[R] The steps of building library in R 2.1.1

[Duncan Murdoch]

On 7/21/2005 12:03 PM, Gabor Grothendieck wrote:

>> 
>> > 2. there is too much material to absorb just to create a package.
>> >     The manuals are insufficient.
>> 
>> The first sentence here is basically a repetition of "the process is too
>> complex".  I think the second sentence is incorrect.  Could you please
>> point out what necessary steps are missing?
> 
> Its not that anything is missing that I am aware of.  Its that there is so much 
> detail one is overwhelmed.  Its not completely the fault of the description 
> since as point #1 mentions the process itself is a key part of the 
> problem.

And what solution do you propose to this problem?  Saying this ought to
be a high priority for the core group is not a solution.  Tell us where
the resources will come from to do this.

>> > A step-by-step simplification is very much needed.
>> 
>> Exactly this has been in the Installation and Administration manual
>> since I put it there in February for the 2.1.0 release.  It's at the
>> beginning of the appendix on the Windows toolset, with multiple
>> references pointing people there.  It's followed by detailed
>> descriptions of each of the steps.
>> 
>> If you think it could be further improved, please submit improvements.
> 
> That is easy to say but, in fact, if anyone does this they are not met
> with a receptive atmosphere.  The excellent post describing the process
> that started this out (even if there are some small errors) is just one 
> example.  

I don't think that post was written with the intention of putting it
into the manual.  It would still take a fair bit of work to do that:

 1. deciding where it fits and what to replace,
 2. correcting the errors,
 3. writing it in texinfo format.

I'd be happy to talk with someone who volunteers to do that.  (I'd
suggest the volunteer should do number 1 first, so as not to waste a lot
of time on versions that don't fit.)

Duncan Murdoch