that deserves a whole separate opera. why does moving the cursor to start/end almost always place it one character off? same when trying to use touch on text for it.
This. I feel like this is due to small shifts when lifting the finger. For the life of me, I cannot get myself to lift it like the testers at Apple do, and so I wished the product itself could go through learning phase; studying intentions and what actual touches happen.
Apple devs are probably doing something with the touch returned by the touchUpInside callback. This is an extremely, extremely common bug in iOS apps, not just with Apple developed software.
Devs love the symmetry of their touch handling code and often have the finger-down, finger-moved, and finger-up callbacks from the system all call the same handleTouch function they wrote. As you can tell, however, the touch from the finger up callback is often better discarded or handled differently otherwise you get these sort of bugs
They are once you know them. I still haven't found great sources of telegram channels. Only ones being surfaced are from the funnels "how to run k5s on a baked potato? I just published guide on my tg, which you can find on my linktree that is in my ig profile!"
That's what I keep thinking about when I see those "once I started taking Claude, I lost 95% of my developement time" posts. Are they really making 20x the software, or are they and their customers simply believing that, watching all those endless streams of green checkmark and rocket emojis?
Is there a chart of music that people like, or at least a good approximation of what people like? Probably last.fm was something like that at the very beginning.
Are not music charts a list of songs people are expected to like, or else?
At some point it stops making sense. You cannot use "the good model" just for the hard bits without basically hand writing you own harness. Even then, it will need full, uncached context.
Feels like consulting a premium lawyer to ask how much time is it.
pole-vaulting records improve incrementally too. and there is finite distance left to the moon. without deep understanding and experience and numbers to back up the opinion, any progress seems about to reach arbitrary goals.
So I tried it and it is worse that having random dude from Fiverr write you code — it is actively malicious and goes out of it's way do decieve and to subtly sabotage existing working code.
Do I now get the right to talk badly about all LLM coding, or is there another exercise I need to take?
Properly securing LLMs goes agains branding, I guess. "this tool is like getting new intern every 15 minutes! they read and write fast and know a lot of stuff, but can accidentally attack or sabotage you if they get distracted! oh, and they work remotely only!" doesn't sound like a good pitch
I have been asking if the business would be happy to employ an extremely gullible insider with a short memory, who sometimes just makes things up, with no fear of any legal repercussions or being fired, to work on important stuff.
Has anyone noticed how poorly tools like claude code (the main one I tried) themselves are working? You'd expect software from company with an infinite AI allowance to be unattainably excellent, instead it lags, hangs, flickers, and feels like unpleasant mvp mess.
I hear at every corner people telling, how they can 100x now, and if my AI use is not laying prime code it's my skill issue. But where is this excellent AI generated software? Do you maybe have some examples you can share?
Currently it sounds like it's been through an allpass/comb filter. Complex parts, while spectrally there, do not make much sense as a real sound. Probably audio analog of the "finger salad" of early image models. I do not count of being able to tell one from another in a few months.
reply