Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
urbanists.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
We're a server for people who like bikes, transit, and walkable cities. Let's get to know each other!

Administered by:

Server stats:

530
active users

urbanists.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.7

#datavis

13 posts7 participants0 posts today
Public
Paul R. Pival (he/him)

Important research!

The Pour-igin of Species pudding.cool/2025/04/wine-anim

"We used ChatGPT Vision to identify animals on close to 1,500 wine labels to see if we could predict the price and quality of a bottle based on the animal on the label. "

The PuddingThe Pour-igin of SpeciesWe used ChatGPT Vision to identify animals on close to 1,500 wine labels to see if we could predict the price and quality of a bottle based on the animal on the label.
#Datavis
Continued thread
Public *
sport of sacred spherical cows

Updated #CDC estimates show we'd pretty much been in a JN.1.11 soup since Dec, until late March, when LP.8.1 took majority.

Data collection continues to be a low priority nationally, as exactly zero regions have enough data for CDC to plot.

Two children of LF.7.7, dot1 and dot2, broken out by CDC, together with PA.1, a grandchild of MC.10.

Raj's dashboard was last updated a week ago today.

❖ #ThisIsOurPolio #Covid #Covid19 #SARS2 #variants #CovidIsNotOver #CovidIsAirborne #dataviz #datavis

Chart: Estimated U.S. Variant Proportions by Common Name Sources: Centers for Disease Control, CoV-Lineages, NYITCOM Research, Daniele Focosi, World Health Network, others [ beadsland on Ko-fi ] Reskin of CDC's Variants Nowcast, and any significant variants in GISAID not broken out by CDC. Five bar-style tree-charts, for fortnights through April 13–26. Legend of last fortnight, organized by subheadings of color-grouped families and convergent clusters. Percentages overlay each color key, reflecting share as of most recent tree-chart. Essentially all WHO-Vaccine-Target JN.1. LP.8.1 accounts for over two thirds; JN.1 + FLiRT for near a quarter. For fortnight of 3/30–4/12, packed bubble charts fill single-variant tiles, indicating diversification of those variants (per GISAID) that wouldn't otherwise be apparent from CDC's Nowcast. Legend: LP.8.1 [purple]: 69% - LP.8.1.1 / NY† & other LP.8.1 JN.1 + FLiRT-KS [browns]: 1% - XEC.4 ¾% - XEQ 10% - other XEC JN.1 + FLiRT [reds]: 6% - LF.7.7.2 4% - LF.7.7.1, LF.7 & LF.7.2.1 / PC 2% - LB.1.3.1 / NL JN.1.11 [greens]: 3% - MC.10.1, MC.28.1 & MC.19 2% - PA.1 ½% - KP.3 2% - MC.1 & other KP.3.1.1 / MC XEK [blue]: ⅜% - XEK Other [greys]: ⅙% - JN.1.16 ⅒% - NB.1† & other WHO-Vaccine-Target JN.1 ¼% - Other (not specified) † Dagger variants are not broken out by CDC, yet represent a significant share of recent GISAID sequences.
Continued thread
Public
Mark Dingemanse

this page 👆 is also to remind myself to think about visuals early.

sometimes I find myself talking or writing about a finding and I realise our paper actually doesn't do the best job of visualising it, so I create something new

2 examples: L how we map colours onto vowels sounds, R three common metaphors in language development. Both made after the papers (2019/2021) were published

learning from this, I now try to think ahead and make visuals clear & reusable in the first place

layered diagram showing, first, a plot of vowel space (by F2 and F1 formants on x and y axes) with 14 roughly equidistant vowel sounds plotted in a triangle. The plot is tilted in a way that makes the left side closer to the viewer. Second, a colour circle partly overlapping with the first plot, with the same kind of tilt and with grey lines showing how it projects onto the vowel space plot. In combination, it is possible to see how individual vowel sounds like /e/, /a/, and /u/ map onto regions of the colour circle like light green, red and blue.
slide showing three key metaphors from the language development literature with relevant references: bootstrapping (Pinker 1984, Gleitman 1990), scaffolding (Bruner 1978, Imai et al. 2008), bridging (Radwanska-Williams 1994, Perniss & Vigliocco 2014). Below the text are three illustrations in black and white, crayon style: a boot with a bootstrap at the back end; a scaffold; and a bridge.
#datavis
Public
UCLAB Potsdam

🔥 »I Want a Better Catastrophe« made it onto the short list of the Information is Beautiful Awards in the “Places, Spaces & Environment” category.

