Contextual Inquiry-Group:100 Proof
From CS 160 User Interfaces Sp10
- Esther Cho: co-ran contextual inquiry interviews, participated in interface design brainstorming
- Calvin Lin: participated in interface design brainstorming, wrote up Problem & Solution Overview and Task Analysis Questions
- Brandon Liu: co-ran contextual inquiry interviews, participated in interface design brainstorming, created interface mockups
- Nathaniel Baldwin: co-ran contextual inquiry interviews, wrote up Target Users and Contextual Inquiry sections, participated in interface design brainstorming
- Matt Vaznaian: participated in interface design brainstorming, created interface mockups, wrote up Interface Design section
Our rationale behind our choice of target users is fairly straightforward. Our app caters toward beer lovers and is designed to be used while the user is out drinking beer at a bar – so we simply went to one of these bars (namely, The Trappist, in Oakland) and talked to the people there.
Person one was a female in her thirties, a beer professional who had been working in the specialty-beer industry for about four years. She considered herself somewhat of a beer expert. She samples beer at least five times per week as part of her job. She’s interested in trying beer of a wide variety, especially Belgian ales. She’s also interested in the specific aspects of a beer as put forward by the Beer Judge Certification Program (bjcp.org).
Person two was a male in his early fifties who was currently unemployed. He samples beer perhaps three times per week. He considers himself a beer enthusiast, not aficionado. He prefers beers with (relatively) lower alcohol by volume, so that he can enjoy a greater quantity without getting to drunk. He prefers maltier beers, and dislikes aggressively hopped beers. He is very interested in locally-produced beers, to reduce his carbon footprint. He places importance on how well specific bars maintain their equipment (e.g. the lines that run from the kegs to the draft nozzles).
Person three was a male in his late forties, who didn’t give an occupation. He considers himself a “hop-head”, and greatly enjoys hoppy tasting beers. He also said that he was not especially skilled at tasting beers, just enjoyed it casually. He estimated that he goes out to drink roughly every week or two. He says he likes to “plumb the depths of a category” of beer – picking one type (e.g. American IPA) and trying many different examples of that type. He likes to know historical trivia about different beer types. He’s interested in the subtle tastes behind beers.
Problem and Solution Overview
Our group aims to solve the problem of remembering and detailing the subjective experience of tasting a beer. Extensive drink menus make it difficult to remember particular beers and decide on new beers to try. The application we propose will address these issues by providing a quick and simple interface for recording one’s thoughts while tasting a beer without being intrusive to the drinking experience. Then, based on user input, the app will recommend drinks with similar qualities.
Contextual Inquiry - Interview Descriptions
As previously mentioned, our interviews were conducted at The Trappist, an Oakland bar specializing in high-end beers. We began by asking our interviewees to describe the process they go through when first arriving at a bar. They all described the process of scanning the menu for beers that appealed to them. Accounts of this were fairly similar. They each mentioned having a style of beer in mind, and looking through the menu for beers of that type based on their names and descriptions – both by style of beer and by keywords in descriptions. Person two also said he likes to ask the bartender for a suggestion, though he said it can be frustrating when at a busy bar and the bartender lacks time to give him attention.
After that, we watched them sip their beers, and asked them to comment on their thought process as they formed impressions of the brews. A number of beer-related aspects were brought up here in addition to overall impression: aroma , hoppiness, maltiness, bitterness, sweetness, yeasty taste, “mouth-feel”, aftertaste. Each of the interviewees also expressed that they formed an impression of what I will call “je ne sais quois” – a taste impression of characteristics of the beer that are hard to fit into a particular category. Person one in particular talked about picking up flavor hints that were evocative of other types of food, mentioning that a beer she was drinking reminded her slightly of oranges.
We then asked our interviewees how they remembered details of the beers that they tasted, they all admitted it was difficult. Person one said that it was simply a matter of lots of practice (as she was a professional). Person three said that he remembered beers that he liked because he tends to re-order them, though that sounded a bit circular to us.
