-
Notifications
You must be signed in to change notification settings - Fork 34
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Testing the time-use of API 32 version of the app #871
Comments
On the permission page, when the user clicks "Fix" to disable battery optimization, only the apps for which battery optimization is turned off are displayed. If the user searches for Openpath and it doesn't appear, they must change the filter to "All apps" before finding it. This process can be confusing for the user initially. |
The "Scan program OPcode" button is not functioning on our device, which has resulted in our inability to scan the QR code. As a result, we had to manually enter the token. |
wrt
Unfortunately, this is what the Android API supports.
And that is the intent I am launching I do have text in the permission screen telling users what to do. For more details, please see the this comment and the ones just below: |
Which version of the app? Which version of the OS (android/iOS)? "Scan program OPcode" works on my test Pixel 4. |
We intended to test the Places feature, but to our surprise, the activities we attempted to add did not appear at all. As a result, we are unable to test this feature while it does not work. |
In the title of the issue I mentioned API32, so it is for Android. The version of app is 1.1.7. |
When adding activities, it should be verified that the end time is not earlier than the start time. Currently, with the app, it is possible to add an activity that starts at 9 am and ends at 8 am on the same day, which is incorrect. |
Please add a start time, end time, and date to each Place card, similar to what we have on Trip cards. This is necessary because each Place card needs to be associated with a specific date, as we only have one date for each activity. As can be seen in the attached screenshot, we have a trip on March 17th and the next trip is on March 20th, we need separate Place cards for days in between. In this case we need four place cards. The first Place card should have a start time of 7:16 PM (end time of the trip), an end time of 12:00 AM, and a date of March 17th. The second Place card should have a start time of 12:01 AM, an end time of 12:00 AM, and a date of March 18th. The third Place card should have a start time of 12:01 AM, an end time of 12:00 AM, and a date of March 19th. Finally, the fourth Place card should have a start time of 12:00 AM, an end time of 10:31 AM (start time of the trip), and a date of March 20th. Without this, it may be confusing to add activities to a single Place card spanning multiple days. |
It it better to have 'Add Activity" and "add trip details" buttons to be styled in the same way. |
It appears that the comments I wrote in #840 (comment) have been addressed in the iOS app, but not in the Android app we are using. |
@MaliheTabasi android 1.1.7 is not the most recent version. The most recent version is 1.2.0. Please check the email titled "Upgrade to API 32; new link" sent on March 16, upgrade to 1.2.0 and let me know if you still see these issues. |
@shankari I have used the link you shared in the email with the title "Upgrade to API 32; new link" and it redirected me to install version 1.1.7 of the app. Please recheck the link you sent on March 16. |
@MaliheTabasi you are right; I said in the email
I sent out the link to the Google Play internal testing channel to the entire technical coordination list asking interested parties to send their gmail ID, similar to the Apple ID on TestFlight. This email was sent on 20th March. I also brought it up at today's meeting. My apologies if that wasn't clear 😄 and I hope we can get you set up on the internal testing channel soon so we can test against the correct version. |
Thanks for the clarification. I will send you the gmails that need to have access to the latest version of the app. |
@MaliheTabasi just wanted to highlight that the label screen is not loaded day by day. instead, the data is loaded week by week. So in general, you cannot assume that the first trip is the label screen is the first trip for the user. The user can try to load additional trips by clicking the download button on the top right (please see video below). scrolling_more_places.movWhen all trips have been downloaded, the download button turns into a check mark. |
@shankari Thank you for the explanation. If the trips are displayed on a weekly basis, it would be helpful to provide some indication to the user on how to view all their trips and not just those for the current week. It may not be clear that by tapping the download button, the user can access all their trips. Additionally, there is no issue with showing a place at the top of each week's trip list. A place card can be displayed at the top of each week's trip, starting at 12 am on the same day and ending at the start time of the first trip of the week. |
@MaliheTabasi do you have a concrete suggestion for "it would be helpful to provide some indication to the user on how to view all their trips and not just those for the current week"? We currently display the time range for the trips at the top and we have a button with a very clear download symbol next to it. We don't have a lot of screen space, and we don't have the ability to display hover text since this is a phone app.
