Evaluation feedback

Last week I had an evaluation. All functionalities were tested by asking the test persons to run over some tasks, meanwhile using the think-aloud protocol. Afterwards they were asked to fill in a questionnaire based on the USE questionnaire. It can be found here.
Below are the results:

General
9 persons of which 7 male and 2 female.
Ages from 18 to 24.
5 master degrees, 4 bachelor degrees.
7 persons with a computer sciences background.
6 persons have a smartphone, 2 have used one before, 1 has no experience with it.

Questionnaire
The USE questionnaire can be divided into 4 factors:

• Usefulness
• Questions 1 to 6
• Overall score (on a scale from 1 to 7)  5,0555
• Ease of Use
• Questions 7 to 11
• Overall score (on a scale from 1 to 7)  5,1777
• Ease of Learning
• Question 12
• Overall score (on a scale from 1 to 7)  6,1111
• Satisfaction
• Questions 13 to 15
• Overall score (on a scale from 1 to 7)  5,2963

Positive remarks

There were some positive remarks that came up with many of the test persons:

• 7 persons mentioned having your course schedule with you is a very useful utility. Moreover, the ability to show on a map where a class is held and calculating a route to that class. 2 of these persons found it useful especially for new students, but less for the older ones.
• 5 persons found it useful to see which of your friends are around, for group assignments, course lectures, etc...
• 4 persons found the functionalities handy, having most of what a student needs in one single application, in stead of being spread out on several websites. They also found the menu easy to use, since it is limited to the main big functionalities.

Negative remarks

There were also some negative remarks though. These remarks can be associated with one or some factors:

• Design
• Sometimes the way to accomplish something is not the shortest.
• The check-in functionality was somewhat hidden. 7 persons had difficulties trying to check in.
• Users were expecting some kind of feedback when there were no courses or events in the next week.
• Functionalities
• User location is not precies enough.
• Long loading times (NOTE: the evaluations were done mostly with EDGE)
• (Reverse) geocoding an adress from the events doesn't work all the time.
• Automatically select current location as starting location.
• Users expected some search function for events.
• General smartphone remarks
• Users are not eager to tap on map markers.
• Users tend to instantly tap OK when dialogs are shown, without reading the message/instructions first.

Having analyzed the evaluation data, I already have some possible solutions in mind. But whether I will do another iteration will depend on how well and how quickly I record the study material for the exams.
Also note, that since the market release, the application was downloaded and installed 16 times, and 3 of these users have filled in a questionnaire. I will do a deeper analysis of this 'online' data later...

Market release.

Yesterday I uploaded my application to the Android Market.
It can be downloaded here.

On the page is also described what functionalities are currently supported,
and what I am still working on at the moment.

This week I am planning a last evaluation of my application.
If you are interested and want to participate you can choose a time slot on my doodle.

Status update

Last week I spent on fixing some bugs in my implementation along with making the thesis poster.
This poster can be found here.

Currently I am working on the implementation of KUL timetables in my application.
Because this was not in my paper prototypes yet, I first created a mockup/storyboard which was then reviewed by my supervisors.

So at the moment my application can do the following:

• show campus map
• search campus buildings, get building info, and locate them on the map
• get routing information to the buildings and display the route on a map
• check in at the buildings (through facebook places)
• show nearby and recent buddy check-ins and display them on a map
• search buddies and show profile and recent check-ins

Easter presentation + new planning

During the easter holidays I and the other students of our group had our second thesis presentation. My presentation can be found here.

Although less than at the presentations in december we received some helpful feedback. Most important was the further planning since we are at a crucial time in our thesis work. I myself have decided, in accordance with my supervisors, to delay my planning. This is because I have a couple of other courses that involve projects, and I will probably not get my thesis text finished in time. After all, if I have to rush it,  not only the quality of both my implementation and thesis text will suffer from it, but also all the work for other courses. So here is my new planning:

• 26/04 - 02/05: thesis poster
• 03/05 - 09/05: adding KUL timetables to the application
• 10/05 - 16/05: adding KUL events to the application
• 17/05 - 23/05: pc room reservations, favorites
• 24:05 - 30/05: evaluation
• 31/05 - 30/06: exams
• 01/07 - 15/07: thesis text ch. 2 literature study
• 16/07 - 22/07: thesis text ch. 3 analysis
• 23/07 - 05/08: thesis text ch. 4 implementation
• 06/08 - 12/08: thesis text ch. 5 evaluation
• 13/08 - 19/08: thesis text ch. 1 intro + ch. 6 conclusion

The reason I take 2 weeks for chapter 2 is to anticipate on problems for getting started, don't knowing how to get started, what to write exactly. The reason for 2 weeks for chapter 4 is because that one will be a big chapter. Furthermore, the intro and conclusion are chapters that should grow throughout writing the other chapters. The deadline is 24/08 so I have foreseen a couple of days at the end for adjusting the final text. Off course, I will try to finish the text faster than according to this planning.

