Getting help in R

---

# GETTING HELP IN R

---

# Me and R

- Studying for a PhD in Statistics at the University of Warwick

- Getting into R:

- 2014: first R course as part of Statistics MSc 
  - 2015: first taste of `ggplot2` 
  - 2016: first R package 
  - 2017: first community involvement (useR!2017, [Oxford R User Group](https://r-oxford.github.io), [rainbowR](https://rainbowr.slack.com), [@R_LGBTQ](https://twitter.com/R_LGBTQ))
  - 2020: first [TidyTuesday](https://github.com/rfordatascience/tidytuesday) contribution
  - 2021: first ...
  
---

# Overview

### Looking for answers
- in R
- online

### Asking questions
- Typology of questions
- Reproducible examples and `reprex` package
- Where to ask
  
### Resources

---

# LOOKING FOR ANSWERS

---

# In R

---

# Within R - Help with functions

```r
# help with a specific function
help(function_name)
?function_name

# quick reminder of the function arguments
arg(function_name)

# see an example
example(function_name)

# see the source code
function_name
```

---
# Within R - `help.search`

```r
# when ?function_name fails (e.g. don't have library loaded)
help.search("function_name") # note quotation marks
??function_name

# for when you can't quite remember the function name
??something_like_function_name
```

---

# Within R - Help with package

```r
# `help`
help(package = "package_name")
# Help panel in RStudio will give links to all documentation
# and help pages for that package

# find/browse vignettes for installed (or specific) packages
browseVignettes()
browseVignettes("package_name")

# use auto-completion in RStudio to see what functions 
# are in a package
?package::
```

`dplyr` demo

RStudio Help home demo

---

# Online

---

---

class: center, bottom
background-image: url(images/google-broom.png)
background-size: contain

---

class: center, bottom
background-image: url(images/google-broom-r.png)
background-size: contain

---

class: center, bottom
background-image: url(images/rseek-broom.png)
background-size: contain

---

background-image: url(https://upload.wikimedia.org/wikipedia/commons/f/f7/Stack_Overflow_logo.png)
background-size: 90%

---

---
# Searching online

- In Google, include "r" and package name in your search and avoid extra words

- Check how recent the result is

- Use [rseek.org](https://rseek.org) to search R documentation, discussion forums, blogs

- Can also try [symbolhound.com](http://symbolhound.com/) if search term contains symbols

- [Ecosia](https://www.ecosia.org) for ethical search

- If trying to get help with an error, include the entire error message

- Use tags in [StackOverflow](https://stackoverflow.com)

- Check a few answers, read comments
  - Reputation (e.g. number of up-votes) can be a good guide

---

# Issues on GitHub page

- Check if it's a known problem awaiting a fix

- commenting means you'll be notified of a fix

---

background-image: url(https://imgs.xkcd.com/comics/wisdom_of_the_ancients.png)
background-size: 75% 75%

---

## Always have a really good stab at finding an answer before asking a question

---

# ASKING QUESTIONS

---

# Typology of questions

---

## Typology of questions

From [this](https://masalmon.eu/2018/07/22/wheretogethelp/) excellent blog, 'Where to get with your R questions' by Maëlle Salmon.

- **Problem**

- Is this a bug? Why is this code not working? How can I successfully install the `rJava` package?

- **Question**

- Is there a tool in R for doing foo? How do I learn about `bar`?

- **Debate**

- What is the best plotting library in R? Should I learn base R or tidyverse first?
---

# Reproducible examples and `reprex`

---

background-image: url(https://tenor.com/view/late-night-seth-lnsm-lnsm-gifs-seth-meyers-gif-9705583.gif)
background-size: 65%

---
class: middle, center

# Reproducible

# Minimal

# Verifiable

---

# reproducible

- You need to capture everything

- `library()` calls at the top
    - all necessary objects.

---

# minimal

- Strip away everything that is not directly related to your problem. 
  
  - Small, simple data objects (usually much smaller than what you're actually working with).
  
