Docker runs better on Linux. And Linux doesn't reboot unless told to do so. And you can SSH into it rather than run some sort of long winded GUI. Oh, and it doesn't phone home all the time with 'telemetry'.
Kinda cuts into their revenue streams if they can't release anything better/faster/stronger.
Also for all the hype for AGI, or whatever the current tag is, it is getting stopped dead in it's track if they can't go beyond private release to the public.
And if they do release these to the public, are the releases going to be so brain dead due to brain-blocks, is it going to be worth it?
What is the whole point, then, of making these models bigger/better?
I wish these articles would provide some idea as to the development environments. Are these captive large C++ environments, or general 'everyone does this' web development with React, or something in between?
You could try running notepad++ in wine, the windows emulator
But I suppose that doesn't really answer your question. Linux is/was a compendium of a bunch of little tools you string together. Many/Most/All of the items in Notepad++ can be found as smaller independent tools.
The real addicts will use console tools like VIM or EMACS to do their editing (not really GUI, but powerful).
As an alternate alternate, try using Visual Studio Code (VSC). Not quite as light weight as Notepad++, but very functional.
On Linux, I'll use KWrite to do my basic editing. I'll revert to VSC when heavy editing is required.
I'm not a Rust programmer, yet, but this is enticing. This appears to be building a language to be compiled with LLVM as a standalone executable.
However, in C++ there is a mechanism via libraries such as Boost::Spirit where a parser can be built inline and compiled inline (no separate compilation step) and used within the C++ application on input streams.
Is this a capability available within the Rust ecosystem?
Will these intro AI systems then mature to be senior devs who can then mentor more junior AIs? Then we won't need any devs? Isn't that the end goal, AI runs the company and we can all go fishing?
To go along with this, the ACM has a recent article on Automatically Translating C to Rust. It gets into the challenges of 'understanding code and structure' so that the end result reflects the intent of the code, not the actual execution paths.
Is there an easier way to get around all the complicated cookie selection? I don't care if they have 183 trackers. Do I need all those? Are the important to me? I suppose they are important to them. Isn't there just a 'no to all' or at least a 'just the bare minimum for state management'?
'The trade, dubbed a “box spread,” carried a kind of mystique. By combining two opposing options positions — one bullish, one bearish — Yang built a strategy that mimics a fixed-rate loan: upfront cash now, repayment at a set date, and a locked-in cost in between.'
Deny All, Accept All, but I never (except in a handful of cases) see the Accept Required. Let us admit that there are cookies required for maintaining state within a web site and account.
I default to Deny All, but click on Accept Required when I see it (trusting that it does do what it says it does)
I guess training LLMs on works of fiction/sci-fi would not be of net benefit. No distinction between reality and perceived reality. Considering LLMs have a hallucination problem as it is.
How would cmake enter the picture? if you have nested CMakeLists.txt files, cmake appears to build within nested directories for 'add_subdirectory' invocations. Does that help with the cache directory conflict?
Probably not, based upon the extent to which enterprises have been integrated with Windows. Enterprises use a lot of functionality which is not really visible to the casual user. All that functionality would need to be retrofitted into Linux and we get right back into the whole bloat cycle again.