The flowchart by @jona combines audio narration with interactive elements as an invitation to join #AndrewBoyd on his narrative path and explore our predicament on your own: flowchart.bettercatastrophe.co

You can find all other short-listed entries here: informationisbeautifulawards.c

flowchart.bettercatastrophe.comI Want a Better Catastrophe · FlowchartA flowchart for navigating our climate predicament
#IIBAwards#BetterCatastrophe#Flowchart
Public *
Marian Dörk

2-year professorship in »Information Visualization & Management« (Oct 2025–Sep 2027, W2) at #FHPotsdam with focus on data visualization, urban complexity & interdisciplinary research.

This is my parental leave replacement - feel free to reach out to me or @boris if you have any questions.

Applications already due on 9 May!

Job ad: jobs.b-ite.com/jobposting/a595
(this is in German, but teaching can be in English!)

jobs.b-ite.comVertretungsprofessur (W2) »Information Visualization & Management«
#AcademicJobs#DataVis#DataViz
Replied in thread
Public
arrrrrmin

@anneroth
Der Mehrheit (alle die nicht afd wählen) wird da ein Niveau vorgesetzt das unterirdisch ist (afd wird nur schwurbeln und falsch informieren), die Ausgaben um das in Echtzeit wieder einzufangen (also den eigenen Auftrag zu erfüllen), werden ja nicht berücksichtigt.

Dazu mal eine einfache #datavis anhand echter Daten von Caren Miosga mit Alice Weidel. So lange dauert es idR bis es einen Faktencheck gibt, der wenn er überhaupt gelesen würde, zu spät ist:
chaos.social/@arrrrrmin/114116

chaos.socialarrrrrmin (@arrrrrmin@chaos.social)Attached: 1 image Unabhängig davon, ob Faktenchecks tatsächlich gelesen werden, quält mich die Frage: Wie lange hat man als Redaktion wohl Zeit, einen Faktencheck zu erarbeiten? Hier ein Versuch das Aufmerksamkeitspotential für einen Faktencheck über Bluesky bzw Mastodon Posts abzubilden: #datavis #dataviz #d3js
Public
Jumping Rivers

Announcing the Jumping Rivers Data Science Gallery!

We are thrilled to release a new Data Science Gallery to showcase our expertise with data dashboards using Shiny, D3 and other web technologies.

Read the full details in out latest blog post!

#RShiny #datascience #datavis #D3
jumpingrivers.com/data-science

jumpingrivers.comGalleryExamples of dashboard and data-based web applications made at Jumping Rivers using Shiny and other web technologies.
Continued thread
Public
sport of sacred spherical cows

Updated #CDC estimates show we've pretty much been in a JN.1.11 soup since Dec, until late March, when LP.8.1 took majority.

Data collection continues to be a low priority nationally, as exactly zero regions have enough data for CDC to plot.

No new variants broken out, yet CDC seems to have resources to tailor their color key: hcommons.social/@beadsland/114

Raj's dashboard was updated on today.

❖ #ThisIsOurPolio #Covid #Covid19 #SARS2 #variants #CovidIsNotOver #CovidIsAirborne #dataviz #datavis