Progress update

I haven't written anything on my blog for the past 2 months.
This is mainly because I am focusing on the implementation of the application, and have lost sight of my blog.

What have I done in the past 2 months?
Well first of all, I have written a scientific paper about the application. This paper can be found here.
Second, I started implementing the application for Android. The implementation will be done in 3 iterations which are each followed by an evaluation.

I have completed the first iteration along with it's evaluation. This first iteration focused mainly on the design of the graphical user interface. Almost the entire application was designed, all with hardcoded data. This design was then evaluated by 4 persons aged 18 to 25. The evaluation was done on an actual device, from the computer science department. The test persons had some time to explore the application shortly after which I asked them about their general impression. These are the results of the evaluation:

• For some screens, it was unclear that some text fields, input boxes, etc. were clickables leading them to an other screen.
• At the screens where I used tab widgets, it was unclear which tab was the selected one.
• Some screens had too many buttons, forcing users to do unnecessary extra work.



Choosing bad colors

• There was some inconsistency regarding the buttons. Since Android smartphones have a back button on the device itself, users can simply go back to the previous screen by clicking this device button. On some screens I added an extra cancel/back button, but not for all screens. The test persons advised me to either add them everywhere, or either nowhere, forcing users to use the device button.

Too many buttons

The first two remarks were mainly caused by choosing bad colors for the screens.

The last two remarks were caused by bad, inconsistent design of the graphical user interface.

Furthermore, since I only got the device shortly before the evaluations and didn't have time to test it myself, it showed that the buttons, text, etc. were too small when shown on the screen of the device.

At the moment I am working on the second iteration: fixing first iteration issues, adding a Google map, building locations, route information, a check-in functionality along with buddy locations (by integrating with Facebook, Foursquare and others). I will post more on this when I finish the second iteration and it's evaluation.

Paper: Mobile Learning: Context and Prospect

This paper is a white paper of the ELI online focus session about mobile learning.
It contains a lot of links to other useful papers, and can be a good good guide for my further literature study.

The paper can be found here.

Evaluation of second paper prototype

Below are the results of the evaluation of my second paper prototype.

# persons: 5

Age: 19 to 24

# years studying in Leuven: 1 - 5 years (1 person already graduated)

All persons followed different studies

2 persons had experience with multitouch of which 1 also with Android

Problems/suggestions (by test persons) while running scenario's:

- Problem: not clear how to switch between different map views (events, buddies, none)
- Suggestion: being able to view buddies, events on other locations (campus maps)

=> Possible adjustment: adjust the filter options, and add icons on map for quick switching between map views

General results:

Purpose of application clear?                                           8,8 / 10

Interaction with application clear?                                     8,4 / 10

Would you use application yourself?                                 8,2 / 10

Does the application have an added value for students?      8 / 10

Second paper prototype

Here is the second paper prototype of my application.
On the following figures you see from top to bottom, left to right:

• A scenario for calculating a route for user defined start and destination
• A scenario for searching and checking in at a building
• A scenario for displaying buddies on a map according to the user defined filter options
• A scenario for searching and displaying an event

Progress update

As I posted in my previous blog, I was intending to finish the evaluation of the second paper prototype. But due to a cold I wasn't able to meet up with some of the test persons. So I have decided to delay the evaluation until after I've had my last exam.
At the moment I am studying for exams so I am not doing that much for my thesis right now.

