Recent searches
Search options
@albersonomiranda@fosstodon.orgSo I was in the process of my first submission to CRAN and it's been removed (before due date for fixing) because I didn't apologize to prof. Ripley for not testing on Fedora 36 (?).
I didn't realize that the process was that personal. It's embarrassing. #rstats
@albersonomiranda I am sorry about this!
I would ignore this email completely and resubmit the fixed package.
This said, if you want to use Rust code CRAN, brace yourself, it won't be easy. Try to make sure that your package works with a (very?) old rust compiler as well.
@gaborcsardi @albersonomiranda not to mention Fedora 36 is 16 months past EOL....
I'm dealing with a similar issue with my Rust packages. But I'm fortunate to have the understanding Uwe
@gaborcsardi @albersonomiranda might be worth documenting here https://github.com/cranchange/cran_histories/discussions
@josi @gaborcsardi @albersonomiranda Uwe is great. Firm about rules and expectations but still courteous and professional.
@albersonomiranda sorry to hear that. Note also that CRAN is not accepting submissions for two weeks and then there is usually a spike (all current packages with a deadline got delayed until the 18th). I hope you get the package back on CRAN soon you can ask for help in R-package-devel mailing list (also on how to navigate CRAN issues 
).
mmh, 
should I track packages archived before its deadline?
@Lluis_Revilla maybe eventually. Right now I don't feel safe even sending anything to the mailing lists
@albersonomiranda … why do we accept this as a community, again?! I must have missed that memo.
@klmr @albersonomiranda do we accept this? I don't think I've ever bothered to get a package onto CRAN for this exact reason.
@albersonomiranda Seriously, this is so toxic.
@albersonomiranda Ripley is an asshole.
@albersonomiranda Ripley complaining about others' manners is exceedingly ironic
@albersonomiranda ouch, sorry. Please don't give up. If it helps at all I never see BDR on https://stat.ethz.ch/pipermail/r-package-devel/ . There was a whole thread *about* a similar situation to which he never responded: https://stat.ethz.ch/pipermail/r-package-devel/2023q4/009997.html (note however that Martin Mächler had to spend time expurgating the thread ...)
@bbolker just spent some time reading that thread and I don't know if I feel better or worse for not being the only one
@albersonomiranda @bbolker there's even a word for that: you've been Ripleyed... (I swear I don't make that up, I've seen people using this in the bird site).
My first submission was similarly very complicated... I made a "mistake": I replied to his e-mail instead of picking up the cran address somewhere else and using it to reply to the whole cran submission teams (they've improved this since then, now I think a reply-to-all does the job): I got something like "WHY ARE YOU WRITING ME YOU'RE NOT ALLOWED TO WRITE TO ME"... (There was absolutely no way to know that, I'm pretty sure the CRAN wasn't in cc of his e-mail)
@HydrePrever @albersonomiranda @bbolker Per Os Keyes's 2015 blog post https://ironholds.org/dont-use-the-mailing-lists/ , this situation has been a problem within the R ecology since 2011 and probably earlier.
While checking this out I also noticed that https://contributor.r-project.org/rdevguide/ReportingBugs.html#disagreement-with-a-resolution-on-the-bug-tracker says nothing about what to do if someone seriously oversteps.
I see there is an open issue https://github.com/r-devel/rcwg/issues/34 (cc @HeathrTurnr) to create a Code of Conduct for R. Perhaps a productive avenue to formalize expectations for everyone.
@brainwane @albersonomiranda @bbolker @HeathrTurnr the blog note is perfect. I got recently, after roughly 15 years of working with R and maintaining various packages, a "do your homework"...
@brainwane @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
Just to speak to the "... and probably earlier" point, and the etymology of "ripleyed" discussed elsewhere in the thread:
```r
fortunes::fortune("ripleyed")
#>
#> And the fear of getting Ripleyed on the mailing list also makes me think, read,
#> and improve before submitting half baked questions to the list.
#> -- Eric Kort
#> R-help (January 2006)
```
Amazing. Thank you.
So, the influence of this one person's behavior has been skewing #rstats norms (and causing survivorship bias) for at least 18 years.
I'll be the first to admit that I am an outsider to R. I've never written any. But I wish @R_Contributors well in alleviating this problem, because this is the kind of norm that causes me to not only avoid an environment but warn others away as well.
@brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr Oh yeah, he is absolutely a missing stair. I’ve been using R for work since maybe 2011 and probably one of the first things I learned, passed down from someone else whisper-network style, was “avoid the mailing list.”
@emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
Having been "Ripleyed" many years ago, and learned that fortunes::fortune("Ripley") has many useful comments, it is worth checking that the person in question is responsible for a lot of positive features. In particular, R and packages compile on multiple architectures simply, with the same results. For comparison, try compiling RStudio.
