Live data from Hacker News

Google Timer is back

google.com

21–30 of 99 posts

Re: Google Timer is back

#21
post #14
post #9

Earlier quoted context omitted.

A bug is a bug. "Why would anyone ever do XYZ" is a separate discussion. And if you write software for a living you know that there is always that one user who will do exactly XYZ.

Nope. Undefined behavior on obviously incorrect input isn't a bug in my world. ( might be in other circumstances )

I mean, the query is “timer $period”. It’s correct in its syntax, a period’s a period. It’s not asking for “timer TIMMY; DROP TABLES”. That’s incorrect input.

Re: Google Timer is back

#22
post #3

and .... it's buggy try https://www.google.com/search?q=timer+724+years click on start, looses most of the time

Some kind of overflow, it resets timer to 40 hours. Same thing happens with 1 year, except timer is reset to 60 hours. Another interesting observation is that "timer 723 years" did not bring timer up, but "timer 722 years" did. Update: "timer 100 hours" works, but "timer 101 hours" does not. That is 100 hours is 144,000 seconds, does not seem to be anything special about this number.

It can just only handle two digits in the hours place after the first tick. Any other digits get removed. 1 year becomes 60 hours because it turns 1 year into 8760 hours, then removes the "87" and leaves the "60"

Edit: this seems like it's just a display bug and the original time is still preserved. If you search "timer 100 year 1 second" then it formats to "876000h 00m 01s", then the next tick shows "0s", then the next tick shows "99h 59m 59s"

> Another interesting observation is that "timer 723 years" did not bring timer up

That is weird. Even weirder that "timer 723 year" (singular) does seem to work.

Re: Google Timer is back

#23
post #9
post #4

Earlier quoted context omitted.

When would you ever need a timer for 724 years?

A bug is a bug. "Why would anyone ever do XYZ" is a separate discussion. And if you write software for a living you know that there is always that one user who will do exactly XYZ.

Sure, a bug is a bug, but the original remark was that the software was "buggy". I read "buggy" as riddled with pretty obvious bugs. If we're calling software "buggy" for having any bugs, then, I guess, nearly all software is "buggy" (and the word looses its usefulness).

Re: Google Timer is back

#24
post #3

and .... it's buggy try https://www.google.com/search?q=timer+724+years click on start, looses most of the time

..and curiously, with the little speaker icon 'on', it produces no audio alarm for me, (firefox 103.0.2). Whereas this[1] one produces a quite notable sound. [1] https://www.online-stopwatch.com/countdown-timer/

These days browsers make sure there's user interaction before audio plays. The other website you linked requires a click to start the timer, therefore meeting that requirement. If Google made this timer muted by default, and you had to click unmute, it would likely work.

Ran into a related bug on a site I was working on the other day and had to learn this browser quirk the hard way.

Re: Google Timer is back

#27
post #3

and .... it's buggy try https://www.google.com/search?q=timer+724+years click on start, looses most of the time

Is it really buggy? Is it a feature that you need a timer of 724 years ... in a web browser?

Re: Google Timer is back

#29
post #14

Earlier quoted context omitted.

Nope. Undefined behavior on obviously incorrect input isn't a bug in my world. ( might be in other circumstances )

You must be a C programmer.

Haha. I was thinking along similar lines as kuboble, but then I'm someone in the process of trying to learn C.

I agree, upon further thought, that at least an error message would be better for the user (e.g. "Maximum timer duration exceeded").

Re: Google Timer is back

#30
post #9

Earlier quoted context omitted.

A bug is a bug. "Why would anyone ever do XYZ" is a separate discussion. And if you write software for a living you know that there is always that one user who will do exactly XYZ.

Sure, a bug is a bug, but the original remark was that the software was "buggy". I read "buggy" as riddled with pretty obvious bugs. If we're calling software "buggy" for having any bugs, then, I guess, nearly all software is "buggy" (and the word looses its usefulness).

Anything over 100 hours gets recalculated modulo 100, so 101 hours becomes 1 hour. But then the page title, which is meant to show the current countdown in the tab display, gets updated to show the full time, but weirdly formatted. So at least two bugs.

Try "timer 101 hours". The page title will (very quickly) show "10:05:9x" where x is the 10s digit of the seconds. While the page itself claims it's a one hour timer (59 minutes and some seconds). So the actual time is in there somewhere, just showing up wrong in both places.

It also doesn't seem to let me enter 30 minutes as a timer, but most other minutes times work. So, yes, it's buggy.

EDIT: It's not actually modulo 100 hours, I should note. The display is showing it that way, but since it has the real time stored internally it still counts down. If it were actually modulo 100 hours then it would zero out a 100 hour (or 200 or 300...) countdown immediately, instead these wrap to 99 hours, 59 minutes, and 59 seconds on the first tick.

EDIT: For grins I'd started a 101 hour timer a bit over an hour ago before stepping away from the computer. Despite the display silliness, it actually will run for 101 hours. After the first hour the displayed time rolls back to 99:59:59 and the page title updates correctly. I also noticed that there's a progress bar below the timer, it correctly shows the progress within the 101 hour timer the whole time.

Post reply on HN