Outreach: Open Science Values and Practice - petermr/CEVOpen GitHub Wiki

Open Science Values and Practice

Material collected for a presentation to INYAS 2021-08-13 (@PeterMR)

Background

invitation from INYAS to @petermurrayrust

... as part of our INYAS' mid-year meeting (online meeting), we are working towards organizing a session on "Open Science – Challenges, Possible Solutions and the Way Forward". We wish to work towards understanding the various issues revolving around Science publications and look at potential solutions to make science free for all. These are issues that concern all academicians, researchers and the community at large, but are very rarely discussed. And hence, we truly believe that this is a great opportunity to initiate this discussion.

... The session is tentatively scheduled to be held on an online platform on 13th August, 2021 (in the post-lunch session - after 3 PM IST). The duration of your talk is intended to be 30 minutes + 10 mins for Q&A. This will be followed by a few other talks and then a 30 mins panel discussion (total session duration 2 hours) .

Submitted abstract

"Open Science: Values and Practice" [0]

Peter Murray-Rust [1], Shweata N Hegde[2,3], Ayush Garg[4]

Science increasingly plays a part in tackling many of the world's problems - gathering knowledge, analysing it objectively, creating new artefacts and procedures, and disseminating the results. Many of the UN's Sustainable Development Goals (SDGs) have a need for science, applied for the benefit of the world. Yet current scientific practice is often inequitable and neocolonialist, weighted in favour of rich (Northern) nations with barriers which exclude the others.

But in the digital century science can belong to everyone, regardless of nationality, age, geography, language. Open science can be inclusive and meritocratic. Openness means:

  • available to all
  • deliberately transparent
  • reusable (with acknowledgment)
  • immediate
  • discoverability Digital allows much greater sharing of resources - computation, and access to equipment. For hardware and reagents, Open scientists are often able to bring down costs dramatically.

This presentation will concentrate on creation of, and access to, scientific knowledge. The values are critical:

  • scientists need to want to share knowledge. Ranganathan's laws of the library: "save the time of the reader" and "every reader their book" are technically possible
  • the sharing has to be frictionless. Logins, gatekeepers, create massive friction.
  • paraphrasing Ranganathan: "the knowledge system [library] should be a growing community" But too few in modern science think of this. The values have been warped by greed and vanity [5] and the Global North thinks more of career progression and corporate profits. Scientists in the Global South can either publish for free or read for free but not both. Prices (not costs) of 3000 USD-12000 to publish a paper are created with no regard to a global science community.

We must reset our values to put the world first and fast - if we do not, then global problems will overwhelm us. Open Science is a key part of our toolset. But "Open" is a broad and often misused label ("openwashing"). True Openness - such as in Open Notebook Science - brings major benefits:

  • inclusivity. The whole world can be involved in knowledge creation and dissemination
  • Continual improvement. Ideas and data can be shared, developed, tested and grow much faster than in a single person's computer.
  • immediacy. Open Science is conducted in full view of the world.
  • completeness and permanency. Openly created science does not get mislaid, or miscopied, or corrupted.
  • speed.

We will illustrate many of these principles with our projects with extract raw text and synthesize it into semantic knowledge:

Open Science and Open Source share many values. We have built Open Source software to extract and process science from published papers and have worked to make this frictionless. The system pygetpapers and pyami can search for and automatically download many hundreds of Open Access articles - "in minutes". These are made semantic and annotated with Wikidata/Wikipedia-based "dictionaries" available in many different languages. The design gives control to the reader, rather than the provider and allows many types of analysis and display. We hope to give a short demo and we are keen to get feedback.

