[R] New-user support package - suggestions?

[Uwe Ligges]

Andrew Robinson wrote:

> Dear Community,
> 
> This is largely a repost with some new information.
> 
> I'm interested in developing a package that could ease the
> command-line learning curve for new users.  It would provide more
> detailed syntax checking and commentary as feedback.  It would try to
> anticipate common new-user errors, and provide feedback to help
> correct them. 

Re. "anticipate", see

install.packages("fortunes")
library("fortunes")
fortune("anticipate")


> As a trivial example, instead of
> 
> 
>>mean(c(1,2,NA))
> 
> [1] NA
> 
> we might have
> 
> 
>>mean(c(1,2,NA))
> 
> [1] NA
> Warning: your data contains missing values.  If you wish to ignore
> these for the purposes of computing the mean, use the argument: 
> "na.rm=TRUE".


Attention. If you give help like this, you will implicitly teach users 
not to read the manuals but trust on R's warning/error messages and 
suggestions.
Some students won't ask the question "why do my data contain NAs" but 
will start using na.rm=TRUE, even if the error was in a prior step and 
no NAs were expected at all.


> I'm interested in any thoughts that people have about this idea -
> what errors do you commonly see, and how can they be dealt with?
> 
> I have funding for 6 weeks of programming support for this idea.  All
> suggestions are welcome.
> 
> Cheers,
> 
> Andrew


Your project sounds very ambitious. I anticipate you will be 
disappointed after 6 weeks of "programming", because you won't have 
achieved very much. I'd rather try to spend 6 weeks of time for some 
more promising projects...

Just my first thoughts, just go on with your project if you are convinced.

Best,
Uwe Ligges