Voting resources, early voting, and poll worker information - VOTE. ... Adafruit is open and shipping.
0

Volunteering in the COVID-19 era
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Volunteering in the COVID-19 era

by flounder on Sun Sep 06, 2020 12:03 pm

I have been a volunteer tutor for the last decade in our local school system. I had to stop last year because my age (73) is just one of my five COVID-19 comorbidities. So I began to think of what I can do to allow me to resume volunteering once schools are back in session come January of 2021.

I tutor 3rd and 4th grade math. I need to see what the student is writing, or what the visual question is in the workbook, or their homework, or their exam. So a simple Skype link is not going to be adequate.

The connection must be fully automated and not require any action on the part of the student. There is only so much you can expect from a nine-year-old, and the technology has to be fully under my control from home.

Right now, I am planning a ”research prototype”, which will be far more elaborate than the ”production” version.

There are some simple problems. For example, a webcam and microphone. A screen. A speaker. This is nominally very easy. But I don't want to require a $500 laptop to make it real. What I really want is a RasPi with its webcam, an HDMI screen, and a network connection. Well, that sounds simple enough.

Now things get interesting. My plan is to include a RasPi Hi-Q camera looking down on the desk top (no, not the “desktop”, I mean the real surface of a real desk). Probably with an area conveniently taped off, and all the student needs to know is that for pages to be visible they need to be within that area. The overhead cam will be a fixed-focal-length lens and the focus will be on the desktop. And, ideally, the depth-of-field is deep enough that a 1/2” thick textbook is still in focus. Where is the best placement? Is it different for left-handed and right-handed students? I don’t know. So I want the research prototype to have a 24”x24” xy platform whose position I can control remotely. If I can identify the unique “sweet spot” for the camera, then the production model will not need an xy platform. If there are two, only a single axis will be needed.

Zoom will also be available. This may be needed if I need to watch closely what the student is doing. Ideally, there may be a single optimal zoom setting, and the “production” model can use a fixed-focal-length lens, which may require a change in the height of the camera to get the right coverage. Of course, in the prototype, if I have xy motion I will need pan/tilt control to keep the work area visible. Even if I need a single-axis control on the production model, there will (I hope) be only two “sweet spots”, and the tilt would be the same, and a simple mechanical “pan” can be easily implemented at the two endpoints.

I plan to implement “zoom” by a rubber wheel friction-coupled to the zoom ring (so it doesn’t try to rip it off or unscrew the lens at the endpoints. So I now have a 5-axis camera: x/y/p/t/z. I will probably throw in focus if necessary, particularly if I need to refocus on a tight zoom. So now I have x/y/p/t/z/f. The product ideally will have none of these, or just a single axis, call it ”x”. Depending on what I discover, I may end up with x/z/f, or x/z or z/f, or maybe just z (z being zoom). If I had all the answers, I wouldn’t need a prototype.

I need to switch the image from the webcam to the picam, but keep the microphone active.

I can use steppers considerably smaller than NEMA-17.

I do not foresee any technical problems attaching six stepper control hats to the RasPi. And I see no problems sending G-code-like commands to the RasPi. I can write a parser in my sleep. I am starting my 58th year as a programmer, and have a PhD in Computer Science from Carnegie Mellon University.

Now here's where things start getting hard.

The obvious connection would be wifi. Which is certainly how I will test it. But I have been advised that since I am neither student nor employee, my chances of getting onto the school district’s wifi are close to zero. And, in addition to the technical, legal, and political problems of getting anyone even to talk to me, the IT czar for the school system resigned at the end of the spring term, so even in the unlikely chance that the school board might grant permission, there is nobody available to make it happen.

So I am looking at Fona as a solution. This raises questions I do not know how to answer. For example, is the 3G version going to give enough bandwidth to support bidirectional video? If not, what frame rate might I hope for? Will it be tolerable, or even useful? Do I really need an AT&T connection, or can I use my T-Mobile account? How do I call the remote phone, given that I know its number? Do I need a Fona at my end, can I use my iPhone’s personal hotspot? These are all beyond my experience.

I plan to write the RasPi code in CircuitPython, and write the tutor-host code in either Java or Python so it can run on Windows, MacOS or Linux, and maybe we can host the tutor side on a RasPi as well. I want mine running on Windows or macOS, since I have dual-monitor support on both, and both monitors on Windows are 31” 1920x1080, and I can use one of them as the second Mac monitor.