We asked our interviewees what aspects of a beer would be important to them to remember. Price and quantity for that price were a unanimous concern, tied to the particular location they were at. The above-listed taste qualities came up as well. All three mentioned, in some form, a desire to remember how a beer stacks up against other beers that they consider similar, or at least what beers a particular beer reminded them of. They also expressed a desire to be able to take a photo either of a beer or of something related to the tasting experience (e.g. their friends drinking the beer). Person one thinks the format (draft vs. bottle) and glassware used are important. She also mentioned that if she was drinking with friends, she would like to remember which beers they liked, for possible later gift-giving purposes. She also mentioned wanting to remember a beer's "drinkability" - whether she could enjoy it easily, or if it was more of a "sipping" beer. Person two was very concerned with the beer's origin (as stated, he preferred to drink locally-produced beers that did not require large amounts of transportation). He also mentioned, unprompted, that he would like to get updates about local bars (such as new beers that are being offered) and was interested in having and sharing information about how well a particular bar maintained its facilities, as noted. Person three was interested in beer history, as mentioned. He also was interested in the "five senses" of taste and how a particular beer triggers them. In addition, he mentioned wanting to remember what he thought a beer would be good for - as in a good summer / winter beer, or what foods it might pair well with.
Task Analysis Questions
1. Who is going to use the system?
Identity: Beer enthusiasts
Background/Skills: Varies across user group: drinkers still developing their palette to experts with a developed palette.
Values, Likes/Dislikes: Don’t want to take time to write notes. Want to remember particular beers. Likes tasting a variety of beers. Appreciate the quality in a beer, how it’s made, where it came from. Care about bar quality.
Personal Characteristics: Wide range of people; all socioeconomic groups and ethnicities, all legal ages, no specific education.
2. What tasks do they now perform?
Evaluate and form opinions on different beers based on various characteristics such as aroma, flavor, and aftertaste. Criteria varies from person to person. Users rely on memory and repetition to recall experiences and drink details. Users talk to bartenders and friends for recommendations and other information.
3. What tasks are desired?
Fast way to recall and retain experiences users have had with specific beers. Recording thoughts and detailed information on beers while tasting and evaluating based on subjective and customizable criteria. Receiving customized drink recommendations based on user’s preferences.
4. How are the tasks learned?
At minimum, users simply use their subjective general opinions on drinks as input. Through experience, users will develop an understanding of beers and their various characteristics, and be able to distinguish between beers based on quality. Training only supplies information; skills are developed through experience. Interface must align with accepted beer vocabulary and natural evaluation process so users will immediately understand the interface.
5. Where are the tasks performed?
Bar, restaurant, pub, beer festival. Social environments. Loud environment, lighting varies. Not common to use phone for extended periods of time.
6. What’s the relationship between user & data?
- Personal data accessible only from the same phone. Data is stored locally on the phone. User has control over own data. User can choose to share data with friends over the Internet.
- Common data: beers similar to one another. Can be used concurrently. Cannot be modified by individual users.
- Must have your phone in order to access data.
7. What other tools does the user have?
“Expert knowledge” of the bartender, knowledge and experience of friends, personal experiences, books, the Internet, camera.
8. How do users communicate with each other?
App is meant for personal data lookup and storage. Social feature allows group of friends to input data on one phone. Sharing feature allows user to send data to other app users across the Internet.
9. How often are the tasks performed?
Varying frequency based on how often users drink and how often they try new drinks. Some drinkers are more adventurous than others – try new drinks often and thus would use the app often to record data and discover new drinks. Others stick to their favorites and would have less reason to input their thoughts or find new drinks. Also depends on how much a user cares about logging their drinking habits – expected to vary.
10. What are the time constraints on the tasks?
Tasks must be quick and fluid – users don’t want to spend more than a few moments inputting data. Real users noted that while they'd like to have the information and memory aid notes would provide, they still didn't have lots of patience to take notes. In general, performing all tasks must be quick since users would not want to spend an extended period of time using a device. Must keep social context in mind - people want to be interacting mostly with each other, not with the device.
11. What happens when things go wrong?
Return to previous point if app closed on accident in middle of a task. Be able to easily edit and remove all data entries. Must be able to quickly correct mistakes during entry, i.e. typos, bad photos, bad voice recordings, mis-hits with finger. Be able to go back in steps if user realizes mistake later in the process. Be able to cancel and start over.
Analysis of Tasks
1. Leave additional info about a drink. This is not the actual beer impressions process but a way to put down any additional info a user would like to remember about a beer. We can't always assume what a user wants to know about a drink so allowing a place for a user to put in their own info accommodates for that.
How difficult is it to currently perform the task with existing tools and applications? Medium, since taking notes often boils down to typing in descriptions.
How difficult would it be to perform the task using your proposed application? Easy, since our interface won't focus on having to type in information. We'll make use of tags and sliders.
2. Search for a beer in different categories or new beers to drink. Users need a way to find their beer through a list but by allowing beer to be searchable by other characteristics of a beer, users can also explore new kinds of beer when looking for a beer with a specific quality.
