From a recent post by a fly.io employee in their forums [1], in response to being asked "do you expect it to be ready for production use this year?"...
>Coming very soon, we’ll be moving to public alpha (which in our world, comes after private beta :smiley:).
What benefits have we gained from generative AI (really ML) that currently outweigh the cost of researching developing, and running them? Or do we stick with an expected value based on what generative AI may be able to do and the next step technologies that come after it?
The article is making pretty clear arguments for costs of generative AI, and raising the author's opinion that it isn't worth it. Just claiming it is in fact worth it without anything to support that isn't super helpful.
Last November, DeepMind released the results of a generative AI model that created theoretical molecular structures for over 2 million undiscovered synthetic materials. Within days, materials science researchers were able to confirm 700 of them. The sheer number of these new potential materials discovered is greater than has been created in the rest of human history combined. These are materials that can be used in manufacturing, energy production, and other objectives that are critical not just for advancing human society, but avoiding the impending crisis we are already facing.
Similar AI endeavors have been underway for medicine and human health.
The author is making extremely shallow, flawed arguments that hinge on an ignorant (or possibly, deliberately narrow-minded) understanding of what generative AI is, how it is already being used, and the magnitude of what is already being achieved with it.
It will be interesting to see how many of those, if any, pan out to have a meaningful use at scale. If I remember right, those 700 or so were synthesized in a lab but I don't think we know much beyond that.
We'll see if any of them end up being viable as far as manufacturing and material availability go, and whether they're better replacements for existing tech like batteries. The hope is that we'll have Jarvis inventing a new material for Iron Man's suit, but we could always end up with an endless pile of technically feasible but functionally useless materials.
A few billion moderately interesting images, and a collection of slightly weird nearly-free marginal cost interns who studied literally every subject but somehow still only act like freshly minted graduates with no real-world experience.
With EVs the focus becomes how wonderful they are because they do not burn fossil fuels in their engines. Great, but what about what all the other issues (including non renewable issues in the rest of the supply chain involved in building the EV)? They're greenwashed away - no need to discuss public transit and densification. EVs will fix everything and we don't need to change our life on any significant way.
They are a way to continue going down the wrong path and feel good about it.
it's a not a waste of time but imo it's also not going to be disrupting. it could be a good sounding board when you know the subject otherwise it's mostly useless. it does makes things convenient but how much value it add, that's yet to be seen.
>Also, author left out how Lula was in jail for corruption before he became pres again.
You left out a few things, like the corrupt judge that sentenced Lula during an election where he was the absolute favourite, and months later was made minister by the genocidal president he helped elect.
Thankfully a modicum of sanity is being restored and the corrupt judge is now being investigated.
Bullshit. Moro vice is vanity not corruption. Lula was freed not because he is innocent but because the mafia judges nullified the process. Crime took a significant hit when Moro was minister of justice, reduction in homicides, effective border control. No wonder that criminals and their leftists supporters hate him. The current government is incapable of fighting crime and it is thriving to the dismay of millions of daily victims. I'm ashamed of being Brazilian because of people like you.
You are factually wrong. The homicide rate is still declining as of 2023. Also worth it to note that the military dictatorship was responsible for the explosion in urban crime and violence in the 70s and 80s, which persists to this day.
The feeling of shame is mutual and I'll add disgust. I decided to leave Brazil after Bolsonaro was elected and I heard gunshots and death calls from neighbors, savages like yourself.
The things that made the world shocked and disgusted about Bolsonaro - the promotion of violence, the ties with the milícias (mafia), the contempt with human life and dignity, the hate of the natural environment - foreigners did not understand that those were exactly the things his supporters, like you probably are, loved about him.
Bolsonaro routinely said things like "we should have murdered 30,000 people (during the dictatorship)". Those statements don't lose votes, they gain votes from an appalling, bloodthirsty, evil minority like the commentor above.
> The feeling of shame is mutual and I'll add disgust. I decided to leave Brazil after Bolsonaro was elected and I heard gunshots and death calls from neighbors, savages like yourself.
You're wrongly assuming I'm a Bolsonaro supporter because in your world anyone that criticizes your idol must be a Bolsonaro supporter and evil. The truth is that both Lula and Bolsonaro are evil beyond doubt but most of their supporters are victims of polarization and disinformation rather than bad persons.
Dude, Brazil was always a savage place. Obviously it's not a good thing but you're very privileged if this is the worst you have faced and you were able to leave the country. The reality is that most Brazilian face worse and can only dream of emigrating.
> Bolsonaro routinely said things like "we should have murdered 30,000 people (during the dictatorship)". Those statements don't lose votes, they gain votes from an appalling, bloodthirsty, evil minority like the commentor above.
What do you expect from a population heavily traumatized by crime and violence where yearly homicides far exceed 30K?