My ultimate goal is to produce a high-quality, inexpensive tutoring workstation. I plan to release it under CC-BY-SA-NC, so that anyone can build one. The goal is to present this to the local volunteer support groups, including our local chapter of Remake Learning, The Mentoring Partnership of Southwest Pennsylvania, local contacts for mentoring.org, and the like, and say that they can go to mid-rollers and say, “Look at the good you can do here. For a mere $N00” (and my hope is that I can keep N <= 5), “You can get one of these tutoring stations in a school that needs it. Not only do you get a tax writeoff, you help students in need, you keep mentors safe, and right here is where we put the plaque that says, ‘This tutoring station was made possible by a grant from Your Name Here’.” In addition, these organizations can lobby local school boards to allow and supply WiFi access, reducing the cost even further. And ideally they can round up geeky volunteers to put the stations together, and rope makerspaces into helping, because I do not want to manufacture these things.

This is more than one person can do. So what I need is to collect some like-minded people to join me and help write some code, help me with the technical issues, and ideally form an Adafru.it/discord discussion group.

flounder
 
Posts: 406
Joined: Wed Sep 18, 2013 9:10 pm
Location: Pittsburgh PA

Re: Volunteering in the COVID-19 era

by kevinjwalters on Mon Sep 07, 2020 2:33 pm

It's an interesting idea. I've had a single brief conversation with a fellow volunteer about remote volunteering for 11-18 age group. This currently hasn't gone anywhere partly because this is a very new solution for over-loaded schools. We didn't get to mulling over the software/hardware needed for this. The issue of visibility of workbooks and writing does present some challenges.

On the non-hardware side of things, both of us concluded that a likely and reasonable requirement here is to ensure the overall service has the ability to securely and robustly log the audio/video (including any screen sharing) for review as a safeguarding feature. Having said that, if the solution can genuinely only be used from school premsies then this might not be a requirement.

Peronsally, I gravitate towards headsets for audio to deal with feedback and clarity, and dual earpiece ones for noisy environments. This may not work so well now on the student end if the devices are shared.

kevinjwalters
 
Posts: 733
Joined: Sun Oct 01, 2017 3:15 pm

Re: Volunteering in the COVID-19 era

by flounder on Tue Sep 08, 2020 2:46 am

It would not be used "from school premises" by me; however, it would be used (in my image) only on school premises by the students. I would hope that the school would undertake whatever precautions are necessary to reduce infection, and I suspect that headphones would not really work in that context. My goal is to start prototyping in the next week or two, but probably not get it done in time for January. My life is already full. However, I would be happy to co-opt you onto the project, if only for someone to bounce ideas off of.

flounder
 
Posts: 406
Joined: Wed Sep 18, 2013 9:10 pm
Location: Pittsburgh PA

Re: Volunteering in the COVID-19 era

by kevinjwalters on Wed Sep 09, 2020 10:42 am

I can certainly give feedback and ideas. In normal times, I would also be doing code clubs with 8-10 year olds too. All of the volunteering I've done has been small groups of 2-4 children rather than 1:1 and the conditions vary from quiet to very rowdy and sparse to densely packed in.

I also do some seasonal volunteering. That has a professional manager who surveys the sites and we stick to commodity hardware for that as that service is relatively critical and needs to work out of the box with an installation over a limited number of days. 4G is used as second choice/backup for Internet provision but reception is checked for viability and best spot in the site surveys.

I've looked briefly at 3G a long time ago (pre-4G) and my (unreliable) recollection was substantial packet loss and latency all over the place. It reminded me of a return to dial-up. This suggests some early feasability tests on multiple sites would be useful.

A lot of the rooms in schools we go into do not have mains power at the desks we sit at, btw. The school's computers tend to be placed around the edges of the room with occasional desks protruding into the room a bit, i.e. cabling is running under those desks. There's one library location which might have power boxes in the floor but I don't think it does. I use battery power if I take my laptop for demos. The exception for power at desk would be the science labs but we are rarely in those.

kevinjwalters
 
Posts: 733
Joined: Sun Oct 01, 2017 3:15 pm

Re: Volunteering in the COVID-19 era

by flounder on Sat Sep 12, 2020 12:46 pm

Every room I have tutored in has multiple outlets and I only need one desk surface, not a "student desk" but a regular office desk or other flat table, and every room I have tutored in has a suitable flat table somewhere in the room, usually near an outlet, so this is not the problem. The problem is going to be bandwidth.

flounder
 
Posts: 406
Joined: Wed Sep 18, 2013 9:10 pm
Location: Pittsburgh PA

Please be positive and constructive with your questions and comments.