my mac app didn't open at all for 4 months and i had no idea
there's no changelog for PopTask anywhere, the app store "what's new" box is read by roughly nobody, so from now on every update lands in this thread, small ones as replies, bigger ones as their own post
what's shipped since july:
ππ½ the app learns your times now, correct a suggested time once and it remembers for that kind of task, so if your gym is an evening thing, "gym at 6" starts meaning 6 pm for you .. weekdays and weekends stay separate
ππ½ it picks up your shorthand and your usual typos too, and a name it doesn't recognise gets left exactly as you typed it π¨π»βπ» all of that follows you between iphone, ipad and mac through your own icloud
ππ½ a speed pass, so the suggestion card lands while you're still typing instead of after you stop
ππ½ and one i'm not proud of .. the mac app wasn't launching at all on MacOS Sonoma, had been that way since april, and i only found out because someone sent me a crash log instead of deleting the app [ a big shoutout to Chris π ]
Β
what's next:Β
at launch someone here [ narek ] made the point that voice input is useless if your hands and eyes are both busy, because you can't check what it heard .. it should read the task back, that's on the list
also on it: microsoft calendar, and proper two-way google calendar
Β
which of those three would you actually use?
or is there something more obvious i'm missing π
Replies
i would probably do Google Calendar next. once tasks and calendar start talking to each other properly, the time suggestions become way more useful.
@jeremy_loomisΒ that's the strongest case anyone's made for calendar and it's not the one i expected π
Answering the question at the bottom first: read-back, easily, and it is not close. Calendar integrations are table stakes that people will wait for. Voice input you cannot verify is a feature that quietly teaches users not to trust the app, and that damage does not show up as a bug report.
But the actual story here is the four months, and I would not file it under embarrassing. It is the most useful thing in the post.
Nothing was broken in a way anything could see. The app did not crash on your machine. Downloads kept happening. There was no error rate to watch, because a launch that never happens does not emit anything. Your dashboard, whatever it is, could not distinguish "nobody opened it today" from "nobody could". Those two render identically as a flat line, and the flat one looks calm.
Which is why Chris is not just a nice person, he is your only instrument for that class of failure, and instruments that depend on someone being unusually generous are not instruments.
The cheap fix is not crash reporting. It is a heartbeat: the app phones home once on successful launch, and you alert on the absence, not the presence. Zero launches on Sonoma for a week is a signal you can act on. Zero crashes on Sonoma for a week is the same data wearing a reassuring face.
Same shape as the changelog thing you opened with, honestly. The app store "what's new" box being read by nobody is also a silence you had no way to measure.
this is the best thing anyone's said @rabnoor_sΒ π«
the flat line is the part that got me .. i'd been telling myself "woahh i had no crash data", but the real problem is i had no shape of data that could have told me :') zero launches and zero crashes draw the same picture and one of them is a fire