catgary

7 ms·

61.
by catgary 2y ago
See, I find it the opposite - superficially fussy, but it lets the silliest things slide. I don’t want my function to struggle with type instability in some contexts (perhaps only showing up deep in a physics simulation or motion planning a
62.
by catgary 2y ago
I’m kind of sad, because I think I have completed the journey from advocating for Julia at my first job to becoming a full-blown hater. I have complaints about the type system and the general failure of a good autograd library to emerge as
63.
by catgary 2y ago
I kind of wish the effort spent on Roc was put into productionizing/commercializing Koka (which Roc draws a lot of inspiration from) rather than starting an entirely new language. Like I get starting from scratch when it comes to Haske
64.
by catgary 2y ago
Look, I agree that someone pitching category theory to programmers should make more of an effort to introduce terms/use common language when possible. And you can look at Gordon Plotkin’s papers to find some good, hard CS papers that t
65.
by catgary 2y ago
It might help the programmer but confuse the quantum physicist or algebraic topologist. It’s an interdisciplinary field and we do our best (well, I’m out of it so it’s not really “we” anymore). The abstract differentiation stuff that makes
66.
by catgary 2y ago
In general, I’d say there’s too much superficial flexibility but not enough control. - I wrote this elsewhere: I find their approach to memory management/mutable arrays really hits the worst of both worlds (manual memory management and
67.
by catgary 2y ago
So, I’ll admit that I’m not a fan of multiple dispatch a la Julia. I much prefer typeclasses and explicit union types. Also, I found Julia is really the worst of both worlds of garbage collected languages and manual memory management becaus
68.
by catgary 2y ago
I get it, but once you’ve compiled your program down to a computational graph…well those are basically morphisms in a monoidal/multi category, no? And then you can have a stack of MLIR dialects that represent different levels of abstra
69.
by catgary 2y ago
I’m sorry, but I’m going to disagree with you on that. Can you point to any of the language designers who had a background in programming language theory? The closest thing I see is Bezanson’s work on technical computing, which seems laser-
70.
by catgary 2y ago
Julia was a scientific computing language made by scientific computing experts. They did a great job on some things, but whiffed a few major decisions early on.
71.
by catgary 2y ago
Diffusers is pretty well done (I think the author was basically rewriting some Julia libraries and adapting them to JAX). I can’t imagine it being too hard to adapt most SciML ODE solvers. For simulations, JAX will choke on very “branchy” c
72.
by catgary 2y ago
The ODEs stuff in Julia is nice, but I think diffusers/JAX is a reasonable backbone to copy over whatever you need from there. I do think Julia is doing will in stats and has gotten some mindshare from R in that regard. But I think a r
73.
by catgary 2y ago
Yeah, Jax with Equinox, jaxtyping, and leaning hard on python’s static typing modules + typeguard lets you pretend that you have a nice little language embedded in python. I swore off Julia a few years ago.
74.
by catgary 2y ago
Julia has really lost the differentiable programming mindshare to JAX. I’ve spent weeks or months getting tricky gradients to work in Julia, only to have everything “just work” in JAX. The quality of the autograd is night and day, and goes
75.
by catgary 2y ago
This is a nice project, but “You only linearize once” is more-or-less the type theory version of “Reverse Derivative Categories”, so JAX really does this already.
76.
by catgary 2y ago
The goal is to probably avoid duplicating efforts when llvm-symbolize already exists. There’s obviously a snarky comment to make here about Go developers and duplicating efforts.
77.
by catgary 2y ago
I am an ML scientist, my company and several others are using IREE to deploy our models to edge devices. It is the most promising technology in this area. Those academic publications are a sign that the people involved actually know what th
78.
by catgary 2y ago
AMD already has major ongoing projects with OpenXLA/IREE. Lots of established engineers/researchers, and it’s in collaboration with Google/AWS. Hotz is delusional if he thinks that he can do better by ripping off Karpathy’s t
79.
by catgary 2y ago
Yeah, AMD is already pouring a lot of support into OpenXLA/IREE, which has a lot of well-respected compiler engineers and researchers working on it, and companies like AWS are also investing into it. I don’t really think TinyCorp has a
80.
by catgary 2y ago
An arbitrarily nestable list is a tree, no?
81.
by catgary 2y ago
Yeah, this was my general impression after a brief, disastrous stretch in robotics after my PhD. Hell, I work in animation now, which is a way easier problem since there are no physical constraints, and we still can’t solve a lot of the pro
82.
by catgary 2y ago
Provincial governments were a mistake - the average Canadian fundamentally misunderstands the division of powers and responsibilities between federal and provincial governments, and this is remarkably useful for bad actors. I personally thi
83.
by catgary 2y ago
The convoy was largely funded by right wing American groups, so it’s not surprising they received favourable media coverage.
84.
by catgary 2y ago
Ehh, I think a surprising amount of the Quebecois’ problems were self-inflicted by letting the Catholic Church run people’s lives, and the Quiet Revolution helped a lot. Like, it wasn’t the anglos bullying people’s grandmothers into having
85.
by catgary 2y ago
I love Bill Lawvere, but he punched the president of his university in the face, he was genuinely lucky to find a position in academia afterwards.
86.
by catgary 2y ago
Are you talking about the militant separatists who had already committed mailbombings and escalated to assasinating a government official and kidnapping a foreign diplomat on Canadian soil in 1970? It’s also important to note that afterwa
87.
by catgary 2y ago
If someone tried this protest in the US someone would have been shot night one.
88.
by catgary 2y ago
Yeah are we forgetting Alberta truckers who were planning to murder RCMP officers, and were straight-up terrorists?
89.
by catgary 2y ago
There had already been one standoff between local residents and truckers, I remember there being chatter that the next weekend groups were going to coordinate in their neighborhoods and drive out the convoy on their own (using baseball bats
90.
by catgary 2y ago
I think it’s worth noting that the weekend before this, local residents of Ottawa had basically “stormed” one of the trucker convoy camps to unblock a road. There were genuine concerns that the residents of Ottawa were ready to take matters

More ›