[R] Find in R and R books
spencer.graves at structuremonitoring.com
Tue Nov 23 03:20:35 CET 2010
Hi, Mike, et al.:
On 11/22/2010 5:43 PM, Mike Marchywka wrote:
>> Date: Mon, 22 Nov 2010 17:28:57 -0800
>> From: spencer.graves at structuremonitoring.com
>> To: gwo at well.ox.ac.uk
>> CC: r-help at stat.math.ethz.ch
>> Subject: Re: [R] Find in R and R books
>> Other people like R Site Search
>> (http://search.r-project.org/nmz.html), which is available via the
>> standard R function "RSiteSearch".
>> For me, the fastest literature search on virtually anything
>> statistical is the "findFn" function in the "sos" package.
>> (Disclaimer: I'm the lead author of that package, so I may be biased.)
>> "findFn" sorts search results to put the package with the most matches
>> first. The print method opens a table of the results in a web browser
> Again, I have in past taken docs for various things like R, rendered
> html to text and used things like grep and built my own indicies.
> However, your facility does seem in that line of thought.
> Personally I haven't had a problem with google scholar or
> probably even citeseer would return good hits, R is not
> a common english word so I think google can make use of it.
Thanks for this.
For me, anything that mixes math with worked examples is vastly superior
to either alone, because I no longer have to puzzle sometimes for hours
over a single line or page of mathematics: I can try a variety of
examples and walk through the code line by line until I understand. In
that way, I find R packages much more intelligible than theoretical
treatises. This is especially true when the R package comes with a
vignette or companion documentation with script files working the
examples (e.g., like the "scripts" subdirectories for "nmle" and "fda").
>> with hot links to the individual matches. "sos" comes with a vignette,
>> which includes an example of the "writeFindFn2xls" function. This
>> writes a "findFn" object to an Excel file with two sheets: The second
>> is all the matches found. The first is a summary of the packages found
>> with extra information not available via RSiteSearch.
>> Hope this helps.
>> On 11/22/2010 3:19 AM, Georg Otto wrote:
>>> Alaios writes:
>>>> Also when I try to search in google using for example the word R inside the search lemma I get very few results as the R confuses the search engine. When I was looking something in matlab ofcourse it was easier to get results as the search engine performs better.
>>>> What are your tricks when you want to find some function that provides some functionality?
>>> To search R-specific sites the best place to go is this one:
>>> R-help at r-project.org mailing list
>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help