GOSH 2022 SESSIONS Documentation - THURSDAY Block # 2 (11:00 am)

Gatherings GOSH 2022
1

GOSH 2022 SESSIONS

*Note: As the body of each Forum Topic is limited to 32000 characters, the post of the sessions has been split in different Topics according to the Blocks originally defined in the program. This way it is better distributed and easier to manage.

Block # 2 (Thursday / Jueves, 11:00 am)

Environmental monitoring and conservation

Date: 2022-10-27
Time: 11:00
Place: outside
Facilitator: Shannon Dosemagen, Alison Parker
Notetaker: Katie Hoeberling

Participants

  • Andres, open climate
  • Luis Felipe
  • Mara, synthetic bio
  • Isaac, synthetic bio
  • Amanda, wildlife tracking
  • Lenin, enviro health (PM in Lima)
  • Guillermina, cit sci./enviro sci.
  • Lara, local organizing
  • David, org that funds scientists, underwater robots
  • Rachel, water monitoring
  • Shannon, water quality research, EnviroDIY
  • Elio, bioengineering, insects
  • Martin, open data, civic tech, water quality, evidence → courts
  • Valeria, teach science prototypes
  • Diego, biotech hardware
  • Clint, high school students
  • Gaurav, synthetic bio
  • Laura, visual artist, intersections b/n art + molecular bio
  • Josh, conservation, camera traps
  • #22 people#

Notes:
Look at themes from roundtable on OSH + low-cost tools in environmental monitoring
What is happening outside the US?
Choose 3-4 key messages, work through them in smaller groups
Building trust b/n communities and gov’t, and tools and their data
Interconnected resources
Coming together for tool evaluation standards
Trust
Will communities, lawyers trust devices enough? Gran Morelos canal, Mexico to connect airports and create industry hub, using Raspberry Pi to collect data, Martin is part of group creating dashboard
Using water from river to cool gas, heating up rover
turbidity, heat, pH, conductivity
Water quality; jazz festival effect → more people, more pollution in Lake Geneva
open data related to PM in Lima
When you choose a low-cost sensor, try to export the data, can’t compare data from other parts of the world
Use Chinese sensors, European sensors; different platforms; standards in Peru
water quality research lab, works w/ fishery and bug people who use expensive instruments for testing
But want to do real time monitoring, which gets very expensive
Used to use low-cost sensors, but people started questioning the data. Had to use expensive sensors for people to take their data seriously
have an amazing record from high-quality instruments at very different costs, can validate cheap sensors
road salt for melting snow/ice → washes into streams, town discovered their stream was saltier than ocean at several points in winter, reduces stream health
power company funded them to do controlled lab tests, then apply in streams
If you can point to research that others have done, even if you can’t do it yourself, gives your data some meaning
Can’t just show raw data, how can your data be backed up by other data/studies they can trust
Can you recommend validated low cost sensors – list on website, other groups maintain lists
point to list of validated sensors
Lake Victoria, Kisumu in 3 countries, high polluted
fish farming is a large source of cash, pollution interferes with this
innovation hub came up with sensor to monitor pH, not necessarily low cost but easy to use because doesn’t need internet, took 5 years to test prototype
desparate to see what works
gov’t not very involved in this conversation
Take a baseline before changes (e.g., before power plant is constructed)
Make sure your research is on policymakers’/agencies’ radar
Sparks
Establishing a common language to truly understand needs
German standardization body: established std. for OH documentation, history of standards, how to advance standards that are needed
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Environment / Conservation Biology - SubGroup about “Interconnected resources and pathways to shared solutions”

Date: 2022-10-27
Time: 11:00 (Panama local time in 24-hour format!)
Place: Outside Tables
Facilitator: Shannon and Allison
Notetaker: Gourav

Participants

  • Gourav
  • Clint
  • Laura
  • Isaac
  • Lara
  • Allison
  • Josh
  • Guillermina
  • Tama
  • Andres

What is needed to make progress into making interconnected resources and pathways to shared solutions. What can be do to make the community be more interconnected
What are the interconnectedness of the people currently in the session group?

