Ben Bolker
@bbolker.bsky.social
Ecology, evolution, epidemiology, statistics (mixed models). McMaster University, Hamilton, Ontario https://math.mcmaster.ca/bolker
Thinking about vibe-coding something to take BibTeX & (1) remove interior braces (2) Brace-protect [& case-correct] Proper/CAPITALIZED keywords (from a def file with regexps) (3) brace-protect/italicize species names (ditto). Anyone done this already? Like this: web.archive.org/web/20010707...
(Thanks to Rolf Sander for the shell script and for corrections.)
web.archive.org
November 11, 2025 at 1:01 AM
Thinking about vibe-coding something to take BibTeX & (1) remove interior braces (2) Brace-protect [& case-correct] Proper/CAPITALIZED keywords (from a def file with regexps) (3) brace-protect/italicize species names (ditto). Anyone done this already? Like this: web.archive.org/web/20010707...
follow-up: I guess you're referring to conclusions arising from Friston et al (2022) doi.org/10.1038/s415... ? That paper looked cool from a modeling point of view but I guess I'm not surprised it's non-identifiable. Can you point me to a published/public critique?
Dynamic causal modelling of COVID-19 and its mitigations - Scientific Reports
Scientific Reports - Dynamic causal modelling of COVID-19 and its mitigations
doi.org
November 10, 2025 at 1:20 PM
follow-up: I guess you're referring to conclusions arising from Friston et al (2022) doi.org/10.1038/s415... ? That paper looked cool from a modeling point of view but I guess I'm not surprised it's non-identifiable. Can you point me to a published/public critique?
Scraped data from the plot ... gist.github.com/bbolker/d6a6... Scraping is imperfect because the plot is a fairly crappy JPG (hard to distinguish overlapping points). Dashed line is 156 particles per liter, nominal normal/superspreader cutoff (points fall on either side due to scraping error)
November 9, 2025 at 5:46 PM
Scraped data from the plot ... gist.github.com/bbolker/d6a6... Scraping is imperfect because the plot is a fairly crappy JPG (hard to distinguish overlapping points). Dashed line is 156 particles per liter, nominal normal/superspreader cutoff (points fall on either side due to scraping error)
deep cut! `fortunes::fortune("dog")`: "Firstly, don't call your matrix 'matrix'. Would you call your dog 'dog'?
Anyway, it might clash with the function 'matrix'." -- Barry Rowlingson, R-help (October 2004)
Anyway, it might clash with the function 'matrix'." -- Barry Rowlingson, R-help (October 2004)
November 7, 2025 at 3:06 AM
deep cut! `fortunes::fortune("dog")`: "Firstly, don't call your matrix 'matrix'. Would you call your dog 'dog'?
Anyway, it might clash with the function 'matrix'." -- Barry Rowlingson, R-help (October 2004)
Anyway, it might clash with the function 'matrix'." -- Barry Rowlingson, R-help (October 2004)
I think you have an extra close-parenthesis there: `gg0 <- ggplot(mtcars, aes(wt, mpg)) + geom_point(size=2) + theme_minimal(); gg0 + annotate(geom='text', x=4, y=32.75, size=6, label= "hat(beta)==(X^t*X)^{-1}*X^t*y", parse = TRUE)` works for me ...
November 3, 2025 at 9:30 PM
I think you have an extra close-parenthesis there: `gg0 <- ggplot(mtcars, aes(wt, mpg)) + geom_point(size=2) + theme_minimal(); gg0 + annotate(geom='text', x=4, y=32.75, size=6, label= "hat(beta)==(X^t*X)^{-1}*X^t*y", parse = TRUE)` works for me ...
for `annotate`, this seems to work: `annotate(geom='text', x=4, y=32.75, size=6, label="'Concentration of'~CO[2]", parse = TRUE)`
November 3, 2025 at 8:53 PM
for `annotate`, this seems to work: `annotate(geom='text', x=4, y=32.75, size=6, label="'Concentration of'~CO[2]", parse = TRUE)`
try the revised version at gist.github.com/bbolker/5ba6... ? (tl;dr we now need `sapply(s, function(x) parse(text=x))` rather than `parse(test=s)`. I expect this should be back-compatible with ggplot2 < 4.0.0 but haven't tested ...)
scientific notation with plotmath superscripts in ggplot
scientific notation with plotmath superscripts in ggplot - scientific_10.R
gist.github.com
November 3, 2025 at 8:12 PM
try the revised version at gist.github.com/bbolker/5ba6... ? (tl;dr we now need `sapply(s, function(x) parse(text=x))` rather than `parse(test=s)`. I expect this should be back-compatible with ggplot2 < 4.0.0 but haven't tested ...)
Perhaps OP was referring to this (somewhat) famous quotation?
Singer & Pincus. 1998. “Irregular Arrays and Randomization.” PNAS. doi.org/10.1073/pnas....
Singer & Pincus. 1998. “Irregular Arrays and Randomization.” PNAS. doi.org/10.1073/pnas....
November 3, 2025 at 1:59 AM
Perhaps OP was referring to this (somewhat) famous quotation?
Singer & Pincus. 1998. “Irregular Arrays and Randomization.” PNAS. doi.org/10.1073/pnas....
Singer & Pincus. 1998. “Irregular Arrays and Randomization.” PNAS. doi.org/10.1073/pnas....
OK, a bigger challenge ... (I haven't thought this through so maybe it doesn't make sense ...) what is the *longest* initial homogeneous binary sequence that arises from the 4.2e9 (= .Machine$integer.max*2) possible 32-bit integer seeds ... ?
November 3, 2025 at 1:53 AM
OK, a bigger challenge ... (I haven't thought this through so maybe it doesn't make sense ...) what is the *longest* initial homogeneous binary sequence that arises from the 4.2e9 (= .Machine$integer.max*2) possible 32-bit integer seeds ... ?
did you find these seeds by thinking carefully about/reverse-engineering the Mersenne Twister implementation used in R, or by brute force, or ... ??
November 3, 2025 at 12:37 AM
did you find these seeds by thinking carefully about/reverse-engineering the Mersenne Twister implementation used in R, or by brute force, or ... ??
thanks! this looks good github.com/plotly/plotl... (Posit will be supporting a migration of the documentation, and support for the R plotly package will continue ...)
Plotly is retiring its R documentation · Issue #2456 · plotly/plotly.R
I am opening this issue as a potential avenue for some thoughtful, respectful community discussion about the current state of the plotly R package, it's relationship and support by the Plotly organ...
github.com
November 2, 2025 at 12:35 AM
thanks! this looks good github.com/plotly/plotl... (Posit will be supporting a migration of the documentation, and support for the R plotly package will continue ...)
what do you think about these options bsky.app/profile/bbol... ? (Also: etherpad.org Apparently also www.authorea.com (owned by Wiley), although amusingly this was one of the first public documents that came up ...)
November 1, 2025 at 11:55 PM
what do you think about these options bsky.app/profile/bbol... ? (Also: etherpad.org Apparently also www.authorea.com (owned by Wiley), although amusingly this was one of the first public documents that came up ...)
oh, now I see that's a derivative ...
November 1, 2025 at 11:15 PM
oh, now I see that's a derivative ...
my only issue with git+quarto is that it's not live, which (1) makes collaboration harder [git conflicts, ugh] and (2) makes things harder for non-technical collaborators (nice to enable them to edit in a friendly WYSIWYG web front-end ...)
Agree with your point about Google.
Agree with your point about Google.
November 1, 2025 at 10:46 PM
my only issue with git+quarto is that it's not live, which (1) makes collaboration harder [git conflicts, ugh] and (2) makes things harder for non-technical collaborators (nice to enable them to edit in a friendly WYSIWYG web front-end ...)
Agree with your point about Google.
Agree with your point about Google.
collab editing: www.overleaf.com (pricing model annoys me, LaTeX-centric, now pushing AI assist [ugh] - although I like the Git backend); hackmd.io (markdown-specific); google docs; typst.app/docs/web-app/ (still fairly new) ?
HackMD: Your Collaborative Markdown Workspace for Knowledge Sharing
HackMD gives you a real-time Markdown editor for collaborative work. Working with Markdown files in HackMD is simple, straightforward, and fun.
hackmd.io
November 1, 2025 at 10:42 PM
collab editing: www.overleaf.com (pricing model annoys me, LaTeX-centric, now pushing AI assist [ugh] - although I like the Git backend); hackmd.io (markdown-specific); google docs; typst.app/docs/web-app/ (still fairly new) ?
Agree that these are horrible. What is your preferred set of workspace tools (especially for collaborative writing/commenting, where Git(hub)+quarto/markdown is awkward)? Google stuff? Git? Discord/Slack/Basecamp/Zulip?
November 1, 2025 at 10:27 PM
Agree that these are horrible. What is your preferred set of workspace tools (especially for collaborative writing/commenting, where Git(hub)+quarto/markdown is awkward)? Google stuff? Git? Discord/Slack/Basecamp/Zulip?
B' includes evolution of resistance? (I really like Bonhoeffer, et al 1997. www.ncbi.nlm.nih.gov/pmc/articles... for this) ... also Yates et al 2007 doi.org/10.1371/jour... (grad students found a minor error while working through the paper in class, authors were very gracious!)
Human immunodeficiency virus drug therapy and virus load
Analysis of the short-term dynamics of human immunodeficiency virus (HIV) type 1 infection in response to drug therapy has elucidated crucial kinetic properties of viral dynamics in vivo (D. D. Ho et al., Nature 373:123-126, 1995; A. S. Perelson et ...
www.ncbi.nlm.nih.gov
November 1, 2025 at 9:44 PM
B' includes evolution of resistance? (I really like Bonhoeffer, et al 1997. www.ncbi.nlm.nih.gov/pmc/articles... for this) ... also Yates et al 2007 doi.org/10.1371/jour... (grad students found a minor error while working through the paper in class, authors were very gracious!)
¿Maybe? it's being cannibalized by glmmTMB's ordbeta capabilities? (Could be more attractive to analysts who want ordbeta without taking on the hassle of Bayesian machinery?) ATM "glmmTMB ordbeta" gets 31 GScholar hits (20 from 2025), "ordbetareg" gets 51 (28 from 2025) ...
November 1, 2025 at 9:02 PM
¿Maybe? it's being cannibalized by glmmTMB's ordbeta capabilities? (Could be more attractive to analysts who want ordbeta without taking on the hassle of Bayesian machinery?) ATM "glmmTMB ordbeta" gets 31 GScholar hits (20 from 2025), "ordbetareg" gets 51 (28 from 2025) ...
thanks, interesting and useful. I will look into github.com/HannaMeyer/C... ...
GitHub - HannaMeyer/CAST: Developer Version of the R package CAST: Caret Applications for Spatio-Temporal models
Developer Version of the R package CAST: Caret Applications for Spatio-Temporal models - HannaMeyer/CAST
github.com
October 31, 2025 at 10:26 PM
thanks, interesting and useful. I will look into github.com/HannaMeyer/C... ...
(3/2, sorry) on the other hand Mushagalusa et al 2024 doi.org/10.1186/s400... muddies the water somewhat (spatial blocking not needed if sampling is random or systematic (does that ever happen?) ?) Do you have thoughts/opinions/suggestions for further reading?
Random forest and spatial cross-validation performance in predicting species abundance distributions - Environmental Systems Research
Random forests (RF) have been widely used to predict spatial variables. Several studies have shown that spatial cross-validation (CV) methods consistently cause RF to yield larger prediction errors co...
doi.org
October 31, 2025 at 12:27 AM
(3/2, sorry) on the other hand Mushagalusa et al 2024 doi.org/10.1186/s400... muddies the water somewhat (spatial blocking not needed if sampling is random or systematic (does that ever happen?) ?) Do you have thoughts/opinions/suggestions for further reading?