How difficult is it to currently perform the task with existing tools and applications? Medium, since existing apps don't have structured ways of searching for beer.
How difficult would it be to perform the task using your proposed application? Easy, since we'll let the user create their own categories.
3. Rank beers differently in different categories. Beers have many different characteristics associated with them and we want users to be able to rank beers separately within the context of any of these characteristics.
How difficult is it to currently perform the task with existing tools and applications? Hard, since existing apps only have support for scoring from 1 to 5 stars.
How difficult would it be to perform the task using your proposed application? Easy, since the user can simply drag and drop.
4. Visualize tastings of a particular beer. After accumulating many tastings, reviewing that data can be a daunting task. By visually seeing the data, users can remember a particular tasting and compare two different tastings as well as easily assess the beer's characteristics.
How difficult is it to currently perform the task with existing tools and applications? Hard, since existing apps don't have features for individual tastings.
How difficult would it be to perform the task using your proposed application? Easy, since one beer has support for multiple tastings, and users can visualize these.
5. Do a quick, full evaluation of a beer. Because users are out to drink beer with friends, leaving an evaluation on a beer should not take up much of their time. One interviewee had even said that he is too impatient to take notes, but that it also would have been faster to remember a beer if he had. By easily allowing a quick but full evaluation of a beer, the app does not detract from the beer drinking experience, yet yields detailed impressions of the beer being consumed.
How difficult is it to currently perform the task with existing tools and applications? Hard/nonexistent: the only way to take notes about taste is by using the keyboard.
How difficult would it be to perform the task using your proposed application? Easy: they can do it with one hand and in real-time. they don't have to fill in information later.
6. Synthesize multiple users' impressions, all on one phone. When users go to bars with friends, we wanted the application to include the user and his friends as a way of social interaction. One interviewee mentioned that when he goes to bars with friends, they each order a beer but take turns trying it. Knowing this, the task for the app is to be able to record different people's impressions on different beers.
How difficult is it to currently perform the task with existing tools and applications? Hard: nothing like this exists.
How difficult would it be to perform the task using your proposed application? Easy, since the app will set up these social features based on the # of users and # of beers the group inputs.
Our application will focus on the following functionality: Inputting a thorough review of a beer, searching through previously added beers, organizing them by various categories and ranking factors, and allowing a group to do reviews of multiple beers at once. When a user brings up the main screen (Task 1, Fig. 1) they are presented with four button options: [give a] new review of a beer, search your beers, organize tastings, and [start a] group evaluation. The next section describes each task in detail.
New review of beer: When a user presses the green button on the main screen (Task 1, Fig. 1) they are brought to the General info tab (Task 1, Fig. 2). Here they input basic information about the beer they are trying, which includes the name of the beer, where they are drinking it, the price, overall rating from a scale of 1 to 10, and if the user wants they can tap the square icon in the top left corner to take a snapshot of the glass/bottle for future reference. When are done with this page they can click the Details tab on the bottom to take them to the Details screen (Task 1, Fig. 3). Here the user will be entering and reviewing more specific information about the beer. From the top to the bottom you have brewery, country of origin, a slider for beer color with values between light and dark with gold in the middle, a slider for aroma with values between mild and dark with medium in the middle, a text box for optional input of a specific kind of smell, and a place to enter a specific season for the beer. Pressing the third tab on the bottom brings users to the Taste Wheel screen (Task 1, Fig. 4). This interactive tool lets users slide their fingers over the different tastes as a way of inputting how they feel about them. For example, if a beer is very bitter the user would slide their finger over the bitter section of the wheel. The longer they hold it there tells the app how “bitter” the beer tastes. Finally, tapping on the rightmost tab brings users to the Tabs screen (Task 1, Fig. 5). At the bottom of this screen a paper sack is displayed, surrounded by different tags that each having their own unique colors. Users can drag tags into the sack to associate them with the beer. Having tags allows for various sorting and organization functionality later.
Search your beers: Pressing this button from the main screen simply takes you to a long list of your previously entered beers, allowing you to view, edit or delete them (Task 2, Fig. 6).
Organize Tastings: Pressing this button brings the user to a page dedicated to organizing and sorting their beers (Task 3). Users can sort their beer relative to other beers in the same category, for example: by origin, or by sweetness. There's also a "visualization" of a specific aspect of a beer, made using a sparkline graph.
