Notices by tef (tef@mastodon.social)

huh "do what you love and you'll never work a day in your life" was actually a warning about burnout

@thomasfuchs 98% of software development is excel functions by volume heh

i lamented to a coworker "we only had performant software because computers were slow, and had quality software because distribution was expensive and neither of those are true anymore"

meanwhile, as investor story time moves away from "we only hire the best and smartest 10x" to "i generated 70,000 loc before breakfast" it feels like the last remaining social or market forces for quality are evaporating

zoom, enhance

similarly, i've heard a few times that "we might cure cancer", and sure enough some brute force computation can fold proteins fast

but in practice it is more likely these tools will be used to fabricate experimental results, push dietary supplements and other snakeoil cures

and more coarsely, ai isn't pouring funding into the CDC, ai isn't reversing the destruction of the FDA, and is more than likely going to be used to justify those things

i've heard a few times that "waymos will make streets safer" so i went and looked up sf's traffic fatality statistics and they're pretty much identical

i mean, there is a slight increase over the last two years but there's sufficient variance to avoid suggesting a trend

as i understand it, waymos tend to take people off busses and other forms of transit, rather than out of their own cars

so i'm doubtful it will lower deaths on the road, just the number of busses

i don't want to be all "you are not immune to propaganda" but a lot of these arguments prey on optimism and hope that technology can lift people up

but when you start to examine the rhetoric, like "what if <imaginary circumstance where the tools are useful>"

or "bad thing? that's a lack of training and dicipline"

it just feels like gun logic in a new outfit

i said elsewhere but there is a distinction between things being commodotized and things being democratized

sure enough there's a new low cost bidder in town and things are a little more affordable to some

but we all know those prices aren't sustainable, and i'm pretty sure that when the prices go up, and the adverts appear, the world won't be much better off for it happening

i give it a year before your representatives are cutting library funding "who needs it when we have ai"

sure enough machine translation has reasonably proven itself as a mostly public good, albeit at the expense of the translation industry

so i am aware that good things can come with bad prices, but i haven't really seen much good and i am seeing a lot of bad things

it literally breaks my heart that the public web now sits behind a proof of work system, forcing strangers to mine coins to buy access to webpages

because a bunch of tech companies are desperate for an poison-free training set

it feels like a lot of the arguments i hear boil down to "what if none of the bad things were happening right now, and instead, good things happened instead"

and sure, if that were true, things would be good

but, well, all of the bad things are happening already and none of the good things are any closer to appearing

and i'm just not confident "wait and see if everything reverses course" is a sensible way to evaluate the impact of new technologies

we're destroying the open web

we're burning down the closest thing i've ever seen in my life to the library of alexandria

and people are explaining to me how warm it keeps their hands, and maybe, in the future, the ashes will contain the secrets of the universe

the worst bit? i still like machine learning, i still think stochastic approaches can have benefits

but if i wrote software that pushed vulnerable teenagers to suicide, or enabled people to sexually harass strangers with pornographic forgeries

i would take a step back from the keyboard and ask my good buddy hans, "are we the baddies"

or at least, i hope i'd ask those hard questions

"yes your problem is capitalism" yes

arguing about the technology in a vacuum is a debate club tactic, a way of denying the real consequences and moving the focus to imaginary benefits

i am not capable of putting on the pom poms and cheering for supply-side economic arguments that paper over the real world harms

i don't care if tests pass or if it works on your machine, i care how it acts in production

finding myself repeating "fuzzers are stochastic, with enough cpu time you will always find a bug, results are a usually demonstration of resources not algorithms"

like, hacking, along with spam and fraud are the sorts of activities where things only need to work 0.1% of the time to be successful

it isn't a demonstration of clever code or tooling but the uncompromising effectiveness of sheer brute force heh

who called it "stochastially generating code from a spec" and not "the irrational unified process"

realising i have spent more than a decade yak shaving a specific problem and all i have is a field full of half shaven yaks and an ever growing pile of discarded shears

found myself typing 'win32' deleting it and writing 'win64' and then deleting it again and writing 'windows'

@whitequark people are used to saying "A for Apple" or "B for Bus" to spell things out here, but as someone who does know the nato phonetics, i have confused many a call centre

@fanf it's notable that growth stops around 2014 and then declines from around 2016 onwards

before hitting the llm part in 2023 and a huge nosedive

it makes me wonder what moderation changes or policy happened around that time, but at the same time it might just be novelty wearing off and people who previously contributed moving on

around the same time growth plateaued

in hindsight, i shouldn't be surprised "chatgpt kills teenagers" rings hollow

given the whole three decades of school shootings and nothing meaningfully changing for the better, more dead teenagers on the pile doesn't change much about anything