This project is to design a revolutionary, intuitive, and absolutely unique interface for a currently non-existent system. This system will be constructed as a Senior Design Project at South Dakota State University. When it is completed it will have the ability to mix beverages with any combination of the sixteen liquids contained inside it. As there is no current system this will be a complete and one-off design from nothingness.

Interviews

All three of the following interviews were prefaced by the same header or background information. I gave the users a very general sense of what the project was so that there answers would not be influenced by my preconceived notions.

My name is Derek Fernholz and I am building a Computerized Automatic Beverage Mixing system, in other words, a "Computerized Bartender". This system will use a touchscreen to allow users to choose what drink they would like from the available drinks, customize said drink, add a new beverage to the database, among other things. Once the drink is chosen and confirmed the machine will take over and gather the required ingredients from the 16 different liquids in the system to concoct the selection.

Interview 1

Interviewer: Derek Fernholz
Interviewee: Kristian Peterson
Age: 23
Culture: American
Computer / IT experience: Little to Average

1. Q. Is there anything you need further explanation or clarification on?
   A. What Happens if Liquid gets spilled on the touchscreen? How user friendly will it be? Will I need a deep knowledge of computers to be able to use it?


2. Q. Have you ever seen such a system in the "wild"?
   A. No, never, this is a new concept to me, and an exciting one at that!


3. Q. Do you think that this type of system has valid real world application?
   A. Definitley! It would increase efficiency and decrease cost (by having to staff fewer postions or have as many people working at a given time) and wait time in an establishment that serves mixed drinks to its patrons.


4. Q. Would you like to use such a device?
   A. Yes, it would enable me to try a variety of different drinks independently (I wouldn?t have to rely on another person to make them for me). It would give me the chance to taste what a specific drink is supposed to be like because the proportions would be precise after eliminating the human factor.


5. Q. What potential problems do you as a user foresee for such a system?
   A. Having a certain liquid required for a favorite drink not available within the chosen 16 ingredients. Another problem could potentially be, letting people enter their own drinks, making sure they don?t mess up any preset options or break the program in any way.


6. Q. What are some important features or things you would like to see implemented in this project?
   A. Accuracy of the liquids being dispensed in the glass (making sure all the liquid that is supposed to be in the drink physically gets into that drink). Making sure all electrical components are safe from any potential spills or accidents within the system.


7. Q. Do you have any questions you would like answered at this time?
   A. I'm thirsty! When is this system going to be ready?

Interview 2

Interviewer: Derek Fernholz
Interviewee: Andrew Boerema
Age: 21
Culture: American
Computer / IT experience: High to Expert

1. Q. Is there anything you need further explanation or clarification on?
   A. There will only be 16 different ingredients to choose from?


2. Q. Have you ever seen such a system in the "wild"?
   A. No I haven't. This seems like quite a novel idea.


3. Q. Do you think that this type of system has valid real world application?
   A. Yes, I do. I think that having a machine like this would cut down on the costs to bars for their staffing. Also, it would allow for faster service time.


4. Q. Would you like to use such a device?
   A. Yes, I would. Especially if I didn't have to tip it.


5. Q. What potential problems do you as a user foresee for such a system?
   A. The issue of a robot not really checking an ID for the correct age. Also, a robot can't as easily recommend a drink to you. Finally, if you are in a high-volume setting, the machine would perhaps run out of ingredients quickly


6. Q. What are some important features or things you would like to see implemented in this project?
   A. Perhaps a larger selecting of ingredients, but I understand the implementation issue. Maybe have a suggestion system based on drinks that the user has enjoyed in the past


7. Q. Do you have any questions you would like answered at this time?
   A. Is this a paid survey?

Interview 3

Interviewer: Derek Fernholz
Interviewee: David C Freitag
Age: 22
Culture: American
Computer / IT experience: Above Average to High

1. Q. Is there anything you need further explanation or clarification on?
   A. No, everything seems to be quite clear in the above description.


2. Q. Have you ever seen such a system in the "wild"?
   A. No, this seems like a unique and innovative system.


3. Q. Do you think that this type of system has valid real world application?
   A. Yes, I think it would be very useful in bar applications. It would make it easier for bartenders to keep up with customer orders, increasing efficiency to maximize profit margins and customer satisfaction.


4. Q. Would you like to use such a device?
   A. Of course I would!


5. Q. What potential problems do you as a user foresee for such a system?
   A. Perhaps, if certain drinks aren't in the system it may cause difficulty or frustration for the user.


6. Q. What are some important features or things you would like to see implemented in this project?
   A. A house favorites section, maybe based on a system that would count the number of times a certain drink is poured. Thus giving you a list of perhaps the 10 most ordered beverages. Making the drinks scalable, so that you can make different sizes of the same drink and still have them be proportional.


