You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 8 Next »

GR2 - Designs

Scenario

Katie, an MIT Physics junior was working late in J-Lab, as one does, and when she got back to EC she was too tired to cook. Having worked hard for the last several hours, she decided that she deserved a treat, some Thai food. Pepper Sky's seemed like a great call, but she knows from experience that her $11 curry won't meet the $15 delivery minimum. She could order something else, but she is strapped for cash and it would be more fun to eat with some other people. She remembers that great website those kids in that UI class made that perfectly fits this problem, so she loads up Hungr.

She logs in and brings up the new order menu. She clicks the Facebook Connect button and chooses some friends in EC and Senior House. She adds that she wants Thai and is thinking Pepper Sky's, leaves the default join the order time and clicks Go. Across EC and at Senior House, people's new email sounds go off and they see the subject: "Katie is hungr-y for some Thai, are you?". Some of her friends, Kerry, Janet, Tim, and Todd think "Thai is exactly what I need to make this pset suck less", open up the email, and click the link. Other friends have already eaten and don't bother.

After a few minutes pass, every interested is on the choose restaurant screen. There is some discussion about some other Thai restaurants and some nominations, but in the end everyone decides that Pepper Sky's still sounds great. Katie hits the go button and everyone gets the select food screen. Katie adds her chicken curry and Tim adds a scallion pancake.  Neither Janet nor Todd are that hungry, so they are looking to split something. Janet adds an order of Duck Choo Chee from the menu and pushes a "half" button. Todd sees the half indicator and thinks that Duck sounds great so he pushes the half button too. Todd adds a Thai Iced Tea and Janet and Katie hit the "me too" button. The order has come together and every body has indicated they are done, so Katie hits go and Foodler opens up with the order all filled out. The main hungr interface switches over to show what everyone owes and has options to inform everyone when the food arrives.

When the food arrives, she hits the button and everyone gets another email "Katie says the food's here. Hope you are still hungr-y". Everybody comes over, pays up, and enjoys the deliciosu Thai food. They are real happy and full. 

Designs

Design 1:

  1. First, Katie needs to invite friends to an order. She can do so by searching through her Facebook friends via dropdown menu or by typing in an e-mail address of the desired person. If desired, she can choose a restaurant or type of restaurant from the searchable drop down list of restaurants in her area. In this case, she chooses “Thai food”. She can also optionally add a message to send along with the invitation. She can view the people she's added in the “friends invited” box on the right side of the screen.
  2. Next, she is directed to the page to choose a restaurant. Upon accepting the invitation, her friends can also view this screen, and their current status will be displayed in user info tab on the left. This will show what friends are currently viewing the page, which haven't responded yet, and those who declined. Tim really doesn't want Mexican food tonight, so he filters out Mexican restaurants. Janet wants Thai food as well, so she also selects restaurants with Thai food, and the multiplier next to Thai food is increased to show that multiple people favor this option. As Katie and her friends are offering up suggestions about food, a list of constraints is displayed on the right. Once everyone has agreed on a restaurant, they can select that restaurant as their current restaurant (option on list of restaurants) and proceed to order.

 

  1. Now, everyone is looking at the same ordering page. They can clearly see who has chosen food and who hasn't on the right side of the screen. Each food item will have a “me too” button near it. In addition, on the right side of the screen will show a list of people willing to share items. For example, when Janet requests a half order of Duck Choo Chee, her request will show up on the right as “Janet is willing to share Duck Choo Chee with one other person.” Tim was slow in choosing his food, so his name remained highlighted in red for quite some time.
  2. Finally Katie gets to finish her order. She records a tip amount in hungr and everything is sent over to Foodler for finishing. She double checks everything and submits the order. She can also view an itemized list of what Janet, Tim, and Todd owe her. When food arrives, she can push the painfully large “food is here” button.

  1. At this point, Janet, Tim, and Todd are redirected to a page showing how much they owe, what they ordered, and where and when the food will arrive.
  2. Lastly, an optional perk would be to show users past orders so that former organizers can keep track of who owes them money.
General Comments on Design 1:

Learnability: To start up an order, it is quite natural for people to search through list of Facebook friends, as everything is connected to Facebook nowadays. In addition, the general ordering interface will be very similar to that of Foodler (and can be used exactly as such at a base level). Our interface will simply provide simple add-ons that people can choose to use (“me too” “I want to share”). In addition, payment will be the same as usual for online food ordering.

