AlexErrant

React DX is hot garbage. Words cannot express how much I LOATHE hook rules. Coming from a Solid JS background, where reactive primitives are just Javascript functions... I groan every single time I run into (yet another) hook rule.

I have to conditionally render empty fragments because React can't handle conditional hooks. It's the stupidest thing ever. "Oh hey let me allocate memory for this hook that will almost certainly never be used except under edge conditions! Sure, React can do conditional components, but conditional hooks are just too much for us!"

> does Gmail have any "Display un**ed" option?? Er, no.

On the email, click the dotdotdot, show original. https://imgur.com/a/ymjtfCI

Forgive my probing, but why consider Windows/Mac specific APIs then?

I'm building a voice-controlled app, so it requires transcription. The problem is that the app talks back, so some amount of AEC is needed. It looks like your "generic" AEC is ML powered, but your planned platform specific AEC may use non-ML tech (i.e. looks like you're considering Windows/Mac specific APIs). Have you had a bad experience with AI powered AEC?

Heh, I recently reported a bug to a pretty big healthcare company where I could simply increment the integer ID in the URL and see other patients' info.

Didn't get so much as a "thank you". At least they fixed it. But I'm sure they have other vulns given how stupid this one was...

> we created Patchwork—a web-based collaboration environment for malleable software... storing both user data and software code in Automerge documents. On top of that, it adds version control utilities like history views and simple branching. These tools apply to any document in the system—whether a piece of writing, or code for a software tool... Eventually we also plan to release Patchwork as an open-source tool

What milestones would you like to hit before open-sourcing it? As an outsider, it looks like it has a LOT of features, and I wonder if there's feature creep. Still, version control for everything is a tall order, so perhaps it needs plenty of time to bake.

> “I came up with a problem which experts in my field would recognize as an open question in number theory—a good Ph.D.-level problem,” he says. He asked o4-mini to solve the question... o4-mini presented a correct but sassy solution

I wonder who gets first author credits on that paper.

> Guess the missing word to complete the following sentence:

> Everyone respects him, he's the _ _ _ _ _ in this business.

Apparently the answer is "daddy"... this is not a good question or answer. Good luck with the launch, but you might want a native speaker to audit your question bank.

> We do not know whether the embryo models recently created from stem cells could give rise to living people or, thus far, even to living mice.

So it doesn't even work in mice... how about we get that working first. Then maybe grow chickens/cows for meat. Then write "revolutionize medicine" headlines.

Of particular note:

> Parker said that, as a federal contractor, the company has been reviewing all of its programs and initiatives in response to Donald Trump’s executive orders that direct federal agencies and contractors to dismantle DEI work.

Which is good because "ASCVD is a disease that begins in childhood; hence, primordial prevention is an important target for improving cardiovascular morbidity and mortality later in life."

> Elevated LDL-C and triglyceride levels have a positive correlation with atherosclerotic lesion prevalence that persists from childhood through early adulthood... Follow-up data from the Young Finns cohort after 12 and 27 yr also demonstrated positive correlations between elevated childhood serum cholesterol and triglycerides to elevated levels in adulthood... Children from the i3C cohort with high and borderline-high total cholesterol have 1.5 to 2.13 times the risk of both fatal and nonfatal cardiovascular events in adulthood than children without. In addition, i3C children with high and borderline-high triglycerides have 1.69 to 2.47 times the risk than children with normal triglycerides.

https://journals.lww.com/jcrjournal/fulltext/2022/11000/prim...

It's stage 3... and "the existence of stage 4 is basically a formality"

> [stage 3] basically means "finished, pending editorial nit review---but since multiple implementations haven't happened yet, there's a reasonable chance that we'll discover something is broken, and need to fix the normative content

https://github.com/whatwg/meta/issues/336#issuecomment-24874...

It's stage 3... and "the existence of stage 4 is basically a formality"

> [stage 3] basically means "finished, pending editorial nit review---but since multiple implementations haven't happened yet, there's a reasonable chance that we'll discover something is broken, and need to fix the normative content

https://github.com/whatwg/meta/issues/336#issuecomment-24874...

Here's Webkit's tracking issue https://bugs.webkit.org/show_bug.cgi?id=281223

No movement since its creation on October 10, 2024

This exists https://github.com/lyonbot/solid-dockview

The Clean Label Project did an AMA on Reddit, and people raised issues regarding their reporting, methodology, and data.

I see they haven't learned anything in the intervening 7 years.

https://www.reddit.com/r/IAmA/comments/64mpty/we_are_the_cle...

