[R] (Off-Topic) Time for a companion mailing list for R packages?

Stephen H. Dawson, DSL @erv|ce @end|ng |rom @hd@w@on@com
Thu Jan 13 15:50:51 CET 2022 

Good comments, Ivan.

I also have found StackOverflow to not be too helpful in a structured 
conversation. I have had it and miss success with GitHub interactions.


*Stephen Dawson, DSL*
/Executive Strategy Consultant/
Business & Technology
+1 (865) 804-3454
http://www.shdawson.com <http://www.shdawson.com>


On 1/13/22 9:01 AM, Ivan Calandra wrote:
> This is indeed a very valid question.
>
>
> IMHO, I still think that these kind of package-specific questions should have a space on the main R mailing list, as long as these questions are about how to do something (coding), and not bug or feature requests. Regarding tidyverse and ggplot2 (and the likes), many answers actually provide ways using different packages (how many times have I seen the same question answered with base R, data.table and dplyr?). This is a great way to learn.
>
> And honestly, how many only use base R? Of course, everything can be done in base R, but packages also belong to R, that's the way it works and why it works so well. Reinventing the wheel is rarely a good solution.
>
>
> Regarding other help platforms, GitHub is great for the package maintainers that spend time on solving issues, but I myself feel bad to ask there "how to do that" (help is not always good).
>
> I gave up on StackOverflow. Everytime I tried, I either got treated as an idiot by persons who didn't even read my questions (with answers like "do it" to questions like "I don't know how to do it"), or I couldn't comment on my question to make it clearer because I was new (which is ridiculous, it is when you are new that you need to be able to clarify the most!).
>
> If people think it's tough to ask a question on the R mailing list because beginners are not well-handled, they should try SO, it's much worse!
>
> This is to say we should keep the R mailing list!
>
>
> My $0.02
>
> Ivan
>
> --
> Dr. Ivan Calandra
> Imaging lab
> RGZM - MONREPOS Archaeological Research Centre
> Schloss Monrepos
> 56567 Neuwied, Germany
> +49 (0) 2631 9772-243
> https://www.researchgate.net/profile/Ivan_Calandra
>
>
> ________________________________
> From: R-help <r-help-bounces using r-project.org> on behalf of Duncan Murdoch <murdoch.duncan using gmail.com>
> Sent: Thursday, January 13, 2022 2:25 PM
> To: Kevin Thorpe; Jeff Newmiller
> Cc: R Help Mailing List
> Subject: Re: [R] (Off-Topic) Time for a companion mailing list for R packages?
>
> Currently help for contributed packages is available on StackOverflow,
> package-specific web sites and Github.
>
> I rarely read package-specific (e.g. RStudio) web pages, and have only
> posted questions there a few times, with generally unsatisfactory results.
>
> Most package developers (including tidyverse ones) respond very
> helpfully on Github, but to post there you need to already have a very
> good idea of where the problem lies.  It's not an appropriate place for
> beginners to ask for help.
>
> That leaves StackOverflow.  It gets way too much traffic from people who
> don't pose their questions well, but it has the advantage over a mailing
> list that questions can be edited and improved (or deleted) after the
> fact.  We should be sending beginners there.  And if you want to read
> questions and help people, apply N95 filters to the questions:  e.g. I
> mainly read questions that have been unanswered and undeleted for an
> hour or more.
>
> Duncan Murdoch
>
> On 13/01/2022 7:44 a.m., Kevin Thorpe wrote:
>> This is an interesting issue and something I have been thinking about raising with my fellow volunteer moderators.
>>
>> I honestly don�t know what the best solution is. Personally, I would loathe having to check multiple web-forums/mailing lists to find an answer. New users often do not appreciate the subtleties (i.e. RStudio is not R) and will continue to post here. The frequent reply to questions outside base R that inform them they are off-topic could come across as unfriendly. That could have the side effect of making the community appear elitist. Folks are also often referred to package maintainers but not all maintainers are equally responsive to queries about their packages. In summary, it can be very hard for novice users to get the help they need.
>>
>> I appreciate the desire of many to keep the focus of this list narrow, yet despite the narrow mandate there are many readers who can answer non-base R questions, which is probably one of the reasons we see the questions. I wonder if there would be an appetite to create a new list, R-package-help, that has a broad mandate (as suggested by Avi). Naturally there is no guarantee that specific questions about some esoteric package will be answered, but that�s a different problem. On the other hand, why not expand the mandate of R-help rather than going to the trouble of creating a new list? Like I said, I don�t know.
>>
>> Thanks for raising the issue.
>>
>> Kevin
>>
>>
>>> On Jan 12, 2022, at 11:24 PM, Jeff Newmiller <jdnewmil using dcn.davis.ca.us> wrote:
>>>
>>> TL;DR The people responsible for tidyverse don't think much of mailing lists.
>>>
