I need talking points on whether or not this is possible.

I am a public health researcher, so bear with me.

I work on a research project with Wii Fit. My team wants to get the participants’ data from Wii Fit. We only want to get the participant-specific data on Wii Fit—nothing else.

I suggested that we should just have the participants give us a copy for the game save data on an SD card and then examine it in house. Seems easiest to me.

Not being a programmer, I have no clue what is and isn’t possible. I need to tell them specifically why it isn’t possible.
Or, if it is possible, I need to tell them what is involved and why it would take too much time to do it. Or, that it would have serious privacy concerns for our review board.

Anyway…..so, let’s say we install the homebrew channel on all 400 Wiis.

1. Could an app be written for us to get the participant specific data of Wii Fit (time used, etc)?

2. If that is possible, wouldn’t the information just display on their Wii? How could even we access it remotely? Let’s assume most participants do not have internet connections.

3. How could we avoid looking at other “private” information? Could we even avoid looking at the data of other family members who use Wii Fit?

Thanks in advance--I really appreciate it. I just need some discussion points on this topic for my team.

Marie



Edited 1 time(s). Last edit at 02/11/2010 02:05AM by marieresearch.

If you don't mind me asking I'd like to know a little more about your position in this research project. You say you're a public health researcher and that you are working with a team. However, you sound to be the only person vouching for a "homebrew" solution but unfortunately you also seem more than a little intimidated by the technical details of this project (which is perfectly okay by the way).

What I'm wondering is what is the goal of this project (what do you expect to learn through your research), your exact position in the project, and do you have someone more able to work with computers (a programmer for example) that is part of your team?

Anyways, that is just a little curiosity you have sparked, I apologize if I have offended you.

Onto your questions:
1)The Wii Message Board records the time a game was played so you can already use data from that to see the time the game was played. You do not need any homebrew to view it. However, you can not see how much of that time the player was actually physically exerting themselves (in theory, you could leave the game on the pause screen for 4 hours and you would record 4 hours of playtime). I'm not sure if Wii Fit itself even tracks more specific data.

2)How about having the participants record their data and send it to you "snail mail" style? That's the easiest way I can think of.

3)Yes, if you can get data from Wii Fit and understand what it means, it should be trivial to pull the information apart and avoid the information you dont need.


On a side note, have you looked into Wii Fit Plus which records more and different types of data than it's big brother Wii Fit?



Edited 1 time(s). Last edit at 02/08/2010 05:19PM by Arikado.

I am research assistant. I am very intimidated by the technical details, as this is not my area of expertise. We are using the Wii Fit Plus with a disparate population to test its affect on health indicators. We already have clinical tests in place at the 1, 3, and 6 month (blood, weight, etc).

My team decided that they not only want the clinical data, but they also want the data from the Wii Fit Plus itself. This was not included in our original research proposal to our federal review board. In short, there are serious concerns with public health privacy and HIPPA when it comes to "getting" unconsented information from a participant (even something as small as a Wii Fit Plus usage data).

As far as your snail mail suggestion, it is complicated because having the participants "record" their data on paper and mail it via snail mail or something changes the parameters of the study. For example, if you have to manually record how often you work, you tend to work harder. So this in essence skews our study

So, basically, my teams wants some solution where the Wii Fit Plus data is just uploaded from the participant's Wii to a server here on site. I don't think this is possible, but I don't know enough about it to articulate quite why. They seem to think that "hacking" the Wii and writing an app will be the solution. There is a serious problem of getting only the exact data that we need though. We cannot be "looking" at data from people other than our participants. For example, if my daugher sally uses my Wii, we cannot "look" at Sally's data. I suspect that a homebrew solution will not limit our privacy invasion--it will actually create more of a problem.

Anyway, I am not sure that a homebrew solution exists to this problem, but I can't articulate why to my team because I am not a programmer (if something did exist, I would hire a programmer).

Can you help me explain to my team why this homebrew solution doesn't exist?



Edited 1 time(s). Last edit at 02/11/2010 02:06AM by marieresearch.

well yes you can look at the data for wiifit with homebrew. you can also have a homebrew app that sends that data to anywhere via the internet connection. but how the hell is the homebrew app gonna know which data in the wiifit save you want? the app will have to parse the entire save file and find whatever info you want out of it. so it will see the save data for everybody. but it can choose not to display or transmit that data anywhere. you will have to have a list to choose from, or you will need every person to use the exact same name to do wiifit so you know which profile to use. and the stats you want to get will be hard coded.

so if looking at peoples' names in the game is not poking into their privacy too much, then yes, what you want is possible. you can even run it right from bannerbomb. so you just put bannerbomb on a sd card with your app and leave it in the wii. then have the person play the. and when it is time to get the stats, you have them go to the menu that triggers bannerbomb, that starts your homebrew, and they select the game profile that they want to send you info about and BAM.



Edited 1 time(s). Last edit at 02/08/2010 05:59PM by giantpune.

Thanks for explaining. That is really helpful. I really do appreciate it.

My concern is that less than 10% of our target population has an internet connection. Hell, we don't even know that any of them are going to have TVs. (Note: I did NOT design this study).