> I predicted the 2008 crash

> In August 2006, I reported that a collapse of the global banking system would take place in 2008 due to a complex convergence of financial and energy crises. My source for this revelation was a sensitive interview with a senior advisor to the US Department of Defense, with extensive access to the US military, intelligence and political establishment. “The source… said that leading US financial analysts privately believe that ‘a collapse of the global banking system is imminent by 2008,’” I wrote at the time.

So it wasn't his prediction - his un-named source predicted it...

It's literally in quotes, i.e. not his own words. So he could easily deny it being his prediction. But since it came to pass, he gets to claim credit...

Not sure if you're affiliated with patlynk, but after entering some details I get matched to 724 studies, the first of which wasn't in the nation I said I was in, and the next few of which didn't give any info on how to enroll. Best info I got was their mailing address. Frankly until the UX improves I'm advising against anyone signing up.

What issues did you have with FTS5 that made you switch to Tantivy? Does using Tantivy mean you have to load the entire Sqlite db into memory?

I ask because in my own project I have my own search query grammar [0] written in Lezer that more or less "transpiles" [1] to SQL+FTS5 that's been uh, tedious to build.

    [0] https://github.com/AlexErrant/Pentive/blob/main/shared-dom/src/language/query.grammar
    [1] https://github.com/AlexErrant/Pentive/blob/main/shared-dom/src/language/query2sql.ts

I'm not a believer in that timeline. There's a large distance between superconductivity in the lab and commercial application (since you specified "changes the world immensely").

E.g. MRIs still use NiTi (critical temperature of ~10 kelvins), discovered in 1962, for a number of reasons (this is in spite of MgB2 having a critical T of ~39k, ReBCO with a critical T of ~90k, and BSCCO with a critical T of ~108k):

> In this paper, we analyze conductor requirements for commercial MRI magnets beyond traditional NbTi conductors, while avoiding links to a particular magnet configuration or design decisions. Potential conductor candidates include MgB2, ReBCO and BSCCO options. The analysis shows that no MRI-ready non-NbTi conductor is commercially available at the moment. For some conductors, MRI specifications will be difficult to achieve in principle. For others, cost is a key barrier. In some cases, the prospects for developing an MRI-ready conductor are more favorable, but significant developments are still needed. The key needs include the development of... [omitted]

https://pmc.ncbi.nlm.nih.gov/articles/PMC5472374/

Unfortunately, it probably won't be as simple as "step 1 discover material, step 2 manufacture, step 3 profit".

(I'm not OP - I'm on the Charybdis.)

I prefer a trackball because I can momentum-flick my cursor from one side of my 4k screen (at 100% zoom) to another side monitor pretty easily... as long as I keep the ball bearings clean. I've been thinking about using a metal trackball for more momentum, but haven't gotten around to it yet.

I have a "sniper" key I use to decrease sensitivity for more precision cursoring... but rarely use it (maybe once a month?). I do not have thumb fatigue.

Happy Charybdis user here. If you ask, I'm sure Quentin would be willing to build a Charybdis with a trackball on both sides - he was very accommodating of a change I requested (I wanted ball transfer units). He already supports having the trackball on either side.

Norwitz and Layne differ on the lean mass hyper-responder phenotype. Norwitz likes referencing his meta-analysis of 41 RCTs, but pay attention to its conclusion:

> A substantial increase in LDL cholesterol is likely for individuals with low but not high BMI with consumption of an LCD, findings that may help guide individualized nutritional management of cardiovascular disease risk. As carbohydrate restriction tends to improve other lipid and nonlipid risk factors, the clinical significance of isolated LDL cholesterol elevation in this context warrants investigation.

"warrants investigation"??? Come back when you have hard outcomes, i.e. reduced MACE. His meta-analysis of RCTs concludes that some people on keto have increased LDL! Norwitz admits in another video: "No, I'm not going to say that high ApoB is fine... ApoB is necessary but not sufficient [to cause ASCVD]" https://youtu.be/270ZyfSGLkE?t=484

So, what "humiliation"? If high ApoB (synonymous with LDL for our purposes) is bad, why would you go on a diet that causes it?

Layne has a large lay audience, and he has to stick to his main message: that w/r/t diet, calories are all that matter. Lowcarb/lowfat/vegan/whatever is fine. Platforming someone who's obsessed with LMHR is going to be too much biochemistry for Layne's audience. And for an effect that's relevant to perhaps only 2% of people? Low-carb advocates have their own YouTube channels; there's no reason Layne, as a center-aisle scientist, needs to engage.

