ssivark

62 ms·

781.
by ssivark 6y ago
In answer to the post's lament that popular & well-regarded languages seem not to emerge from academic work, I'd like to point out that the Julia programming language (released in only 2012, around the time of this post) whose
782.
by ssivark 6y ago
Taking a step back, mathematics just happens to be our most developed tool/perspective for understanding patterns . So, to the extent that programming languages are concerned with modeling certain patterns in programs or domains or
783.
by ssivark 6y ago
Sounds fascinating! (I have a background in theoretical physics) How do you go about finding the right kind of projects to work on, or the right kind of environment that appreciates what you bring to the table? And how often do you come acr
784.
by ssivark 6y ago
Thanks for a lucid explanation of the underlying mess. I now have a much better perspective. PS: I particularly enjoyed the unification/gravity analogy btw :-D >.<
785.
by ssivark 6y ago
GPUs are typically useful for training (due to massive parallelism), but not for inference .
786.
by ssivark 6y ago
They’ve indicated a willingness to sponsor a million dollars a year and set up a committee. The committee may decide that the best way to that goal is to direct that funding towards a community fork led by one of the original maintainers of
787.
by ssivark 6y ago
I hope they find a way to join forces, and sooner the better.
788.
by ssivark 6y ago
”[...] The issues for writings are not necessarily the same as those for software. For example, I don't think you or I are obliged to give permission to modify articles like this one, which describe our actions and our views. But ther
789.
by ssivark 6y ago
To rephrase... because the models predicting model failure (often) use the same assumptions/knowledge as the original models, so they both fail at the same time once assumptions become invalidated. "intertemporal heterogeneity in
790.
by ssivark 6y ago
I wonder if this opens the platforms to adversarial use cases, such as gate crashing the Q&A session with such keywords, eventually spoken out at some ultrasound frequencies so that humans can’t hear it, but machines will (if fed raw au
791.
by ssivark 6y ago
While your comment on biases is correct, “sloppy” is a bit harsh; I would think of this as a best-effort attempt to reach out to the community through known popular channels. I’m sure there are tons of Emacs users who go about their work wi
792.
by ssivark 6y ago
Please say more!
793.
by ssivark 6y ago
While I agree with the argument that it’s better to let the semantics drive the formatting, I wish we could get over formatting discussions already, with the codebase recording the structure, and the displayed format being a matter of edito
794.
by ssivark 6y ago
The talk by Sid Kasuvajhula is an absolutely fascinating presentation about extremely flexible object-oriented interfaces, organized in a multi-hierarchial manner -- so very nicely exploiting the compositional structure inherent in the cont
795.
by ssivark 6y ago
I appreciate the challenge they face, and this attempt at a solution. If their retail arm does succeed better than anticipated, that would be a welcome bonus and richly deserved success that could augment the FOSS investment. In a sense thi
796.
by ssivark 6y ago
Or, to play a recent meme, “no coding today... because us-east-1 is down.”
797.
by ssivark 6y ago
> but at this point in history I can go and learn about pretty much all of it whenever I want. You could learn any of it, but most certainly not all of it. And that’s a subtle but crucial difference. As the number of pieces/inte
798.
by ssivark 6y ago
> And this isn't a kind of scientific complexity like how biology is and always has been too complex for us to understand. This is the things that you and I use and rely on being too complex to understand. The same is true for biolo
799.
by ssivark 6y ago
That kind of caching is an interesting solution to incrementally building a database instead of spending hours up-front indexing. So the tool is ready for immediate use. Quite nifty :-)
800.
by ssivark 6y ago
I love that we’re seeing fast & flexible solutions for personal search. I’ve recently been playing with Recoll for full-text-search on content. Since it indexes content up front, the search is pretty fast. It can also easily accommodate
801.
by ssivark 6y ago
To add... > ”the key observation is that as long as you are right, being misunderstood by most people is a strength not a weakness.” This is (implicitly) only applicable to competitive/adversarial endeavors and not collaborative e
802.
by ssivark 6y ago
Official report from the NSF: https://www.nsf.gov/news/news_summ.jsp?cntn_id=301737
803.
by ssivark 6y ago
Sounds super exciting; thanks for pushing the boundaries on something new and potentially so much better! Please don’t get discouraged by comments pointing out the current limitations of the platform — it just means that the core ideas are
804.
by ssivark 6y ago
For those who don’t know, check out the whole series of comics by Rakhim. They’re pretty good! :-D
805.
by ssivark 6y ago
Wow, astroid seems quite close to what I had in mind. I'll definitely give it a try -- thanks! Seems also like it is not actively maintained currently, but I'm hoping it's probably feature-complete and stable enough to be in
806.
by ssivark 6y ago
Is there anything that has a clean and responsive viewer that supports HTML/formatted email with images, with an efficient keyboard-driven interface like mutt/notmuch/etc? Ideally, I would like my mail storage to be separated
807.
by ssivark 6y ago
1/r^2 comes from the fact that space is three dimensional. No amount of beam collimating can change that (unless we have some fundamental breakthrough/change in our understanding of physics). Better beam coherence could keep impro
808.
by ssivark 6y ago
It’s about the second order effects of what it takes to keep an ecosystem healthy over a longer period. We might pay attention to individual “winners”, but it takes a healthy “scene” incorporating variety and scope for evolution & innov
809.
by ssivark 6y ago
For TV and radio there’s a finite resource (radio spectrum) which is then allocated by government permission. OTOH, it’s much easier to set up a website and publish some content or even host conversations, than it is to set up a TV station.
810.
by ssivark 6y ago
Sounds reminiscent of the data processing inequality https://en.wikipedia.org/wiki/Data_processing_inequality

More ›