"Survey on a stick"

I've ingested a lot of caffeine today, and I just had a thought ...

One of the instruments that I'm currently working on is deployed on tablet PCs so that researchers can perform door-to-door household surveys in a neighborhood.

Another instrument I'm developing will be used in an office setting, with the survey respondents visiting the researcher's office. The researcher will be using either their desktop PC or laptop to conduct the survey.

In both cases, I gather up all the files that make up the Blaise instrument and install them on the target tablet/laptop/desktop computer. After that, of course, is the task of getting the data off the computer for analysis.

Earlier today someone asked me to copy the Blaise instrument files onto a USB stick so they could install the survey onto a computer at another physical location they were visiting later in the day.

It just occurred to me (I'm a little slow) that theoretically they could've just left the instrument files on the USB stick, plugged the USB drive into the computer, run the survey, saving the data to the USB stick, and eject the USB stick when they were done. All the Blaise survey files (and the data) would stay on the USB stick, and nothing would have to be installed on the hard drive of the computer. And no data would have to be extracted from the computer, since the data was already on the USB stick.

It's been awhile since I've written anything other than the simplest DOS batch files. What would it take to "autorun" a Blaise survey on a USB stick when you plugged that USB stick into a computer? Maybe I'm looking at it the wrong way, and a batch file may be the wrong way to go. There may be better solutions.

Any thoughts on the simplest way to place a Blaise survey (and its data) self-contained on a USB stick and "autorun" it when you plug it in?

Or is there a fatal flaw in my thinking, in which case I may finally have to switch to decaf.

Never had to deal with this type of data collection but opens new doors
Here some info that can help.

http://en.wikipedia.org/wiki/Autorun.inf

Hello Bill,

I just poured in my pint of coffee and have yet to swallow it.
I provide our surveyors with a stripped version of Blaise. I.e. i put all the blaise-files of the questionnaire in a folder (i keep only one (1) questionnaire in one folder), together with the programs DEP.EXE and if necessary MANICMD.EXE (we use primarily SPSS for the analyses).
I also put a batchfile (if necessary 2 batch files) to go with it.

Those batch files are generic so that they are not specific for (the name of) a certain survey and can be used in any folder as long all the blaise files and those programs are in that same folder.

Here are the two batch files:

Questionnaire.bat

@ECHO OFF
FOR %%a in (*.bmi) DO .\DEP.exe "%%a"

Blaise2Ascii.bat

@ECHO OFF
FOR %%a in (*.msu) DO .\manicmd.exe "%%a"

DEP is called for every *.BMI-file in the folder end if needed MANICMD is called for every *.MSU-file in the folder.

In many companies and probably with many individuals, the autorun features is switched of and a USB-stick or CD-ROM or so will never autorun.

With those batch-files on a USB-stick a surveyor only has to plug it in, wait until it is recognized and available, open the folder on the USB-stick with the Windows Explorer and double click on the 'Questionnaire' icon. All the files that are produced will remain on the USB-stick including the temporary ones. Nothing will come let alone remain on the host computer.

Of course after finishing the questionnaire he/she MUST first exit DEP.EXE and neatly unmount the USB-stick (right click, eject and waiting until the message 'This device can safely be removed' appaers.
We don't want corrupt files...

So, and now for my coffee....

Greetings,

Kees de Boer

============================

Quote BillMietelski:
I've ingested a lot of caffeine today, and I just had a thought ...

One of the instruments that I'm currently working on is deployed on tablet PCs so that researchers can perform door-to-door household surveys in a neighborhood.

Another instrument I'm developing will be used in an office setting, with the survey respondents visiting the researcher's office. The researcher will be using either their desktop PC or laptop to conduct the survey.

In both cases, I gather up all the files that make up the Blaise instrument and install them on the target tablet/laptop/desktop computer. After that, of course, is the task of getting the data off the computer for analysis.

Earlier today someone asked me to copy the Blaise instrument files onto a USB stick so they could install the survey onto a computer at another physical location they were visiting later in the day.

It just occurred to me (I'm a little slow) that theoretically they could've just left the instrument files on the USB stick, plugged the USB drive into the computer, run the survey, saving the data to the USB stick, and eject the USB stick when they were done. All the Blaise survey files (and the data) would stay on the USB stick, and nothing would have to be installed on the hard drive of the computer. And no data would have to be extracted from the computer, since the data was already on the USB stick.

It's been awhile since I've written anything other than the simplest DOS batch files. What would it take to "autorun" a Blaise survey on a USB stick when you plugged that USB stick into a computer? Maybe I'm looking at it the wrong way, and a batch file may be the wrong way to go. There may be better solutions.

Any thoughts on the simplest way to place a Blaise survey (and its data) self-contained on a USB stick and "autorun" it when you plug it in?

Or is there a fatal flaw in my thinking, in which case I may finally have to switch to decaf.