  - Preferably built-in, familiar data sets, e.g. `mtcars` or  [others](https://stat.ethz.ch/R-manual/R-devel/library/datasets/html/00Index.html)
  
  - use [datapasta]([https://github.com/milesmcbain/datapasta) when working with your own data, or data outside of R
    - `datapasta` demo
  
---

# verifiable

- Actually reproduces the problem!

- Explain steps (using comments or rmarkdown)

- Include error messages, and possibly output of `sessionInfo()`

- Include what you've tried to fix it

---

class: inverse, center, middle, special-bold
  
# Can be a lot of effort, but often the process of coming up with a good **repr**oducible **ex**ample (**reprex**) allows you to identify (and possibly solve) the problem

---

# The `reprex` package

```r
install.packages("reprex")

# Included as part of the tidyverse
install.packages("tidyverse")
```

The goal of a reprex is to package your problematic code in such a way that other people can run it and feel your pain. Then, hopefully, they can provide a solution and put you out of your misery.

---

# Basic `reprex` steps

- write some R code
- copy it
- run `reprex()` (or use RStudio addin)
- check locally
- paste into target

---

# `reprex()` demo

---

# All things `reprex`

Jenny Bryan's `reprex` RStudio workshop:

- [Video](https://rstudio.com/resources/webinars/help-me-help-you-creating-reproducible-examples/)
- [Slides](https://speakerdeck.com/jennybc/reprex-reproducible-examples-with-r)

---
class: inverse, middle, center

# Where to ask?

---
## Where to ask **problem** questions

- For bugs/issues related to a particular package, file an issue (usually GitHub)

- CRAN package page, `BugReports` url (e.g. [dplyr](https://cran.r-project.org/web/packages/dplyr/index.html))
  - Bioconductor page, `BugReports` url (e.g. [treeio](https://bioconductor.org/packages/release/bioc/html/treeio.html))
  
- [StackOverflow](https://stackoverflow.com) (using tags, e.g. [r], [tidyverse], [ropensci])

- Community forums:

- [RStudio community](https://community.rstudio.com) for questions relating to RStudio IDE, tidyverse, package development and plenty more
  
  - For questions related to open science, reproducibility, documenting data, extracting data out of something (a PDF! a PNG!), package development, rOpenSci’s [discussion forum](https://discuss.ropensci.org) might be the right venue.

---

## Where to ask **question** or **debate** questions
  
- [RStudio community](https://community.rstudio.com) or [rOpenSci’s discussion forum](https://discuss.ropensci.org) (see above),

- [R epidemics consortium](https://www.repidemicsconsortium.org/forum/)

- [R for Data Science](https://www.rfordatasci.com) online community, [Slack](https://rforsatascience.slack.com) and [twitter](https://twitter.com/R4DScommunity). 
  - Learners/mentors/community/office hours.

- R [mailing lists](https://www.r-project.org/mail.html): 5 main and 22 special interest

- Google groups: e.g. [ggplot2](https://groups.google.com/forum/#!forum/ggplot2) and [manipulatr](https://groups.google.com/forum/#!forum/manipulatr) for tidyverse data manipulation queries

- Twitter [\#rstats](https://twitter.com/hashtag/rstats)
  
---

# Twitter, R and #rstats

- [\#rstats](https://twitter.com/hashtag/rstats) twitter is a friendly place
- Twitter for rstats book: [https://www.t4rstats.com](https://www.t4rstats.com)

Good for:

- Easily/quickly answered questions (e.g. is there a package to do foo?)
- Debate questions

Not good for:

- Tagging package maintainers
- Too much code
- Problem questions (unless sharing a link to the question elsewhere)

---

background-image: url(https://media.giphy.com/media/l3q2Lz5yuEFUXX3IQ/giphy.gif)
background-size: 65%

---

# Don't be shy or embarrassed to ask questions!

# It's OK not to know something!

# It's OK to get something wrong!

---

# [recipes example](https://github.com/tidymodels/recipes/issues/542)

---

# [vignettes example](https://github.com/rstudio/rmarkdown/issues/1119)

---

# [fonts example](https://stackoverflow.com/questions/63357522/purchased-font-in-hugo-academic-not-showing-on-all-modern-browsers-computers)