Is there a solution that doesn't require an internet connection that would be easier than just having them copy the actual game save file onto an SD card?

I don't even know what I mean honestly, so don't make fun of me. I am really out of my comfort zone here.

You mentioned "Let’s assume most participants do not have internet connections."

So here's the message that you send to your research team:

It is not possible to get someone's information, period, if they do not have an Internet connection, without them physically giving it to you.

It's impossible "hack" a Wii to magically have an Internet connection. Much like it's not possible to hack my radio into an HDTV. You mentioned privacy, which is very important, I will agree. If your research team wants to somehow anonymously gather their data, then they will require, at the very, very least, an Internet connection.

The only other way to get their data would be for them to literally send it to you, either via a message on the Wii message board (Internet again!); sending you an SD card that they have backed up their saves on (still have to parse it, as mentioned above); or by Arikado's suggestion of snail mail.

It sounds to me that you have a rather demanding research team. One that thinks that computers are "magic". I've met a few people like this.

Don't get me wrong, I am intrigued by the idea and I'm all for improving people's health, I would love to see this project come to fruition.

My suggestion was to have them physically give us the gamesave file on an SD card, go back to our review board and just make them sign another consent that says "I understand that the researchers will only look at my Mii file and will delete other Miis etc etc". I think that this is the easiest way.

I honestly feel that the quality of data that we get is not going to be worth messing around with 500 Wiis, hiring a programmer, and praying that the participants have an internet connection and can upkeep their Wii.

Please answer this:
I assume that the warranty is voided in these circumstances, too. Correct? Like, going in and messing with the Wii voids the warranty?



Edited 1 time(s). Last edit at 02/11/2010 02:07AM by marieresearch.

Quote

I honestly feel that the quality of data that we get is not going to be worth messing around with 500 Wiis, hiring a programmer, and praying that the participants have an internet connection and can upkeep their Wii.

I think that says it best as far as talking points. What your team is looking for IS possible, but like you said, without internet connections and the resources to hack and code for 500 Wiis, it SO won't be worth it. The internet thing is the biggest crutch because now you would rely on an sd card to collect the data (doable) and then physical transport to the server (doable). Both can be done but probably not worth all the work.

If you are the Wii expert and the one who would be expected to code/make/acquire homebrew to do the above, you might just want to tell them not plausible. need more info, time, money, and internet connections :D

and yes, nintendo would not like the changes that would have to be made to the Wii to do this, so unless you own the 500 Wiis in question, that's another form that people would have to sign saying they understand they are "illegally" altering the state of their Wii (i put illegal in quotes to not have to start a debate on the legality of all this... Nintendo doesn't like it, nuff said)

I applaud your questions and conversation on this forum (as you said it is out of your comfort zone).

I think you'll find the main members of this forum to be a little more mature than your average internet forum so please feel free to ask any questions that you may have and we'll do all we can to help.

You know, I've found that money is a great motivator, and a wonderful deterrent. Quote them a price.

Let's see, first of all you'll need to hire a programmer, or even a team. Anyone know the going rates? I'll throw something out, and anyone can chime in for more accurate numbers:


- $20/hr for a 50 hour job 20*50 = $1,000
- Overhead of 25% 1000*0.25 = 250
(When do projects come in on time and on budget? We must account for this)
- Insurance for broken Wii's 200*400 = $80,000 insurance policy
(Since they are modifying their Wii's, if they some how "brick" it, your team will be responsible for replacing every single broken Wii)

Plus whatever you get paid for doing this. Thoughts?

Yes, thank you all. Everyone has been very helpful and kind—not making fun of my lack of Wii knowledge.


Basically, I want to talk them out of this plan. Okay, so to summarize, here is what I am coming to:

“Without a reliable internet connection, we would need to use SD cards and have the participants physically hand over the data. Two options exist:

1.The participant saves the gamesave file to a SD card. The participants hand us over the entire gamesave file on the SD card. It contains their data as well as their family’s Wii Fit Data.
a. Concern: We should not be accessing their family’s unconsented data.
i.Solution: sign another waiver
2. We hire a programmer to write program that would parse the data to just the participants data, download the data to an SD card. The participant hands us over the SD card with the parsed data.
a. Concern: voiding the Wii warranty and providing the participant with a hacked Wii
b.Concern: Is this worth the time and effort?”

Am I summarizing the technical parts correctly?


Next,

Dancingninja....I was just going to ask that....How long would it take for someone to write this program and then alter the Wiis?? You think about 50 hours?



Edited 1 time(s). Last edit at 02/08/2010 07:34PM by marieresearch.

In some of your posts I believe you mentioned that you have the people you're studying come in to your facility so you can perform tests on them. Since you note that most of your target population in your study won't have internet connections or even TVs (and least likely of all Wiis) is it possible for you to have your test subjects come in and let you monitor them playing the game?

I should clarify: we are purchasing Wiis and giving them to the participants as "compensation" for the study. They are a group of minority, low-income participants who will be recruited from from a clinic. If they don't have a TV, I will probably have to disqualify them from the study, or buy them a TV.

The details are still being ironed out.

