It’s rough at first but you will learn the baby’s rhythms and preferences. If you track their sleep and wake up times (I did it the old fashioned way in a notebook) you’ll see a pattern emerge pretty quickly, and then it gets easier because you will figure out how to work with it.
Every baby is different so most of the advice you find won’t work, but if you try enough things you’ll eventually find something that works consistently. Or you might just luck out and get a good sleeper.
The big tip I have for you is to understand wake windows. Babies can get too tired to sleep(!) so you need to make sure to put them to sleep roughly 1-1.5 hours after they last wake up.
Follow a routine every day. I posted elsewhere in this thread what worked for us. It was tough when they were infants because neither of ours slept through night till about 2. The routine saved us.
try co-sleeping, and also a comfortable baby-carrier that allows you to carry the baby around while keeping your hands free so you can work. the most difficult from babies not sleeping is that they are not supposed to sleep alone. see attachment theory. the other advice, if you can follow it, is to sleep yourself every time the baby sleeps. again, co-sleeping makes that easier.
I dunno, we found that our kid slept slightly better moved to his own room at 5 or 6 months old. Although that meant maybe 4 wakings rather than 5. Now he's nearly three years old and sleeps solidly for 10 or 11 hours. My guess is that food and metabolism have a big part to play.
My mutant power is the ability to put babies to sleep. Before I had my own I'd put other people's kids babies to sleep easy peasy. It's something I've been able to do since I was a teenager.
I use cloth diapers, but modern disposable diapers can hold a lot, a lot of pee. Significantly more than any cloth diapers can. This means a lot less blowouts with disposables.
I think this is something we'll start to see which is something like a Mandela-Effect, but from LLM results. When we had deterministic search - everyone could see the same result, but now using LLMs knowledge becomes a training and seeding issue. Two people can confidently be given completely different information, so in both cases perceived as true.
Gemini answers with 3 different champions dating back to 2024 and the list of events that the matches were played at. None of the results mention this guy.
"It is fundamental to language modeling that every sequence of tokens is possible."
This isn't true, is it? LLMs have finite number of parameters, and finite context length, surely pigeonhole principle means you can't map that to the infinite permutations of output strings out there
The most famous criticism of Flash was "Thoughts on Flash" by Steve Jobs, which said among other things that it's too inefficient. He did cite inconsistent hardware acceleration for H.264 that was a real performance drawback of Flash for video in particular, and was also complaining about the power usage for interactive Flash content in general. Jobs was right at the time from what I can tell, but somehow the end result was even slower stuff. People did keep repeating the line that Flash is slow.
I also remember people citing performance as a reason YouTube switched from Flash to HTML5. Searching those blogs now is giving a lot of 404s. Like I said this should've helped since it's video, but somehow YouTube immediately got slower anyway back then. Back then I installed an extension to force it to use QuickTime Player for that reason.
The proprietary and insecure parts were real problems too. I'm fine with the decisions that were made, but this was a drawback.
Is there a reason why you chose to post this comment for free, without rewards, knowing full well it's going to end up in the training data of some LLM in the future?
Well, the way intellectual property works, anything I write on the internet is, by default, all rights reserved. Different website's policies will impact this, of course, and different laws (and quirks like "fair use") as well, but in general, if I write a snippet of code like:
printf("%p\n", 0xbeefbeef);
/* insert awesome new compression algorithm here */
Then no, I'm not providing it for free. In fact, all rights are reserved. Don't see a license? Then you don't have the right to use it e.g. to build a product.
The question was about a comment you posted on this specific site, whose terms[0] say:
> By uploading any User Content you hereby grant and will grant Y Combinator and its affiliated companies a nonexclusive, worldwide, royalty free, fully paid up, transferable, sublicensable, perpetual, irrevocable license to copy, display, upload, perform, distribute, store, modify and otherwise use your User Content for any Y Combinator-related purpose in any form, medium or technology now known or later developed.
Matt Levine pointed out in a past article that the real danger of insider trading are company insiders being incentivised to damage the company to make a quick buck.
> insiders being incentivised to damage the company
I'd like to emphasize that this incentive doesn't have to be an accidental find by the insider either: The "market" can end up facilitating anonymous crowd-sourced bribery by enemies or competitors, who create the potential for profit knowing that eventually an insider will take the other end of the implied deal.
Every time I see someone dismissing these kinds of issues--especially someone whose salary depends on not-understanding it [0] --I imagine how their tune would change if the shoe was on the other foot. For example, if someone created a "prediction market" where people could anonymously bet on unusual deaths or serious injuries of... prediction-market executives.
Artificially influencing a stock's price is illegal by itself, and there's probably about a half dozen other charges that could be tacked in this scheme, possibly including extremely serious ones like wire fraud which gets tacked on pretty much every crime involving digital tech.
Maybe there should be a maximum betting limit the same way there should be an election contribution limit — let's say something like 10x the federal minimum wage or whatever so if you are betting under USD 75.5, it is A ok but once you cross this number, we require public disclosures, no hiding behind LLC, natural persons only, KYC, the whole shebang.
Actually, now that I think about it, let's get rid of the minimum, there should be no minimum, all bets even five cents must be fully disclosed and attributed to natural persons, no hiding behind "corporations are people, my friend" nonsense.
Image generators can make reasonable-looking raster images. LLMs are good at coding. But drawing SVGs sits at the worst of both worlds.
reply