Hello everybody! I’m excited to participate in my second Summer Workshop (last year was my first). I’m looking forward to catching up with those of you I met last year and others whose paths I’m finally crossing with again (for example, Deb Linton…Hi Deb!!), and getting to know everyone else. So much expertise and awesome projects being described in this forum!
About me: I’m an assistant professor of biology at SUNY Geneseo, in western New York state–a teaching-focused, primarily undergraduate institution. I’m a community ecologist by training, and also consider myself to have an education specialty. In addition to introductory biology lab and lecture courses, I teach a majors-required ecology course and an elective biostatistics course. Students learn to use R in both of the latter courses. Like Derek Sollberger and others, I’ve been teaching them via RStudio and R Markdown.
Some recent specific quant bio activities/projects of mine (all initiated by attending the 2018 Summer Workshop!):
A new stats curriculum called Cats Teach Stats. The philosophy of this curriculum is to use cute cat videos and other cat-related data from the Internet to reduce the anxiety associated with doing mathematics and statistics. We think it’s not just for cat people. I’d be glad to share links to our resources if anyone’s interested and/or interested in working with us catisticians.
This past spring, I led the Make Teaching R in Undergraduate Biology Less Excruciating (Make TRUBLE) FMN, in which participants developed modules in R with Swirl. It was super fun!
One of my general goals for participating in next week’s workshop is to learn lots about teaching Data Science, because my institution is working on creating an interdisciplinary minor for it.
Seeing a lot of people mentioning R Studio. It's a great tool for many
reasons, which makes it all the more frustrating that I'm excluded from
using it. At last check several months ago, R studio was not accessible to
assistive technologies like screen readers (text-to-speech output). Second
hand information suggests that the R Studio powers that be are rather
unresponsive and uninterested in this problem. I'd be very interested in
talking with folks about possible alternatives or building a coalition of
users to force some movement on this.
I know this is just an accessibility issue but given the wide use of the
software, it's big. It is a barrier to engaging students who use AT in data
science, and it's a barrier to me adopting all the cool things you all do
with it in the future. I'd love to chat with people.
On Wed, Jul 10, 2019 at 12:41 PM Suann Yang @ QUBES
wrote:
---- Emailed forum response from aohasley@gmail.com
Did you all have a project space or group on QUBES you are using to share
your Cats Teach Stats stuff? I can't wait to see what you've done with it
since last time!
On Wed, Jul 10, 2019 at 1:41 PM Suann Yang @ QUBES
wrote:
---- Emailed forum response from mathprofcarrie@gmail.com
Happy to talk accessibility and R! The person I know best in this space is Jonathan Godfrey. Here’s a link to one paper of his https://journal.r-project.org/archive/2013-1/godfrey.pdf. As far as I understand, you’re right that RStudio is not accessible to screen readers. However, Godfrey still suggests that R is the best statistical software for blind and low vision folks. He has a more general website on this topic https://r-resources.massey.ac.nz/StatSoftware/ I am also very well acquainted with the RStudio developers, so we could talk about what features might be useful for people who use screen readers. My sense is that many of the features that make people like RStudio are visual affordances.
Looking forward to meeting you in person in a few days! I asked my lab about this question, and I wanted to share some of the responses I got in case they are helpful to anyone:
1) There is an update from an RStudio employee about a month ago stating "The RStudio IDE itself is not yet accessible via screen readers. We know this is unacceptable, and I am actively working on improving that for the next major release (1.3) of the RStudio IDE, but unfortunately that will not be ready this fall; more likely early in 2020."
2) Apparently Visual Studio Code can be set up to run R code similarly to RStudio, and they have thought a fair amount about accessibility, including screen readers
On the RStudio front (mainly reiterating what Amelia and Ellen said), I had the opportunity to talk with the RStudio points of contact yesterday since they are here at the useR conference, and they said that they are still working on those accessibility issues. The cynicism is that different colleges and different businesses may have different definitions of "accessibility". Perhaps if we form a clear proposal of what is needed at colleges, it would streamline how RStudio should proceeed.