Here's something though, I still had to post on this blog:

About a month ago, december 15th to be exactly, the faculty of engineering had a demo day at their department.
I went there to look at some projects made by the students whose major was computer science.
This was extremely helpful and interesting since their project consisted of making Mobile Toledo for smartphones.
The ideas and features behind these apps were much in line with my own. For example, each of the groups (there were 3) implemented the ability to search for some building, display it on a map, and some could ask for directions.
One also had the ability to ask questions to the teacher or reverse, the teacher asking questions to the students. (this is in line my classroom app idea, but I'm not going to do this, since I choose for campus app)
The apps on overall looked quite good, although the interaction felt quite clumsy. Off course they were still just prototypes and they probably didn't do a proper evaluation. But as I said, overall looked quite good.

I've read the blogs of the groups, and had a look at the wiki for the course.  I don't think this will help me any further for my thesis, but it was very interesting going to the demo day, I enjoyed myself there!

Evaluation of first paper protoype

Below are the results of the evaluation of my first paper prototype.

At the moment I am already evaluating a second prototype (which is adjusted according to these results).

The results of the second paper prototype will be posted at the end of this week/beginning of next week.

# persons: 5

Age: 20 to 22

# years studying in Leuven: 1 – 4 jaar

All persons followed different studies

3 persons had experience with multitouch of which 1 also with Android

Problems/suggestions (by test persons) while running scenario's

- Problem: not clear how to get a map starting from first screen

- Suggestion: a simple and quick way for switching between different maps (buildings, buddies, events)

- Suggestion: possibility for having favorite buildings, buddies and events

=> Possible adjustment: adding 'Maps'- and 'Favorites'-tabs to the menu

- Problem: target-icon not clear

- Suggestion: searching route with starting and end location

=> Possible adjustment: chaging target-icon to 'Locate me' and adding functionality for searching a route from some starting to end location

- Suggestion: coloring campuses with available PC rooms green, other campuses orange/red

- Suggestion: possibility to search for rooms other than PC rooms

=> Possible adjustment: adding coloring

General suggestion/remark: Use more colors in the paper prototype

General results:

Purpose of application clear?                                         8,5 / 10

Interaction with application clear?                                     8 / 10

Would you use application yourself?                              8,4 / 10

Does the application have an added value for students?     8 / 10

First paper prototype

Finally I was able to scan the first paper prototype of my application.
On the following figures you see from top to bottom, left to right:

• A scenario for searching for a certain building, displaying the route there from the user's current position, and checking in at the building
• A scenario for displaying some buddies on a map according to some filter options chosen by the user
• A scenario for searching for an available PC room
• A scenario for displaying some events on a map

New planning

In the presentation I suggested a new planning which was approved by the teachers and advisors, especially because I work in an iterative way, and not a waterfall kind of way.
I will try to keep me to this schedule as best as possible:

• 13/12 - 19/12: evaluation paper prototype iteration 1
• 20/12 - 26/12: paper protoype iteration 2 (with feedback of iteration 1)
• 02/01 - 03/02: exams
• 04/02 - 13/02: getting to know the development software I will work with
• 14/02 - 06/03: implementation iteration 1 (with hardcoded data)
• 07/03 - 13/03: evaluation iteration 1
• 14/03 - 20/03: adjusting iteration 1 (with evaluation feedback)
• 21/03 - 10/04: implementation iteration 2
• 11/04 - 24/04: evaluation iteration 2
• 25/04 - 08/05: final implementation (with evaluation feedback)
• 09/05 - ........: writing report + presentation

Christmas presentation

Last tuesday I had a presentation about the work I have done so far.
The slides of the presentation can be found here.

The feedback and comments and remarks I received from both teachers as other students were really helpfull.
I will explain them a bit more:

• One remark was that I did not explain why I had chosen for this kind of application.
  I will adress this issue by searching for papers that explain the problems that students face on campus, along with results that support the idea of creating this sort of applications, one that (partially) solves these problems.
• Also I need some more and better reasons why I chose for Android OS.
  That is why I will make a deeper comparison between the different mobile OS's in the next couple of weeks, and will make a final decision based on this comparison.
• They pointed out that I have to make sure that students keep using the check-in functionality, otherwise my application will lose a great part of its intended purpose.
  I will look for some kind of reward-system and look at how Foursquare and Waze solve this (both applications were suggested to me after the presentation, although I already have been using, and thereby know Foursquare)
• In my presentation I talked a bit about privacy concerns when integrating with Facebook.
  One solution for this is a so called social network path which limits the number of friends to 50.
• Because I want to display extra information about users (on buddy info page), I will probably have to keep profiles for every user. This will give me a very large database to deal with.
  So either I drop out some unnecessary or either I implement some profile page for every user.
• As a last remark one student said I could be an advantage to have the functionality where you cannot just check in at buildings and classrooms, but also send a request to some buddy to ask him where he is.
  I will consider this in the next couple of weeks, and see whether the persons wich test my paper prototype have the same remark.

