[Jeremy]

Hi Danil,

This is an issue with the native-Ibex Question controller. Here’s a workaround:

Add a file named global_z.css to your project’s Aesthetics folder:

.PennController-myController.Question-Question ol {
    list-style: none;
    display: flex;
}
.PennController-myController.Question-Question li {
    min-width: 5em;
    pointer-events: none;
}

Then you can use this in your script:

newController("myController", "Question", {
    q: Question_asked,
    as: ["кон","коне","коня"],
    randomOrder: true,
    hasCorrect: 2
})
    .center()
    .print()
    .log()
    .wait()
    .remove()

Replace myController both in the css file and in your script if you want to use a different name

Jeremy

[Jeremy]

Hello Tyler,

I do see 22 submissions, but there is no rows in the database for the 22nd one, so you will necessarily be unable to download it

I went and checked if the rows were still being processed for addition to the database, but unfortunately I couldn’t find anything. I don’t know what happened (maybe a bot tried to send an empty submission?)

I apologize for the inconvenience, let me know if you want me to do delete the empty submission from the records

Jeremy

[Jeremy]

Hi,

You might be interested in the explanations on this page

Your code creates two Video elements both named "28e", so when PennController creates the second one, it has to rename it to give it a unique name. You will need to give a different name to each Video element, and accordingly call wait on the appropriate one depending on the value of the Var element

Jeremy

[Jeremy]

Hi Ecenur,

Thanks for the feedback. This is actually an issue that was raised in the past. Unfortunately there is not much one can do about it, as it takes place on the participant’s end and I don’t know of a way to make the script detect the presence of such a translation feature

Jeremy

[Jeremy]

Hi Jack,

Glad to read that you were able to download your results in the end. When download stalls for more than 3min, you can consider that the connection with the server is lost. Try downloading again at least 15min later, making sure to refresh the page first (otherwise it will try to use the stale connection)

Jeremy

[Jeremy]

Hi Jet,

This is not something you can currently do on the farm, as there is no way to retrieve information from past results from a current run of the experiment

The closest you could get is check the value of __counter_value_from_server__ which, in the absence of a SetCounter in your experiment, will get incremented only when you receive a submission. However, it could still be that multiple participants open your experiment before the counter gets incremented to the cap value, and all those participants will still take your experiment and submit their responses

Jeremy

[Jeremy]

Have you identified that the problem only occurs when you use a single csv file, but that it doesn’t occur when you use 3 csv files?

If you think having the table in main.js would help, you could use AddTable after ResetPrefix:

PennController.ResetPrefix(null)

AddTable("myTable", `column1,column2,column3
cell1row1,cell2row1,cell3row1
cell1row2,cell2row2,cell3row2`)

Replace "myTable" with whatever name you want to use, and make sure you also reference that name as the first argument of your Template command. Note that spacing and linebreaks matter between the two ` characters in AddTable, so follow the same format as above

Jeremy

[Jeremy]

Huh, this is a very weird problem indeed

Do you recognize those words that are absent from the csv files as words that *used* to be in the csv file at some point? I can’t see a pattern in the browser information of the participants that encountered the problem vs those who didn’t, but the farm is currently having glitches where it can save multiple copies of the same file: you only see one file listed in your project when editing it, but internally there can be several versions of it and the one that gets served to a participant can be random

My suggestion is you make sure you have the csv files that you want on your computer, then delete all csv files from your project on the farm (refresh the page after you delete a file and make sure it does not reappear in the project tree) and after that, upload the csv file(s) from your computer to your project

Let me know if the problem persists

Jeremy

[Jeremy]

Hi Ecenur,

I think you might have been working on the project as I was taking a look at it, because the debugger once threw an error about not finding a csv table, which was consistent with the copy of your project I created (which had only one csv file) but then I refreshed the page and no longer saw the error, and the copy I created after that showed three csv files

I am not sure I understood your description of the problem: what do/did the “really weird mixed-up sentences” consist of? Do you mean that some row.questions come/came from lines in the table that had different values in their group cell?

Jeremy

[Jeremy]

Hi,

Use randomize("label") to get a randomized series of trials labeled “label” where you would just use "label" to get a non-randomize series:

PennController.ResetPrefix(null)

var blocks = [randomize("block1"),randomize("block2"),randomize("block3"),randomize("block4")];
fisherYates(blocks);

Sequence("training", ...blocks)

Jeremy

[Jeremy]

Hi,

1) This sounds like a problem related to the saving issues that have been extremely frequent recently. For example, I just tried changing "test" for "woman" in a copy of your experiment and it worked as expected. Maybe try copying the content of main.js, pasting it in an external editor, deleting the file that’s in the Scripts folder in your project, make edits in your external editor, then recreate the file in your project and paste the content from your external editor

2) I am able to run your experiment without changing its current code, and I do see a trial labeled “fin” that contains the content of endform.html. Maybe your unsuccessful experience results from the same saving issue as above

Jeremy

[Jeremy]

Hi,

This sounds like an issue with the client-server communication, where the client (your browser) was able to retrive the list of submissions, but not the corresponding rows (except for the ones in the preview, ie. those of the most recent submission). When that happens, your page “sees” the list of submissions in the cache but does not realize the rows are missing, so it won’t try to download them again if you just click on “Download” again. You need to refresh the page to clear the cache so your browser tries downloading the rows again (and hopefully the same problem does not happen again)

Jeremy

[Jeremy]

Hi,

If you know where the critical word is in advance, just list its index in a column in your table (I am assuming you’re generating the trials using Template) and use .log on the closing parenthesis of newTrial() to report that value as an additional column in each results line of the corresponding trial. Then you can read that column from your results file during your analyses to apply the correct calculations

Jeremy

[Jeremy]

Hi,

If you don’t want to halt the script until the participant presses a key, remove .wait() that you’re calling on the Key element

Jeremy

[Jeremy]

Hi,

PennController preloads resources, which means images should be fully preloaded before PennController executes any print command on them. The .log command you call on an Image element will report the timestamp corresponding to when the .print command on that same element was executed, so that should be when the image was completely rendered on the screen

Jeremy

[Author]

Posts