Millisecond Forums

> It seems that there might be some codes in our study that are not compatible with PC + Inquisit 5 Web

That's theoretically possible, although the 404 error is strange -- 404 indicates that some file that the script requires does not exist on the server. This can, for example, happen if you have the script load different files (e.g. videos in different formats) depending on the platform used (OSX vs Windows) per conditional <include> elements, and there's either a mistake in the <include> elements pertaining to Windows or the files referenced in thos <include> elements have not been uploaded.

If you can post a link to the study in question, I should be able to tell you what specifically is wrong with it.

Thanks!
Sure here it is: (but is has pictures so you can't just run it...is this okay?)

and the web study link:
http://research.millisecond.com/ydunham/batchautocatinverstion.web

Thanks! So, I checked the scripts and files uploaded to the server and additionally ran the web experiment under Windows. It's all looking and working fine, to be honest. The logs, however, indicate that several revisions were made to the batch script and/or the files after the initial registration and uploads on December 4.

For example, the logs contain this error:

"Unable to open file 'http://scripts.millisecond.com/ydunham/batchautocatinverstion/race inquisit5 normal&inverted.iqx'. HTTP status code 404"

And that appears to be true, because -- neither a file called "race inquisit5 normal&inverted.iqx" exists on the server, nor does the currently uploaded revision of the <batch> script reference any script called "race inquisit5 normal&inverted.iqx":

<batch>
/ file="gender inquisit5 inversion.iqx"
/subjects = (1 of 2)
/ groupassignment = random
</batch>

<batch>
/ file="race inquisit5 inversion.iqx"
/subjects = (2 of 2)
/ groupassignment = random
</batch>

So, what is _probably_ happening for you is that the erroneous, outdated <batch> script is still cached either locally on your Windows computer or some network cache server in your institution's network is still erroneously serving the wrong / outdated file, thus triggering the error.

What you can do:

(1) If the problem is a locally cached file, clear your Windows system's temporary and temporary internet files. You can do this via Windows onboard means, or you can use an app like CCleaner for convenience (I recommend using the "portable" version): https://www.piriform.com/ccleaner/builds

(2) If the problem is a network cache, all you can do is wait for said cache to expire at which point it will start fetching and serving the correct files.

Thanks! I just did some work to figure out the reason too...I found that it might be because of the file names (it has "&" in the names) so I changed it...
This might be why you didn't encounter any error--after I changed the names and upload the changed files, I could run on PCs too!

Ah, yes -- that makes sense and explains why I didn't see or get any errors.

So...is it true that in Inquisit 5 Web, PCs can't read file names that have a "&" in it? At least it's not a problem with Inquisit 5 Lab on PCs...

Yes, that would seem to be the problem here -- the explanation, roughly, is this: "&" is a character that has special meaning in the context of a URL, if you will. It is, for example, used to indicate the presence of query parameters as in:

www.example.com/search?parameter1=foo&parameter2=bar

Now, the OSX built-in libraries are smart enough to figure out that this is not what is meant by the "&" in the file name and escape / convert the character to a format that will not mess up the download URL. The Windows libraries are a bit dumber.

In an offline context, where no URLs are involved, this however doesn't matter, which is why things work just fine under Inquisit Lab.

I see! This makes a lot of sense!
Thank you!