Added SPM8 revision 6313 (compatible with MATLAB >= 2012b), couple to…

[rcwelsh]

Adding SPM8 revision 6313 to the repository. I verified this worked with FirstLevel script. This also will work with upstream branch spm8Batch/update_to_work_with_spm8_R6313 coming as the next pull request.

@UMPsychMethodsCore/developers : Who is going go give +1 on this and see it all the way through to public?

[dankessler]

For the most part this looks fine, but what's going on with the Utilities tree?

I see a number of files introduced in

Utilities/gPPPIv13.1/

[rcwelsh]

I added that as well, some people might find useful. It’s the generalized PPI package from McLaren at Harvard.

-Robert

On Nov 2, 2015, at 2:13 PM, Daniel A Kessler notifications@github.com wrote:

For the most part thi slooks fine, but what's going on with the Utilities tree?

I see a number of files introduced in

Utilities/gPPPIv13.1/

— Reply to this email directly or view it on GitHub https://github.com/UMPsychMethodsCore/MethodsCore/pull/376#issuecomment-153128281.

[dankessler]

That's fine, but I'd rather track Utilities using the approach laid out on the wiki.

Unless gPPPIv13.1 is specifically dependent on r6313 I'd rather unbundle.

[dankessler]

So I guess ignore the part where I say "That's fine" :P

[rcwelsh]

but then if i follow the wiki for gPPI seems like the SPM8 R6313 and the SPM8 should stay separate :-) ?

On Nov 2, 2015, at 2:28 PM, Daniel A Kessler notifications@github.com wrote:

So I guess ignore the part where I say "That's fine" :P

— Reply to this email directly or view it on GitHub https://github.com/UMPsychMethodsCore/MethodsCore/pull/376#issuecomment-153132629.

[dankessler]

Not sure I understand/follow. I agree that SPM8 and SPM8_R6313 are separate, and that moreover gPPPI is another separate entity (that I presume depends on having access to some version of SPM8, but not a specific release candidate).

SPM release tracking predates the "Utility" workflow; the best thing we had back then was the "core" branch to track these sorts of changes.

Later, when we were about to bolt on BNV for some visualizations we were doing I realized that the existing workflow didn't really fit for a tool that was for the most part externally developed but that sometimes underwent small patches by us for batching or other functionality, so I (with input from other members of the dev team) outlined the Utilities workflow.

Fast forward to today, and SPM is still kind of unique in that we maintain multiple versions simultaneously (which is different than what we do for the other Utilities), which is why I still think of the two workflows as meaningfully distinct.

[rcwelsh]

Okey @dankessler , this should be good to go. I removed the gPPI from Utilities, did the local commit and then pushed to robert_verse.

[heffjos]

This is being merged into core_beta. Do we first want to merge it into core_alpha?

[mangstad]

I suspect, per discussion in #378 that we may be simplifying our proposed workflow. I think removing the alpha channel makes the most sense, because essentially our personal local repositories on whatever machines we're doing the development on serve as the unofficial alpha channel.

[heffjos]

I did not think anything was officially decided from that discussion yet. We also still have pending pull requests from alphas to betas (#361 and #337) and local branches to alphas (#381, #367, and #366).

[mangstad]

Sorry, I wasn't suggesting it was decided, just that I think that's the most likely course of action for the future.

[dankessler]

This looks fine to me. Since everything is going into beta anyway I'm not going to worry about the target, and since it's only touching spm8_with_R6313 I can't imagine it's controversial