jmull

5 ms·

1.
by jmull 5d ago
It's not the M5 Ultra itself, but the M7s or M9s that will do the damage. 99% of people will use whatever AI is free. The sophisticated, heavy users that are willing and able to pay a lot of money the ones that will be interested in co
2.
by jmull 6d ago
AI is like honey to these magical thinkers.
3.
by jmull 6d ago
I hate that AI makes this kind of vacuous article appear, on the surface, to be credible enough that it makes it in front of my eyeballs.
4.
by jmull 7d ago
Allocation happens. You can leave it to the compiler, the runtime, or let the code control it. The automatic solutions are usually pretty good and usually the right place to start. But if performance is a priority, you want options. BTW, “p
5.
by jmull 8d ago
> The security benefit can't be overstated. Security benefit appears strongly overstated. Compared to using a password manager (including generated secure passwords for each account), there isn't a significant security benefit.
6.
by jmull 8d ago
> refuses to allow me to authenticate to it because it wants a passkey I don't have... it wants me to use an Android phone I (also) don't have And yet you love passkeys? How much will you love them when you're locked out o
7.
by jmull 13d ago
Open source is collaborative, and isn't exactly a niche thing, even at commercial companies.
8.
by jmull 13d ago
Is that a settled question? China isn't exactly struggling right now.
9.
by jmull 13d ago
> But don’t be the out of touch Kung Fu master, heir to lifetimes of tradition, that gets mauled by an amateur MMA fighter. Seems like a fundamentally poor analogy. Software development is collaborative, not competitive.
10.
by jmull 14d ago
Yeah, this 100% looks like an effort to use fear to create a regulatory moat.
11.
by jmull 14d ago
I can't help but notice your counter arguments are "lol, okay" and "These are some 2024 arguments". Not exactly convincing stuff.
12.
by jmull 16d ago
Tailwind templates give you a quick way to get good, functional, customizable components. LLMs give you a quicker way to get good-enough, functional customizable components. You have to pay for LLMs too, but you aren’t getting the LLM just
13.
by jmull 17d ago
I used to be a fan of smaller phones -- my current one is a 12 mini. I'm getting this thing. Changes in eyesight change your priorities.
14.
by jmull 17d ago
Nah. Tailwind provides an actual simplifying layer for the (over-)complicated practice of front-end web development. By being well-though-out, well-designed, and very well documented, it actually made web app styling easier/simpler (wi
15.
by jmull 22d ago
It's a weird article. Despite the title and some of the text, much of the article makes the point that LLMs are next-token predictors, where the predictions are based on both training data and various reinforcement learning technique
16.
by jmull 22d ago
> You can just watch or read the mainstream news for a few days and see for yourself, really. I read mainstream news, but I don't know what you're referring to. It's not like they are perfect, but I do see accountability w
17.
by jmull 23d ago
I do. I’m open to being convinced, but if you’re claiming mainstream media has no accountability, I think you need to substantiate that. I realize there are information bubbles on the internet where this is accepted as truth, but unexamined
18.
by jmull 1mo ago
One way to gauge how close to viable local inference is the quality of the arguments against it are. $100/mo for ten years is $12K. $200/mo is $24K. When you look at the valuations of the companies providing these tokens, you don&
19.
by jmull 1mo ago
I think this is an attractive chimera, but will never be popular and will only be used much in specific domains. The thing is, trust is a massive optimization. If you believe code from X is competently made and not malicious, you can acce
20.
by jmull 1mo ago
> Here are some conclusions I’ve reached through experimentation: ...a bunch of stuff specific to certain models the author is currently using. Experimentation is good for technical leaders, but you want to be focusing on approaches and
21.
by jmull 1mo ago
There’s essentially zero doubt that universal healthcare would save over $1T per year. How do we know? The experiment has been done multiple times, all over the world. In the worst case (Sweden), healthcare is a bit short of $5K per capit
22.
by jmull 1mo ago
It’s really not experimental though. Assuming their math and design are good, this is pretty strong, as these things go. But there were a lot of other things that were different between the groups that could reasonably be affecting results.
23.
by jmull 1mo ago
The most important thing is to default to the user’s system setting. Then you (and your users) only need the toggle, if at all, for edge cases. (If the user cares, they likely will have set system settings the way they like them, and if the
24.
by jmull 1mo ago
It’s in TFA, just a few sentences in. Isn’t the article about how to best deal with the UI for letting the user indicate that the system should decide? Which covers your case.
25.
by jmull 1mo ago
There isn’t though. Garbage collectors aren’t magical black boxes with no levers, and you can use them how you like. You address performance problems relating to GC the same as you do anything else: measure, analyze, optimize, repeat.
26.
by jmull 1mo ago
Not sure I understand the question, but it generally works like this: Once you measure, you'll find a small fraction of the code is taking a large fraction of the time. When you zoom in on trouble-spots you may find, e.g. that the GC i
27.
by jmull 1mo ago
Game developers are always trying to push the boundaries. The only game engines without performance issues are ones hardly being used.
28.
by jmull 1mo ago
GC languages typically have features of the language and/or standard library that make GC the default, not the only option.
29.
by jmull 2mo ago
I think the consternation here isn’t warranted. If a large enough proportion of incoming students need an adjustment period, surely it’s better to have one than not. Also, it’s misdirected. If incoming students aren’t prepared for college,
30.
by jmull 2mo ago
Read the comment I replied to. It claims the researches must have somehow controlled for correlation (whatever that’s supposed to mean). Which they haven’t. There are a bunch of other comments arguing there’s something beyond correlation sh

More ›