---

# Helping others

- e.g. StackOverflow [[r]](https://stackoverflow.com/questions/tagged/r) and twitter ([#rstats](https://twitter.com/hashtag/rstats))
  
- **Generate content**

- blog posts ([distill](https://themockup.blog/posts/2020-08-01-building-a-blog-with-distill/), [blogdown](https://bookdown.org/yihui/blogdown/), [hugodown](https://hugodown.r-lib.org))
  
  - Share code (e.g. [#TidyTuesday](https://twitter.com/hashtag/tidytuesday), [#FunctionFriday](https://twitter.com/hashtag/functionfriday), [gists](https://gist.github.com))
  
]

- **Engage** in community forums

- Become a mentor in the R for Data Science [community](https://www.rfordatasci.com)
  
- **Contribute** to open source packages ([however small!](https://github.com/tidyverse/modelr/pull/55))

- **Document** code

- package help pages, vignettes
]
---

# RESOURCES

---

# Resources: Asking questions

- StackOverflow guides and question on what makes a good question
  
  [https://stackoverflow.com/help/mcve](https://stackoverflow.com/help/mcve)
  
  [https://stackoverflow.com/help/how-to-ask](https://stackoverflow.com/help/how-to-ask)
  
  [https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example](https://stackoverflow.com/help/mcve)

---

# Resources: `reprex`

- reprex package on tidyverse site

[http://reprex.tidyverse.org](http://reprex.tidyverse.org)

- reprex package on GitHub

[https://github.com/tidyverse/reprex](https://github.com/tidyverse/reprex)
  
- Jenny Bryan webinar

[https://rstudio.com/resources/webinars/help-me-help-you-creating-reproducible-examples/](https://rstudio.com/resources/webinars/help-me-help-you-creating-reproducible-examples/)
  
- Excellent blogpost (with gifs!)

[http://www.njtierney.com/post/2017/01/11/magic-reprex](http://www.njtierney.com/post/2017/01/11/magic-reprex/)

---

# Resources: `tidyverse` help 
- tidyverse guide to getting help

[http://www.tidyverse.org/help](http://www.tidyverse.org/help/)
  
- tidyverse guide to contributing

[http://www.tidyverse.org/contribute](http://www.tidyverse.org/contribute)
  
---

# Resources: Blogposts
      
- Common errors in R

[https://github.com/noamross/zero-dependency-problems/blob/master/misc/stack-overflow-common-r-errors.md](https://github.com/noamross/zero-dependency-problems/blob/master/misc/stack-overflow-common-r-errors.md)
  
  tl; dr: Most errors in R are due to looking for something that isn't there.
  
- Where to get help

[https://masalmon.eu/2018/07/22/wheretogethelp/](https://masalmon.eu/2018/07/22/wheretogethelp/)
  
---

# Resources: Slides

- Slides made with [xaringan](https://github.com/yihui/xaringan)

- Slides styled with [xaringanthemer](https://github.com/gadenbuie/xaringanthemer)

- Fonts from [Google Fonts](https://fonts.google.com):
  - Header font in **Cabin**
  - Body text font is **Montserrat**
  - Code font is **Fira Code**

- Template available: [ellakaye/emk-talks](https://github.com/EllaKaye/emk-talks)
---

.large[
- When faced with a question/problem, have a good look for an answer in R itself and/or online

- Don't be shy to ask questions!

- Write reproducible examples

- Give thought to where you ask/post your question

- Give back by helping others
]
---

.large[
<i class="fas  fa-link "></i> [bit.ly/emk-getting-help-in-R](http://bit.ly/emk-getting-help-in-R)

<i class="fas  fa-paper-plane "></i> [E.Kaye.1@warwick.ac.uk](mailto:E.Kaye.1@warwick.ac.uk)

<i class="fab  fa-twitter "></i> [@ellamkaye](https://twitter.com/ellamkaye)

<i class="fas  fa-laptop "></i> [ellakaye.rbind.io](https://ellakaye.rbind.io)

<i class="fab  fa-github "></i> [github.com/EllaKaye](https://github.com/ellakaye)
]