Until the LMHR has been shown to have reduced MACE, which Norwitz admits we don't have evidence for https://www.youtube.com/watch?v=gxRLRYEQaEs&t=971s I don't see the point of Layne platforming him to just talk about biochemistry and confuse people.

> This user experience is mostly good, as users typically care about domain-level filtering (commits between these time ranges) versus the arbitrary traversal of pages

If you're using SQLite, you can use temp tables as a hack to get arbitrary traversal:

Imagine an MS Excel-style spreadsheet that may have 10k+ rows that a user may search/sort on; I virtualize it and query for data in blocks of 100 rows. Being Excel-style, you can arbitrarily scroll any distance down the spreadsheet. OFFSET has terrible perf when skipping 15k+ rows, and users might want to jump to the bottom of the spreadsheet (or anywhere, really). I handle this in two steps:

    1. Return the first 100 results of a query ASAP.
    2. In a nonblocking manner, add all the relevant ids for that query to a temp table.

When a user tires of the first 100 results and starts randomly scrolling, I can use cursor pagination with the temp table's rowid (WHERE rowid > 15000 ORDER BY rowid LIMIT 100) by taking advantage of the fact that the spreadsheet's row number is now the same as the temp table's rowid. Building this cache is relatively expensive; it takes ~10 seconds on my devbox with ~15k rows. However jumping to the bottom of that result set takes ~200ms, so I'm happy with it. (Note I'm on a fork of wa-sqlite, i.e. it's running in the browser - "real" sqlite is much, much faster.)

Heh, I do 2hrs of z2 4x a week (8hr total) with a Charybdis and inclined treadmill. I use two wrist rests rubberbanded together though; the height of the Charybdis is significant and I don't hover my hands. I'm usually programming, though when I run into more complex problems I usually switch to less cognitively burdensome tasks. Or I just stare out the window and let my gut sit on the problem... which is surprisingly effective in coming up with solutions.

https://imgur.com/a/YpQElUs

I clamped a piece of wood to the desk's edge to have something to push against, as having a point of stability is nice. You can see the back of the treadmill's control panel - I took a regular treadmill, dismantled it, and mounted the panel to my adjustable desk. The treadmill is propped up on blocks to get a ~12% incline; my usual pace is around 3.1-3.4mph, depending on the season/humidity. Sweat... is significant. I usually have a towel under my wrists to uh... collect the arm-runnage >_>

I'm using ball transfer units for the trackball. They're nice, but (as with all trackballs) you should clean the BTUs if you want to keep the smooth flicking. I use a 4k tv as my monitor, so flicking from one corner of the screen to the laptop screen all the way on the other side is really nice.

I'm engaging in pedantry here, but computer science has proofs. When we have hardware level security vulnerabilities, I trust my device as far as I can throw it. Proofs are only sound and valid in maths.

https://securityintelligence.com/news/apple-m-series-chips-h...

Everything Apple's done yields verifiable security - but it's not "provably secure". The two are distinct, and when you try to sell to me with bad language I get squinty-eyed. Especially since "confidential computing" already exists on x86/AWS, and I struggle to see the difference. It just sounds like Apple marketing to me.

> don’t totally discount the idea of designing a system that has provable security properties

They're only as provable as your assumptions/givens. Given a hardware vuln, where is your security now?

Not to dunk on you, but your misuse of "provably" makes me discount what you said.

If we could "prove" security, we would. Proving security in a networked environment? Hahaha - there have been successful attacks on airgapped envs.

Not sure why I'm responding to a throwaway (are you an AI?), but this is patently false.

Holistically: https://pmc.ncbi.nlm.nih.gov/articles/PMC5563491/#S1

More pointedly:

> Five healthy rhesus monkeys, three male and two female, weighing from 2.4 to 6.1 kilos, were fasted for periods of 2 to 4 days... With one exception, the monkeys developed a ketosis within 2 days

https://www.sciencedirect.com/science/article/pii/S002192581...

Note that Attia's article is about rhesus monkeys.

Okay gave my tldr more nuance. It was clearly too short.

TLDR: AGI doomers who talk about extinction are baseless.

At 10m, he gives 3 doomer scenarios: nukes, bioweapons, and nanotech.

He asks: "Is engineering an act of intelligence alone?" He spends ~15m supporting the answer of "lol no". (He also says that science is harder than engineering, and AI will need science to kill us all.)

This isn't to say he's e/acc - at 35m he mentions how AI can continue racism, economic dislocation, classism, etc.

His conclusion is we need to focus on the real problems, not the "human extinction" ones.