny or informative .

Advancement in AI research seems to be the only thing at this point.

I haven’t had an issue using Python with LLMs where I have to decide “Should one use pip, poetry, or uv?” Since there is enough training data using pip or just choose that since it is the most boring solution and many of the commands map to uv since uv has a superset of features. Not that go is a bad solution honestly I would just say use what you know best.

This is true but the point of YC is when that they will fund things that can fail it’s why VC exists .

Why would I use this instead of using iMessage for Business that is the official way and is more robust and doesn’t violate ToS. If you get shut down I have to redo this setup using Apple ?

At least in my experience depending on how serious the company was about security plugin downloading would be prohibited.