Hi. Can I ask for more context on your reply?
We're talking about a desire to address discouraging behavior.
I read your reply as implying a belief that our norms for addressing that behavior should vary based on whether the person has also made useful contributions. And that we should have 2 standards for "basic manners" in an open source project, 1 for people who have made sufficient contributions to it, and 1 for those who haven't.
Please do tell me if I'm wrong!
(I'm quoting the "basic manners" line from the email quoted in https://fosstodon.org/@albersonomiranda/112892922378067813 by @albersonomiranda .)
I recognize that you might be making a different argument than the one I was inferring, and I do want to know if you are! For instance, maybe you thought I needed to be informed of the stature the person holds within R, because that will make it more difficult to persuade people that a norm needs changing.
@brainwane @albersonomiranda @emjonaitis
I was making a different point. Professor Ripley (https://www.stats.ox.ac.uk/~ripley/) was for many years the most prolific contributor to core R code. See http://romainfrancois.blog.free.fr/index.php?post/2009/10/09/celebrating-R-commit-50000 from 2009.
Paul, I recognize that he's been a major and prolific contributor to R for decades; that's not in dispute.
But I'm not clear on why you're bringing that up in response to Erin's and others' observations and experiences.
So, could you please be more specific? Do you think we should use the fact of his positive contributions when evaluating the behavior we've been discussing? If so, how? What standard ought we apply?
@brainwane @albersonomiranda @emjonaitis
I recognise that standards of behaviour in voluntary projects have changed over the years, with codes of conduct. Getting older (and major) contributors to abide by such standards in voluntary projects is not entirely simple. Recognising contributions is necessary as part of this process. I didn't read that in the thread.
Oh that is very helpful - thank you. Now I understand your reasoning. I was working under the assumption that everyone in the thread is already firmly aware of the magnitude of those prior contributions.
@brainwane @PaulBivand @albersonomiranda @emjonaitis
Given their contributions to R, I would see no reason for them to stop continuing to contribute code to R.
The issue is with their interactions to those who would like to also contribute to R, and the lack of respect for others time and efforts in those endeavors.
@rmflight @brainwane @PaulBivand @albersonomiranda @emjonaitis I think there’s no past or future contribution he could make that justifies the community accepting his rude and aggressive communications to folks who also want to contribute to R.
@benschneider @rmflight @brainwane @PaulBivand @albersonomiranda @emjonaitis if having to deal with newbies, human error, basic standards of politeness makes someone so irritable, maybe they need to take a well-deserved retirement
@PaulBivand @albersonomiranda @emjonaitis Thanks for this conversation, which I've now written about at further length:
https://harihareswara.net/posts/2024/double-standards-and-volunteer-behavior/
@PaulBivand @emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
But that's just excusing the person's behavior because of what they've accomplished / contributed.
We really need to stop that shit (I can think of lots of examples in science where we do this).
How many people have been turned away from contributing *anything* to #RStats because one member of CRAN decides on the regular to act like an entitled acerbic twat towards others?
@rmflight @PaulBivand @emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
At that time I feel I am probably the only one that has a good interaction with him. (more than 10 years ago)! Albeit it was not about an R packages but about spatial stats that I did not understood fully (still prob the case 
). Anyway he is prob not on mastodon so we can rant as much as we want if we can't find a proper place to discuss I do not see how it can improve
My suggestions on how to improve the situation https://social.coop/@brainwane/112910212407022869 involve a few potentially good places to discuss it:
https://github.com/r-devel/rcwg/issues/34 and https://github.com/RConsortium/r-repositories-wg . Hope that is helpful!
@rmflight @PaulBivand @emjonaitis @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
@rmflight @PaulBivand @emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
It's not just one person either. It really is a marvel that R has the supportive community it does, when some of the old guard have been actively hostile to newcomers for decades.
https://stat.ethz.ch/pipermail/r-help/2015-December/435044.html
@plantarum @rmflight @PaulBivand @emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @bbolker @HeathrTurnr wow... That's a whole another level of online trolling
@plantarum @rmflight @PaulBivand @emjonaitis @brainwane @yjunechoe @R_Contributors @albersonomiranda @bbolker @HeathrTurnr mmmh yes (superb Usenet vibe in this one), but is Rolf Turner a member of CRAN?
(Anyway that illustrates well the "don't go to the mailing list"... I'm only on R pkg devel. Uwe Ligges is incredibly nice, patient, and helpful, among others.)
@PaulBivand @emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr
As somebody who spent the better part of a decade moderating Mozilla's communities, I can tell for sure that the "but they made a lot of positive contributions" argument is a disaster.