Group Evaluation: Pressing this button brings you to the Group Evaluation page (Task 6). To start an evaluation, you first enter all the names of the people and all the names of the beers around the table. When pressing Start Evaluation, you are taken to a matrix of screens where each person has their own column and each beer has its own row. The ratings for group evaluation are done using the taste wheel for easy input and interaction (Task 6.2). At any point you can end the evaluation with the End Evaluation button and all the input will be stored. The inputs from different users and tastings can then be visualized, adding a "social game" aspect to the application (different people's impressions of the same beer)
1. Easy: Alex goes to his favorite beer joint, the Hartmann Lounge. He sees that they have a new beer on tap, called Agrawala. He orders it and as he takes the first few sips notices it has a really distinct flavor unique to IPAs he has tried before. He remembers that he just downloaded a great new beer app that will let him make a great review. He opens up the app and presses New Review of a Beer. This takes him to a page to fill out basic information. Alex goes through each tab at the bottom, rating the smell, color, marking that it is a seasonal to the winter, Using the color wheel to effectively convey his pallete's feelings, and adds a couple tags for referencing the beer later. When he is done he presses Save in the top right corner and continues to enjoy his new found beer.
2. Medium: Matt enters the Jupiter and wants to see what he's tried there before. He can open up the "rankings" screen and see, in an order he created, what beers he had tried in the past, and which ones are his favorites. If he's in the mood for a sweet beer, he can get information aggregated from previous tastings, so he can determine what beers are sweet or have a sweet aftertaste. (Task 3 sketch)
3. Hard: Calvin, Esther, Nat and Brandon all go to the Trappist together, where they get the sampler with 4 beers. They enter each of their names and the particular aspects that they most enjoy, for example, Brandon:hoppiness, Calvin:sweetness, Esther:fruity flavor, Nat:Maltiness. The app creates 4 pages and 4 beer evaluation pages. (Task 6 sketch) They can each take detailed notes on the beer's tastes using the taste wheel, and pass the phone around. When they're done, they can compare notes. (Task 4-5 sketch)
Analysis of Approach
The first and foremost of the iPhone's affordances is mobility. This is advantageous to users because they would already have their iPhone with them on the go, and it would be more practical than, say, bringing their laptop or a notepad.
The iPhone also affords users the ability to input information via touch screen. Although it may seem no better than bringing pen and paper to jot notes, the touch screen allows for faster input simply through touch and drag. This allows users to input their evaluation of a beer quickly into their phone using one hand while tasting their beer with the other without breaking their locus of attention from where it should be - the enjoyment of their beer. A quick interface such as this would encourage users to use the application to take notes rather than give up on note-taking entirely due to lack of patience - a serious concern that we came across in our interviews.
Another affordance the iPhone brings is that it allows users to accumulate large amounts of data while still being compact. One issue with writing out notes on paper is that after a hundred tastings, a user would have accumulated a large amount of data that would be bothersome (at best) to carry around for reference. The iPhone allows users to have their information on hand at all times and quickly retrieve a particular entry by flexible search, rather than having to flip through paper notes.
One obvious alternate solution to the problem of remembering beers is a pen-and-paper notepad; we've already clearly outlined how this is less than ideal. Another potential solution to the problem is found in other applications on the market at the moment, and consists of essentially the ability to rate a beer on a scale from 1-5, and then type in additional notes by hand. This, unfortunately, is little better than the pen-and-paper scenario in terms of input speed. While a section for notes is an important feature of a feature-rich application for beer note-taking, relying on it heavily is likely to be cumbersome enough that it would result in the aforementioned eschewing of note-taking. Our approach allows users to quickly give their beer impressions and later to easily view, in a graphical format, myriad aspects of their impressions on the beer's different characteristics.
Pros of our approach include a quick way to input detailed and measurable information about a beer in an unobtrusive manner. Our approach is also heavily cross-referenced and categorized, to allow the data to be browsed and arranged in whatever manner most makes sense to the user. Additionally it is highly portable.
One possible con to the approach is the (hopefully slight) learning curve involved in learning our "beer wheel" input method. Similarly, the first time several people sit down for a group evaluation, each individual (other than the phone's owner, presumably) may have to learn the interface, possibly annoying the participants. Finally, you might spill beer on your phone. On the first two points, our aim is to provide as clear and intuitive an interface as possible, so that people will hopefully learn quickly enough that any difficulty in entry will be outweighed by the bonuses of having their impressions for posterity. The final issue is probably beyond our control.
jupiter's stuff - http://www.jupiterbeer.com/jupiter/beer.htm
paper sack - http://www.clker.com/clipart-14581.html
right arrow - http://www.clker.com/clipart-16561.html
iPhone GUI PSD 3.0, Teehan+Lax http://www.teehanlax.com/blog/2009/06/18/iphone-gui-psd-30/