Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Migrate "R Known issues" from SciComp wiki to /computing/cluster_software & overhaul ? #187

Closed
ptvan opened this issue Nov 15, 2018 · 12 comments
Labels
scientific computing scientific computing domain related issues

Comments

@ptvan
Copy link
Contributor

ptvan commented Nov 15, 2018

Proposed Domain
Computing

Content Summary
There's a useful "R issues on Rhino" page (https://teams.fhcrc.org/sites/citwiki/SciComp/Pages/R%20Known%20Issues%20and%20Changes%20Omnibus.aspx) but AFAIK it's not linked anywhere. It's still being referenced/used (a postdoc in my lab found it this week), so maybe we can migrate it to https://sciwiki.fredhutch.org/computing/cluster_software/#packagesmodules-for-python-and-r and update with any additional changes ?

Local Content Expert(s)
Anyone in SciComp, really.

@ptvan ptvan added the scientific computing scientific computing domain related issues label Nov 15, 2018
@vortexing
Copy link
Collaborator

That seems like it could go directly into a compdemo. Can we do that?!??!?! @fizwit You edited that page last. What do you think?

@ptvan
Copy link
Contributor Author

ptvan commented Dec 11, 2018

@bmcgough or @fizwit, what do you think about merging this (and other R tidbits) into the page referenced above ?

@bmcgough
Copy link
Collaborator

I think a separate page for the R issues as we hope these are transitory and will be solved as bugs and workarounds are addressed. Search will be how people find these things as they will be facing error messages from R in most cases. While it is not exactly a howto or demo, it does sort of fit there. Perhaps we consider an FAQ-style page template? That seems like the place for info like this, and an FAQ would give us a place to link stuff we don't know where else to link (make up a question and link it!).

@atombaby
Copy link
Contributor

One thing to keep in mind is that we really don't have transitory bugs since we keep old (i.e. buggy) versions of R around. There are, of course, other issues that may be fixed within an EB release. In the current page it's not clear which "bugs" apply to which releases.

@vortexing
Copy link
Collaborator

vortexing commented Dec 12, 2018

This strikes me as a twofold issue. One being "Bugs in the current Default version of R" and the other being "Bugs in historical versions of R still supported".

Perhaps we begin a Resource Library entry that is organized by R version, and under each version we list the known bugs. This would require the current Sharepoint content to be broken down by version, which may not be possible, so we may need to have at the bottom a section of "Historical bugs". And we might just have to be ok with that.

Edited to add: I'm hesitant to go down the FAQ avenue. Let's pick a theme, like "Known R Build Issues", make a compdemos page for it, and then organize within it to allow it to collect all of the random items about R that exist. We could then, as needed, make a similar page for other topics to collect thematically related FAQ's. As these evolve they could move into article content.

@vortexing
Copy link
Collaborator

@bmcgough and @atombaby what do you guys think about this issue? Can we resolve it in some way? What portion of known bugs do we want to document for people?

@vortexing
Copy link
Collaborator

@ptvan do you think moving some of this content to your R Tips and Tricks demo might be of interest?

@ptvan
Copy link
Contributor Author

ptvan commented Feb 25, 2019

That sounds good to me !

@vortexing
Copy link
Collaborator

@ptvan Shall we resolve this one way or the other?

@atombaby
Copy link
Contributor

We're going to track build-specific information in the index for the build description (FredHutch/easybuild-life-sciences#177)

@fizwit
Copy link
Member

fizwit commented May 30, 2019

I see two types of R issues that can be documented. General issues that effect multiple versions of R and/or R issues relating to our environment. And specific issues that would only relate to a single release. To handle the case of documenting issues with a release, I have added an Know Issues
section of the software release docs in Easybuild life sciences.

More general issues with R could be documented here Available R Modules

My suggestion moves software issues out of SciWiki and puts them with the software documentation.
Scientific Software

The applies to Python and other tools.

@vortexing
Copy link
Collaborator

I think this is good. I like @fizwit 's solution. I'm calling it!!!!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
scientific computing scientific computing domain related issues
Projects
None yet
Development

No branches or pull requests

5 participants