The blast radius of a toxic community leader is huge and mostly invisible. For every complaint you hear there are dozens or hundreds of people who saw what you tolerate and silently moved on.
@emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr this makes me incredibly sad and upset
@emjonaitis @brainwane @yjunechoe @R_Contributors @HydrePrever @albersonomiranda @bbolker @HeathrTurnr Another important reason to avoid the lists is that the archives are forever. Modern forms of communications typically have some sort of GDPR button that wipes your presence off the slate. The only way to wipe you off the mailing list archives is the owner going through them semi-manually. Your enemies will be looking through them years from now to stock up on ammo against you.
@HydrePrever @albersonomiranda @bbolker the fact that we can all guess who sent that with such little context really says something.
@HydrePrever @albersonomiranda @bbolker
I had a similar incident with Prof R. He sent me an email from his academic email account (I assume by accident) to which I replied to his question. His response was a huge rant about me sending a html email to CRAN. After I pointed out I was replying to his personal email address he was pretty rude and never addressed the issue which was the point of his first email.
Because of things like this I now don't contribute my free time to R.
@HydrePrever @albersonomiranda @bbolker
fortunes:: fortune("ripleyed")
@albersonomiranda @bbolker that thread made me sad. There were plenty of complaints about the manner in which the poster made their complaint, but pretty much nothing addressing the substance.
I'm sorry you're experiencing this. I'm extra sorry that some members of our community think it's OK to have to deal with this sort of behavior.
CRAN is valuable. By having your tested package on CRAN, you're helping all future versions of #RStats. We need to be better about this as a community!
@jonthegeek
Stories like this and other rumors about CRAN and the toxic submission process is why I don't think I would ever submit a package there.
Sorry (?)
@MrHedmad @albersonomiranda @bbolker it sucks that this happens, but, in my experience, it isn't the norm. I understand why it drives people away, though. I don't know how to fix it, but I'd like to find a way.
Acknowledging that it's a problem is the first step, though, so thank you for voicing your opinion! Hopefully if it becomes clear that his attitude is a real problem, something can be done.
@jonthegeek @MrHedmad @albersonomiranda @bbolker there is 0 transparency in CRAN. In Julia, Python, and Rust packaging ecosystem you can see it all on github. You can make a discussion that other people can participate in. In our community we accept being talked down on in private.
First step would be to make the submission process actually transparent imo.
Then we can address forcing passing tests on an OS that is past its end of life....among many other qualms
Based on my experience with other package management platforms and other similar situations among open source teams, may I share some free advice?
Norms for acceptable behavior need to change, and the package management system needs better financial and digital infrastructure so that the platform is not dependent on overworked volunteers.
Steps toward improving norms: https://github.com/r-devel/rcwg/issues/34 approve a Code of Conduct, and have private conversations with respected senior figures to work out whether they also want improvement.
Steps toward improving CRAN's digital/financial infra:...
... skim the minutes and the issues at https://github.com/RConsortium/r-repositories-wg and get an idea of how receptive CRAN is to offers of help. I bet it's possible to get directed funding toward a support specialist like the one PyPI recently hired https://pyfound.blogspot.com/2024/03/announcing-pypi-support-specialist.html , as well as workflow and codebase improvements to CRAN itself.
A very useful 2016 piece by @christi3k discusses "Misalignment of shared values and the painful process of re-alignment" of group norms in open source. "If the shared values that were implicit when a community forms are not made explicit as the community grows, you end up with divergent thinking about what the shared community values are."
Thus my "private conversations with respected senior figures" recommendation.
@albersonomiranda hard to imagine why the R community has a bad reputation amongst muggles.
@albersonomiranda I wasted a bit of time of CRAN volunteers as well on my first submission. I also got a very unwelcoming reply. Since then I've been able to maintain several pkgs there.
Part of CRAN maintainers responsibility is to deal with humans who submit packages to CRAN. I do not understand why someone would accept being a CRAN volunteer without accepting some need to care towards newcomers who don't behave as expected because they/we do not know better.
@zeehio @albersonomiranda @adamhsparks How does one actually go about becoming a CRAN volunteer? I see the discontent (and have felt it myself) but maybe this is a “be the change you want to see in the world” situation.
A big part of the value of CRAN is in the human review, which means actual humans, flawed like the rest of us, are involved.
@AlexAxthelm @zeehio @albersonomiranda @adamhsparks I would like to know too
@AlexAxthelm @zeehio @albersonomiranda @adamhsparks hahahahaha great question-- they don't take volunteers! There's a working group trying to change that, but it's several years in and I don't know if there's been any movement.