>>> IANAMLA (I am not mailing list admin) and I know some people get kind of heated about these things, but my take is that this list _is_ about R so to be on topic the question needs to be about R and how to get things done in R. Since contributed packages are almost by definition creating capabilities linked with specific problem domains or domain-specific-languages (DSLs), and there are thousands of these, it isn't practical to support questions framed within those DSLs here. It seems perfectly legitimate IMHO to mention such packages here, as long as the question does not hinge on that package, and even to offer small solutions to posed R problems using such packages. Others may disagree with my perspective on this. Unfortunately all of this this subtlety is usually lost upon newbies, much to the detriment of this list's reputation.
>>>
>>> The responsibility to setup and manage support for contributed packages belongs to the package maintainer. In the case of tidyverse, the general opinion of those people seems to be that web forums avoid the "only unformatted info can be shared" nature of traditional mailing lists, so mailing lists have AFAIK not been built or tended.
>>>
>>> Unfortunately, they also try to "allow all topics" as much as possible in those forums to minimize the appearance of unfriendliness to beginners, but my impression is that this leads to such a wide range of topics that many posts don't get answered. I have certainly found it to be just too much quantity to sift through, and I really am selective about which portions of the tidyverse I work with anyway, so I don't hang out there much at all.
>>>
>>> On January 12, 2022 7:27:20 PM PST, Avi Gross via R-help <r-help using r-project.org> wrote:
>>>> Respectfully, this forum gets lots of questions that include non-base R components and especially packages in the tidyverse. Like it or not, the extended R language is far more useful and interesting for many people and especially those who do not wish to constantly reinvent the wheel.
>>>> And repeatedly, we get people reminding (and sometimes chiding) others for daring to post questions or supply answers on what they see as a pure R list. They have a point.
>>>> Yes, there are other places (many not being mailing lists like this one) where we can direct the questions but why can't there be an official mailing list alongside this one specifically focused on helping or just discussing R issues related partially to the use of packages. I don't mean for people making a package to share, just users who may be searching for an appropriate package or using a common package, especially the ones in the tidyverse that are NOT GOING AWAY just because some purists ...
>>>> I prefer a diverse set of ways to do things and base R is NOT enough for me, nor frankly is R with all packages included as I find other languages suit my needs at times for doing various things. If this group is for purists, fine. Can we have another for the rest of us? Live and let live.
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: Duncan Murdoch <murdoch.duncan using gmail.com>
>>>> To: Kai Yang <yangkai9999 using yahoo.com>; R-help Mailing List <r-help using r-project.org>
>>>> Sent: Wed, Jan 12, 2022 3:22 pm
>>>> Subject: Re: [R] how to find the table in R studio
>>>>
>>>> On 12/01/2022 3:07 p.m., Kai Yang via R-help wrote:
>>>>> Hi all,
>>>>> I created a function in R. It will be generate a table "temp". I can view it in R studio, but I cannot find it on the top right window in R studio. Can someone tell me how to find it in there? Same thing for f_table.
>>>>> Thank you,
>>>>> Kai
>>>>> library(tidyverse)
>>>>>
>>>>> f1 <- function(indata , subgrp1){
>>>>>       subgrp1 <- enquo(subgrp1)
>>>>>       indata0 <- indata
>>>>>       temp    <- indata0 %>% select(!!subgrp1) %>% arrange(!!subgrp1) %>%
>>>>>         group_by(!!subgrp1) %>%
>>>>>         mutate(numbering =row_number(), max=max(numbering))
>>>>>       view(temp)
>>>>>       f_table <- table(temp$Species)
>>>>>       view(f_table)
>>>>> }
>>>>>
>>>>> f1(iris, Species)
>>>>>
>>>> Someone is sure to point out that this isn't an RStudio support list,
>>>> but your issue is with R, not with RStudio.  You created the table in
>>>> f1, but you never returned it.  The variable f_table is local to the
>>>> function.  You'd need the following code to do what you want:
>>>>
>>>> f1 <- function(indata , subgrp1){
>>>>     subgrp1 <- enquo(subgrp1)
>>>>     indata0 <- indata
>>>>     temp    <- indata0 %>% select(!!subgrp1) %>% arrange(!!subgrp1) %>%
>>>>       group_by(!!subgrp1) %>%
>>>>       mutate(numbering =row_number(), max=max(numbering))
>>>>     view(temp)
>>>>     f_table <- table(temp$Species)
>>>>     view(f_table)
>>>>     f_table
>>>> }
>>>>
>>>> f_table <- f1(iris, Species)
>>>>
>>>> It's not so easy to also make temp available.  You can do it with
>>>> assign(), but I think you'd be better off splitting f1 into two
>>>> functions, one to create temp, and one to create f_table.
>>>>
>>>> Duncan Murdoch
>>>>
>>>> ______________________________________________
>>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>>> and provide commented, minimal, self-contained, reproducible code.
>>>>
>>>>       [[alternative HTML version deleted]]
>>>>
>>>> ______________________________________________
>>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>>> and provide commented, minimal, self-contained, reproducible code.
>>> --
>>> Sent from my phone. Please excuse my brevity.
>>>
>>> ______________________________________________
>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
> 	[[alternative HTML version deleted]]
>
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> 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 mailing list