kannanvijayan

12 ms·

91.
by kannanvijayan 1y ago
Oh I understand the problem. But the real issue isn't some external replacement, which is what tends to get focused on. The real problem is that it's happening from the inside out, and it's happening everywhere. The reality
92.
by kannanvijayan 1y ago
It's playful language in observation, not prescriptive. The key distinction is that the caste system is impose on the individual, which when you think about it is actually more analogous to national or religious identity. These new fr
93.
by kannanvijayan 1y ago
When I was a fresh immigrant child comments like these used to make me feel more uncomfortable than they do now. How I perceive these comments has changed over time. What I want to address about this comment is the implicit identity associ
94.
by kannanvijayan 1y ago
Whatever copilot defaults to doing on vscode these days. I didn't configure it very much - just did the common path setup to get it working.
95.
by kannanvijayan 1y ago
I've been hacking on some somewhat systemsy rust code, and I've used LLMs from a while back (early co-pilot about a year ago) on a bunch of C++ systems code. In both of these cases, I found that just the smart auto-complete is a m
96.
by kannanvijayan 1y ago
I haven't had to in a while as I've lapsed into IDE usage as of late, but my vimrc is something I committed to memory a long time ago. set tabstop=4 set shiftwidth=4 set expandtab set showmatch set nohlsearch set ba
97.
by kannanvijayan 1y ago
I tried once, but I'm not sure what terms to throw in there. Visible matter, estimated dark matter.. anything else? I think my estimate came out to less dense than the required threshold but it was a while back now and cobbled togethe
98.
by kannanvijayan 1y ago
I don't have enough of the background to speculate about the numbers. Dark energy feels "big" if we think of it in terms of the actual energy it would take to accelerate the universe away from itself at the rate that we see.
99.
by kannanvijayan 1y ago
Interesting reading - this is the first thorough response I've gotten to some of these question. Will check out the reading material.
100.
by kannanvijayan 1y ago
Yeah I was referencing the event horizon as the most meaningful measure of size. And whether the density is fixed over time or not doesn't affect the question. Let's take the universe at its current average mass/energy densi
101.
by kannanvijayan 1y ago
I thought about this part. I'm not sure we can link apparent size from outside the event horizon to apparent size from inside. Apparent distance is something that's affected by relative frames of reference and the frames of refer
102.
by kannanvijayan 1y ago
Oh I have so many questions on this topic. I've often wondered about this. I don't have any direct physics training, but it's something that felt really plausible after I learned that the mass of a black hole is linearly pro
103.
by kannanvijayan 2y ago
LLMs are reasonably competent at surfacing the behaviour of simple programs when the behaviour of those programs is a relatively straightforward structural extension of enough of its training set that it's managed to correlate together
104.
by kannanvijayan 2y ago
> They must abide by treaties You're somewhat mistaken with this assessment. The modern convention is that any treaty can be annulled simply by one of the parties declaring "what kind of idiot would have made this treaty"
105.
by kannanvijayan 2y ago
America had nothing close to a violent transfer of democratic power for a long time too. Not sure if you've noticed, but a lot of things are happening that "are unthinkable to happen" recently. I don't think this argume
106.
by kannanvijayan 2y ago
Ah, sorry for the misinterpretation! Also, we may have overlapped on the team :)
107.
by kannanvijayan 2y ago
Apologies if this response seems aggressive - this is just a topic I'm very passionate about :) I think technically in languages like scheme, the opportunity would be to optimize other sorts of dispatches. The classic dispatch mechani
108.
by kannanvijayan 2y ago
WarpMonkey doesn't get rid of the profiling stuff - the profiling is inherent in ICs - we keep hitcounts and other information for various paths taken through code (including ICs) and use that to guide compilation later. Warp's un
109.
by kannanvijayan 2y ago
Well on dynamic languages the ICs do give a nice order of magnitude speed-up by themselves, since the guard eliminates a whole hashtable (or linear) lookup instead of (in this case) a single memory indirection. But yeah - on spidermonkey we
110.
by kannanvijayan 2y ago
The thing is, sophisticated interfacing, GC, and multithreading - assuming they're developed and deployed in a particular way - only apply in the cases where you're applying it to use cases that need those things. The core comput
111.
by kannanvijayan 2y ago
It's really difficult to compare the JVM and wasm because they are such different beasts with such different use cases. What wasm brings to the table is that the core tech focuses on one problem: abstract sandboxed computation. The ma
112.
ExBoxing: Bridging the divide between tag boxing and NaN boxing(medium.com)
1 points by kannanvijayan 2y ago|0 comments
113.
by kannanvijayan 2y ago
Fair point. In that particular circumstance I had no desire to learn the details of the system - the need of the day was to get in there, get the shims in, and get out to the other code that mattered. If I hadn't worked for more than
114.
by kannanvijayan 2y ago
I can chime in with a similar anecdote. I use co-pilot extensively in "fancy tab completion" mode. I don't use the conversational features - just the auto-complete to help my coding along. I specifically found it very usefu
115.
by kannanvijayan 2y ago
That's a very good point. There are certainly more examples of laws that go un-enforced to reinforce your point. For example jaywalking laws are routinely flouted with little enforcement. Likewise buggery laws are routinely ignored
116.
by kannanvijayan 2y ago
> I look forward to the time when I can have as much high-quality (to me) fiction to read as I like, because it's all generated by LLM. Some time after that, I'd love to see the main Star Wars sequence done properly. I won&#x27
117.
by kannanvijayan 3y ago
It wouldn't just work for javascript and the way it's used, though. It modifies semantics in a way that would affect the behaviour of existing programs in production environments. It would break things that currently work, and i
118.
by kannanvijayan 3y ago
I left the spidermonkey team a while back but was there when this issue came up and I was strongly against implementing support for it. For me it wasn’t about it being “too hard” (implementation difficulty wouldn’t have been that bad actua
119.
by kannanvijayan 3y ago
I’m surprised no one has brought up the fact that this code has quadratic time complexity when the underlying algorithm could be linear. And the fact that it’s allocation behaviour is horrible, leaving behind ‘count-1’ garbage objects, coun
120.
by kannanvijayan 3y ago
I find architecture diagrams particularly useful for interviews of high-level candidates. I'll structure the first half as an informal conversation where I pick up some technical work they've done and get them to talk about it in

More ›