Efficiency: Linking up to Facebook will allow suggested completions of names, as will providing a list of restaurants. This will help organizers suggest things quickly and easily. Selecting restaurants will be facilitate by filter options embedded in the menu (“want” “do not want”) next to restaurants and food types. In addition, allowing people to order simultaneously cuts down on menu reading time and allows for more instant feedback regarding choices.

Visibility: This interface prioritizes important things, including showing a list of people included in the order as well as constraints imposed upon restaurant selection. Showing the constraints allows the user to know whether or not a certain restaurant will be “allowed”. In addition, familiar “vote up/vote down” buttons will perform the obvious action of showing the other users a list of preferences in the right pane.

Errors: Users will have the option to delete items that they don't want from their order, as well as the option to back out entirely. Essentially, one can undo any action up until the time that the order is submitted. If this design goes without a chat box entirely, the users may run into a communication issue if the interface doesn't take into account every desired communication. For example, Janet could have ordered half the duck, but Tim could have clicked the “other half” button first, leaving Todd still wanting to share the duck. In this model, without the chat box, there would be no way for Todd to communicate to Tim that he specifically wanted to share the duck with Janet (ie optimizing food order could be tricky).

Design 2:

  1. William wants to invite his friends into an order.  He logs in to the page through Facebook connect, and is brought to the create order page.  Here he is able to type in his friends' names (with suggestions provided by facebook), as well as the e-mail addresses or phone numbers of other people he wants to invite to the order.  Once he he has selected at least one friend, the arrow leading to the next box lights up.
  2. Here he is able to choose the restaurants he wants to seed the list with, and can search by name (with suggestions provided via foodler) and in addition filter these by category and price.  Adding one or more restaurants lights up the arrow for the next box.
  3. The Get Started box holds a large "Send!" button and a box to enter an optional message.
  4. Content in any of the boxes can be modified at any stage until the form is sent out.

  1. Heather receives the e-mailed link from William, and deciding that she's up for Thai food, clicks it.  She is automatically identified by the unique url.
  2. She chooses which restaurants she wants to order from, and briefly thinks about suggesting somewhere else before deciding that Pepper Sky is good enough for now.
  3. William, when everyone has finished, clicks the selection button for the restaurant.  The arrow heading to the next box lights up, indicating it's time to order.
  4. Heather, seeing this, clicks the view menu button, which brings up a floating div containing the restaurant menu in a foodler-style format.  This menu allows her to view prices, select special options, and specify if she would want to split the meal or eat the whole thing.
  5. The menu dynamically updates to show the selections of the other users, and indicates the split/full status of each item.
  6. At any time, the user can remove an item they've selected.
  7. When Heather is done ordering (having decided to just save some of the rice for later), she clicks the done button at the bottom of the frame.
  8. After everyone has done so, William brings the Order to completion.

  1. Once this is complete, Heather is brought to this screen, informing her how much she owes, and to whom.
  2. She will also receive an e-mail when the order arrives, reminding her of how much she owes.  The e-mail will use dynamically created images and HTML to show the status of their payment.
General Comments on Design 2:

Learnability: The process of adding friends is similar to that of adding friends to things on facebook, primarily because it is processed through a facebook interface.  The glowing arrows provide a guide as to what the process is, and how it should be followed.  A lot of the interface components are similar to those on other sites, e.g. the "doodle-like" restaurant selection frame, so should require little learning.

Efficiency: As with design 1, Facebook will suggest completions of names, and foodler will provide a list of restaurants. This will help organizers suggest things quickly and easily.  Filtering options will make this even faster.

Visibility: The process is well demarcated by the "glowing" arrows.  The internal information storage (such as selected users and restaurants) is visible at every step, and other users' interactions are also easily visible as the system progresses.

Errors: The system is divided into two primary steps, creating the order and placing the order.  In each of these steps the user is able to back up and reverse any of the choices they have made, but they can't usually jump between states.  This is alleviated by making the most important details (e.g. participants, restaurant selections) modifiable by all users in the second stage. For example, sharing the link to the page allows one user to add additional people to the order, even after the ordering has begun.  The second page only allows adding (and not removing), though, as it ensures that the involved parties will never lose track of what happened in the past and lead to their current state.

  • No labels