pres-foss4g-e-2014-07.Rmd

---
title: "Teaching R as a command line GIS"
author: "Robin Lovelace, `@robinlovelace`"
date: "16/07/2014"
output: beamer_presentation
---

## A presentation of 2 halves

```{r, echo=FALSE}
library(png)
library(grid)
```

```{r Halves, fig.height=1.5}
img <- readPNG("figure/pitch.png")
 grid.raster(img) # this presentation: written in R(md)!
```

- Cutting edge research is nothing without communication
- Teaching and outreach benefit from innovation and application

## The two sides of knowledge creation

```{r Sides, fig.height=2.7, echo=FALSE}
img <- readPNG("~/Dropbox/Public/img/ivory-tower.png")
 grid.raster(img) # this presentation: written in R(md)!
```

- Message of this presentation: don't get stuck in an isolated ivory tower or
comfortable teaching bubble!

```{r, echo=FALSE}
# It's a chicken and egg situation: which comes first?
```

## The structure of this presentation

- Inside the Teaching Bubble
   + Problems
   + Solutions
   + Lessons learned   

- Stuck (up) in the Ivory Tower
   + Problems
   + Solutions
   + Lessons learned   

- Transcending the teaching/learning divide

```{r, echo=FALSE}
#    + Input data
#    + A spatial interaction model
#    + Extensions
```

- Conclusion

## A bit of context

## The teaching bubble - in numbers

- 7 R courses, 7 months
- 5 introductory, 2 'advanced'
- Over 150 attendees from many different countries

## Human issues

- Very different abilities `->` different learning rates
- Some people 'getting left behind'
- 'Never touched a command line' `->` frustration
- Not understanding method/applications

## solutions

Extensions
QGIS + make familiar

## Technical issues

- Projectors failing
- R or RStudio not installed
- Packages breaking
- No internet
- GitHub `merge`: confusion and chaos

## Solutions

- Start VERY slowly
- Right balance of content between:
    + lecturing (boring but necessary)
    + worked examples (risk loosing people but fun and interactive)
    + working on your own (can do at home but encourages good questions)
    
## Lessons learned

- The importance of demonstrators

## The importance of demonstrators

## Themes that cut across the divide

- Reproducibility a great benefit in teaching and research
- Good research can lead to good teaching and vice versa
- The balance between spontaneity and stability
- The software MATTERS: take time to describe not only *what* you did, but *how* you did it
    + Details are important - e.g. autocompletion in RStudio


```{r}
summary(cars)
```

## Transcending the teaching/research divide

- Teaching and research needn't compete
- Good research can provide excellent teaching materials
- Innovative teaching can lead to new research ideas and people power
- Teaching can link people up
- Open source software joins people from different place together

## Conclusions

## References