Philipp Birken / is on Nostr

Mathematician at Lund University, works on Scientific Computing for a sustainable world.

I used to be very active on the german Wikipedia, and in Wikimedia Deutschland, with a focus on mathematics and quality control.

Website: https://www.maths.lu.se/english/research/staff/philipp-birken/

Blog (not really updated): https://birchlog.blogspot.com

Last Notes

2024-11-14 16:49:49

- reply

@nprofile…gw7s The whole notion of software as infrastructure still needs to be developed, both in government and academia. Academia needs to adjust criteria for promotion so as to no longer disincentivize software maintenance and development. There needs to be a middle infrastructure layer between email and running an HPC center that is about software development. At the university, national and EU level.
The Research Software Engineering "movement" in the UK is great. More of that!

2024-11-14 16:49:49

- reply

2024-11-14 07:44:43

- reply

@nprofile…gw7s Europe in particular is spectacularly bad at supporting scientific software development and maintenance. This is a prime example of it. There should be (at leat) one european climate model that is considered Infrastructure and funded accordingly and in a long term way.
The US national labs are luckily different, but the election is again proof that relying on them is problematic.

2024-10-23 06:16:45

- reply

@npub14k2…ed68 I don't get it anymore. Isn't the ability to vote of 2.5 million people in North Carolina questionable? That's 16 electoral votes that are unpredictable. Isn't this the whole story of the election or am I missing something?

2024-10-22 18:31:27

- reply

@npub14k2…ed68 In Germany, one would start a fight now. In Sweden, one would sulk. What's the danish way? Getting new and even better ideas and better collaborators?

2024-08-04 09:30:03

Very nice article in #SIAM News by one of the world leading researchers on the #mathematics of #AI, Gitta Kutyniok, on the Mathematics of Reliable Artificial Intelligence. What sources of error are there, what are major positive and negative results, what could the “right to explanation” in the EU AI act mean mathematically?
https://www.siam.org/publications/siam-news/articles/the-mathematics-of-reliable-artificial-intelligence

2024-07-27 07:23:21

I had no idea, this really sucks!!
Taylor & Francis feeds academics’ work to Microsoft AI – Pivot to AI
https://pivot-to-ai.com/2024/07/20/taylor-francis-feeds-academics-work-to-microsoft-ai/

2024-07-21 20:50:19

Marsrover Curiosity should be on Mastodon, and not on X, right?

2024-06-17 14:53:16

At the Nordic Numerical Linear Algebra workshop, this year in Odense, Denmark. The second keynote by Alex Townsend from Cornell University via zoom. Worked very well, and was sustainable.
He says that you don't meet the Kahan matrix in a dark alley. Or otherwise put: Many worst case estimates are not representative of what happens in practice.
Also a very nice presentation of methods to compute the numerical rank of matrices.
#NordicNLA
#NumericalLinearAlgebra

2024-06-02 13:02:39

Mir ist da am Hamburger Hauptbahnhof was zugelaufen. @npub1quy…kauq
Bin gespannt!
https://fediscience.org/system/media_attachments/files/112/547/154/761/391/547/original/137eb1c882ee26de.jpg

2024-05-23 13:49:28

- reply

@npub1mhz…wteh Does anybody have experiences from hub conferences? And possibly a written report that one can use to push the idea?