OpenSesame 4.0 will be released soon! Take it for a test drive now!

sebastiaan

OpenSesame 4.0 Melodramatic Milgram is a major new release with many new features and improvements! 😎 The list is too long to include in this post but preliminary release notes can be found here:

• https://osdoc.cogsci.nl/4.0/notes/400/

OpenSesame 4.0 will be released in the coming months. The exact release date depends on currently ongoing testing as well as on ongoing development of closer integration with JATOS / mindprobe.eu (not included in the current prereleases).

For now, prerelease packages and installation instructions can be found here:

• https://osdoc.cogsci.nl/4.0/download/

Among the new features are 17 complete translations that cover all major languages. The image below shows the Japanese translation.

Enjoy and let us know how OpenSesame 4.0 works for you!

labovich

Thanks! I started working with it and I really like it!

I also found some bugs I'll be happy to report (for example, the consent form doesn't work)

I'll be happy to report them all if you'll give me the link.

sebastiaan

@labovich Thanks! Before reporting the issues, make sure that you are running the latest versions of all packages. (When you start OpenSesame 4.0, an update check will be performed in the background, and after a few minutes an update notification will appear plus an option to perform these updates.)

labovich

Hi Sebastiaan

I am using it for a while now.

It was a while since there were any updates when I started OpenSesame. Is there anything else I should do?

sebastiaan

@labovich No, then you're good. Please let us know of any bugs you encounter, and don't forget to include the relevant version numbers. Thanks for helping us test! 😀

labovich

No problems. Here are the ones I've encountered so far:

(I'm working on Windows 10 (up to date on all updates) and OpenSesame 4.0.0a33 Melodramatic Milgram):

1. "dismiss this message" does not respond

2. clicking on "list available keys" - gives an error (it allowed to send an error message, but here too, "Dismiss this message" does not respond)

3. consent form: always responds as if the Checkbox is unchecked

4. Don't know if it's a but or a feature, but in the top toolbar the very useful icon of switching between working with multiple or a single tab is gone (available from another menu), but the toggle option is there.

I really love the other changes! Thank you for this!

sebastiaan

@labovich Thanks! There was an issue with the updater itself some time ago. Right now the latest version is 4.0.0a43 (not a33). If you download the latest prerelease package then the updater will be fixed and you will get updates to all of the packages. Issues 1 and 3 are actually already solved. Issue 2 isn't though, so thanks for reporting that!

Don't know if it's a but or a feature, but in the top toolbar the very useful icon of switching between working with multiple or a single tab is gone (available from another menu), but the toggle option is there.

Based on user feedback, it seemed that most people simply enable one-tab mode and leave it at that. So that's why the default is now that one-tab mode is enabled and the toolbar icon has been removed.

labovich

Thanks!

The latest I found in the link above was 4.0.0a41. would it work?

sebastiaan

Yes, and then from there you will receive further updates automatically.

labovich

Thanks! I downloaded version 4.0.0a41. It still doesn't do any automatic updates, and I encounter the same errors.

labovich

If there if any way I can get the link to install the newer version, I'll appreciate it.

4.0.0a41 doesn't update automatically, and I still have the same issues.

sebastiaan

@labovich You were right, actually. The updater still didn't work when being launched from the installed version of OpenSesame (which is of course the most common scenario). I just updated the prerelease packages so you can grab those. And then, from now, updates should appear automatically! (And if not, please let me know!)

• https://github.com/open-cogsci/OpenSesame/releases/tag/prerelease%2F4.0.0a27

labovich

Thanks Sebastiaan!

I've downloaded and installed from the link.

At the first time I opened it, there was an attempt to update (I just left it alone, and at some point the command window just closed).

A day after, I rebooted the computer and restarted opensesame. I'm still in the same version. Was this supposed to change?

Also, the "dismiss" still don't respond, and I get the same error message with the consent form.