Core functionality + planning

After the meeting with prof. Duval I was asked to think about what core functionality I would like to implement in my app, and to have a planning until the end of the year.

Core functionality:

• Display maps of campuses of KUL + google map of Leuven
• Search for buildings and classrooms on these campus maps
• Display or describe the path to buildings/classrooms from user's current position
• Managing a buddy list
• Searching for nearby buddies
• Check in at buildings/classrooms
• Display events from KUL or student associations
• Search for nearby events
• Display news from KUL or student associations
• Search for an available PC (by means of non-booked PC classroom)

Planning:

• December: paper protype (+ evaluation)
• February: implementation of first iteration (maps + buddy list)
• April: adding events, news, available PC functionality
• During entire year: literature study
• May: final report + presentation

Choosing a final direction

About a month ago I had a meeting with prof. Duval to show him my ideas.
This meeting was very interesting since I had some feedback on the ideas.
At the end of the meeting he asked me to make a choice:
Either I should concentrate on creating an app for use in the classroom (answering questions, polls, following presentations and marking or taking notes on slides, ...)
Or I should concentrate on an app for use around the campus (with maps, events, buddy list, ...)

After some thought I decided to go for the second one, because I think this idea lends itself more for use on a smartphone. It will probably make more use of the several features that a smartphone offers (e.g. GPS).
Answering questions, polls are things that are also possible an a simple mobile phone with text messaging.
And taking notes or markings on slides while following a presentation is maybe not the best idea to do on a smartphone, because of the small screen and small keyboard.
This could be done on an iPad, but I really wanted to stick with a smartphone, so I chose to go for the campus app idea.

In the meeting prof. Duval also pointed out some apps that where already available on this part.
iStanford
This is a mobile app for people at the Stanford University.
It can show a map, events, news, videos...
http://mobile.stanford.edu/
iPrinceton
This is a mobile app for people at the Princeton University.
It has much of the same features as iStanford.
http://mobile.princeton.edu/
iTU Delft
This is a mobile app for people at the Delft University of Technology.
It has less features than iStanford and iPrinceton but it has at least the ones interesting for me to look at.
http://www.e-learn.nl/2010/02/01/itu-delft-iphone-app-van-de-tu-delft

iStanford: list of some events at the university

 

iPrinceton: searching a building on the map

Since I would like to implement a buddy list, prof. Duval also advised me to install foursquare on my iPod Touch.
This app uses a check-in system where people can check in at places and manage a buddy list.
I used the app for some time, and I have to say I like the concept.
I will probably handle the buddy list in my app in a similar way (with the ability to check in at campus buildings...)
http://foursquare.com/

foursquare: a place nearby where the user can check in

Using the smartphone on campus: use cases

Use case: Finding the way to a classroom

A student (the user) has trouble finding some classroom he needs to go to. He wants to use his smartphone to show him the way.

• Pre-condition: the user is on the campus and has the app running
• Post-condition: the way to the classroom is shown on the smartphone and he can now follow the indicated path
• Main scenario:

1. The user indicates he wants directions to a classroom by touching the 'find classroom'-icon on his smartphone screen
2. The app asks him to enter the building abbreviation and the classroom number
3. The user inputs the info using the keyboard on the screen
4. The app calculates the shortest path to the classroom and shows it on the map

Use case: Searching for nearby friends/colleagues

A student (the user) wants to see whether there are friends or colleagues in his nearby surroundings.

• Pre-condition: the user is on campus and has the app running, he has some friends/colleagues added to his buddy list (done by some use case to add friends/colleagues)
• Post-condition: nearby friends/colleagues are shown on the smartphone
• Main scenario:

1. The user indicates he wants to look for nearby friends/colleagues by touching the 'buddies'-icon on his smartphone screen
2. The app asks him to enter some extra filter options (name, only friends/colleagues, ...)
3. The user selects and inputs the wanted filter options
4. The app searches for nearby friends/colleagues using this filter options and shows the results on the map
5. The user selects one of his friends/colleagues by touching that person's position on the map on the screen
6. The app gives some extra info about the selected person (name, major, current activity, ...)

