Greg Warrington

@GSWarrington

Math Prof at U. Vermont. In parallel universes I am an astronomer, a photographer and a juggler.

Vrijeme pridruživanja: kolovoz 2013.

Tweetovi

Blokirali ste korisnika/cu @GSWarrington

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @GSWarrington

  1. Prikvačeni tweet
    14. ožu 2019.

    New preprint with Bridget Tenner on visualizing the results of instant-runoff (ranked-choice) elections: A zoomable demonstration:

    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    31. sij
    Poništi
  3. proslijedio/la je Tweet

    “Australia, which has used RCV for over a century, once used this system to elect its national Senate. Following the US, states doubled as districts, with three seats to be elected. Of 60 total state-based elections, 55 produced single-party delegations."

    Poništi
  4. 24. sij
    Poništi
  5. 24. sij

    Always discouraging when your paper doesn't get any citations during its first 225 years in print....

    Poništi
  6. proslijedio/la je Tweet
    20. sij

    henry wente: geometer & teacher. he left this world today: 20-jan-2020. he's the reason i'm a mathematician.

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    18. sij

    For those of us who are tenured at R1 math departments, I want to share some things I think we should all be doing. 1/n 1. Directly ask grad students how much they are sleeping, and talk to them about strategies for making time for family, hobbies, and personal health...

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    8. sij

    Theorem: 4 = 2. Proof: Let x^x^x^x... = 2. Then x^(x^x^...)=2, so x^(2)=2, so x=√2. Let y^y^y^... = 4. Then y^(y^y^...)=4, so y^4=4, so y=4^(1/4)=√2. So x = y, and x^x^x... = y^y^y... QED. ????!!! 🙂 (h/t Pete Winkler)

    Prikaži ovu nit
    Poništi
  9. 22. pro 2019.

    Now that it's officially winter, a leaf from last season. Original file is 14000x21000 stitched together from 1100 individual images (b/c focus stacking) using the amazing, free hugin panorama software.

    Poništi
  10. 10. pro 2019.

    Someday I'll get around to making that list of real and fake math terms to quiz people on.

    Poništi
  11. 21. stu 2019.

    Currently, if a municipality wanted to use an accumulation chart to display their official RCV results, they'd need to interface with the innards of the chart. A standardized, full-information format would make things a lot easier. 6/6

    Prikaži ovu nit
    Poništi
  12. 21. stu 2019.

    Currently, to produce our accumulation charts, we had to write our own RCV tabulator. But every locale uses slightly different rules and we haven't spent the time to perfect it; there are slight discrepancies between our results and official results. 5/6

    Prikaži ovu nit
    Poništi
  13. 21. stu 2019.

    My next hope is for a standardized format for the data produced by an RCV/instant-runoff tabulator: The full record of which ballot gets awarded to whom in which round. 4/6

    Prikaži ovu nit
    Poništi
  14. 21. stu 2019.

    There are lots of things you *can't* do if you only know numbers of transferred votes. My paper with Bridget Tenner on visualizing results, now out in the Notices of , requires the ballot-level rankings. Other questions do too. 3/6

    Prikaži ovu nit
    Poništi
  15. 21. stu 2019.

    I'm hoping that places like , who have just started using RCV, will follow this practice (maybe they already do and I just don't know the link). I don't know a good link for the results from . 2/6

    Prikaži ovu nit
    Poništi
  16. 21. stu 2019.

    Thanks to , and for posting the ballot-level data for their RCV elections . This is far superior than simply posting the results. 1/6

    Prikaži ovu nit
    Poništi
  17. 13. stu 2019.

    Here's a thread considering the vote distribution. The declination is assigning a number to each vector of vote shares illustrated below. 6/6

    Prikaži ovu nit
    Poništi
  18. 13. stu 2019.

    Here's the thread from which I pulled the data (I used the '2012-18 avg' column to associate a Dem vote share to each district): 5/6

    Prikaži ovu nit
    Poništi
  19. 13. stu 2019.

    Of the 12 maps I considered, 6 had values close to 0, 5 had values close to 0.15 (~ +1R seat) and one had a value of 0.34 (~ +2R). Only this last could be considered a relatively extreme value in any historical sense. 4/6

    Prikaži ovu nit
    Poništi
  20. 13. stu 2019.

    I'm interested in seeing how the declination measure evaluates the proposals just based on how votes are distributed. This is just one, incomplete aspect of a map, but it's still a starting point and can help indicate which maps merit further consideration as fair maps. 3/6

    Prikaži ovu nit
    Poništi
  21. 13. stu 2019.

    Their full analyses consider important considerations like population deviation, county splits, compactness, and preservation of communities of interest. 2/6

    Prikaži ovu nit
    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·