I look forward to hearing from you. So that we can all follow the discussion:

  • If you post a message on this page, I will reply on this page.
  • If I leave a message on your talk page, I will watch for your response there.



Congratulations on having received the 2020 Faculty Innovation Award from the American Association of Community Colleges! DavidMCEddy (discusscontribs) 01:03, 20 September 2020 (UTC)

Indeed, congratulations DannyS712 (discusscontribs) 01:34, 20 September 2020 (UTC)
Wikiversity is lucky to have someone like you involed in our project. --mikeu talk 01:14, 24 September 2020 (UTC)

Broken PortalsEdit

Dave, The portals linked from the top of the Wikiversity main page are broken. For example, the Portal:Mathematics has two script errors that are announced in prominent red text. I have looked at the code (briefly) and don't see where to fix this. These errors have remained for some time, and are in several of the portals. If you get a chance to look at this and either send me in the right direction or fix the errors I will appreciate it. Thanks! --Lbeaumont (discusscontribs) 18:44, 3 October 2020 (UTC)

@Lbeaumont: The import of Module:Portal on 19 January 2020 broke the template. The revisions don't include the lead function. See [1].

Elementary AlgebraEdit

Is there a way to implement the elementary algebra course? I feel like we would be able to use it as it can help students with algebra courses. We can also use it as a "pre-requisite" to the College Algebra course. If it is possible to create an elementary algebra course, I would be glad to help, although I might need to fit it into my schedule. MattPerfects26 (discusscontribs) 11:07, 13 October 2020 (UTC)

@MattPerfects26: Sure! We can do whatever we have time for. See College Mathematics for one approach. That may already have most of what you are looking for. Otherwise, I'd recommend building a similar course using OpenStax: Prealgebra. -- Dave Braunschweig (discusscontribs) 15:08, 13 October 2020 (UTC)

I created the page God: Rational Proofs of ExistenceEdit

I've read some content on various proofs for the existence of God and felt I could help. But I didn't entirely source the article and wanted to move it to draftspace. I did move it to "draft: God: rational proofs of existence" but I don't want a redirect there from the main article. Can you help me out in removing the re-direct as I am not sure how to do it.--HalMartin (discusscontribs) 14:28, 16 October 2020 (UTC)

@HalMartin: Done. The easy approach for something like this is to add {{Delete|Author request}} at the top of the page.
When you are ready to move it back to main space, it would be best to find an appropriate learning project to add it to as a subpage. While I respect the article's position, see Wikipedia: Existence of God for why it wouldn't be a stand-alone page. I'd recommend putting it under God, but if there's a better place for it from your perspective, please let me know. -- Dave Braunschweig (discusscontribs) 15:02, 16 October 2020 (UTC)

Small wikisEdit

Hello, i am Katerina Sarri, from the greek el.wiktionary. I became your student by reading your Lua lessons, the only excellent introduction I could find for beginners like me. I am indebted. As an example of an ordinary sysop of a small.language.wikiproject, I am here to ask for help, through your experience, your power, your influence.
People like me find themselves in wiktionry/etc by chance, knowing nothing about programming whatseover, but have to complete some tasks: old semifinished modules by authors who have left for some years now, urgent needs, etc.
Small wikis like us, have no official 'global' support for Lua, modules and templates. They are doomed to html, not being able to perform the standard tasks of their project. These tasks of every wiktionary, of every wikipedia, of every wikiproject could be summarized in 10 to 20 kinds of modules for each project, I presume. Had they been offered, the editors would just need to 'change the words' and adjust to their language's idiosyncracies. Copypasting an advanced module from a sister project is not a solution. Our modules cannot follow the steps of the excellent en.wikt modules because they are too complicated.
Sir, I have been trying to find a LuaSupport bureau, or a LuaSchool, simply: Lua help. Many kind people assisted me immensely from en.wikt, from en.WP. But I know how boring and irritating to experts childish questions can be: it feels very uncomfortable to go around wikiprojects seeking help. I was hoping that you know if such a LuaSupport were possible, whom to contact, how to ask for it. Thank you for listening. Sarri.greek@el.wiktionary, Sarri.greek (discusscontribs) 19:44, 17 October 2020 (UTC)

@Sarri.greek: I'm glad you found the Lua course helpful! There used to be a technical advice team. See mw:Technical Advice IRC Meeting. I don't know if they are still active on IRC or not. But a check of the archive would give you names of possible people to contact. Beyond that, I don't know of any Wikimedia Lua support groups. Off-wiki, I'm sure there are Reddit and Stack Exchange / Stack Overflow areas for Lua. But going around wikiprojects seeking help may be the best option. -- Dave Braunschweig (discusscontribs) 19:59, 17 October 2020 (UTC)

RMarkdown vignettes with Wikiversity?Edit

@Dave Braunschweig: Do you have thoughts about what one might do with RMarkdown vignettes that effectively make reproducible the statistical analysis behind, e.g., material I'm publishing on Wikiversity? They would be like appendices to a Wikiversity article that would be in foreign formats.

For example, my Wikiversity article on Forecasting nuclear proliferation is supported by Spencer Graves; Douglas A. Samuelson (2020), "Forecasting nuclear proliferation", Ecfun: Functions for EcdatWikidata Q89780728, which is currently a "vignette" as part of my "Ecfun" package for the R (programming language).[1]

Each such vignette in effect comes in three files: (1) A *.Rmd source code master. (2) A *.R file with only the R code. And (3) a display document that could be in HTML, PDF, MS Word, Open Office or some other format. Users typically will first skim the display document. If it interests me enough, I will access the *.R file and walk through that, testing variations to check my understanding. If I really want to do more with it, sometimes I may also want the *.Rmd file, so I can modify it still further.[2]

However, packages submitted to the Comprehensive R Archive Network (CRAN)[3] must meet size restrictions that are suggesting to me that I perhaps should not be storing too many of these things with that package. I recently wrote code to not display the figures with the output so the HTML or PDF would not exceed their size limit!

So, back to my original questions: First, do you agree that it would be smart and appropriate to make available to the reader of the research I publish on Wikiversity these documents that make that research 100% reproducible?[4] If yes, how would you suggest this be done?

Thanks, DavidMCEddy (discusscontribs) 06:54, 21 October 2020 (UTC)


  1. Spencer Graves (4 February 2020), Ecfun: Functions for EcdatWikidata Q56452538.
  2. The *.Rmd files are text files in a version of Markdown used especially with RStudio, which are easily compiled (or "knitted") to HTML, PDF, MS Word, or other formats. These are similar to Jupyter notebooks in being designed to support mixing narrative with computer code. A year or two ago, I was writing code in both Python and R, and I found I could use both Python and R in the same RMarkdown vignette, when I was unable to do that in Jupyter. I've since heard that Jupyter now supports that. However, I haven't since had a need for that, so I have not confirmed that. RMarkdown is under active development and things like figure numbers with references are only supported with some output types but not others. I'd have to research it to be more specific. And what seemed not to work yesterday may work today and vice versa. The problem with things breaking suggests to me that the RStudio developers have not been super rigorous about developing enough unit tests for what they have done. However, it's free and open source, so I'm grateful that it works and is available!
  3. Comprehensive R Archive NetworkWikidata Q2086703.
  4. And trivially reproducible for someone who can install the free and open-source R and RStudio software and make it work for them.

