isaiah
@isaiahcarew.bsky.social
i'm a software developer. primarily for macOS. primarily in Obj-C. primarily working on https://yourhead.com/stacks
but it hasn't always been that way. i used to write other software, in other languages. and before that i designed microchips.
but it hasn't always been that way. i used to write other software, in other languages. and before that i designed microchips.
i don’t think i could be easily convinced to buy another Bausch. there were fundamental design and manufacturing problems. whatever quality advantages they once had have been nullified.
even if LG has the same issues, it will be no worse. and the price is much easier to swallow.
even if LG has the same issues, it will be no worse. and the price is much easier to swallow.
July 7, 2025 at 9:39 PM
i don’t think i could be easily convinced to buy another Bausch. there were fundamental design and manufacturing problems. whatever quality advantages they once had have been nullified.
even if LG has the same issues, it will be no worse. and the price is much easier to swallow.
even if LG has the same issues, it will be no worse. and the price is much easier to swallow.
no major complaints.
no huge advantages.
the old Bosche cleaned silverware a bit better.
the LG is a bit noisier, but still very quiet. the bosche was just silent.
on the plus side it’s easier to load since everything is configurable.
all the racks and rollers are super smooth.
no huge advantages.
the old Bosche cleaned silverware a bit better.
the LG is a bit noisier, but still very quiet. the bosche was just silent.
on the plus side it’s easier to load since everything is configurable.
all the racks and rollers are super smooth.
July 6, 2025 at 3:31 PM
no major complaints.
no huge advantages.
the old Bosche cleaned silverware a bit better.
the LG is a bit noisier, but still very quiet. the bosche was just silent.
on the plus side it’s easier to load since everything is configurable.
all the racks and rollers are super smooth.
no huge advantages.
the old Bosche cleaned silverware a bit better.
the LG is a bit noisier, but still very quiet. the bosche was just silent.
on the plus side it’s easier to load since everything is configurable.
all the racks and rollers are super smooth.
joke of the day: what do you call a gluten free italian?
hungry!
hungry!
April 3, 2025 at 6:03 PM
joke of the day: what do you call a gluten free italian?
hungry!
hungry!
yikes! i had never bothered to cap the concurrency on this operation queue. it had never seemed to need it.
simply putting a more reasonable cap on the queue concurrency was enough to keep things from dying.
🧵 fin
simply putting a more reasonable cap on the queue concurrency was enough to keep things from dying.
🧵 fin
March 15, 2025 at 3:00 PM
yikes! i had never bothered to cap the concurrency on this operation queue. it had never seemed to need it.
simply putting a more reasonable cap on the queue concurrency was enough to keep things from dying.
🧵 fin
simply putting a more reasonable cap on the queue concurrency was enough to keep things from dying.
🧵 fin
the protocol handler that directly feeds images to the webkit socket gets a few requests then seems to backs off. and they come in like that until the page is done loading.
on sequoia the handler floods with over 60+ requests. i'm not entirely sure what falls over, but something did.
🧵 5 of 6
on sequoia the handler floods with over 60+ requests. i'm not entirely sure what falls over, but something did.
🧵 5 of 6
March 15, 2025 at 3:00 PM
the protocol handler that directly feeds images to the webkit socket gets a few requests then seems to backs off. and they come in like that until the page is done loading.
on sequoia the handler floods with over 60+ requests. i'm not entirely sure what falls over, but something did.
🧵 5 of 6
on sequoia the handler floods with over 60+ requests. i'm not entirely sure what falls over, but something did.
🧵 5 of 6
last night i eventually traced it to a change in how webkit loads images.
or at least, that's my hypothesis.
the "good" machine is still running macOS 12 and whatever webkit was contemporary with that.
🧵 4 of 6
or at least, that's my hypothesis.
the "good" machine is still running macOS 12 and whatever webkit was contemporary with that.
🧵 4 of 6
March 15, 2025 at 3:00 PM
last night i eventually traced it to a change in how webkit loads images.
or at least, that's my hypothesis.
the "good" machine is still running macOS 12 and whatever webkit was contemporary with that.
🧵 4 of 6
or at least, that's my hypothesis.
the "good" machine is still running macOS 12 and whatever webkit was contemporary with that.
🧵 4 of 6
but yesterday i noticed on a different machine that the problem simply didn't appear. not ever.
i spent a couple hours eliminating possible differences. i even shut down cores on one machine so it had roughly the same CPU resources.
still nothing obvious. hmmm… 🤔
🧵 3 of 6
i spent a couple hours eliminating possible differences. i even shut down cores on one machine so it had roughly the same CPU resources.
still nothing obvious. hmmm… 🤔
🧵 3 of 6
March 15, 2025 at 3:00 PM
but yesterday i noticed on a different machine that the problem simply didn't appear. not ever.
i spent a couple hours eliminating possible differences. i even shut down cores on one machine so it had roughly the same CPU resources.
still nothing obvious. hmmm… 🤔
🧵 3 of 6
i spent a couple hours eliminating possible differences. i even shut down cores on one machine so it had roughly the same CPU resources.
still nothing obvious. hmmm… 🤔
🧵 3 of 6
but this is a plug-in and the deadlock is outside my code so there was a limited amount of info available.
what i did find was it didn't seem to be the sort of race condition that i think of as normally causing deadlock. even single stepping through the code triggered it 100% of the time.
🧵 2 of 6
what i did find was it didn't seem to be the sort of race condition that i think of as normally causing deadlock. even single stepping through the code triggered it 100% of the time.
🧵 2 of 6
March 15, 2025 at 3:00 PM
but this is a plug-in and the deadlock is outside my code so there was a limited amount of info available.
what i did find was it didn't seem to be the sort of race condition that i think of as normally causing deadlock. even single stepping through the code triggered it 100% of the time.
🧵 2 of 6
what i did find was it didn't seem to be the sort of race condition that i think of as normally causing deadlock. even single stepping through the code triggered it 100% of the time.
🧵 2 of 6
seems normal to me. you do you.
March 11, 2025 at 11:44 AM
seems normal to me. you do you.
- officially it’s “Compiter Engineering and Information Science” but mostly it’s cal an EECS now.
- i was 10yo and naively thought i could write my own games.
- i run yourhead.com and build small tools for individuals to create web pages.
- i was 10yo and naively thought i could write my own games.
- i run yourhead.com and build small tools for individuals to create web pages.
YourHead Software
YourHead Software
yourhead.com
March 9, 2025 at 7:16 PM
- officially it’s “Compiter Engineering and Information Science” but mostly it’s cal an EECS now.
- i was 10yo and naively thought i could write my own games.
- i run yourhead.com and build small tools for individuals to create web pages.
- i was 10yo and naively thought i could write my own games.
- i run yourhead.com and build small tools for individuals to create web pages.