Dear all,

https://www.quantamagazine.org/ninth-dedekind-number-found-by-two-independent-groups-20230801/

Looks like a job for BOINC IMHO :)

Opinions ?

Best regards,

Samuel

Seem like BOINC would be a perfect fit. I’d crunch it. 😃

Just needs someone who understands the maths well enough and also has the programming skills and is at a university willing to fund it.

Well, Lennart Van Hirtum is studying at Paderborn University and Christian Jäkel at Dresden University of Technology.

What about a collaboration between these two German universities ? How much funding does a BOINC project need, in average ?

Some code by Lennart Van Hirtum is already available on Github: https://github.com/VonTum/Dedekind. If Christian Jäkel could contribute his code (based on that paper: https://arxiv.org/pdf/2304.00895.pdf), this BOINC project could offer two applications, one for each algorithm.

Opinions ?

Best regards, Samuel

How much funding does a BOINC project need, in average ?

I would be interested to know the budgets of some projects. It would need someone to produce and maintain the code, servers to send out tasks and receive and validate them. One day I should probably look at the server side of BOINC to find out more of what is involved. An average is probably not that meaningful given that some projects deal with much greater volumes of data than others.

I think a reasonable analogy would be the NumberFields@Home Project. That, again, is the research interest of a single, now retired, mathematician - Eric Driver. He seems to manage the steady-state operations of the project (submitting new batches of work, for example) by himself, and has time to read and respond to the (few) issues raised by volunteers on the project website.

I think the biggest hurdle for a small-scale project like that is the initial set-up. There are a huge number of different moving parts, each of which would probably be regarded as a separate discipline in a traditional installation: the BOINC data storage and management tools, the website, and the communications and security regime, to name but three. Professional support is probably needed at this point - I think Eric's former employers at the University of Arizona were vital here, and have continued to host his server in the university's infrastructure.

I think anyone thinking of setting up a similar project could learn a lot by seeking out the early posts on their website, and reading about the problems encountered and solutions found.