Stories, tidbits and shark from a San Jose local

https://doi.org/10.5281/zenodo.23353583

Software which promises to let you write slides or docs as code often delivers excellent experiences when rendered to HTML, in exchange for some tedium in codifying layouts in plain-text. But what happens when the output is in the formidable PDF format? In this RAM talk I contend that the ecosystem for reproducibly generating Latex PDFs of docs or slides is still missing out on some of the “it just works” feeling that web-native content workflows seem to have.

https://doi.org/10.5281/zenodo.23353583

I’ll share my experiences cobbling together a sort-of RSE career path in an institution that didn’t really know what to do with me…

https://doi.org/10.5281/zenodo.23353583

Caring for a dog may not be human-centered, but it is user-centered. I’ll share what my reactive and anxious dog has taught me about challenging assumptions, engaging community, and meeting users where they’re at — dogs and humans alike.

https://doi.org/10.5281/zenodo.23353583

Coding agents have divided the research software engineering community. RSEs who value the craft of writing code tend to resist them, and those who mostly want what the code produces tend to welcome them. Both camps frame trust as a question of authorship (who wrote the code, and whether anyone read it). That framing does not scale to an autonomous software factory, in which agents plan, write, test, and review code from a specification faster than people can read it.

Computing and science already depend on tools their users did not build and do not inspect. Few programmers read the assembly their compilers emit, and astronomers do not grind their own telescope mirrors; in both cases trust rests on calibration, testing, and reproducibility. We argue that a software factory is an instrument of the same kind, sitting between the RSE and the code. Agent-written scientific code can then earn trust on its merits rather than on who wrote it, and the argument asks no one to take a model’s output on faith. Those merits are measurable, through validation against problems with known answers, deterministic simulation testing (which replays any failure from the seed that produced it), and independent reruns that reproduce the result. Within an organization that adopts a factory, the RSE works as its instrument scientist (specifying it, calibrating it, and deciding when its output is fit for use).

https://doi.org/10.5281/zenodo.23353583

Wanna hear some entertaining stories from an RSE perspective? I shall regale you with them! For a few minutes, at least.

https://doi.org/10.5281/zenodo.23353583

How does the leadership profile change as organizations mature, and what does US-RSE have to do to continue on its maturity path?

https://doi.org/10.5281/zenodo.23353583

We like streaming TB-sized HDF5 and Zarr assets hosted on S3 (AWS)

But how fast are the most common default access methods??

Get a sneak peek of some of our benchmark results!

https://doi.org/10.5281/zenodo.23353583

As technologists, programmers, and research software engineers, we often look to scientific and engineering-oriented literature to predict how artificial intelligence will change our work, our relationships, and our society; but the fields of literature, film, and television offer several well-reasoned opinions and predictions in the futurism genre. In this presentation, we examine some unexpected examples of futurism in fictional literature, film, and television to find out what there is to learn about science, technology, and society in the age of AI

https://doi.org/10.5281/zenodo.23353583

Blast is a real-time astrophysical transient data processing system that is generating and serving a growing set of data. We are prototyping a storage and distribution system based on IPFS that could pave the way toward a sustainable cyberinfrastructure that supports open science in the age of big data and short-term funding.

https://doi.org/10.5281/zenodo.23353583

Sitting and sugar is the new smoking. The solution, or at least an attempt, is to go on more S.M.O.K.E. breaks (with no smoking but walking instead).

https://doi.org/10.5281/zenodo.23353583

let’s explore the worlds deepest, oldest, and largest (by volume) freshwater lake!

https://doi.org/10.5281/zenodo.23353583