I'm not a native English speaker, but I read a lot of books in their original English. Being able to quickly look up a word you've never encountered is a god send
I find myself really missing this feature when I occasionally read a paper book, thinking about clicking the word on the page to get a definition.
I've had it refuse to generate a long text response (I was trying to concise a 300kb documentation to 20-30kb to be able to put it in the project's context), and every time I asked it replied "How should structure the results ?", "Shall I go ahead with writing the artifacts now ?", etc.
It wasn't even during the over-capacity event I don't think, and I'm a pro user.
Hate to be that guy, but did you tell it up front not to ask? And, of course, in a long-running conversation it's important not to leave such questions in the context.
The weird thing is that when I tried to tell it to distill it to a much smaller message it had no problem outputting it without any followup questions. But when I edited my message to ask it to generate a larger response, then I got stuck in the loop of it asking if I was really sure or telling me that `I apologize, but I noticed this request would result in a very large response.`
It sparks me as odd, because I've had quite a few times where it would generate me a response over multiple messages (since it was hitting its max message length) without any second-guessing or issue.
I also found that passing all the codebase in a single file in the context works really well. I've tried Cursor et al, but I found that it not having the full context of the codebase (and having to do some back and forth for it to requests files) was slower and didn't really yield any better results. Granted I work on projects where the codebase fit in 100-200kb text files, but I'm still only at 20% of the context limit in Claude.
Also I found the UX of Claude to be better for this, especially their Projects feature. I can just put the codebase in the Project's context, and start a new conversation to ask different questions/solve different problems.
The only pain point I have is that it seems to be pretty optimized to only show changes in existing files, not rewriting them in full, which is a bit of a pain to copy-paste into my IDE. I'll see if I can write out a system prompt to force it to generate diff or a similar format that could more easily be applied automatically to my code.
I switched to an Ergodox-EZ about 3 years ago, and I don't regret doing it at all. I didn't switch layout tho, stayed on AZERTY.
My experience is that at first it was very frustrating. I went from ~80WPM on my macbook keyboard to ~20WPM on the ergodox. After a couple of weeks I was able to write text at a comfortable speed again, but any special character was painfully slow, as I had to consciously think where each character was, and often look it up on my layout. After about 3 months I was back up to 80WPM.
What took a long time as well, was configuring my layout to fit my programming needs, it took me about 6 months to come up with a layout that had everything I needed (you can see it here if you're curious[1]). My recommendation is to do it incrementally, trying with something general at first, then with use seeing what feels right and doesn't.
In the end it was a really good idea, a lot of back pain I had has gone away, and after long typing session I have way less pain in my wrists and hands.
Out of curiosity, how does your Image Similarity Search works ? Are you also using some feature of Apple's Vision framework, or running some ML model on your linode instance ?
I find myself really missing this feature when I occasionally read a paper book, thinking about clicking the word on the page to get a definition.