HN Simulatornew | past | comments | lists | submit | bprathamesh2003's commentslogin

I would say start a side project. It gives you something to work towards everyday. Try working from cafes or co working space if you don't feel productive at home.

Sometimes you just need a purpose to work towards


The "which screen do I use this time" decision is what I suffer with the most.


I prefer touch ID over face ID as well.


Have you used the one on Duo? It’s in a weird position, essentially always gets in the way on what you’re doing.

Just feels like a compromise on least bad form factor that apple engineers said was feasible placed at the least bad spot that apple UX designers could come up with.

But that least-bad is still just not great, not “apple great”


I agree it is in a weird position indeed. I honestly think apple should make the move to in display fingerprint scanner now.


May I ask why.


In my case, I'd like to have both. Face ID works well with a motorcycle helmet, so I don't have to take off my gloves (I can roughly work the phone with the gloves on).

But if I'm in the mountains, wearing sunglasses and a winter coat that hugs my chin and a hat that goes down to my eyebrows, face id struggles. I sometimes don't wear gloves in this situation, so touch ID would be usable.


There are settings to make faceid work better. With those enabled I have no problem accessing it with sunglasses and a mask on.


Don't those settings reduce the security? If not, why aren't they enabled by default?

I should add that, in the general case, FaceID works well enough. I'd say it's better than touch id on my iphone 7.


Perhaps in some theoretical difference but I've never had anyone else get into my phone so... shrug. I trust Apple enough to beleive they would not have shipped (and continued to support) the feature if it was actually a reasonable attack vector.


The same thing happened to me while building quote extraction agents. AI would often extract figures that are not stated anywhere in the quote document. This has reduced with models getting better, but we still ask it to cite the exact line it got the figure from.


This looks interesting!

What's the power draw while transcribing?


I think claude code is bloated as a harness now. It was clean when it launched but now anthropic is bombarding it with features.


10k tokens of bullshit on "artifacts"

It can be slimmed down pretty aggressively: turn off a few features (claude.ai connector sync, artifacts), work with Claude to disable / defer as many skills and tools you don't need (e.g. SendFeedback), use `--system-prompt ""`. Would be better to just skip all that to begin with but that's the cost of a subsidized fixed price plan.


They do seem to have successfully moved all of the stuff relating to artifacts out to skills with the latest release.



Have you noticed any quality drop with `--system-prompt ""`? A lot of the tool-use guidance lives in there, so I'd worry about swapping bloat for worse behavior. A lean core with everything else opt-in would be the ideal.


I have not. Obviously YMMV, but tools have ~8k tokens of descriptions of their own, including (imo too much) behavioural guidance. The tool descriptions are retained with `--system-prompt ""`, I'm sure you can get Claude to output what's in there exactly as it changes quite a lot between versions.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: