I'm getting ready for the kickoff of an #NSF POSE grant workshop. So many good papers to reference, so little time. I'd better start a Zotero library now... #OpenSource#Standards#Metadata
NASA SpaceFlight #NSF had a very interesting live interview with Travis Brown, PhD, #Ingenuity's current Chief Engineer. Travis said that in spite of an original though/assumption that Ingenuity's rotors had hit the ground during landing of #Flight72, they now tend to believe that it was rather a very hard landing that damaged the rotors, since there is no evidence anywhere on the regolith of a blade hit.
Can you imagine the amount deadweight loss due to #SciENcv (and ever changing #NSF biosketch and CPS formats in general)? These process changes do not help proposers, and it is unclear they help any panelists either. I am sure it is all about audit trails. So unfortunate.
Update on the National Science Foundation (#NSF) stance on #AI in review and proposal writing.
"A key observation for reviewers is that sharing proposal information with generative AI technology via the open internet violates the confidentiality and integrity principles of NSF's merit review process."
From the summary:
"Scientific articles that are freely available to the public grew by more than seven-fold in volume and increased in influence relative to those behind #paywalls from 2003 to 2022…In 2022, 51% of articles, of which status we know, were in #OpenAccess journals compared to 23% in 2003."
Of all the low down scummy #bank#fees, I think I hate the NSF fee the absolute worst.
Back in the day, it made sense when electronic transactions didn't exist and while cheques could be routed electronically due to their routing number, a human still had to process it. There was cost associated with the bounce.
But $48 for dipping $2 into #overdraft on a $15 purchase these days? Cmon. It costs the bank nothing (ok a few electrons of server power). The true definition of a #tax on the #poor.
If I ran a bank, #NSF#fees (if not abolished entirely):
would only be applied 24 hrs after the infringing transaction if the account is still negative - a $48 penalty because an account transaction came in an hour late = lame.
would scale with the size of the overage. The penalty for overdrafting by $1 << being short $100.
would be charged to a special NSF account. This gets clawed back interest free at <low % rate> from your main account only when you have balance >> <$threshold>.
Sad to have to miss #FITNG2023, but the fantastic Kira Ashton Benazir Neree, & Dany Olivares made it! 🎉
➡️ Benazir received a #FITNG2023
Young Investigator Award, and presents poster 1-D-35 now!
➡️ Kira is presenting poster 1-B-12, also now!
Come say hi! 👋
➡️ Benazir's poster 1-D-35 compares looking & EEG-decoding measures of infants' visual representations, w/ Kira Ashton, Dany Olivares,
Alexis Black, Lauren Emberson, Dick Aslin, & others. This work is based on early data from our ongoing #NSF funded project on the infant visual brain 🧠
A company creating a digital process to monitor farm insect pests is receiving funds to expand its technology for tracking and counting mosquitoes as a general public health threat.
A digital health company with an app for monitoring maternal health is receiving nearly $1 million to add an algorithm for detecting a dangerous pregnancy complication.
US’ #NIH Australia’s #ARC : 'NO’ to ChatGPT for peer-review
US' #NSF and Europe's #ERC mulling it over with working groups.
Concerns:
*Privacy/Piracy “the information becomes part of its training data. ” (why I don’t chatGPT though I really hate writing)
*Error ”AI-written reviews will be error-prone"
*Bias ”against non-mainstream views”
*Boring "lack … creativity “
I'm lurking in a National Science Foundation webcast giving advice to researchers writing proposals, and the answer to the audience question "Any recommendations about wording the broader impacts section for those of us in states with anti-DEI legislation?"
is "We have no idea".
This could be a disaster for science in red states, if we are no longer competitive for federal funding.
RT @quansightai: Now available >> @pandas_dev in the geosciences
See how Quansight and @USC joined forces to (finally) make Pandas available for use in the paleogeosciences with non-nanosecond-resolution datetimes. #NSF#pandas#USC
#NSF 📆 May 9, 2023 Though Ship 25, equipped with tiles and fins, has been extensively tested, and would appear to be closer to flight readiness than Ships 28 or 29, it was constructed last year. As a result, #SpaceX might want to use a newer #Starship with design improvements and learnings from last month’s launch.
Ship 26, not equipped with tiles or fins, has been seen with the #Raptor stand underneath it at the “rocket garden” storage area. Some preparations appear to have been made to the vehicle, and it is still a possibility it could be chosen to fly next. Ship 27, similarly equipped to Ship 26, is also in the storage area, and still could be prepared for a future flight. https://www.nasaspaceflight.com/2023/05/starbase-repairs-ships-future-flights
"the #mainstream media doesn’t cover #SpaceExploration and #SpaceScience with the same questioning 🔍 rigor that they reserve for politics. People writing about #SpaceExploration are mostly cheerleaders 🥳 for the cause, rather than independent observers keeping a watchful eye 👀 on how our national monies are spent."
"For the moment, we recommend that if #LLMs are used to write scholarly reviews, reviewers should disclose their use and accept full responsibility for their reports’ accuracy, tone, reasoning and originality."
PS: "For the moment" these tools can help reviewers string words together, not judge quality. We have good reasons to seek evaluative comments from human experts.
Apart from #quality, one concern is #confidentiality. If grant proposals become part of a tool's training data, there's no telling (in the NIH's words) “where data are being sent, saved, viewed, or used in the future.”