7. Q. Do you have any questions you would like answered at this time?
   A. No. You've been most informative.

Interview Summary

After conducting the above interviews I can say that I firmly believe that my project is unique and a very worthwhile endeavor. Overall the feedback has been very positive and many of the thing mentioned or touched upon just go on to confirm my plans. A new feature was even born from this interview process and I think it will be a fun and interesting option. This new feature is the "House Favorite" feature that David Freitag proposed. This is simply a top listing of drinks dispensed in a predetermined time period. I believe this will be a valuable addition and it will take virtually no effort to implement.

Task Analysis

In my case the tasks will be more along the lines of features that the system will have. I have grouped these tasks by their ranging difficulty.

Hard

• Making PHP code from the interface control actual physical devices. In my case this would be along the lines of a lower level language, like C or Embedded C being executed by PHP.


• Searching - in a broad sense this is not a difficult task, but as we have discussed this should be incredibly user friendly and extremely intuitive. There should be minimal interaction between the user and the system for most of the navigations. This will require extensive groupings and categorizations of the various different concoctions and several different sorting algorithms. Also, this will all have to happen dynamically on call and with little to no load time (almost real time interactions).


• The main crux of the system is the fact that this will actually be a fully working system. So a simple mockup of what it might look like and presentations on what would happen are not feasible. There will be concrete test cases and this interface has to deliver up to or beyond the expectations of the designer and all others involved.

Medium

• The Customizable Drink Option will be a bit of a challenge along the lines of the fact that the user should be able to edit an instance of a drink but not change the default settings of said drink. Also, due to calibration techniques, there will need to be extensive testing to ensure that the machine will be able to deliver the correct amount of fluid consistently.


• The Taste Test Option poses the same problem as the latter description of the Customized Drink Option. It will rely even more heavily on the calibration of the device and even more, the software will have to know what to do to create a one-eight scale version of say an eight ounce drink (or a one ounce sample). This will all have to be done on the fly and in as close to real time as possible.


• Add a New Drink will be probably the easiest of the medium level tasks. At a broad sense I would say it could even be considered an easy task, but when we dig deeper we find that there is a lot that could go wrong so there are multiple things that will have to thought through extensively before implementation. There will have to be some sort of checking to ensure the same exact combinations of liquids isn?t already in the database under say a different name.

Easy

• The Regular Drink Menu will be fairly trivial I believe. Since all drinks will be categorized in the database this will merely be the result of one query to populate this page. The only issue from an interface perspective will be displaying all of the information and more importantly the correct information for the user to peruse. Also the interaction with this page or any of the pages really should be as intuitive and easily discernable as possible.


• The Shot Menu Page will follow suit with the above option of Regular Drink menu except obviously for shots.


• A new feature we have come up with in the last week is a ?House Favorite? Drink Page. This will be done on a counter type system, we will track what the frequency of any particular drink getting ordered at a particular location and then present the say, top ten or so on a ?favorites? type page. This task will be quite simple I believe but will also add to the uniqueness, complexity, and the overall user enjoyment derived from this system.

Low-fidelity Prototypes

Figure 1: An idea for the front "Splash" screen. Figure 2: A mockup of the display drink page. Figure 3: A proposed search interface.

Use Scenarios

This Scenario will take the user from the "Start Page" to Drink Selection

• The start screen contains "icons" for all the various menus or tasks that the system will perform.
• From this screen the user will select the "Choose Regular Drink Menu Page"e option.
• Upon selection the user is taken to said page and presented with more options.
• On this page the user then selects to limit results by category by selecting the "Choose Category" option.
• Next the user will be presented with all of the predetermined Categories that the system holds.
• Upon selecting the Tropical Category the user is taken back to the Menu but now the display limits the drinks shown to the selected category.
• In the search area at the top of the page, the user enters Teq (because they know the drink contains Tequila) and then selects the "Search Based on Ingredient" radio button.
• Now the list of drinks is narrowed to all tropical drinks that contain Tequila as an ingredient.
• Within this list should be "Tequila Sunrise", the user then "Touches" said drink to select it.

This Scenario will take the user from Drink Selection to Pouring

• A user selects which drink they would like and "Touches" it
• Then a menu will be presented to them with all the glass size options available for that drink.
• Next, the user will select which size glass they would like, in this case, a 12 ounce glass is chosen.
• Then the menu changes and the user is given a menu with their selected drink and the default levels of each liquid scaled to their glass size.
• Now the user will place a glass under the "Nozzle" and "Touch" the pour option.

This Scenario will take the user from Customizing a Drink to Pouring

• A user has already selected which drink they would like and what glass size it is to be poured in.
• Then the user will select the "Customize It" option.
• The right side of the screen has now changed to show "Sliders" for each ingredient that are set to the default amount for said drink.
• The user then decides to increase an ingredient, in this case Tequila, in their drink by sliding the level slider to the right.
• The other ingredients, in this case Orange Juice and Grenadine, will then scale down dynamically with respect to eachother and the glass size and the image on the right will be changed accordingly to give a visual illustration of this change.
• Then the user will put the glass under the "Nozzle" and "Touch" the pour option.