Again, this screen is not a diary (either daily or weekly). It is a timeline, similar to most social media apps. You might need to have long-term data to see the full effect. I don't think that adding a fake trip in some arbitrary point of the timeline is a great idea and is not consistent with the high level user interface design. |
@shankari Yes, I do. You can improve the UI and more clearly show that time range to the user. currently, it is like some unimportant text. Instead of a grey download symbol which I believe is not very clear, you can use a button with a label like "display all weeks" which is more user-friendly. Currently, the app shows 8/8 trips while 8 is the total trips within this week. this could be misleading because I as a user might think 8 is the total number of my trips. It is better to display this like 8/20 where 8 is the total number of trips in this week and 21 is the total number of trips and that "display all" can show all the trips. I did not suggest adding a fake trip. We need a Place card which starts from 12AM before the first trip of the week. If you prefer you can first create it between that first trip of the week and the last trip of the previous week and then simply break it into two places. One place that is before 12AM can be shown in the previous week as the last component and another place that can be shown in the current week and starts at 12AM. This approach is totally reasonable such that between any two trips there is a place component. We did not ask you to add a place component in an arbitrary point in timeline. we wanted to have a place component between each consecutive pair of trips. All we are asking you to do is simply break that place into multiple places and everything about this place cards including their start time, end time, date, and location are clear. I also would like to add that the whole logic behind time-use feature is to be able to have a diary either daily or weekly. A timeline can be a diary, as the trips are sorted based on their date and time. |
@MaliheTabasi I think we should move this to an in-person discussion since I think we are arguing at cross purposes here.
Correct. But a diary implies "flipping" pages from one day to another with a clearly defined start of day and end of day. A timeline is a smooth progression of time without any "flipping". That's why "current day" or "current week" does not make sense in the context of a timeline. |
@MaliheTabasi did you also want to make a new survey for at least the time use? Right now, I think we just have a placeholder survey. Once you create it, @AlirezaRa94 can submit a PR to add it to https://github.com/e-mission/nrel-openpath-deploy-configs and change the URL and the labelVars at https://github.com/e-mission/nrel-openpath-deploy-configs/blob/main/configs/stage-program.nrel-op.json#L72 @AlirezaRa94, can you add the new survey xml or json to a |
@MaliheTabasi I thought about #871 (comment) some more and I am not sure that splitting the place is the best option, at least for the common case where the dwell time in the place is < 24 hours long. Note that the common case for people is that they sleep overnight. If you split the place into two, they will have to enter the time use twice (slept from 10pm -> midnight; slept from midnight -> 7am). Since we have a continuous timeline view versus a diary, the user can actually enter a single time use for the entire span, which is clearly more intuitive. I am open to splitting if the place spans 24 hours, but I don't think we should split at midnight for the reason above. Again:
|
@shankari I've just given you my comment as a person who is testing the app. I left the decision-making regarding the icon or text in the button to your UI team as I am not an expert. |
@shankari Yes, that is correct, let me talk to Taha and we will share the main survey with you after his approval. |
@shankari I understand your point. However if we display only one Place between two consecutive trips in two days, that Place card point to two different days, so how can we decide what date to display on it? If a person sleeps from 10PM on 15th March to 6AM on 16th of March, then the end time of the activity is earlier than its start time if we do not consider change in date, which does not make sense. That's why we need to take account of the change in date by splitting Place card into two. If we can come up with another solution to this problem then splitting Place only if it is longer than 24 should be enough. |
If we expect overnight activities to be a common need, the most proper solution is to explicitly ask for the end date in the survey. |
It is clearest to the user when 1 place corresponds to 1 blue card, rather than multiple consecutive place cards. If we really do need to split up multi-day places, I would suggest that we divide the card up into sections, separated by a line. However, I am not sure if the project timeline allows for this anymore. |
Thank you for sharing your thoughts. I completely agree that the proposed solution seems feasible. It's essential to ensure that the user enters a valid end date, taking into consideration that it should not be after their next trip. Moreover, some modifications to the user interface will be necessary to implement this feature. |
As I had mentioned to Shankari, we are open to exploring alternative solutions to address the issue of a single place card being linked to multiple days when only one date is available for inserting activities. We understand that splitting places may not be the only solution and are willing to consider other approaches if they prove to be more effective. |
@JGreenlee @sebastianbarry it looks like the real request wrt "first place" is that the timeline start with a place instead of a trip (if we are displaying places). @JGreenlee I know that the composite trip currently returns only the trip and the succeeding place. It seems like it should be fairly trivial to have it return both the preceding and the succeeding place. We can then maybe change the display timeline to start with the preceding place of the first composite trip. It's going to be a fair amount of work, and will involve end-to-end changes, though. |
I actually do like this solution; it would go something like this I gather:
I do think however @JGreenlee is right:
|
@sebastianbarry the first three of your points are perfectly correct. However, the last point won't really work the way that you/Maliheh want.
