AI summary by TwinMind

Sabine Hossenfelder | Research Made Easy: Evaluate and Rank Your Way | Talks at Google

Talks at Google

50m 37s

Introduction and Background

  • Sabine Hossenfelder: theoretical physicist studying quantum gravity, prolific blogger (Back Reaction blog since 2006, 7+ million views), contrarian author (Lost in Math), YouTube creator (Science Without the Gobbledygook)
  • Currently research fellow at Frankfurt Institute for Advanced Studies, originally from Frankfurt
  • Traveled globally before returning to Germany

PhD Thesis and Early Career Crisis (2000-2003)

  • PhD thesis: particle physics phenomenology beyond standard model
    • Focus: production of tiny black holes at Large Hadron Collider (LHC)
    • Popular topic 2000-2003, attracted attention (CERN lawsuit over world destruction fears)
  • Black hole creation theory:
    • Einstein's gravity (3 space dimensions + 1 time): insufficient energy at LHC
    • With large extra dimensions: gravitational interaction stronger at short distances, enables black hole creation
    • String theory predicts 6 additional space dimensions
    • Required exact size range: too large = already detected, too small = beyond LHC reach

Naturalness Problem and Scientific Crisis

  • Conference presentation question about energy range coincidence
  • Standard answer: naturalness argument - if energy much higher, beyond Higgs boson discovery mass
  • Naturalness definition: theory without very large/small numbers unless explained (e.g., by symmetry)
  • Standard model unnaturalness: Higgs mass should be much larger than observed
  • Literature review revealed: naturalness argument ambiguous, arbitrary, relies on arbitrary assumptions
  • Personal crisis: couldn't believe scientific soundness of argument underlying thesis work

Career Decision and Fellowship Rejection

  • Received prestigious Emmy Noether Fellowship (German Research Foundation, 5-year funding for whole group)
  • Declined fellowship rather than spend 5 more years on work she didn't believe in scientifically
  • Started blogging carefully (2005), years before LHC turned on (2009)
  • Initially thought she was "dumb" for not understanding the argument

LHC Results and Community Response

  • LHC discovered Higgs as anticipated (major breakthrough)
  • Failed to find: supersymmetric particles, dark matter, gravitons, black holes
  • Expected community to abandon naturalness argument
  • Instead saw: denial, "next run" hopes, energy requirement adjustments, eventual push for bigger collider
  • 2014: decided to write book (Lost in Math) to inform public about unsound scientific arguments

Scientific Community Dynamics and Group Psychology

  • Scientists don't sufficiently consider social influences on their thinking
  • Problem: assigning researchers to specific groups (string theory, cosmology, etc.)
  • Group psychology research: people identify with and defend group identity
  • Proposed solution: tag clouds instead of group labels to show research breadth
  • Attempted implementation at Perimeter Institute and Stockholm (IT liked it, administration rejected)

Scimitar Development and Motivation

  • Website/web interface providing database access for custom analysis
  • Addresses problem: thousands of people repeating obviously wrong arguments
  • Community too closed, doesn't listen to outsiders (philosophers could have corrected naturalness problem)
  • Publication/citation pressure drives researchers toward popular topics regardless of scientific merit
  • Personal example: LHC prediction papers still most-cited work, still receives invitations despite being wrong

Scimitar Features - Archive Database

  • Currently live: scimitar.org (staging version shown with updates coming)
  • Keyword clouds: visual representation of researcher's topics
    • Color coding: red = recent work, blue = older work
    • Customizable, downloadable formats
  • Similar authors: find researchers with similar interests
    • Useful for conference organization, collaboration finding
  • Topic matches: find people who worked on specific topics
    • Applications: conference planning, science journalism, expert identification

Scimitar Features - Custom Metrics

  • Archive limitations: post-1994 papers only, certain science areas only
  • Microsoft Academic Graph: covers all science fields, longer time range
  • Addresses over-reliance on simple metrics: paper count, impact factor, citations, h-index
  • Committee experience: people fall back on readily available but flawed measures
  • Custom metrics enable personalized definitions of "good science"
  • Institutional flexibility: specialized vs. broad interests, different research styles

Scimitar Demonstration

  • Author list creation (Turing Award winners 2018: Yoshua Bengio, Geoffrey Hinton, Yann LeCun)
  • Default h-index ranking, cross-checked with Google Scholar
  • Normalized h-index: accounts for field size differences
  • Single-authored papers metric: shows individual vs. collaborative work patterns
  • Custom metric creation: citations per author (divides citations by number of authors)
  • Metric submission and approval system for community use

Scientific Metrics Problems

  • Current system encourages popular topic work, hinders progress
  • People know measures have severe shortcomings but use them due to easy availability
  • Need for metrics matching personal/institutional scientific values
  • Goal: enable better scientific evaluation beyond naive paper counting

Future of Scientific Research

  • Scientists need greater self-awareness of community influences
  • Should be taught in education
  • More mindful use of scientific impact measures
  • Technology can address information filtering and influence recognition

Quantum Computing and Current Work

  • Google's 2019 quantum supremacy achievement: remarkable accomplishment
  • Working on YouTube video about quantum computing developments, company activities
  • Upcoming video timeline: next month or two

Superfluid Dark Matter Research

  • Potential experimental confirmation possibilities
  • Galaxy formation observations: phase transition evidence (different from cold dark matter/modified gravity)
  • Laboratory detection research ongoing with student
  • May be too weakly interacting for feasible detection

Fundamental Physics and Pseudoscience

  • Fundamental physics attracts people responsive to big questions
  • Field addresses formerly religious questions: universe origin, free will, information after death
  • Discipline closest to religion-science boundary
  • Second book in progress: science/pseudoscience/religion/philosophy boundaries
  • Parallel universes example: not provable but not ruled out by science
  • Inspirational value shouldn't be underestimated

Investment in Fundamental Physics

  • Question too complex for quantification (thousand-year payoff timelines)
  • Simpler question: how to best invest current societal funding
  • Answer: not bigger colliders - other investments more promising
  • Better allocation of existing resources rather than funding level debates

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome