While I love this set of prompts, I’ve not seen my clause opus 4.6 give such verbose responses when using Claude code. Is this intended for use outside of Claude code?
Msot do, but Anthropic indicates that theirs is "is not considered a long-term or production-ready solution for most use cases" [0]; in any case, where the OpenAI-compatible API isn't the native API, both for cloud vendors other than OpenAI and for self-hosting software, the OpenAI-compatible API is often limited, both because the native API offers features that don't map to the OpenAI API (which a wrapper that presents an OpenAI-compatible API is not going to solve) and because the vendor often lags in implementing support for features in the OpenAI-compatible API—including things like new OpenAI endpoints that may support features that the native API already supports (e.g., adding support for chat completions when completions were the norm, or responses when chat completions were.) A wrapper that used the native API and did its own mapping to OpenAI could, in principle, address that.
There's nothing as good as VB6 that's developed and supported by *anyone*. It's not a Microsoft only phenomena.
I think programmers started wanting "real" languages (notice the quotes), and henceforth got more complexity and things take longer, although with GenAI, we may be back to the "draw as screen and do this" that we were with VB6. Just now the source generated should be considered the object code, and the prompt is the new source (at least for those types of apps)
It's been super amazing to see how much they could continue to support newer hardware and keep it going considering that I don't believe they have the kernel source.
It wasn't too long ago I saw OS/2 on some ATM machine that was crashed.
I used to love OS/2 back when developing DOS applications (since I could crash the app and not the machine). OS/2 got me interested in "real OS's" and then SunOS in college, etc.
Up til 20 years ago there were a surprising number of ATMs still running OS/2; NCR and Diebold supported old machines for a long time. Especially small market/small regional banks wanted to get the absolute most out of their capex investment. Over the years, I've worked with a couple of those dead-enders on different GRC projects, mostly because I'd actually seen OS/2 before. AFAIK, those vendors stopped supporting OS/2 in the 2000s; I'd be very, very surprised if there were any left now.
I you're interested in how a very "not Unix" operating system is architected, I really recommend Deitels' "Design of OS/2". Very interesting.
It's definitely true that they do not have access to the original OS/2 source - this has been confirmed by people from Arca Noae in various interviews/presentations I've seen. I've never heard a definitive explanation for why, but two reasons are usually speculated:
1) Due to the amount of third party code in OS/2 (most notably, the DOS and Win 3.x layer) that IBM is unable to license out the code, or unwilling to go to the trouble to figure out the legal implications.
As far as I know, yes. There were no changes made to eCS which required source - everything was implemented as drivers, or layers on top of the base OS.
I have not heard or seen any direct confirmation of this anywhere. If you have, I would really like to know. I am looking at a follow-on review and this would be great background info.
> most notably, the DOS and Win 3.x layer
I think what you put in parentheses here is the real reason.
IBM probably still has the source. It seems to be methodical, unlike say Symantec which lost the QEMM and DESQview source.
But IBM and MS co-developed OS/2. MS has joint ownership of this code.
MS has a 50+ year history of being a deeply dishonest and unreliable company. It hates FOSS and only releases what it has to. MS-DOS 4 only got out became someone found it and made it public.
Satnav Nutella has no more understanding of this than the Queen of England. He will do and say whatever is needed to make Number Go Up.
MS releases tiny token gestures to make the incomprehending loud FOSS advocates believe them. Notepad, Calc, ancient DOS releases... nothing that matters.
It won't release Windows 3 because some of that code is still in Windows today.
MS does not love Linux. WSL2 is an embrace-and-extend tactic. If MS had a real clue left then WSL1 would never have been a product: it would have just extended the NT kernel POSIX personality to run Linux binaries.
Remember the core of Windows is the NT kernel and it can natively run OS/2 binaries and Unix binaries.
It doesn't because MS turned it off. NT is a version of VMS with native Unix and OS/2 binary support and a GUI built on Windows 3 code and MS won't let that code out. If it did the ReactOS people could make a ReactOS that was Good Enough. The WINE people could make a seamless one that make .EXEs a 1st class Linux citizen.
MS is terrified of that because it doesn't have the skills to do the equivalent any more, and WSL2 is the existence proof of that. It couldn't even get systemd working in WSL2 until it hired Poettering to do it. Then he stayed there just long enough to get the money and he's off out again.
The reason IBM won't release the OS/2 source, even to Arca Noae, is Microsoft.
Also, around the 25 minute mark, Rosenthal points out that the Workplace Shell source code "no longer exists in one place anymore", so I do think that there are problems with finding all the source for the OS.
I'd love to see the policy on review tools to start with, I know even people who are skeptical of getting "AI Slop" thrown at them by agents at a high rate, getting code reviews from some of the SOTA models definitely can be helpful.
Google found that with Jules years ago at this point, same for other automated tools.
When I first saw the headline though, it sounded like someone was listening to one of my favorite Rush songs.
If that ever gets fixed then I'd look at replacing Sublime (which is still my go-to for quick editing) and then see if it can handle more advanced coding (which one the rotating list of various vscode forks handle today)
So, it's been one year since I use Zed daily, and I didn't encounter this issue, or any other issue for instance, everything is smooth and I never encountered a failure or a crash.
I work on large (everything is relative, though) monorepos, that would probably qualify for this limit, and I remember already did the kind of "workaround" discussed in this issue years ago on this device. I think it's hard to blame the software when the default file limit is so low depending on the languages you work with.
Anyway, if you would encounter this problem, you would have already encountered it with other tools, or else this is fine.
Have you tried BBEdit recently? It's incredibly configurable, AI features are entirely opt-in, and it supports LSPs now. It's my go-to for basically all text editing these days.
Honestly when I saw Okta in the headline, I had assumed the article was going to say they were breached again.
This one is amusing, and as another comment mentioned below, large companies are awful at accepting patches on github. Most use one-way sync tools to push from their internal repositories to github.
Cool, I just had claude code write me something similiar this week to go through my immediate directories and get me this type of information on each one of this (since all of my git repos are under a single dir)
fun fact: check-projects is initially a nodejs script I wrote specifically for my projects few years ago;
My first usage to test out claude code was to generalize this script: cople hours later it was entirely rewritten with Go and and CI on github actions you see now here.
reply