Use case: Searching for nearby events

A student (the user) wants to check whether there are current or future events nearby that might be interesting.

• Pre-condition: the user is on campus and has the app running
• Post-condition: nearby current or future events are shown on the smartphone
• Main scenario:

1. The user indicates he wants to look for nearby current or future events by touching the 'events'-icon on his smartphone screen
2. The app asks him to enter some extra filter options (period, location, sort of event, subject ...)
3. The user selects and inputs the wanted filter options
4. The app searches for nearby events using this filter options and shows the results on the screen
5. The user selects one of the events by touching that event on the screen
6. The app gives some extra info about the selected event
7. The user indicates he wants to find the way to the event by touching the 'location'-icon
8. The app calculates the shortest path to the event and shows it on the map

Use case: Finding an available PC

A student (the user) needs a PC to complete some task for school, or maybe just because he has some spare time. He wants to find an available one quickly without having to check every PC classroom.

• Pre-condition: the user is on campus and has the app running
• Post-condition: the available PC's are shown to the user
• Main scenario:

1. The user indicates he wants to find an available PC by touching the 'free PC'-icon on his smartphone screen
2. The app asks him to enter some extra filter options (location, operating system, ...)
3. The user selects and inputs the wanted filter options
4. The app checks all PC's that adhere to the filter options to see whether it's available (no one is logged on or there has been no activity since 15 minutes or so)
5. The app shows a list with available PC's, sorted by location starting with the ones most nearby
6. The user selects a PC by touching it's name on the screen
7. The app shows some info (name, location, operating system(s), reservations for the classroom, ...)
8. The user indicates he wants to find the way to the PC classroom by touching the 'location'-icon
9. The app calculates the shortest path to the PC classroom and shows it on the map

Using the smartphone in class: use cases

Use case: Asking a question

A student (the user) wants to ask a question to the teacher about the course in the classroom. But he is afraid other students will laugh at him, so he asks it using his smartphone.

• Pre-condition: the user is in the classroom and has the app running
• Post-condition: the question is asked and added to the question list which is viewable by the teacher
• Main scenario:

1. The user indicates he wants to ask a question by touching the 'ask question'-icon on his smartphone screen
2. The screen shows a textline and a keyboard
3. He inputs the question about the course using the keyboard on the screen
4. The question is added to the question list

Use case: Answering a question from a colleague

A student (the user) wants to answer some questions of his colleagues.

• Pre-condition: the user is in the classroom and has the app running
  
• Post-condition: the question list is updated (the answered questions are updated with an answer)
• Main scenario

1. The user indicates he wants to see the question list by touching the 'question list'-icon on his smartphone screen
2. The screen shows the question list, all questions asked by the user and his colleagues
3. The user touches the 'answer'-icon next to a question he thinks he knows the answer to
4. The question list is updated
   (Steps 3 and 4 are repeated until there are no questions left the user can answer to)

Use case: Voting on a question

A student (the user) wants to see the question list and vote on questions that he finds are interesting.

• Pre-condition: the user is in the classroom and has the app running
• Post-condition: the question list is updated (all questions the user voted for have received an extra vote)
• Main scenario

1. The user indicates he wants to see the question list by touching the 'question list'-icon on his smartphone screen
2. The screen shows the question list, all questions asked by the user and his colleagues
3. The user touches the 'vote'-icon next to a question he finds interesting
4. The question list is updated
   (Steps 3 and 4 are reapeated until there are no questions left that the user finds interesting)

Use case: Asking a poll to the students

The teacher (the user) wants to start a poll, u multiple-choice question to check whether the students are paying attention and understand the course.

• Pre-condition: the students in the classroom have the app running
• Post-condition: the poll has been asked, and is sent to the smartphones of the students
• Main scenario:

1. The user indicates he wants to start a new poll by touching the 'poll'-icon on his smartphone screen
2. A poll list is seen on the screen of the user
   (It is more likely that the teacher has created these polls at home, rather than having to make them during his lecture. For this there should be an extra use case for creating a poll, or sending a poll from laptop to smartphone for example)
3. The user chooses one of the polls
4. The poll is sent to the students' smartphones

Use case: Answering a poll

A student (the user) wants to answer a poll that is given by the teacher.

