kblincoe / qualopt_se701 Goto Github PK

On the individual page for each study, the email to be sent out as the invitation is displayed. However, it does not display the proper formatting of the email, as all the email is incorrectly put on to a single line. This page should display the email properly with all the new lines correctly shown.

Currently, the database being used causes an error when QualOpt is compiled for production as no connection can be established with the server. This database holds all the registered users and the GitHub data for sending invitations. The app cannot function without the database connection and must be fixed.

When you are creating or editing a study, the first input box's title is 'Name'. This should be changed to 'Title' to more accurately reflect the language used to describe a study

Currently, there are only five filters for finding participants. However, the GitHub API can access significantly more information than what is currently being used. Add as many filters as possible to more accurately reflect the information available.

Currently, the Twitter sign in gives a response of: "Your request cannot be processed. Sorry, an error has occurred. Status: Not Found (Not Found). Message: Not Found" when attempting to use it to sign in.

Currently at the bottom of every page is "Created by Fraser and Dinal". This needs to be replaced with the University of Auckland.

Then on the information page add an about section saying that the project was originally worked on by Fraser Lewis-Smith and Dinal Wanniarachchi, was further developed by the students of SoftEng 701, and was all under the supervision of Dr Kelly Blincoe.

At the moment to create a new study, you have to fill out every single info field each time. Users would want to be able to copy existing studies so they do not have to fill out the fields if they are keeping information the same for the new study.

On the create/edit study screen, the input box for the incentive is currently just a plain text input. This input should be changed to include two inputs.

One which provides the category of incentive out of the following:
-One-off payment
-Randomly allocated
-Other

This should then be followed by a text input box where the user can enter either the amount of money or the details of the incentive if other is chosen.

As QualOpt saves users' login, emails and passwords, there needs to be sufficient security to ensure none of the information can be accessed by a third party. Investigate how this information is stored and how easy it would be for an unsolicited user to access.

In the edit/create study screen, it is difficult to choose participants as it requires highlighting and cumbersome methods of selecting items from the list.

At the moment the user has to write their email from scratch everytime. It would be useful for the user to be able to save default email templates. A default email template could also be provided for them.

Currently, you can only send emails manually. It would be useful to be able to preset a time and date for research emails to be automatically sent out. This should also include which timezone the time is according to.

When you open the create/edit study page, the top input box should be auto-focused so that the user can immediately start typing in the title of their study.

Currently, when entering an email on the register, reset password and settings pages the only requirement is that the entered string is 5 characters long. However, this is insufficient to ensure the email is in a valid format. It should check that the entered text is in the correct email format containing an @ with characters both before and after the symbol.

It would be useful to be able to save a set of participants chosen for one study so that they can be reused for another study.

On the overall study page, there are three buttons at the end of each individual study (View, Edit, Delete). If you go into an individual study these three buttons are replicated in a similar manner but with slightly different purposes (they are instead: Back, Edit, Send).

However, the buttons are displayed in different ways on each of the pages. The overall study page buttons are joined together and the individual study page buttons are separated. These buttons should be displayed in similar ways to provide unity to the program.

There is a typo currently in the readme.md. The typo reads "impletmented" but should say "implemented".

This picture portrays the issue:

Edit: The estimation for this is very small. I tried to add the label when I created the issue, as mentioned in another comment, but couldn't.

Currently the application is reportedly not working with JRE 9 from the README due to a bug with Spring Boot. An investigation is required into what is causing the incompatibility and for it to resolved.

Problem likely to be library based, not much information known at this point except what I have been told.

It would be useful to be able to have different statuses of the studies based on where they are in the process, i.e. inactive (before any emails have been sent out); active (once emails have been sent out); completed (once the user is finished with the study).

This should be able to be viewed from both the individual study screens and the overall study screen.

Currently, if an email is sent to an incorrect email address there is no way for the user to know this from within QualOpt. It would be useful, after sending the emails, to be able to see which ones were successful and which bounced back.

Once you register an account the social media login buttons remain and the default account message also remains. If the user has just registered an account they would not then login through a different method. Therefore these should be hidden after the user has successfully registered a new account

Such a setup would mean that you can easily have an environment up to develop and test both parts of the application.

Currently, the site has only used mock participant data or no data at all for its database. The site needs to be connected to the GitHub API so that actual GitHub users' emails can be acquired for use in surveys.

NOTE, REALLY IMPORTANT: You must ensure that no emails are sent to actual GitHub users during this project. Make sure that either any emails are sent to fake addresses or that no emails are sent at all.

On the individual page for each study, there is the option to send out the invitations to participate in the study. However there is no confirmation when you click 'send invitation' to ensure the user is certain of all the details they have entered. This could lead to human error and accidentally sending a survey to many people.

Currently, a study can only be deleted from the overall study screen. A user would likely want to be able to delete the study from the individual study screen too so they can have multiple options of where to click from.

If you attempt to reset your password with an incorrect email, there is no response from the site telling you that the email is incorrect. Essentially nothing happens once you click reset. A warning should appear which advises the user that the email is not registered.