The different parts of our conversation are:
Interconnecting Resources + Pathways
Shared Solutions

Needs:
Finding the common language in the context of languages as well as in the context of the interdisciplinary nature of the research that we do. In a way, a biologist and an engineer speak two different languages and we have to translate that to a common language in order for everyone to understand the conversation.
Finding boundary options (a part of social science) that are okay with all the stakeholders
Listening is an important aspect of someone who doesn’t necessarily have the same needs as someone else.
Communication is one aspect but we should “look at our differences differently”. Paulo Tiere also connected to interconnecting.

Opportunities

Boundary Objects - we have translations but we need to translate the idea and the field itself not just the language. Maybe make something like google translate that translates ideas

Instead of trying to establish interconnectedness, we should look at the needs first and actively listen to the stakeholders. When working with communities, the engagement with these communities should be established as very important roles.

Interconnectedness would mean checking up with the stakeholders of the solutions you are providing to get feedback and go through iterative design to further better your product.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

LAB HARDWARE #1

Date: 2022-10-27
Time: 11:00
Place: Large Room
Video-conference: None
Public documentation channel: None
Facilitator: George
Notetaker: Dorry

Participants:

Notes:
Multivoting process was done to select which topics would be discussed in the sesión.

Do/Have:

  • 3D metal printing
  • Soil Respiration Meter
  • Passive Acoustic Monitoring
  • Flow-through Colorimeter
  • Open Flexure Microscope
  • Microscope
  • Incubator (Large door)
  • Aperture Adjustment Knob
  • Pipetting robot
  • Motorized XY scanning stage
  • Portable Gas Detection Sensor
  • RasPi/Arduino based Data Collectiond

Work in Progress

  • UV/Vig
  • HPLC
  • Spec
  • Spinning Incubator
  • DNA electrophoresis (cheap electrode)
  • Micropipettes
  • Digital BPM tube connector (ensure proper connection between internal + external blood pressure monitor)

Want

  • Microbiology incubator
  • Rudimentary Mass Spec
  • PCR
  • Electroporation Device
  • Low RPM Centifuge
  • Position encoder
  • Droplet microfluidics pump
  • Fluorescence microscope electronic control + focus
  • Bioreactor- mammalian cell
  • CO air quality sensor
  • Raman, FTIP, ATR, Maldi-TOF, NMR, HPLC
  • Water Distillation Stills
  • OptoElectrowetting
    ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

AI

Date: 2022-10-28
Time: 11:00
Place: Ñeque
Social media links and hashtags: AI
Facilitator: Constantin - Zalmotek.com
Notetaker: Alexandra - Zalmotek.com

Participants

  • Niko
  • Linda
  • Josh
  • Alexandra
  • 7 students from Panama school with their robotics teacher

Exploratory session
In the last few years, AI has been evolving rapidly, solving lots in tedious or overly complex tasks. We invite you to participate in this exploratory session to find out how we can use AI to our advantage.

What aspect in your day to day work do you hate the most/consider tedious and it could be automated with AI?
Answers:
spelling correction (Grammarly)
AI assistant - general tasks
AI assistant for answering questions from domains we are not interested in and we don’t have knowledge about
note taking, captions
writing, generating new content
personalized recommendations - social media learning about our passions
self-driving cars
speech to text for meetings when too many people are talking
AI for design, Photoshop, correcting pictures, video editing
e.g Disney uses AI + 3D modelling for 360 video generation
animations, making 3D animations look real (cleanup) → Stability - open source tool
searching for a word you don’t know when reading in another language takes a lot of time
long time lapses take hours to edit
field biology, going to places far away is complicated because you have to carry so much equipment to analyze big species - make the equipment smaller
making the BOM and searching for components in stock when designing PCB
research before starting a hardware prototyping project, checking compatibility between components and sensors
video editing: lights + colors are very hard to match manually, maybe an AI could read the whole scene, and decide the right combo and how to imitate from other scenes
when switching from an IDE to another the buttons are not the same/the interface is different, maybe AI could be use to “google” the function you’re looking for
AI for writing code, software copilot
AI in medical areas - language interface.
AI in agriculture - direct integration

Open source tools for AI:
Hugging face (https://huggingface.co/) - open source AI models (speech recognition, image recognition etc)

How do we feel about AI?
Answers:
we should be working hand in hand with AI
AI can fail for you faster - automated testing to discover what works and what doesn’t work
You would rather trust a human than an AI

How can we interface with an AI?
Blocks, nodes - visual programing

Conclusions:
Any type of AI has to be done in a hybrid way in the beginning: human decisions + AI. After it learns it can be used by itself, without human help.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Business Models for OSH

Date: 2022-10-27
Time: -
Place: -
Facilitator: Ryan
Notetaker: Obanda

Participants

  • Rajas
  • Greg
  • Ian
  • 5 participants

Highlihgts:
Rajas: -
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Hardware for Academia

Date: 2022-10-27
Time: 11:00 (Panama local time in 24-hour format!)
Place: Outside
Facilitator: Moritz
Notetaker: Pablo, Juul

Participants

  • Juul
  • Bill
  • Harold
  • Byron
  • Pablo
  • Ernesto
  • Pablo
  • JP

– ernesto: a bit of behavior, bit of molecular biology, high school biology teacher
– bill: panama, innovation stem teacher in Panama
– harold: Singapore, engineer by training. passive acoustic monitoring for wildlife monitoring. never been academic but find them to be very strange lot
– pablo: physics teacher, work in development of open science hardware for teaching science also agro science for land management, water monitoring
– byron: work in developing microchips. Only one university can design chips. trying to involve more people to break the barrier between this one uni and other universities.
– JP: work as research associate at UT Austin. Science, microbiology and health.

Notes:
What do people talk about in Academia? Participants share their experience with open source hardware and proprietary hardware at the academia.
– Juul: 1. transfer office barrier. 2. if no one is selling a piece of equipment then it doesn’t matter if it’s open.
– JP: production after you have a prototype. Scaling.
– Ernesto: difficult to compete at high level using open source hardware. Difficult to get reactives in Panama.
– Bill: not thrilled with how much we depend on corporate tech. better educate educators about open source tools, e.g. how to manage a network and how to teach kids
– Pierre: from Peru. supply chains. We deal with customs in chile. i have to pay maybe 10% more and if you have a small budget then it’s a problem. hard to reproduce any open hardware design. from local supplies you often can’t get the components you need
– Harold: I think what I bump up against most is the fact that academic scientists are the scientists. The way science is practiced now cannot be practiced in any other way than in an academic setting. Scientists are academic firsts and scientists second. subject to an incentive structure that discourages the … science.
– Pablo: universities sometimes use very expensive equipment for research and also for teaching. an open source spectrometer for teaching doesn’t need to be the same. we have had some
struggle with getting equipment for teaching physics. We’ve also developed all the equipment we need for 1st year and 2nd year.
– Byron: Sometimes you have a piece of equipment but the person who knew how to use it has left and no one knows how to use it. They are very worried because they have very expensive equipment no one knows how to use. trying to do reverse engineering to figure out how to use it.
– Moritz: as an academic it’s publish or die. You can’t overestimate how that shapes every action you do. We want to make everything in our lab as open as possible. There are other labs that work on similar things. If someone else can take our hardware design and snipe us by skipping that step. We want to be open but we have to keep it closed for some amount of time in order to get funding for the future.
Motivation: - If we disregard the “publish or die” and perverse incentive structure of academia what would you actually want to be doing
– JP: empower communities to air quality monitoring. We don’t publish because we didn’t do it, we grabbed info from others. We want to publish about the benefits of the open hardware experience.
– Juul: make it easier to build their own labs.
– Ernesto: Transition between academia to make an impact in the general public and that’s a gap, and part of it is funding. In the paper they want to have an impact outside of academia but in reality it’s not happening. It would be useful to convince universities to have a maker space. On the contrary, at high school we have maker spaces.
– Bill: not a researcher. k-12 educator. finding ways for research to happen outside of academia is important. As a K12 educator we are restricted in what we teach based on what the universities want/expect the students to know. they have a specific skill set they expect
– Pierre: in some courses for undergrad students. They have not only proprietary tools but also open source tools. After discussing with students That’s not always the case. Often students are only trained in a proprietary tool so that’s all they know. If you teach them both they can pick whichever tool they think is appropriate and often I see them pick the open choice. In terms of research in the institute where I work they try to develop these hardware tools for understanding human microbiome to understand single cell interactions. Sometimes when you want to perform new and complex experiments then you need a lot of money to buy the equipment. If you could put all of the equipment in a small space and control it remotely. That’s why we are trying to design this microfluidic design that could help. We need to tell the story of successful projects like Arduino.
– Harold: pass
– Pablo: The only way to teach science is by using open science hardware. We need the people who use the equipment to understand how it works. Sometimes we also use very expensive equipment where cheaper and open equipment could do the job. Also repairability.
– Byron: pass
– Moritz: Agree with you on cutting costs. IN the electrofish project we’re building hardware to talk to a fish (open hardware that you can buy with open hardware you put on top of that). Buying one of the units for scientific instrumentation would cost 16k euro per unit. It’s not even that this thing is exactly what you need. E.g. We need 4 channels but the only unit available has 128. We want to have hardware that is exactly what we need for our s. Specific use case. Goal: Cutting costs and limiting resource usage (don’t use a bigger piece of equipment than needed). How to build a maker space that would live even if you get funding for 3 years with no warranty you will continue after that. We need a certification from the IT department to connect our equipment to the grid.
Harold: part of the perverse incentive structure is that no PI wants to be part of low cost project. They want to be part of a high budget project. So let’s make OScH projects expensive.

