[Bioc-devel] biocLite

[Martin Morgan]

Hi Wolfgang, Simon --

I like the spirit of these suggestions...

On 09/14/2011 07:06 AM, Simon Anders wrote:
> Hi
>
> On 09/14/2011 02:07 PM, Wolfgang Huber wrote:
>> currently,
>> biocLite("GreatNewPackage")
>> will produce
>> package ‘GreatNewPackage’ is not available
>> if called, say, from R 2.13 if GreatNewPackage is only in R-devel.
>>
>> To help our less sophisticated users, would it be possible that biocLite
>> advised them of the existence of the package in other than the relevant
>> releases (incl. the devel branch)?

a little concerned about the extra expense of looking up multiple 
repository indexes, and not conveying enough instruction so that it 
sows more confusion than it reaps; maybe pointing to a web page of 
available packages (either current, like 
http://bioconductor.org/packages/release/bioc/ or a new page that says a 
few words about how the R and Bioc releases are coordinated and with 
links / search facilities / instructions for discovering GreatNewPackage?

> In addition to this suggestion:
>
> How about if biocLite warned users if called from an older R, saying
> something like "Warning: Installing from Bioconductor release 2.5. Newer
> versions of the package(s) might be available from the current release,
> Bioconductor 2.8. To use it, update R to version 2.13".
>
> This might reduce the number of questions on the mailing list from
> people using outdated package versions.
>
> The warning might be triggered either by calling biocLite(), or already
> when sourcing biocLite.R.

Yes, prompting lightly with R-2.12, strongly before that, for instance.

Thanks for the suggestions, we'll try and incorporate.

Martin

>
> Simon
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel


-- 
Computational Biology
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N. PO Box 19024 Seattle, WA 98109

Location: M1-B861
Telephone: 206 667-2793