Jamie McCusker (she/her)

Data Science Researcher at RPI now available @[email protected]

[Jamie McCusker (she/her) @jpmccu]

@JanglesLad any tips for "media training" for scientists? So much of "Don't Look Up" could have gone differently if the scientists had some communication/media training.

[Judea Pearl @yudapearl]

Glad to finally see a paper on "explainability" https://t.co/mEwdpGbnMx that explains why explainability cannot be achieved through data-centric
thinking, but requires world knowledge about what is to be explained.

[Jamie McCusker (she/her) @jpmccu]

I'm getting kind of annoyed that I can't package binary files in #python distutils or setuptools so that a shell script can find them. data_files is ignored, and scripts wants plain text.

[Jamie McCusker (she/her) @jpmccu]

@devon_code @futureidentity @mentorafrika Knitting was the second programmable metamaterial (yarn was the first). A specific pattern is the program, but selection of the material really makes a big difference too.

[Jamie McCusker (she/her) @jpmccu]

@SethAbramson @threadreaderapp unroll

[Jamie McCusker (she/her) @jpmccu]

@xeimevta Huh, worldcat gives very different results via RPI. Maybe I got the title wrong.

[Jamie McCusker (she/her) @jpmccu]

@xeimevta So, that's great and all, but this seems to be a book that is out of print and not available in any university library in the western hemisphere. Are there any alternative resources you might suggest? I was hoping to actually read up on this.

[Lisa Amini @LisaAmini1]

Heartiest congratulations to @dlmcguinness and @jpmccu for receiving the Semantic Web Science Association Ten-year Award! Well-deserved!! #AIHN #AIHorizons https://t.co/AFFOd6jx3H

[Jamie McCusker (she/her) @jpmccu]

I got to do my best @johndankosky impression today for a discussion panel at our lab meeting. Thanks for showing us how it's done!

[Deborah McGuinness @dlmcguinness]

So happy to announce that my co-authored book on Ontology Engineering with Elisa Kendall is now available! thanks @MorganClaypool https://t.co/pvR8rL8GVZ #rpi #twcrpi #rpiidea @csatrpi

[Jamie McCusker (she/her) @jpmccu]

My TWed Talk from last week, called "Careful with that Predicate, Eugene". It will be rapidly evolving towards a position paper on ontology and knowledge graph evaluation and quality. https://t.co/o5Q7C2rMZz

[Jamie McCusker (she/her) @jpmccu]

@timrdf Is there a call for abstracts somewhere?

[Jamie McCusker (she/her) @jpmccu]

@karencoyle @timrdf The FRIR vocabulary borrows WEMI from FRBR itself and integrates PROV. I expand on this in my dissertation to make a digital signature framework that signs at the expression level. https://t.co/sDAG1bOnk5

[Olaf Hartig @olafhartig]

Dear PhD students, the purpose of the conclusions section of a research paper is *not* to conclude the paper with a summary of what you do in the paper (that's what the intro is for). Instead, it's purpose is to present the conclusions drawn from the work. What do we learn?

[Jamie McCusker (she/her) @jpmccu]

@olafhartig YES! The first thing I do when I review a paper, after reading the whole thing, is figure out what was concluded (in the conclusion), and evaluate backwards from there. Many papers have nothing to conclude, or claim the world but don't back it up.

[John S. Erickson 🐧 @[email protected] @olyerickson]

"Every day in America, about 104 people die from guns, while in Japan it takes about a decade for that many to die from gun violence." https://t.co/Fqki1N9RjE by @NickKristof via @NYTimes

[Jamie McCusker (she/her) @jpmccu]

@TaikaWaititi @chrishemsworth I feel like an opportunity was maybe missed WRT Thor's goat-pulled chariot.

[Jamie McCusker (she/her) @jpmccu]

@ericleamen @brielarson @MCUExchange @screenrant very Kree colors! I'm guessing this is early on, and she's still figuring things out.

[Jamie McCusker (she/her) @jpmccu]

@timrdf @fils You can throw in my dissertation too, which adds in a better graph digest algorithm and digital signatures: https://t.co/AxzoDPJihc

[Jamie McCusker (she/her) @jpmccu]

So who knows how to implement priorities for real in Celery 4? I'm using redis, but that can change for good reasons. The documentation is vague, at best.