The duplicate place object that we get from the server, will really be a duplicate; i.e. identical. There is no question of duplicate trip objects that "fall on different days"; they would not then be duplicates, by default. Consider the following timeline: place (school) -> trip (school -> home) -> place (home, overnight) -> trip (home -> school) -> place (school) There is only one place between the two trips, not one place per day. That is why it is a duplicate in the composite trip list This will get split into the following composite trips (after the server change to return both places)
|
The redundancy of including both start and end place would not hurt us, and it would not be hard to ignore the duplicates when composite trips are received and unpacked on the phone. The only difficult and time-consuming task here would be splitting up places, but I think we can avoid this. If surveys include an "End date", time use can be recorded overnight. This way, we do not need to split the places. |
@JGreenlee Yes, that is correct, however you might need to take care of the the UI changes that are necessary and some validation processes between the start data and time and end date and time for activity insertion. For example, you might want to show both start date and start time of the Place and the end time and end date of the Place on Place card. or you might want to update UI such that each activity aside form title, start time, end time, and delete icon, has a start date and end date as well when it is displayed to users so that they can realise to which date it belongs. |
@MaliheTabasi can we have a quick in-person discussion to clarify what you see in the label screen?
I don't think we can make progress on this until you have a better sense of how the timeline works. For example:
If you have been collecting data since we released the first trip labeling version (March 13), you should have one month worth of data on at least one of your test phones, so we should be able to show you the differences. |
@shankari Yes, definitely. 1- Currently, user can insert an activity with an end time earlier than its start time. Please see the screenshot below where an activity started at 6:49PM and ends at 6:46PM which is not reasonable. I suggest adding a validation to the end time of activity insertion such that one cannot insert activity with end time earlier than its start time. 2- When there is a Place between two trips that are in two different days, I think it is not a good idea to prefill the end time of the activity with the start time of the second trip in next day (because in this case we may prefill activity such that its end time is earlier than its start time). I recommend setting it to 11:59PM, if date is the same as the first trip. The same logic can be implemented if date is the same as the second trip. In that case, the start time can be prefilled with 6AM and the end time with the start time of the second trip. 3- In the current version, when user adds activities with different dates, she cannot see the dates in the Place card and it is vague that to what date each activity belongs. Please see the screenshot below. In this screenshot these two activities are in two different dates but this is not clear from what we display to the user. Can you please add a date what we display as the activity in the place card? @shankari Could you please let me know if you still think we need to chat in person? If yes I'm available on Wednesday or Thursday (Sydney time). |
@MaliheTabasi if you have been tracking updates to the issue, you will see that we already have fixes for the issues that you originally reported (which I think are the issues that you have reported here #871 (comment)).
Please test against version 1.2.2. May I also suggest filing separate issues for each pending issue so that we can check each issue off once it is done? That way we can more easily keep track of the easy issues which are already fixed, and the more complex issues which require additional discussion/design. |
@shankari Thanks for fixing those issues. Sure, I will do so. |
Closing this since we have filed separate issues for the various bugs/new features |
Hi @shankari
We have installed the app and tested the time-use feature. Please find our comment in this issue.
The text was updated successfully, but these errors were encountered: