jll29

37 ms·

661.
by jll29 2y ago
I was looking for something to teach 11-year-olds a couple of years ago, and ended up using Python with a turtle graphics library; this would have been another great alternative. Programming using a language like Lua has the advantage over
662.
by jll29 2y ago
How amazing is life! And once a mouse is conceived, it goes through 28 Theiler states. https://www.ncbi.nlm.nih.gov/projects/Jax_GSA/IMGC2005_Works... Nature orchestrates things perfectly.
663.
by jll29 2y ago
The main thing for journalists is to strictly separate news reporting from editorial opinion pieces and clearly mark which is which.
664.
by jll29 2y ago
> To be honest only BBC comes into mind when I think about a somewhat impartial medium. The BBC is pro-UK government, who fund them. To me only Reuters (disclaimer: I once worked there) and AP come to mind when I think impartial news. Wh
665.
by jll29 2y ago
While we're at it, I drove a small amount of Normandy Camembert through the Northern part of Alsace today, but it was strictly for personal consumption only and legally obtained from Carrefour. Mystery how folks could spend GBP 11 for
666.
by jll29 2y ago
If anyone needs a good (related) startup idea, here's one for free: Complex trip planning for professionals, but in a different way from Ambulate - not hiking trips, but transactions across Web sites: I really hate the way how today yo
667.
by jll29 2y ago
But publishing a public domain paper with a commercial publisher still makes little sense, unless the paper is also _open access_, for without the latter, the public domain document could still be put behind a pay wall. Just because it is o
668.
by jll29 2y ago
“If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention,” I am not a layer not patent attorney, but I have extensive experience as a former
669.
by jll29 2y ago
It would be quite exciting to train a LLM with (OCR scans of) all mathematical journals, pre-prints, time-series etc.
670.
by jll29 2y ago
Most pre-deep learning architectures had separate modules like "language model", "knowledge base" and "inference component". Then LLMs came along, and ML folks got rather too excited that they contain implicit
671.
by jll29 2y ago
links?
672.
by jll29 2y ago
1. GML => SGML => XML 2. rm * 3. JSON 4. rm -rf /
673.
by jll29 2y ago
I like the Rust way of telling the compiler driver with a command line switch which compiler version to run, then backward compatibility as a problem disappears.
674.
by jll29 2y ago
Interestingly, this "official" Pomodoro site doesn't offer a physical timing device on their shop page. It looks like the inventor of the technique is now based out of Dubai (allergic to taxes? The governments that you lean o
675.
by jll29 2y ago
It's a beautiful site and product. However, the original inventor of the Pomodoro technique explicitly advocates a "low tech" approach - a mechanical kitchen timer, because he argued that the tactile and auditory elements (i.
676.
by jll29 2y ago
How about designing a "proper" standard library for Rust (comparable to Java's or CommonLISP's), to ensure a richer experience, avoiding dependency explosions, and also to ensure things are written in a uniform interface
677.
by jll29 2y ago
This is beautifully written, thanks for sharing. I could see myself using some of the source code in the classroom to explain how transformers "really" work; code is more concrete/detailed than all those pictures of attention
678.
by jll29 2y ago
This is an amazing trove of human knowledge - if made digitally accessible, the titles should be on a Web page and the references crawled by Google Scholar. We should eventuall OCR all that stuff to use it to train LLMs. Seen from that comm
679.
by jll29 2y ago
Wishful thinking: I can't wait until 2050 when the ACM will post an article entitled "50 Years of Queries", which will be about the public release of the first 50 years of ACM Digital Library search queries.
680.
by jll29 2y ago
That's a diamond of a quote; are you a student of Kafka's?
681.
by jll29 2y ago
Before you make a deletion request, make a subject data request and see what they have on you; then request deletion; then make a subject data request again.
682.
by jll29 2y ago
Fines for non-compliance with GDPR regarding data of European citizens can amount to 4% of annual revenue: 83(5) GDPR, the fine framework can be up to 20 million euros, or in the case of an undertaking, up to 4 % of their total global t
683.
by jll29 2y ago
The task of devising a novel application model for distributed Web applications is left to the reader as an exercise. (It's not _just_ JavaScript that one should get rid off, but the abuse of the whole HTML+X set of stacks to implement
684.
by jll29 2y ago
I enjoyed how we got taught the two models in the 1990s, and why one has the four layers you need and the "standard" has seven layers instead. The professor asked "How many layers do you count?" - "Seven." - &q
685.
by jll29 2y ago
ChatGPT: roll up this thread, extract all author-title pairs that describe books, and look up the corresponding amazon.com links where I can buy the whole set.
686.
by jll29 2y ago
I second that. Its translations are also the best translations known to mankind (I read the German one, the French is said to be as good or better).
687.
by jll29 2y ago
Please, someone convert this to plain text.
688.
by jll29 2y ago
"Work" as such is not protectable; it lacks the creative element that e.g. a translation of a book into another language exhibits.
689.
by jll29 2y ago
I don't find the language of the article full of "hype"; they describe the history of different forms of imputation from single to multiple to ML-based. The table is particularly useful as it describes what the article is all
690.
by jll29 2y ago
There are plenty of legit. articles that discuss/survey imputation in ML/AI: https://scholar.google.com/scholar?hl=de&as_sdt=0%2C5&q=%22m...

More ›