Links:
GitLab/GitHub repository:
Open hardware for teaching physics: OSF | Hardware libre de bajo costo para laboratorios de física
TU Delft Makerspace: https://www.makerspacedelft.nl/
Open tools for chip design https://theopenroadproject.org/
Open FPGA design https://f4pga.org/
open microproessor architecture https://riscv.org/
ElectroPEN lighter igniter based electroporator

––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Right to Repair

Date: 2022-10-27
Time: 11:00hs
Place: Back of Toucan Room
Facilitator: Nat
Notetaker: Mihnea

Participants

  • Nat
  • Mihnea
  • Mada
  • Goldjian
  • Jay
  • Jason

Notes:
This was an exploratory session in which we have tried to identify what are the main barriers that stand against Open Source and Right to Repair becoming the norm.
The main culprits were rather obvious:
corporate greed
perceived profitability
legitimacy
liability issues
misinformation
The real challenge was trying to find a blocker that only applied to one of the movements and the consensus was that Right to Repair is embedded in the Open Source philosophy.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––-----–

Open Science Hardware - Wikipenda

Date: 2022-10-27
Time: 11:24
Place: Tucan Room
Facilitator: Tiago (wiki nerd), contact info: @lubianat (twitter) tiago.lubiana.alves@usp.br (email)
Notetaker: Nico M. (levadurologist)

Participants

  • Marcela (Buenos Aires).
  • Joaquin (Mendoza, estudiante).
  • Andy (Gamboa, dynalab)
  • Dario (Italy/US, volunteer contributor, works at wiki foundation research team).
  • 6 participants

Notes:
Tiago presenta la Wikipedia con su computadora. Se puede editar fácilmente desde una computadora, pero también desde el celular con limitaciones.
Hablamos sobre la desigualdad de cantidad y calidad de artículos en diferentes idiomas.
Crear una cuenta es muy fácil, no es necesario ni un email.
Entramos al sitio del “Open Design Movement” en inglés, y vimos la diferencia con el artículo en español. El artículo en español tiene menos contenido. La página de “Open Hardware” es más similar entre idiomas.
Wikipedia shapes public discussions a lot!
Each page belongs to categories, shown at the bottom of each article.
It’s better to add to an existing page than create a new one. New pages will have lots of errors and will probably get deleted.
Good first contributions:
Add links to articles.
Edition summaries are important when changes are “controversial”, even slightly.
We browsed articles around open hardware, and learnt how to add links with the editor. We also realised that there are two list articles on open hardware, and proposed a merge.
To create an article about GOSH, it must have independent coverage in reliable secondary sources. For example: mainstream news, scholarly articles (not self-written), books, etc.
There is a “beta” feature of wikipedia to translate pages.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Assessment

Date: 2022-10-27
Time: 11:30
Place: Outside
Facilitator: -
Notetaker: -

Participants
- Andrew
- Dorn
- Gina
- Jenny
- Josh

links

Reference document: An open toolkit for tracking open science partnership implementation and impact: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6904887/

Work document: 2022 10 27 11:30 Evaluation Session - Google Docs