[0] Note - this abstract may be updated at https://github.com/petermr/CEVOpen/wiki/Outreach:-Open-Science-Values-and-Practice and may point to new links. If anyone wants to try out the (Python) software (alpha and beta) it can be downloaded
[1] Yusuf Hamied Department of Chemistry
University of Cambridge, UK
[2] Regional Institute of Education, Mysuru
[3] NIPGR
[4] Global Indian International School, Singapore
[5] Prof P.Balaram [IISC] to Indian Librarians https://www.youtube.com/watch?v=ndtYh_rp7yw&ab_channel=LISAcademyIndia
[6] We thank NIPGR and Dr Gitanjali Yadav for support and sponsoring (virtual) internships

Past talks

Open Science (PMR)

(Indian examples) graphic novel, annotation , magnification anybody can do this low-zero cost restricting this presentation to ONE example-type - informatics (mention broader picture)

timetable

  • 0 PMR values and practice
  • global challenges - health , climate change (IPCC, report).
  • young scientists
  • open - science and knowledge
  • knowledge problems (Global North) neocolonialism
  • Gita/Jenny picture
  • citizen science - examples (Galaxy Zoo, Open Malaria, SBDD-India)
  • intro to CEVOpen - plant, chemistry based science
  • 12-15 Shweata Open Science and CEVOpen overview
  • literature downloading
  • dictionaries - show dictionary, content and scope
  • extracted data
  • community
  • personal journey
  • introduction to CEVOpen - pictorial table of contents - Kanishka's map of invasives
  • 15-20 PMR details of CEVOpen - include data availability - Sagar's scoping
  • motivation (genetic / protein engineering) -> new products, crop engineering (picture) - TIG2RESS(picture)
  • need images from past CEVOpen material (e.g. Ambreen)
  • Sagar's TPS phytochemistry: plant(picture) -> enzyme(picture) + (reactant) -> chemical(picture)
  • scraping literature -
  • snapshot of TPSchem table with annotation
  • community - past pictures
  • intro informatics - automation, big data, machine learning
  • 20-23 Ayush
  • why do we do science
  • how do we do good science - documenting , issues
  • who can do science
  • what is pygetpapers and docanalysis
  • 23-28 informatics infrastructure for Open Science - include Chaitanya and ethics
  • Chaitanya - bio + picture -> classification/ML - picture of your results - acknowledgemnts
  • speed, immediacy, open
  • preprints
  • challenges
  • makespace - hackathons .
  • getting started (specific <-> general). Funding, community management, outreach ... (Safecast) Fake-news, etc x data journalism
  • 28-30 Resources, thanks, ways forward - invite collaboration/volunteers

Shweata material could use more graphics, some from CEVOpen

================

Volunteers OPEN:

  • TASKS
  • commitment
  • length
  • incremental and modular (independence)
  • extension modules - no dependency
  • bite-sized tasks - 5-500 second tasks - gamified - crowdsourcing
  • domain-rules
  • specialist knowledge
  • hackathon - site/s, goals, documentation, technology.

Themes (expand later) ## why do young people do STEM ?

  • driven "from the heart" - you want to do this
  • how does the world, the universe, work?
  • it's new - to boldly go
  • it makes (scientific) sense
  • it expands my potential
  • it's part of something larger
  • it can help others and the world
  • It's a great career
  • it rests on moral/ethical values - we "have to do this"

Shweata presentation

background:

  • what is open science
  • what is CEVOpen

types of Open science activity

  • observation and data collection

    • naturalists
    • health surveys
  • open hardware and materials

    • microscopes
    • enzymes and reagents
    • raspberry PI
  • re-use of existing data

    • Galaxy Zoo - annotated a million galaxies
    • bioscience data, Human genome project; Ensemble browser
    • climate data analysis and visualisation
    • CEVOpen (our example today)
  • theory

  • teaching and learning

    • openscience lab experiments
    • undergraduate projects

equity and justice

  • valuing young people
  • freeing knowledge
  • challenging imbalances
  • value-based science

benefits of doing it right

  • global connectivity
  • near zero-cost
  • community-based exploration

Shweata's presentation

What we CEVOpen do?

My message

What

My learnings

  • You don’t have to be perfect! Everything’s work-in-progress!
  • Anybody with any level of background can contribute. You don't need a qualification
  • Learning skills from each other isn’t a bad thing!
  • Doesn’t hurt to experiment since you document everything, and can revisit them anytime!
  • Sharing your work is a lot easier because your identity is always attached to your work when you document - GitHub