711 karma · joined January 7, 2022
I don't use React (I main Vue), so I can't speak authoritatively. But as an outside observer it feels like React gets progressively more and more "mind-bendy" with every passing year, and when I look at some high-complexity React apps my mind nearly explodes with the amount of crazy complexity that needs be managed to keep the things behaving the way they're meant to.
The only reason the world hasn't collapsed into a giant puddle of N-squared recursive CPU execution is because Excel's limitations place natural boundaries on the blast-radius of any single SME's poorly built software before they are forced to engage someone who has at least SOME idea WTF they're doing.
I say this as someone who is much closer to the SME end of the spectrum than the Software Engineering Professional end.
GDPR request comes in, just delete the record the ID refers to and you're done.
There have been studies done that show therapeutic doses of amphetamines over a multi-year period actually INCREASED the level of dopamine receptor expression in the brain.
It was by a very small amount, don't get me wrong (3-4%). But it absolutely debunked this entirely assumed theory that long term therapeutic amphetamine use would result in some kind of pseudo-addiction.
And yet, it's completely illegal for me to travel to Japan for even a day with my ADHD medication. I could be carrying my script, a letter from my doctor, and have literally only a single days supply, and I'd be risking a very serious prison sentence.
Outside of North America, even in places like Europe and Australia, ADHD is still close to non-recognised. Take Australia for example, if you assume 2-3% of the population has ADHD, and that every single one of them has to see a psychiatrist once a year to get their medication permit renewed so that their normal doctor can proscribe their medication (which is the legal requirement for ADHD in Australia), it would consume 2x the entire Australian psychiatrist workforce to deliver that many consultations... And that's if they were EXCLUSIVELY seeing ADHD patients...
Without my ADHD meds, I have ZERO ability to actually leverage the "gift" part...
Not ideal, but works.
To someone in 1850, the concept of the atomic bomb, a technology that would give humans the power to end all life on earth, would have seemed unfathomably more advanced than their understanding of human capabilities.
Now imagine you're a nation state with that kind of tech. You actually WANT your adversaries to know you've got it, otherwise it's a useless deterrent. But at the same time, you don't want them to know too much about it, because they could copy it.
So something like UAP's should be the excellent model for how to disclose their existence.
Like, I get that most of the headlines about this stuff get clicks by talking about "zomg aliens", but everything I've read on this topic that ever seemed even slightly credible has tended to lean more towards these UAP's being totally terrestrial phenomena, just utilising one specific technological capability that we are not aware humans have developed (the ability to somehow manipulate the relativistic mass of an object within a given space).
IMO that's a much more likely scenario than reverse engineered aliens.
Not impossible, but much more difficult than you might assume.
But the dudes Autistic, it shouldn't surprise anyone that intuitively understanding and anticipating the inner workings of large amorphous social-structures isn't exactly his strongest skill.
IMO the only real involvement OpenAI has had in that movement is suddenly getting REAL hand-wringy Infront of Congress about how dangerous AI is the moment OpenAI no longer held the only set of keys to the kingdom.
You'll thank me later when you encounter dunning, invoicing terms, etc.
P.s But you'll hate me when you encounter pro-rating...
The biggest mistake I see companies making is not properly appreciating the fact that there is in fact 4-5 entirely independent business processes going on inside that monolithic concept they refer to as a "billing system", and they REALLY need to be kept seperate.
IMO the fundamental seperation that is absolutely critical to make is distinguishing between your "entitlements system" that tracks what SKUs you offer and which customers have what SKUs (and in what quantity), the "accounting system" that takes input from the entitlements system and tracks the over-time net balanced owed by the customer, and finally the actual "billing system" which periodically takes the balance from the the Accounting system, zeroes the account back to nil, and generates an actual point in time "bill" for the customer to pay.
So many systems don't properly respect these boundaries and suffer the unbelievably painful consequences.
If I run a department in a company that is responsible for say, customer service, there are a million ways I could improve customer service by doing things like increasing efficiency in "back of house" areas and the reallocating those funds to customer facing areas.
Problem is... I can't do that, because if I drive efficiency in the BOH area I don't get to keep the surplus resources I just created, it gets gobbled back up by the organisation at the next budget cycle... So there's literally a NEGATIVE incentive for me to optimise.
Now, I understand why the organisation feels the need to capture that surplus, there might be other areas of the business that leadership thinks needs those resources more. But what they're failing to understand is they're actually hurting both the team making the efficiency AND the team who needs the extra resources because now everything is just going to remain exactly the way it was.
This seems like a ridiculously easy problem to solve though... Just allow teams that drive efficiency dividends to retain say half of the benefit they generate.
The company still gets a chunk of resources back to redistribute the way they see fit, AND the team generating the efficiency are able to implement the rebalancing of resources WITHIN the team that incentives them to make the changes in the first place.
I would bet serious money that the NET divided returned to the central pool would actually be HIGHER even though they only get half of each block of surplus, because you would be providing SO MUCH more incentive to individual teams to find and implement these efficiency gains now.
That being said, there is an enormous community of extremely passionate breeders around the world who are absolutely dedicated to breeding HEALTHIER dogs, especially in breeds that are known to have issues due to poor breeding.
We have a Boston Terrier we got from one such breeder after spending over two years looking for someone we trusted to do the process right, and we are so happy we did. Not only do we have an amazing pet, but we know we are activity contributing to the process of improving the overall health of the breed, even though it did cost us more to do.
A few things to look for in order to pick a "good" breeder.
1. How many litters do each of their female dogs have over their lives. If a breeder is expecting them to produce more than two (three at the most) litters, that's a red flag.
2. Do the mums deliver naturally, or via C-section. We only learned through extensive research that some breeds (often including Boston's) almost all have to give birth via C-section because they've been so extremely bread. We specifically looked for a breeder who's dogs a born naturally. If the pups are so extreme in your they've been bread that the mum can't get them out naturally... Something has gone too far.
3. Is the breeder a "purist" when it comes to the breed. Now this is going to be the opposite of what most people expect, but you WANT a breeder who is 1930's levels eugenicist when it comes to their breed. Fundamentally, most of the issues from breeding come about through mongrelisation of the breed. Low quality breeders cross-breed dogs to introduce "cool" or "fashionable" new traits, without caring about the million other genetic inconsistencies they're bringing into the mix at the same time.
A great example is with Boston Terriers we learned. The breed spec for them is VERY clear, they have a white base coat with a black or extremely dark brown "tuxedo" style coat covering their lower body and a portion of their front-legs.
ANYTHING that has colours other than those three, is a cross-breed for fashion purposes. There is no such thing as an "albino Boston", or a "patterned Boston" etc etc, these are marketing names people came up with for mongrelised versions of the breed which will almost always have major health issues.
It's terrible when you want broad swaths of input that you're scanning through to discover new things. Text is an extremely low bandwidth communication tool.
Natural language interfaces belong at the periphery, as the interface between the human and the machine. Other than that, I want my computers dumb as rocks, really fast, any totally predictable - which is basically the opposite of what you get from LLM's.
It DOES feel like it "could" be that holy-grail between IM, project management, email, etc.
But I feel like you need a whole team all onboard for it to make sense.
"Oh and if you'd like to talk to someone from outside your company, my god have mercy on your soul if their entire business isn't also subscribed to our premium plan" - someone at Slack, almost certainly.
All we need now is for someone to give every Linen user an externally accessible handle that allows cross-tenant invites or 'thread sharing' and we'll have officially closed the loop.