There is too much information in a small space on the home page. The user would be less overloaded if the information was moved to a separate About page. Also the small about section added in #18 should be moved to this new page.

The information left behind on the home page should be sufficient for a user to still be able to figure out the purpose of QualOpt.

Create a batch file which can be used to launch QualOpt, this would involve running both mvnw and yarn start. This would be a small task however would add convenience for all developers.

If a visualisation could be added, such as a cluster graph of the types of users based on their GitHub information, it could provide a clearer way of selecting participants for researchers. It could also provide an alternative method of filtering if the visualisation could be made interactable. This visualisation could be featured both on the participants' selection screen but also the home screen to help inform users.

When you change your password it doesn’t ask for your current password, which is an important security measure. Need to check the entered current password matches the logged in user’s password, before allowing the user to change their password. This needs to be done in a secure way so that the password information cannot be intercepted via external sources.

Currently, the home page is overcrowded with too much information. #29 will be moving most of the information to another page. So the home page can be re-designed to better fit the goals of the site.

Currently, invitations can only be sent from the view study screen. A user would likely want to send from the overall studies screen too so they can have multiple options of where to click from.

Currently, the pop-up section for creating a study is identical to the section for editing a study. This is correct except the title is currently ‘Create or edit a study’. This should be changed to be ‘Create Study’ if a new study is being made or ‘Edit Study’ if an already created study is being edited.

There is an unnecessary page under settings called Sessions. This page serves no discernible purpose and should be removed from the site.

As per the title, the maven script doesn't run in docker containers because of git's line normalisation making the batch file CRLF if cloned on Windows.

I couldn't get this script to actually run in a docker container till I fixed this, so this minor issue should probably be fixed and updated in .gitattributes?

Currently, all you can send in the email message body is plain text. It would be useful to be able to add attachments to each of the emails.

On the individual study page, the back button is the same colour as the view button is on the overall study page. Back and view do not have a similar meaning and thus it is confusing for the user to have them associated with the same colour.

At the moment the user can only send a single generic email to all participants. It would be more useful to be able to send a custom personal email to each participant about the study. Including their name and other details from the GitHub database.

Information on how to use the customisation should be added to email creation section when making a study. There will be more room for this after it has been moved in #33

Currently, QualOpt is only able to send out mass emails to find participants of studies. After that, it is up to the researcher to correspond with the participant.

It would be useful if each study could have its own page on QualOpt where participants could find more information about the study, as it could possibly contain a FAQ. This would decrease the mail correspondence for the participants as they could just find all information on the site. It would also be useful to be able to upload a pdf to be hosted as part of that page. The researcher should be able to freely edit that page to keep up-to-details about their study.

Currently, the registration email does not send as there is an authentication exception from the Google API. This will need to be fixed before anyone can register their own accounts rather than just using the default account.

Currently, users can't login using their GitHub accounts. GitHub is the key resource and interaction point for this site and thus the site requires an interface for GitHub users.

Once logged in to QualOpt, GitHub users should be able to opt-in/out of studies and limit the number of invitations they receive from researchers. They should be able to restrict what kind of studies they would like to receive emails about by what the incentive is (the redesign in #5 will help quantify the incentives). This should then affect what searches they appear in when looking for participants.

Also, a permanent signature needs to be added to the bottom of every email from QualOpt which links back to the QualOpt opt-out service.

When you set up a study, there is currently no way of filtering the participants to choose from. The Participants page contains the way of filtering the participants.

This filtering of participants should be moved so that it is done for each study. #33 is splitting the study creation into two parts: main info and email. A third section should be created for the filtering of participants. Once the section has been successfully added there will no longer be a use for the Participants page so it can be removed.

On the Participants page, there are currently five text boxes for choosing filters for the database. For filtering quantitative values, such as the minimum number of repositories/contributions, it would be better to constrict the user to the limits of the data rather than giving them infinite range.

Along the menu bar at the top of the site are options for: Home, Studies, Participants and Account. The options Studies and Participants have no use until the user is logged in. They should be hidden from the menu bar until the user is logged in to help declutter the login process.

Currently, the user can only send their email out to all participants listed. There should be an option for sending the email to all the participants or if more participants have been added since the first time it was sent, then only sending to the new ones in the study, who haven't received it.

In the study section, there's the option to edit a study you have already made. However, there is no warning that you have unsaved changes after editing one of the pieces of information. This could lead to a user losing the work they have written if they accidentally click out of the screen

Currently changing the user's password is done on a separate page under options not under settings. Password changing should be moved to the settings page to reduce clutter.

The settings page should then be then redesigned so it is more obvious when you have made a change to a value. This is to ensure you don't accidentally change a value you did not intend to.

Currently, the Google sign in gives a response of: "401. That’s an error. Error: invalid_client. The OAuth client was not found" when attempting to use it to sign in.

Currently, all the information for a study is entered on a single page. This means that the input boxes are all smaller than what the user might require. This is especially true for the email input box.

To improve the workflow, the email information input should be moved to another screen following on after you have entered the study details.

Currently, the Facebook sign in gives a response of "Invalid App ID. The provided app ID does not look like a valid app ID," when attempting to use it to sign in.