@DavidMCEddy: You can certainly include source code in a Wikiversity page. There would be two approaches. You could either include it in a collapsable section or include it as a subpage. I prefer the subpage approach myself, but I don't have a particular reason other than not wanting to require JavaScript functionality for a user to access the content.
If I understand correctly, you have three files you are considering including. I don't think the mw:Extension:SyntaxHighlight extension supports .rmd, but you could tag that as text. R is supported. HTML is also supported, but I'm not sure it's necessary to include the full HTML if you have the source code to regenerate it, particularly since what you would want to display is the HTML result, not the HTML itself.
I would include a link to an environment that can generate the results, if possible, but I would include the code here with the article rather than assuming that another venue will continue to host it perpetually. Keep the article and code together for future users. For an example of how I do this, see Programming Fundamentals/Introduction/Python3.
Also, you may be interested in something I've been working on for Server-Side Scripting. I now have links where users can easily pull their own sample data. You may want to include something like this for your own work. Examples include:
Dave Braunschweig (discusscontribs) 16:51, 21 October 2020 (UTC)
Hi, Dave:
Consistent, I think, with your suggestions, I've added a section on "Appendix. Companion RMarkdown vignettes" to the Wikiversity article on "Forecasting nuclear proliferation". This consists primarily of links to two subpages:
both of which consist essentially of an introduction on how to get R and RStudio if you don't already have them followed by the RMarkdown code between <syntaxhighlight lang="R Markdown"> and </syntaxhighlight>. They also include separate sections giving the companion LaTeX bib.
Thanks, DavidMCEddy (discusscontribs) 04:25, 26 October 2020 (UTC)
@DavidMCEddy: Looks good to me. Very thorough. -- Dave Braunschweig (discusscontribs) 01:02, 27 October 2020 (UTC)
@Dave Braunschweig: What do you think about writing an article for The R JournalWikidata Q15817103 on something like "Wikiversity for teaching materials and reproducible research"?
The article could start with a broad overview of the mission of Wikiversity and it's relationship to Wikipedia. It could cite a few Wikiversity articles that discuss R, including "R (programming language)" and "Searching R Packages". It could cite Peter Binkley (2006), "Wikipedia Grows Up", Feliciter (2)Wikidata Q66411582, noting that people who might otherwise be trapped in their own echo chambers instead collaborate to produce a narrative that people with very different world views can more or less live with. Also, Feng Shi; Misha Teplitskiy; Eamon Duede; James A Evans (29 November 2017), The Wisdom of Polarized Crowds, arXiv:1712.06414Wikidata Q47248083 did a content analysis of all edits to English Wikipedia articles relating to politics, social issues and science from its start to December 1, 2016, and concluded that (a) the best articles tended to have large teams of editors, arguing over important issues, (b) 95% of articles could be improved by attracting more editors, and (c) only 5% of articles showed signs of counterproductive conflict. All this suggests that Wikiversity could become a leading platform for crowdsourcing research. We could cite my articles on "Forecasting nuclear proliferation" with its R Markdown appendices to display it as a venue for prepublication review, inviting the entire world to collaborate in pushing the boundaries of knowledge on the chosen topic. That so far hasn't happened with "Forecasting nuclear proliferation", but it could.
If you wanted to write such, I'd be happy to review it. If you wanted me as a co-author, I would not want to commit to doing much on it until maybe late next year. I'm planning to submit "Forecasting nuclear proliferation" and "International Conflict Observatory" to refereed academic journals, and I have other papers in my head that I want to write before I'd want to take the lead on something like this.
However, I think The R Journal might like something like this. I published an article there several years ago (as Spencer GravesWikidata Q56452480).
And thanks again for all your support since I started publishing material on Wikiversity -- including, most obviously, fixing that Lua error. DavidMCEddy (discusscontribs) 02:25, 28 October 2020 (UTC)
@DavidMCEddy: I'm currently working on a full course that has to be ready by January 15. I'm fortunate that I didn't need to publish to make full professor (community colleges focus on teaching), and what I've published here at Wikiversity has earned national recognition. If you need some content suggestions, let me know, but I'll leave it to you to write the article. Thanks for asking. -- Dave Braunschweig (discusscontribs) 02:36, 28 October 2020 (UTC)
@Dave Braunschweig: maybe international recognition? DavidMCEddy (discusscontribs) 02:43, 28 October 2020 (UTC)