Yeah, the graphs make some really big assumptions that don't seem to be backed up anywhere except AI maximalist head canon.
There's also a gap in addressing vibe coded "side projects" that get deployed online as a business. Is the code base super large and complex? No. Is AI capable of taking input from a novice and making something "good enough" in this space? Also no.
The later remarks are very strong assumptions underestimating the power AI tools offer.
AI tools are great at unblocking and helping their users explore beyond their own understanding. The tokens in are limited to the users' comprehension, but the tokens out are generated from a vast collection of greater comprehension.
For the novice, it's great at unblocking and expanding capabilities. "Good enough" results from novices are tangible. There is no doubt the volume of "good enough" is perceived as very low by many.
For large and complex codebases, unfortunately the effects of tech debt (read: objectively subpar practices) translate into context rot at development time. A properly architected and documented codebase that adheres to common well structured patterns can easily be broken down into small easily digestible contexts. i.e. a fragmented codebase does not scale well with LLMs, because the fragmentation is seeding the context for the model. The model reflects and acts as an amplifier to what it's fed.
> For the novice, it's great at unblocking and expanding capabilities. "Good enough" results from novices are tangible. There is no doubt the volume of "good enough" is perceived as very low by many.
For personal tools or whatever, sure. And the tooling or infrastructure might get there for real projects eventually, but it’s not currently. The prospect of someone naively vibe coding a side business including a payment or authentication system or something that stores PII— all areas developers learn the dangers of through the wisdom gained only by experience— sends shivers down my spine. Even amateur coders trying that stuff try old fashioned way must read their code and the docs and info on the net and such and will likely get some sense of the danger. Yesterday I saw someone here recounting a disastrous data breach of their friend’s vibe coded side hustle.
The big problem I see here is people not knowing enough to realize that something functioning is almost never a sign that it is “good enough” for many things they might assume it is. Gaining the amount of base knowledge to evaluate things like form security nearly makes the idea of vibe coding useless for anything more than hobby or personal utility projects.
> For large and complex codebases, unfortunately the effects of tech debt (read: objectively subpar practices) translate into context rot at development time. A properly architected and documented codebase that adheres to common well structured patterns can easily be broken down into small easily digestible contexts. i.e. a fragmented codebase does not scale well with LLMs, because the fragmentation is seeding the context for the model. The model reflects and acts as an amplifier to what it's fed.
It seems like you're claiming complex codebases are hard for LLMs because of human skill issues. IME it's rather the opposite - an LLM makes it easier for a human to ramp up on what a messy codebase is actually doing, in a standard request/response model or in terms of looking at one call path (however messy) at a time. The models are well trained on such things and are much faster at deciphering what all the random branches and nested bits and pieces do.
But complex codebases actually usually arise because of changing business requirements, changing market conditions, and iteration on features and offerings. Execution quality of this varies but a "properly architected and documented codebase" is rare in any industry with (a) competitive pressure and (b) tolerance for occasional bugs. LLMs do not make the need to serve those varied business goals go away, nor do they remove the competitive pressure to move rapidly vs gardening your codebase.
And if you're working in an area with extreme quality requirements that have forced you into doing more internal maintenance and better codebase hygiene then you find yourself with very different problems with unleashing LLMs into that code. Most of your time was never spent writing new features anyway, and LLM-driven insight into rare or complex bugs, interactions, and performance still appears quite hit or miss. Sometimes it saves me a bunch of time. Sometimes it goes in entirely wrong directions. Asking it to make major changes, vs just investigate/explain things, has an even lower hit rate.
Our pest control put in a bucket called In2Care that has a little net with some powder suspended above the water - mosquito lands on the net, gets the powder on 'em, carries it to the next site and that site gets neutralized. They're designed for commercial campuses but for ~$200/yr it's well worth it for residential also.
They do take a while to take effect, and they do take maintenance, but my experience so far is that they're super effective.
I have a cousin with DS. You have to be committed and have the means to raise a child with extreme needs. Many of them will live with their parents their entire lives and will not develop cognitively beyond their tweens (hence the Britney Spears anecdote above). The ones that do move out tend to have to go to a place that specializes in assisting them. They can also have pretty extreme health issues.
Yes, they can be beautiful people that bring light to others around them, but those others also don't typically get exposed to the behind the scenes struggles of the entire family to cope with this.
Some people are prepared to do this; I don't judge the ones that decide they're not. I would hate for someone to go into it not understanding what they're signing up for.
I think this point would be better made without using the word “extreme” so much. All children bring new challenges; kids with DS often bring more; my child with DS has never, ever been an “extreme” challenge (just like most of the other families with kids with DS we know). There are definitely outliers where the “extreme” applies, but it’s not a helpful way of thinking about DS in general.
From this[1] list of associated complications one can read:
People with Down syndrome are much more likely to die from untreated and unmonitored infections than other people.
Children with Down syndrome are much more likely than other children to develop leukemia
Children with Down syndrome are more likely to have epilepsy [...] Almost half of people with Down syndrome who are older than age 50 have epilepsy.
And from this paper[2]:
Clinical research and longitudinal studies consistently estimate the lifetime risk of dementia in people with Down syndrome to be over 90%. Dementia is rare before the age of 40 years, but its incidence and prevalence exponentially increase thereafter, reaching 88–100% in persons with Down syndrome older than 65 years. [...] In a longitudinal study of adults with Down syndrome, dementia was the proximate cause of death in 70% of cases.
Saying they can have extreme health issues does not seem excessive given the above IMHO.
It's interesting about the leukemia one. They're also more likely to survive it than children without Down Syndrome and less likely to get a second cancer.
Aside from that, it is actually hard to paint an accurate picture of today with historical data for people with Down Syndrome as the childhood Trisomy 21 strategies have improved and been implemented in the past 20-30 years. 60 years ago kids with Trisomy 21 were moved into institutions. Kids 30 years ago got some basic treatments to keep them alive. Now kids get all kinds of screenings for hearing, vision, thyroid, heart conditions before problems develop. Turns out it's very difficult to grow, learn and thrive when your thyroid doesn't work, or your cardiovascular system wasn't circulating enough oxygen.
There are more struggles for sure, including intellectual disabilities, but many more kids are doing significantly better than their past generations. It costs more, is more work, but like the parent poster said, my experience certainly isn't extreme. We go to more doctor's appointments, have IEP meetings, and she's in speech therapy. She's generally been pretty healthy, happy and very active.
It was scary when she was born. We were given a pamphlet with a list of things similar to your first link. The reality though is she's more likely to have those than the general population, but some of those things are very rare. 100x very rare is still rare. Having all of those issues would be even more rare. The greater point though is that any kid can have those issues too.
To add to what magicalhippo said about the extremes of medical issues, the extremes of parenting seem appropriate. "Average" parenting follows a trajectory of intense parenting of a newborn, and end at light/no parenting of an adult. For an overwhelming majority of families with kids with DS, the intense parenting requirement last long and more progresses slowly and the trajectory plateaus at around the tween stage, where a significant portion of your day, every day, is dedicated to managing and caring for your child. I would say that spending tens of thousands of additional hours, likely up until your own death, caring for an adult child would count as extreme needs.
As the percentage of adults of ordinary abilities who fail to launch continues to rise, I wonder if we'll stop seeing this as a deficit specific to DS and other intellectual disabilities.
You can pop the handle off that Ridgid one and there's a hex bit you can hook up to your existing drill. It's admittedly a bit wonky to operate like that, but for a single family home it's mostly workable.