I'm more worried about a minority in Brazil which openly support a socialist revolution with mass executions and all. I'm worried about seemingly educated individuals supporting a corrupt president that is friends with dictators and evil regimes like Putin, Maduro, al-Assad, Gaddafi, Xi, Iran, Hamas.
> You are factually wrong. The homicide rate is still declining as of 2023.
I'm not. In 2018 and 2019 when Moro was Minister of Justice we saw unprecedented 2-digits decline in homicide rate. In 2020 after he was fired homicide rate increased. Further declines were timid in comparison and crime is thriving despite homicides falling because the current system rewards crime and the current government doesn't care about crime.
> Also worth it to note that the military dictatorship was responsible for the explosion in urban crime and violence in the 70s and 80s, which persists to this day.
Moro anti-crime measures weren't by no optic authoritarian. I don't see how those events can be relevant to this discussion besides you wanting to make false implications.
The job location thing is a bit weird, with just Europe, NY and SF. That is basically meaningless except to show the post-covid decline of SF as an onsite location.
Most of the job ads I see are US only or US/Canada.
They've been posting the same ad for a year now. I've been through their recruitment too, with a similar "we're looking for better people" bland rejection in the initial stages.
Having worked for AWS, I respect the "high bar" thing. Fine. But lets be real for a second: Aha! isn't Amazon, and the past year was the strongest employers-market in the last what, 20 years? I have a hard time believing their engineering needs are consistently surpassing their candidate pool.
Either they're wasting everybody's time and not hiring anyone, or they have an insane churn, or they're the next Nvidia and have been consistently hiring dozens of people. I'll go with the simplest explanation.
Greytext is already doing a good job here, but it would be nice to see some additional quality indicators specifically for the job board. I've been toying with the idea of a "freshness" indicator for scenarios like this.
Looking at the author's resume, they've only been at AWS since 2020. It's unlikely they have a very good understanding of how the culture has changed over time.
It’s true, I’ve only been here since COVID and from what I understand 2020-2021 were exceptionally good years. Lots of growth, fully remote, and plenty of new people and ideas.
Well, another option is it’s true and it’s monotonically getting worse. From my time at aws, it appears to be that’s the case. It was better X years ago, and it will be worse at X+n years for all values of n and X.
Same at Google - was there from 2016 to 2023. And i strongly believe everyone at BigCo underwent a step-change in 2023.
I do give the VP in TFA some credit for trying to give shelter and time. Google went rancid, you'd go to work one morning, get an unannounced event on your calendar, and be informed as a group you were locked out of everything except the internal job posting site and had 8 weeks to find a job. Meanwhile everyone had negative headcount. I've heard at least two dozen stories like that personally.
Another effect is best communicated by a quote: "there's a fuck you got mine attitude." The internal peer support group i was a location lead for was overwhelmed by people with just bonkers stories of aggression and antisocial behavior from middle management punching down. I heard so many stories that wouldn't have gone down at even the strangest and most inexperienced startups I worked at. (Hiring friends with 0 experience, outright lying, pushing out people)
The bullwhip effect of management not really needing to learn to capital-M manage and now being asked to really exposed a lot of issues. Breaks my heart. (I left because of this)
It is true. Amazon was always brutal, but two-pizza team six-pager culture was a great place to rule in hell as opposed to serving in Heaven as a generalist at Google. Two-pizza team six-pager culture died sometime during the last decade. At a guess, when they imported so many AI academics and they brought all the toxic worst practices of academia to bear on it.
The Applied Scientist title at Amazon is the single worst thing they ever created: 15% higher comp and stock if any L10 or up decides you are one, a 100% political position that set everyone against each other just like separating the bonus for the success of Google+ from the rest of Google did in 2011 (was there for that idiocy myself).
>the only way I track traffic trends on this blog is with vbnla, a hacky little Ruby script that summarizes my Nginx access log.
This is literally analytics. I didn't read the rest of the post, so if you're arguing against things like GA, fine, I'm with you. But any kind of analytics, even GA, ultimately does the same thing, just with a ton more complexity and additional tracking sources.
The post is about browser analytics; this is in the second sentence. It's possible you skipped that to get to the part about the script.
(I disagree that this adds any complexity or tracking sources. It's a 70 line script that took me about half an hour to write, and has required no significant changes since I wrote it ~5 years ago.)
Not the original commentor but I've never seen the term browser analytics used before. That certainly doesn't help. Maybe client-side analytics would've been clearer.
Re: added complexity, I was referring to analytics solutions like Google analytics: it adds a ton of sophisticated tracking stuff, but is ultimately a server (well, more like several thousand) somewhere with logs registering visits, not so fundamentally different from your script. I also do a very similar thing (also in Ruby!). Apologies for the lack of clarity, English can be so ambiguous.