Chart: Estimated U.S. Variant Proportions by Common Name Sources: Centers for Disease Control, CoV-Lineages, NYITCOM Research, Daniele Focosi, World Health Network, others [ beadsland on Ko-fi ] Reskin of CDC's Variants Nowcast, and any significant variants in GISAID not broken out by CDC. Five bar-style tree-charts, for fortnights through Mar 30–Apr 12. Legend of last fortnight, organized by subheadings of color-grouped families and convergent clusters. Percentages overlay each color key, reflecting share as of most recent tree-chart. Essentially all WHO-Vaccine-Target JN.1. LP.8.1 accounts for over three fifths; JN.1 + FLiRT for near a quarter. For fortnight of 3/16–3/29, packed bubble charts fill single-variant tiles, indicating diversification of those variants (per GISAID) that wouldn't otherwise be apparent from CDC's Nowcast. Legend: LP.8.1 [purple]: 64% - LP.8.1.1 / NY† & other LP.8.1 XEC [browns]: 2% - XEC.4 15% - other XEC JN.1.11 + FLuQE [greens]: 5% - MC.10.1 & MC.19 1% - MC.28.1 2% - XEQ & other KP.3 3% - MC.1 & other KP.3.1.1 / MC JN.1 + FLiRT [reds]: 4% - LF.7 2% - LB.1.3.1 / NL ¾% - LF.7.2.1 / PC JN.1.11 + FLiRT [blues]: ½% - XEK 0% - KP.2.3 Other [greys]: ⅓% - JN.1.16 ⅒% - other WHO-Vaccine-Target JN.1 0% - Other (not specified) † Dagger variant is not broken out by CDC, yet represent a significant share of recent GISAID sequences.
Public
UCLAB Potsdam

Two new papers on #Storytelling & #DataVis to be part of special issue of IEEE CG&A on #CriticalDataVis:

Critical Interactivity: Exploration and Narration in Data Visualization
by @framoriniii Manuela Garretón @jona Nadia Zeissig @SabineDG @fidel @dielindada @ikyriazi Andrea Scholz & @nrchtct
doi.org/10.1109/mcg.2025.35446

Towards Collective Storytelling: Investigating Audience Annotations in Data Visualizations
by @tobiaskauer @nrchtct & Benjamin Bach
doi.org/10.1109/mcg.2025.35479

#OpenAccess#FHPotsdam
Public
boB Rudis 🇺🇦 🇬🇱 🇨🇦

On Monday, I have 80 minutes to inform & inspire a #datavis class at UNH.

So, datavis-fam, if *you* had the same opportunity, what "must have" knowledge, tooling, approaches, words of wisdom, etc. would *you* want to convey?

Continued thread
Public *
sport of sacred spherical cows

Updated #CDC estimates show we've pretty much been in a JN.1.11 soup since Dec, until late March, when LP.8.1 took majority.

Data collection continues to be a low priority nationally, as only one region (NY/NJ) has enough data for CDC to plot.

No new variants broken out, yet CDC seems to have resources to tailor their color key: hcommons.social/@beadsland/114

Raj's dashboard was updated on Sunday.

❖ #ThisIsOurPolio #Covid #Covid19 #SARS2 #variants #CovidIsNotOver #CovidIsAirborne #dataviz #datavis

Chart: Estimated U.S. Variant Proportions by Common Name Sources: Centers for Disease Control, CoV-Lineages, NYITCOM Research, Daniele Focosi, World Health Network, others [ beadsland on Ko-fi ] Reskin of CDC's Variants Nowcast, and any significant variants in GISAID not broken out by CDC. Five bar-style tree-charts, for fortnights through March 16–29. Legend of last fortnight, organized by subheadings of color-grouped families and convergent clusters. Percentages overlay each color key, reflecting share as of most recent tree-chart. Essentially all WHO-Vaccine-Target JN.1. LP.8.1 accounts for well over half; JN.1 + FLiRT for over a quarter. For fortnight of 3/2–3/15, packed bubble charts fill single-variant tiles, indicating diversification of those variants (per GISAID) that wouldn't otherwise be apparent from CDC's Nowcast. Legend: LP.8.1 [purple]: 55% - LP.8.1.1 / NY† & other LP.8.1 XEC [browns]: 21% - XEC.2† & other XEC 2% - XEC.4 JN.1.11 + FLuQE [greens]: 6% - MC.10.1 & MC.19 3% - MC.28.1 2% - XEQ 4% - MC.1 & other KP.3.1.1 / MC ⅘% - other KP.3 JN.1 + FLiRT [reds]: 4% - LF.7 2% - LB.1.3.1 / NL JN.1.11 + FLiRT [blues]: ½% - XEK 0% - KP.2.3 & KP.1.1.3 / LP Other [greys]: ⅖% - JN.1.16 ⅙% - other WHO-Vaccine-Target JN.1 0% - Other (not specified) † Dagger variants are not broken out by CDC, yet represent a significant share of recent GISAID sequences.
ExploreLive feeds
About