The responses for this thread have me even more excited to interact with all of you over the next week! I'm really glad to hear that accessibility is on the radar for the IDE. The comment regarding Visual Studio Code is a direction that has never occurred to me. That's a great example of the value of throwing things out there to the QUBES and BIOQUEST communities and seeing woh's expertise can help. It's also a great example of things we as instructors can do to prepare ourselves for students who might not be able to use our favorite tool when we are teaching.
I'm also very appreciative of the references people are sharing and because assembling resources is a goal of a couple things I'm working on with several folks, keep 'em coming.
I totally agree that the R language itself can be extremely accessible once you find an environment you can work in. There are even packages for things like producing tactile output files (brailleR) and a few implementations of sonification of figures. The language and the IDE should be discussed somewhat separately in this context. For example, I used plenty of R in my graduate work. However, I used it primarily via the the stock windows GUI that comes with it. That's fairly accessible. However, you don't get any of the lovely functionality that comes with R Studio. That includes things like Github compatibility built-in, practically automatic documentation, etc. So, I had to do things like track all my code (and notes on why I did what I did) manually in Word. If you don't think that made my work less efficient, we should chat. So, while I agree that a lot of R Studio's popularity is driven by visual affordances (excited to see that word come in here!), there's a great deal of underlying functionality that I'm missing out on too.
Lastly, I'm a bit dismayed to hear that there's some cynicism in the R Studio dev camp about not being able to please everyone's definition of accessibility. Yes, different groups may have some different wants, but there is one hell of a big common denominator that's still not being met. There are good guidelines out there for building desktop apps that work with various assistive technologies. There are also good design guides out there for things that just make a GUI good for everyone that go beyond technical things like APIs, object attributes, and so on. So, when I hear the complaint that there's just no consensus, I'm a bit unimpressed. There's definitely some. To be clear, I'm not disparaging the developers themselves. I just have a problem with the culture that put them in the situation they're now in, playing catch up and trying to figure out what accessibility, usability, and good design all mean, after the product is well established already.
So, thanks to everyone. I think this can be a really fruitful conversation and we should definitely keep it going at the workshop. I'd love to get a sense of what accessibility might mean for those using it for education, as well as ways to make curricula that rely on R accessible to those who can't use R studio for whatever reason.
Suann Yang @ on
Hello everybody! I’m excited to participate in my second Summer Workshop (last year was my first). I’m looking forward to catching up with those of you I met last year and others whose paths I’m finally crossing with again (for example, Deb Linton…Hi Deb!!), and getting to know everyone else. So much expertise and awesome projects being described in this forum!
About me: I’m an assistant professor of biology at SUNY Geneseo, in western New York state–a teaching-focused, primarily undergraduate institution. I’m a community ecologist by training, and also consider myself to have an education specialty. In addition to introductory biology lab and lecture courses, I teach a majors-required ecology course and an elective biostatistics course. Students learn to use R in both of the latter courses. Like Derek Sollberger and others, I’ve been teaching them via RStudio and R Markdown.
Some recent specific quant bio activities/projects of mine (all initiated by attending the 2018 Summer Workshop!):
One of my general goals for participating in next week’s workshop is to learn lots about teaching Data Science, because my institution is working on creating an interdisciplinary minor for it.
See you all on Sunday!
Report abuse
Andrew Osborne Hasley @ on — Edited @ @ on
Report abuse
Carrie Diaz Eaton @ on — Edited @ @ on
Report abuse
Amelia McNamara @ on
Happy to talk accessibility and R! The person I know best in this space is Jonathan Godfrey. Here’s a link to one paper of his https://journal.r-project.org/archive/2013-1/godfrey.pdf. As far as I understand, you’re right that RStudio is not accessible to screen readers. However, Godfrey still suggests that R is the best statistical software for blind and low vision folks. He has a more general website on this topic https://r-resources.massey.ac.nz/StatSoftware/ I am also very well acquainted with the RStudio developers, so we could talk about what features might be useful for people who use screen readers. My sense is that many of the features that make people like RStudio are visual affordances.