What makes this difficult is that it is a "study" not a program. So, really the goal isn't to help them lose weight. The goal is to *see if* they do lose weight. So certain factors...such as researcher monitoring, manual data reporting....throw off the validity of the study itself. The study is to see if the Wii Fit Plus has a positive or negative health impact.

Kind of a spin off of the Hawthorne Effect.... That is why manually having the participants write down what they do would affect the outcome of the study, which is on the Wii Fit itself.



Edited 2 time(s). Last edit at 02/08/2010 09:50PM by marieresearch.

Enough is known about the Wii Fit Plus data format to extract "body test" data. This is probably the most important part of the save file, as it contains date, weight, height, and BMI, among other things. These tests do have to be run by users, however.

The best thing to do would have them send in their SD cards and extract the data.

I'm actually not sure how long it would take. I just ran a high ballpark, and I'm glad I did, as I shall explain below...

There is a way for them to submit their data to you and now brick their Wiis. If they transferred their save file to the SD card via the Wii's save manager, then no homebrew would be required, seeing as all you want is the Wii Fit data.

Here's where the 50 hours comes in: I do not know how much work would go into extracting specific data from the SD card. With homebrew you would need to do the work via the homebrew, at the risk of their Wii, a risk that they do not deserve (there have been reports that non-modified Wiis have had trouble, would Nintendo give them a replacement if there was homebrew on there? Most likely not). Without Homebrew you would have to be parsing the data outside, I'm not entirely certain how that would work, as when I took a look at my save file it didn't look pretty.

However, I'm no expert on save files! If someone else thinks that they can do that, then I would love to be wrong about this.

I quoted 50 hours because I don't know how long it would take, and I certainly would like to come in early than late.

Another thing you might mention to them, if you are trying to dissuade them from going down this path: Based on past experience administering a human research study, the purpose, scope, and research methods (incl. data collected and method of collection) typically have to be VERY explicitly defined when the protocol is submitted to IRB - and if your research happens to involve sponsorship from any governmental agencies, they may have their own approval requirements as well. (I'm assuming your study participants are unaffiliated volunteers and not e.g. employees, which is another can of worms on its own.)

Going from measured-in-lab clinical data to also scavenging their home Wii saves for data which will come to bear in your experimental results - signed waivers or not - is a significant change to the scope of the study, and would almost certainly require the IRB's attention if not an entirely new submission/review/approval process. Depending on your in-house resources, this could cost considerably more than actually hiring the homebrew programmer to code up something to extract and pseudonymize the save data itself.

(On top of that, many of the folks comprising the IRB panel are probably not "computer people" and some might never have even seen a Wii before. Likely they are going to require some serious convincing as to how exactly it is assured that guy-you-hired-from-the-internet's homebrew code [you'll have to explain that too] is correctly interpreting the savegame data, pseudonymizing the records if required by your study, isolating and protecting non-study data and non-particpant data, etc...)

One last question: So how do you look at the data on a save file without homebew? So, say I just have a regular Wii and I save the gamesave file to an SD card. Is there something that "translates" the data so my computer can read it?


I mean, we are already getting their blood, vitals, height and weight at 3 month intervals in the clinic. That should be enough.

All in all, I believe that the quality of data that we get is not going to be worth the risks, time, and effort to modify the Wii, go back to the IRB, and just create a whole bunch of extra work.

Well, thanks everyone. I really appreciate all of the talking points that you have given me. You have been thoughtful in your answers and advice, and I think it will certainly help me convince my team NOT to go down this path.



Edited 2 time(s). Last edit at 02/11/2010 02:08AM by marieresearch.

you can turn that .bin file that is the save file created by your system menu into something else. using the keys of your wii, you can run it through sehger's tool and it will be turned into a group of files that look as it does inside the wii (decrypted). these files would be comparable between saves from all wiis, but you will need the keys from each wii to do this.

you cannot compare the .bin file with anybody else's .bin file due to the fact that they are both encrypted with different keys.

other ways to get the unencrypted save data is
1- read it right from the nand with your own app
2- dump it to a sd card with a homebrew app like waninkoko's save game extractor
3- dump the complete nand and just dig through it to get the right files
etc

If you are doing clinic monitoring, there is no reason to use Wii Fit's data. Good call.



Edited 1 time(s). Last edit at 02/10/2010 01:26AM by WikiFSX.

Quote
WikiFSX
If you are doing clinic monitoring, there is no reason to use Wii Fit's data. Good call.

Indeed.

And you asked a very dangerous question to the wrong people.

You asked: "Is X possible?" to a bunch of techs. And, everything is possible. But the amount of effort required and quality of the result are not something that techs consider when you come with an "is possible" question. As the "is possible" question is an thought exercise. And we love those. We love our most technical advanced solution for any problem.

I was kinda waiting to a solution involving the USBgecko :P

Yeah, I have to agree with Daid. An "is-possible" question is seen as a challenge.

The only thing that's worse is "that's impossible".

In any case, I hope that we were of some help. If you need any of us to "talk nerd" to the research team (complete with unintelligible technobabble) in order to convince them not to do this, just ask.