I’ve been thinking about this for a while and I don’t know if there is any solution and/or if I make any sense.
So let’s say I’m binging a show on a Friday evening, but I happen to watch the last 3 episodes during the night. Then those 3 episodes are registered as watched on a Saturday, which is technically true. However, for me it still feels like the same Friday evening/night as I was on a streak and not went to sleep.
Not a big issue per se, but then in the analysis what you watched recently but also your overall stats these are added to the factual date and gives the illusion I watch more stuff on a Saturday instead of a Friday.
I don’t binge often but usually finish watching the last thing of my day after midnight. I would still like the timestamp to be correct but the aggregation of the data to be matched with when I started my watch streak, if that makes sense.
Maybe you can allow people to set a time offset in their settings for the statistics?
Let me know what your opinion is as I don’t have any other good suggestions as a feature or setting.
Is it because you want to use the Check In feature while watching? If not, you can just select Other Date and manually enter any day and time you want.
Me, personally, I don’t care about the exact time I watched something, just what day and in what order I watched something. So on Trakt I just set everything to the date I want, then set the time to 00:00. Everything shows up in my history in the order I marked them as watched for that day.
So if you reach the end of the day, set everything you watched after midnight to 23:59. If you need to make sure everything is in the right order you can increment the seconds between them.
Yeah, no, that’s the thing. I personally do care about the exact timestamp and am very strict with immediately marking it as watched when I saw it (or synced by Plex).
I would only like to see a change in the statistics when aggregating the data.
Eh no even just after midnight Saturday is Saturday and not still Friday. Facts above feelings
if you care so much about feeling like after midnight is still the day before, then change the time a little if it’s actually just a minute or two after midnight and set the day to the day before? you can’t have it both ways
I second this request. I often watch things over midnight and manually backdate my watches, which is totally possible but a hassle.
The sentiment is similar to how some TV guides list times beyond midnight as 24:00, 25:0, 26:00 etc. Sure, it’s technically another day but most people will say they’ve watched something on Saturday night instead of Sunday morning if they’ve technically watched it at 00:30 on Sunday.
I don’t care so much about the statistics, just having a custom time when the date separators inserted in the History would be enough to make me happy and my Trakt live much easier.
This is how Japan denotes opening & closing/starting & ending times in a lot of cases. The supermarket near my apartment stays open until 1AM, but the time on the door reads “10:00 ~ 25:00”. The logic is that while the clock has reset, the waking day hasn’t. It makes the most sense for a work culture like Japan, but people all over the world work and play past the 24th hour of a day all the time.
Perhaps instead of asking that the day markers move, there can be some new kind of marker for something like sessions. Then, in addition to grouping by clock date, you can also group by sessions and you wouldn’t be limited to just one session per day either. You could have a morning session, an evening session, a binge party session… Maybe even allow users to label the sessions to sort and search for them later. You would see your views grouped like so on the normal History page, or you would still see groups by calendar date but also a colored overlay that showed you where sessions began and ended.
I’m a little confused about why you’re so strict about having the watch time be accurate, but want to see inaccurate data when it’s aggregated. The fact is you didn’t watch those things on Friday, you watched them on Saturday, even if it still felt like Friday night.
I’ve started backdating my watches that finish within 10 minutes or so after midnight, because the majority of that watch actually did happen on the previous day. But to have a change in the stats that basically says you watched things on a day when you didn’t doesn’t totally compute for me
Perhaps a better solution to this would be to have a different time at which the day “resets”. Like for example, I wake up at 5 pm and go to bed around 11 am. It would probably be ideal for me to have statistics count my next day as having started sometime in the middle of my sleep session.
This is a problem that can crop up in other apps too, like a calorie counting app I use, which automatically flips to the next day at midnight even though I eat my “lunch” at like 3am. So I can understand the frustration with something like this, even if the statistics are technically accurate.
Don’t trying to reinvent the wheel… it works fine.
I would prefer more detailed statistics, including which Actor or Director is my favourite through how much I watch his movies, and how many hours I spend with him.
@Codex13 not sure how to explain it better, the accuracy is more like a log entry. I personally find it interesting when I watched something, visited a place, took a picture, … So I like that trakt allows me this, and it bugs be out that I didn’t discover trakt sooner. However, when looking at a dashboard you aggregate metrics, similar to a business report, you visualise the data to your specific needs. In theory, I could export my data and do my own stuff with it but then I have a secondary place for metrics and I like my trakt dashboard which shows “Last 30 days at a glance”
@morphinapg Good suggestion, a personal setting for an offset or day reset timer would fix my issue as well
you can’t have it both ways 