Agreed, these are my vibes too. It feels much better to do planning and strategy and architecture etc. with Opus 4.7 than GPT-5.5. GPT just feels like a robot that gets instructions and does exactly that. Opus feels like an almost human that sometimes has actually good ideas and pushes back on bad ideas.
So for now its planning/architecture/strategy -> Opus. Pure coding -> GPT.
Helps with agentic coding that GPT is much roomier with the tokens you get.
Obviously you have multiple agents justify why they picked a certain response and then create another agent that picks the solution with the best justification.
I also have a small private app that technically could have been a PWA.
It’s not a PWA because the UX is just always inferior. Even though we’ve come really far in browser UIs, the browser is still very clunky compared to the smoothness of a native app.
No I can confirm this. I am at least an average C# dev, with 16 years of experience.
I have built a very nicely responsive real-time syncing iOS app in what amounts to a weekend of time. (I only have an hour here and there, young kids) I had zero iOS/Swift development experience prior to it.
I can also confirm that this wouldn't have been built if it weren't for Claude Code. It's "just" an improved groceries app, that works especially well for my wife and me.
Without LLM's, and with just an hour here and there, I wouldn't have done the work to learn the intricacies of iOS and Swift dev, set up the app, and actually tweak and polish it so it works well -- just to scratch the itch of a bit better groceries handling.
"My favorite is always the billion dollar mistake of having null in the language. And since JavaScript has both null and undefined, it's the two billion dollar mistake." -Anders Hejlsberg
"It is by far the most problematic part of language design. And it's a single value that -- ha ha ha ha -- that if only that wasn't there, imagine all the problems we wouldn't have, right? If type systems were designed that way. And some type systems are, and some type systems are getting there, but boy, trying to retrofit that on top of a type system that has null in the first place is quite an undertaking." -Anders Hejlsberg
> "My favorite is always the billion dollar mistake of having null in the language. And since JavaScript has both null and undefined, it's the two billion dollar mistake."
> -Anders Hejlsberg
Why can't all-1s be null? E.g. a small int goes from the range 0-255 to the range 0-254, but we get a really useful property with no out-of-band Nullable overhead.
With signed ints it even leads to symmetric ranges in the negative and positive directions.
The FORTH-83 standard changed FIG-FORTH's official value of TRUE from 1 to -1 so all its bits were set. That was a rough transition like Python 2=>3, but worth it. It also defined /MOD integer division to be floored (rounded towards -infinity instead of zero like FIG-FORTH), which was also a tough change but the right one, especially for graphics.
>4. For various reasons the definition of all divide functions general effect is that quotients are floored instead of rounded toward zero. This should cause no problems for most pre-existing application software. The new divide functions are marginally slower than the old (a few machine cycles under most circumstances). The side-effects of the redefinition for floored divide can be counter-intuitive under some circumstances. For example, in FIG-Forth the operation
-40 360 MOD
>would return the obvious answer (-40) on the stack, while 83- Standard Forth will return the answer 320!
>5. The true flag returned by all logical operations has been changed from the value 1 (in FIG-Forth) to the value -1 (in Forth-83, all bits set). If your code used the 0 or 1 returned by a comparison in an arithmetic operation, you will need to interpolate the operator ABS after the logical operator. This is a particularly difficult problem to look for in your source code. However, we feel that this mutation in the 83-Standard was beneficial as it allows the returned true/false value to be used as a mask for AND.
You're kind of missing the point. Turbo Pascal has been dead for a lot longer. Or is it?
The point is that TypeScript and C# are extremely similar for a good reason, not a coincidence, and that Anders Hejlsberg knows what the fuck he's doing and talking about, and has been implementing amazing groundbreaking well designed languages and IDEs for a very long time. Turbo Pascal was so great it flummoxed Bill Gates, so Microsoft sent a limo to recruit and hire Anders Hejlsberg from Borland, then he made Visual J++, Windows Foundation Classes, C#, and TypeScript.
>Scott MacGregor of Microsoft said that Bill Gates "couldn't understand why our stuff was so slow" compared to Turbo Pascal. "He would bring in poor Greg Whitten [programming director of Microsoft languages] and yell at him for half an hour" because their company was unable to defeat Kahn's small startup, MacGregor recalled.
>"According to the suit, Microsoft also offered Mr. Hejlsberg a $1.5 million signing bonus, a base salary of $150,000 to $200,000 and options for 75,000 shares of Microsoft stock. After Borland's counteroffer last October, Microsoft offered another $1.5 million bonus, the complaint says."
C# is nominally typed, which, in practice, leads to safer code and less type gymnastics. Of course you can avoid the type gymnastics with "any", then you you're sacrificing safety.
reply