Report abuse
Suann Yang @ on
Hi Carrie! Yes, we have a couple of places to see where Cats Teach Stats is at:
Report abuse
Ellen Bledsoe @ on
Hi Drew,
Looking forward to meeting you in person in a few days! I asked my lab about this question, and I wanted to share some of the responses I got in case they are helpful to anyone:
1) There is an update from an RStudio employee about a month ago stating "The RStudio IDE itself is not yet accessible via screen readers. We know this is unacceptable, and I am actively working on improving that for the next major release (1.3) of the RStudio IDE, but unfortunately that will not be ready this fall; more likely early in 2020."
https://community.rstudio.com/t/accessibility-of-r-with-screen-reading/33260/4
2) Apparently Visual Studio Code can be set up to run R code similarly to RStudio, and they have thought a fair amount about accessibility, including screen readers
https://code.visualstudio.com/docs/editor/accessibility
Also, thanks for the other references, Amelia!
Report abuse
Derek Sollberger @ on
On the RStudio front (mainly reiterating what Amelia and Ellen said), I had the opportunity to talk with the RStudio points of contact yesterday since they are here at the useR conference, and they said that they are still working on those accessibility issues. The cynicism is that different colleges and different businesses may have different definitions of "accessibility". Perhaps if we form a clear proposal of what is needed at colleges, it would streamline how RStudio should proceeed.
Report abuse
Andrew Osborne Hasley @ on
The responses for this thread have me even more excited to interact with all of you over the next week! I'm really glad to hear that accessibility is on the radar for the IDE. The comment regarding Visual Studio Code is a direction that has never occurred to me. That's a great example of the value of throwing things out there to the QUBES and BIOQUEST communities and seeing woh's expertise can help. It's also a great example of things we as instructors can do to prepare ourselves for students who might not be able to use our favorite tool when we are teaching.
I'm also very appreciative of the references people are sharing and because assembling resources is a goal of a couple things I'm working on with several folks, keep 'em coming.
I totally agree that the R language itself can be extremely accessible once you find an environment you can work in. There are even packages for things like producing tactile output files (brailleR) and a few implementations of sonification of figures. The language and the IDE should be discussed somewhat separately in this context. For example, I used plenty of R in my graduate work. However, I used it primarily via the the stock windows GUI that comes with it. That's fairly accessible. However, you don't get any of the lovely functionality that comes with R Studio. That includes things like Github compatibility built-in, practically automatic documentation, etc. So, I had to do things like track all my code (and notes on why I did what I did) manually in Word. If you don't think that made my work less efficient, we should chat. So, while I agree that a lot of R Studio's popularity is driven by visual affordances (excited to see that word come in here!), there's a great deal of underlying functionality that I'm missing out on too.
Lastly, I'm a bit dismayed to hear that there's some cynicism in the R Studio dev camp about not being able to please everyone's definition of accessibility. Yes, different groups may have some different wants, but there is one hell of a big common denominator that's still not being met. There are good guidelines out there for building desktop apps that work with various assistive technologies. There are also good design guides out there for things that just make a GUI good for everyone that go beyond technical things like APIs, object attributes, and so on. So, when I hear the complaint that there's just no consensus, I'm a bit unimpressed. There's definitely some. To be clear, I'm not disparaging the developers themselves. I just have a problem with the culture that put them in the situation they're now in, playing catch up and trying to figure out what accessibility, usability, and good design all mean, after the product is well established already.
So, thanks to everyone. I think this can be a really fruitful conversation and we should definitely keep it going at the workshop. I'd love to get a sense of what accessibility might mean for those using it for education, as well as ways to make curricula that rely on R accessible to those who can't use R studio for whatever reason.
Report abuse
Pat Marsteller @ on
Might want to contact Sarah A. Orlofske
Department of Biology
Northeastern Illinois University
She developed a food web cases with some R components.
Report abuse