Is it all because there was no update?

labovich

Hi Sebastiaan

Playing a bit more with the file, I've learned that all of the forms gave the same message of "variable does not exist". When I re-created the forms in the current version the experiment ran without any problems.

I guess the forms in the first version of opensesame-4 I worked with were maybe incompatible with the current version. Anyway, it's fixed now. Thanks for the update!

sebastiaan

@labovich That's probably correct, but since OpenSesame 4.0 should be fully backwards compatible that's still a bug. Could share an experiment that shows this issue (i.e. that runs fine on OpenSesame 3.3 but gives this variable-does-not-exist error on 4.0)?

labovich

sure:

color_word_tali_ready_v3.osexp

labovich

and here is a link to another experiment that works in v3 but not 4 (the two versions are attached):

opensesame files to share - OneDrive (sharepoint.com)

sebastiaan

@labovich Thanks! I fixed the first issue, related to forms, in 4.0.0a50. I cannot open the second experiment though. If that still provides errors after the update, can you attach it or share it through a different link?

labovich

Thanks Sebastiaan!

I made a lighter version of the experiment so I can upload it here:

Newexperiment.osexp

If there is a solution of how to run this in v4 I'll be happy to hear (the opensesame course I built with V4 begins next week...)

labovich

I cannot find the link to 4.0.0a50 (only to 49)

sebastiaan

If there is a solution of how to run this in v4 I'll be happy to hear (the opensesame course I built with V4 begins next week...)

You're right that this runs in 3.3 but not in 4.0. However, this rather seems unexpected behavior from 3.3 because logically speaking this shouldn't work: you're referring to a variable col in the table of new_loop before this variable has even been defined, because that only happens in the loop table of nested_loop which is executed as part of new_loop.

I'm sure there is an idea behind this structure, but I don't see it at the moment. What kind of behavior were you trying to implement?

As an aside, the .set() and .get() methods for getting and setting variables have now been fully removed. This has already been deprecated for years, and I felt that 4.0 was a good moment to remove this kind of deprecated functionality that was mainly weighing down the code.

I cannot find the link to 4.0.0a50 (only to 49)

Assuming that the updater is now properly fixed, an update should appear after a while! (But updater is terribly slow, mainly because conda is so slow.)

labovich

Hi Sebastiaan

I'm still with .47, there were no updates, but I'll try to restart my computer...

Maybe I attached the incorrect file. I'm attaching a file that works in V3 but not 4.

Maybe (almost certainly) I should modify it to fit V4 but I don't know how...

Thank you again for all your help, I really appreciate it!

fun_with_Flags.osexp

labovich

I think that something is blocking my updates:

I'll try running opensesame as admin.

I hope that it's not the university's protections

sebastiaan

@labovich Indeed, it seems that the connection is blocked by the operating system, possibly by the Windows firewall.

Maybe I attached the incorrect file. I'm attaching a file that works in V3 but not 4.

It was already the case for the previous version of the experiment that it works in 3.3 but not 4.0. My point was that it should actually not work, so in that sense it's not a bug to be fixed in 4.0, and I was curious about what the logic behind this loop structure was.

labovich

Hi Sebastiaan

Re. the updates, I'll be happy to get a link until I can get help with making windows unblocking it.

About the loop structure - it's very simple (I thought) - using the cells of a main loop, you refer to a nested loop, just so it would take some property from the nested loop to present (e.g., a specific small animal you present, when all you care is that the animal is small). The nested loop is the one that runs the trial sequence, that begins with defining the value of a variable that break the nested loop after a single trial. That way, the higher sequence with the main loop begins again and so on...

This "bug" is very useful. Without this functionality I won't be able to run many of my studies. I was even able to use this method to have several levels of such nested loops: An experiment with 5 Stroop trials (only one of them, a random one is incongruent) followed by a geometry problem that, although I don't care what the actual problem is, I controlled whether the response to it should be 'right' or 'wrong'.

I'm sure there are ways to do it by what you will call a "simple" code. But for many of us who don't code, it's impossible.