• Pre-condition: the user is in the classroom and has the app running
• Post-condition: the user has answered the poll and the poll is updated
• Main scenario:

1. A message pops up indicating that there is a new poll from the teacher
2. The user touches the screen at the message and the poll shows up
3. The user answers the poll by touching the right answer on his screen
4. The choice is sent back to the teachers smartphone and there the poll is updated

Taking notes on slides: use cases

Use case: Taking notes on a slide

A student (the user) wants to take notes relating to the current slide that is shown.

• Pre-condition: the user is in the classroom and is following the presentation of the teacher on his smartphone
• Post-condition: the new notes are stored at the current slide
• Main scenario:

1. The user navigates to the current slide on his smartphone (if this slide wasn't on top yet)
2. He indicates he wants to add extra notes to the slide by touching the screen at the 'add notes'-icon
3. A 'window' pops up in the style of a notebook, along with a keyboard on the screen
4. The user types in the notes using the keyboard on the screen
5. He confirms the notes by touching the 'store notes'-icon
6. The application stores the new notes with the current slide and returns to the slideshow

 Use case: Sharing notes with a colleague

A student (the user) wants to share and exchange notes about some presentation with a colleague.

• Pre-condition: the user and his colleague both have the presentation on their smartphone
• Post-condition: the notes are exchanged so both students have each other's notes along with their own
• Main scenario:

1. The user indicates he wants to share his notes with a colleague by using the 'share notes'-feature of the application
2. His smartphone gives a list of possible presentations to share notes of
3. The user selects a presentation
4. He types in the mobile phone number of his colleague or selects the colleague's smartphone by Bluetooth or Wi-Fi
5. The colleague sees a message whether he wants to share notes on his smartphone and accepts
6. The notes relating to the presentation are exchanged between the smartphones and integrated in the according presentation

Simplenote

Related to my previous post, I went to look at some good note taking apps.
I especially searched for one which had good sync and sharing features.
Simplenote seems one of the best apps that provides both.

First, file (or 'note') synchronization is done. The app keeps an online version of your notes.
These notes are accessible with any browser so you can make changes on the notes at home on your desktop, laptop, etc... All versions will stay synchronized.

Notice the 'Web Syncing' option

Second, the app supports sharing. Notes can be easily be shared by e-mail, by publishing them as web pages, and you can even invite other Simplenote users for collaboration.
This would be very handy for students that have missed some classes for various reasons, or just to collaborate with your colleagues.

Example of a shared note (the wi-fi icon)

Off course, this app would have to be adjusted so that also the slides (see previuos post) are viewable.

Some reviews of the Simplenote app:
http://www.macworld.com/appguide/app.html?id=89982
http://www.macworld.com/article/153627/2010/08/simplenote_sharing.html

Taking notes on slides on a smartphone

I was browsing some simple apps that could help students in class,
when I came up with the following idea.

Suppose the teacher is using slides (e.g. PowerPoint) in his lecture.
These slides are mostly availabl on the teacher's homepage, blackboard or somewhere on the net.
Now as a student you feel these slides are not 'complete', maybe because there is not enough information on them, or the teacher is adding some extra info to his lecture, that is nor on the slide nor in the course book.
So you want to take some extra notes on one or more slides.

I have found some note taking apps (Notes is standard on iPhone, Simplenote, Evernote, WriteRoom, etc...)
Which one of  these is the best note taking apps I cannot conclude, since I am not able to test them.
But I think it depends on your personal preferences and needs anyway.
However, none of these implement a feature where you can stick notes to some or more slides.
(I read something about sticking a note to some image, but I feel it's not the same, since it's just a single image, probably not containing any text)

So I searched for an app that presents slides in a manner that is suited for smartphones.
Keynote (+ Keynote Remote) is a good app for iPhone for creating and giving presentations:
http://www.apple.com/iwork/keynote/#flexible
This app is intended for the presenter, however, the creator of the presentation. Not the ones following the presentation.
The app allows the creator to add notes for himself to the slides, and it has lots of other features.
With some simple adjustments I believe that this app could be used by students as well to take notes related to one or more slides of the presentation.

Example of a slide with extra notes

Here is a short video that presents just the features needed for the above idea:
http://www.youtube.com/watch?v=nBzXUWgl9T0