ssivark

25 ms·

331.
by ssivark 3y ago
But if that’s the kind of research that racks up tens of thousands of citations, does that mean all the researchers in the field are simply LARPing?
332.
by ssivark 3y ago
Oof, that article is brilliant. Seldom do I see such an honest perspective, with penetrating insight into the essence* of the matter.
333.
by ssivark 3y ago
I have at least one complaint with the numpy model: When you chain a sequence of vectorized operations on arrays, loop fusion would save you from allocating memory for each intermediate variable, and the round trip time of moving it from RA
334.
by ssivark 3y ago
> In my "physics of biomedical imaging" bubble, people are hardly doing state-of-the-art ML, but rather expensive forward models for which computing a gradient is cumbersome. I’d appreciate any pointers to the literature; curio
335.
by ssivark 3y ago
Twiddla?
336.
by ssivark 3y ago
This article by Vannevar Bush inspired JCR Licklider to write his "Libraries of the future" [1] which in turn inspired the young researchers working on DARPA computing projects, who then went on to invent a large fraction of moder
337.
by ssivark 3y ago
> but to fill out the rest of the architecture matrix I have been personally buying GPUs. That's been sufficient for covering most architectures, but we will need a sponsor if we are to acquire CDNA 2 and CDNA 3 hardware. This seems
338.
by ssivark 3y ago
Self-hosting is definitely not easy, but it’s the indirection more so. You’re excited about X and want to gather a community around that, and you definitely don’t want to spend a lot of time on Y (mechanics of running a forum). It’s good
339.
by ssivark 3y ago
This is of course a very popular idea among economists (Eg. Paul Krugman has advocated something like this for a long time), because it allows them to have their cake and eat it too. But I fail to see any shred of evidence backing it up — i
340.
by ssivark 3y ago
The spirit of the claim is to reduce the number of expressions by 10x while only sacrificing the “unnecessarily complicated” functionality. It is a question of system design that demands judgement, and not just a raw code length compression
341.
by ssivark 3y ago
> applied AI in areas of marketing, development, technology. Those topics seem too broad (and shallow?) for anyone to curate the firehouse effectively. There’s too much noise unless one is more selective.
342.
by ssivark 3y ago
> pretty sure beats both Julia and Mojo: Would the C compiler automatically exploit vectorized instructions on the CPU, or loop/kernel fusion, etc? It’s unclear otherwise how it would be faster than Julia/Mojo code exploiting s
343.
by ssivark 3y ago
Are you talking about static typing? If so, that sounds fundamentally incompatible with the basic worldview behind Julia — that you want to delay binding/dispatch till runtime, and leverage multiple dispatch, for high composability &am
344.
by ssivark 3y ago
Tysm; this is good stuff for me to ruminate on. One of the challenges for me has been switching to this very empirical orientation (I’m far more used to a conceptual understanding driven mode). If & when you do get around to fleshing o
345.
by ssivark 3y ago
> Experiment noise has taught me so much over several thousands of manually-run (yes, I'm a masochist, it's a personal preference) experiments converging in a few seconds. Is there any resource that makes legible how to go abou
346.
by ssivark 3y ago
Nice example. At that point there's a subtle shift in emphasis from quantum gravity as the subject of the article to " my experience talking with a leading quantum gravity researcher" as the subject of an article. What the
347.
by ssivark 3y ago
IIUC, the nostr communication protocol stands on its own, which should allow you to “bring your own payments” to the mix, or even engage without hooking up any payments till you see a compelling need?
348.
by ssivark 3y ago
If a link is established, would it mean that workers who get tinnitus correlated with wearing noise canceling headphones all day to focus in “open office” environments might have standing to claim damages/recompense under OSHA?
349.
by ssivark 3y ago
And what kinds of tools do you think will be used to the assemble data center racks where you wish to “upload” yourself? :-P
350.
by ssivark 3y ago
> Within our lifetimes, books with every person in town's name, number, and address used to be delivered to everyone's home. And our opinions towards that changed precisely because some people started abusing that data to do ta
351.
by ssivark 3y ago
It’s illegal, in many places, to sell your vote. Extending that example to a general principle — if you have any decision making power/responsibility over others, then allowing it to be subverted (for pecuniary or other benefit) is a d
352.
by ssivark 3y ago
> The real solution is not to homeschool. It is to force ALL kids, regardless of age, location, and income to be forced into the same school system. Want to improve things? gotta make it better for everyone. But what if there is a fundam
353.
by ssivark 3y ago
It most definitely is — I don’t think I’ve stated anything to the contrary?
354.
by ssivark 3y ago
Yes, because somebody might innovate an education model where administrative salaries are 90% of the cost but students learn well, and earn well. An 8% cap would close off such avenues for progress. In general (going back to basic cyberneti
355.
by ssivark 3y ago
Does this have Julia bindings?
356.
by ssivark 3y ago
Thanks for the response, and thanks for all the effort towards an awesome WM :-) I can totally appreciate your priorities. My use of Wayland (and for a lot of users , I suspect) is mostly just flowing with all the upstream distribution defa
357.
by ssivark 3y ago
Is there a Wayland compatible WM analogue of Awesome?
358.
by ssivark 3y ago
On a related note, folks who found this interesting might also like to read “There’s plenty of room at the top” where computer scientists showed how orders of magnitude performance is left unused when a naive high-level implementation is no
359.
by ssivark 3y ago
In a field as young as software, three decades is too long for almost any software tool to remain frozen. Many new ideas might have built up — especially as the world around you (and needs) changes. You might keep writing extensions for you
360.
by ssivark 3y ago
Meditating on the converse statement is also an interesting thought exercise: A vector is (just) a (cached) function (evaluation).

More ›