Live data from Hacker News

We were promised Strong AI, but instead we got metadata analysis

calpaterson.com

201–210 of 427 posts

Re: We were promised Strong AI, but instead we got metadata analysis

#201

Earlier quoted context omitted.

There is a patch thanks to contributor Marx but everyone keeps pressing "remind me later".

Probbably due to the memory of the deaths of millions of people: https://en.wikipedia.org/wiki/Mass_killings_under_communist_... You get bit by a viper in the bathroom and your going to avoid that bathroom for a while.

> This article has multiple issues

Re: We were promised Strong AI, but instead we got metadata analysis

#202

Earlier quoted context omitted.

It would be interested to know how this process looked like. I'd imagine it was very gradual and consisted of little steps each leading to the next one, and when they finally started to serve ads they were probably thinking it's not that bad. I believe crossing that line made paved the way for considering things like tracking the whole web morally acceptable, maybe even positive.

The DoubleClick acquisition was the inflection point, even obviously at the time. Before that AdWords was reasonably in-line with Google's standards and culture.

We ran a DoubleClick ad engine on the Clear Channel websites right around that time period. The front-end was a Java Web Start application, ugly, slow, but very intuitive to use for ad-targeting. I wrote several extensions for it using their plugin-api and the ad engine worked really, really well. Had several calls with the DART team and they were all smart people delivering a necessary service to keep websites free to use. Hard to believe that something so simple and useful could become the tool for evil that it has.

Re: We were promised Strong AI, but instead we got metadata analysis

#203
post #134

Earlier quoted context omitted.

From the article > 12-moth plan > 6-month plan Why would a dating site have a 12-month plan, and why would a user of a dating site want a 12-month plan? Not only would you hopefully want to be off the site within 12 months, as soon as you found someone compatible, you would hopefully delete the app, but you've unnecessarily paid for months you will (hopefully) never use. I don't understand why anything but month-to-m…

The cost of 12 months of a dating site is trivial compared to the benefits of finding the right person. If someone offered you a soulmate if you gave them a couple hundred dollars, you'd take it in a second, right? Paying ahead actually aligns your incentives better, because the site is no longer incentivized to drag you along single month after month to keep you paying.

Effectively, you're not paying for "12 months" despite the label, you're paying for a significant chance at finding a soulmate? If that's the case, why not label it as such?

Re: We were promised Strong AI, but instead we got metadata analysis

#204

Earlier quoted context omitted.

Yep. Money is the one unpatchable zero-day for every platform and service on earth.

There is a patch thanks to contributor Marx but everyone keeps pressing "remind me later".

Let me know when we reach post scarcity and no longer need a rationing device such as money.

Re: We were promised Strong AI, but instead we got metadata analysis

#205
post #197

Earlier quoted context omitted.

Or, maybe, "Everybody is altruistic until they have shareholders." The idea that companies should only be beholden to shareholders that has taken firm hold over the past 50(+/-) years doesn't look to be a good one, in hindsight.

Past 50 years, really? "It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest." - Adam Smith, 1776

> The idea that companies should only be beholden to shareholders

This concept was popularized by Milton Friedman with his Shareholder Theory. OP is not wrong.

https://en.wikipedia.org/wiki/Friedman_doctrine

Re: We were promised Strong AI, but instead we got metadata analysis

#206

Earlier quoted context omitted.

> I literally append "site:reddit.com" Stop saying that! The more people repeat that on HN, the more we risk advertisers realizing it and doing reddit-targeted SEO and reddit will be useless too! (I'm only half joking) Also, it's funny, I was just reading the thread about malaria eradication and DDT-resistant mosquitoes; the problem of SEO is eerily similar (any countermeasure is eventually defeated by evolution).

You are 10 years too late for that concern. I often wonder if there are any real people commenting, submitting, or upvoting left on Reddit.

This is emphatically not the case in my experience.

Re: We were promised Strong AI, but instead we got metadata analysis

#207

Earlier quoted context omitted.

Google page ranking depends on how long people spend on the webpage, so if you have to wade through longform text, it mistakenly assumes that the content is quality. Note that most recipe blogs are fake. Someone wrote the original content long ago. Then, someone else hired a copywriter off a freelancing platform to change the words of the text just enough to avoid copyright violation, then put up a new copycat websit…

> Google page ranking depends on how long people spend on the webpage, so if you have to wade through longform text, it mistakenly assumes that the content is quality. Is this true? I have a hard time believing it because it's a pretty naive assumption. A site that lets me get what I'm after quickly is generally going to be what I want.

I worked for a startup obsessed with vanity metrics. They would freak out if the average time spent per article dropped down, even if our conversions increased. Just an odd bunch of people. They went down with the company while doing nothing but micro-optimizing for bad metrics.

Re: We were promised Strong AI, but instead we got metadata analysis

#208
post #134

Earlier quoted context omitted.

From the article > 12-moth plan > 6-month plan Why would a dating site have a 12-month plan, and why would a user of a dating site want a 12-month plan? Not only would you hopefully want to be off the site within 12 months, as soon as you found someone compatible, you would hopefully delete the app, but you've unnecessarily paid for months you will (hopefully) never use. I don't understand why anything but month-to-m…

The cost of 12 months of a dating site is trivial compared to the benefits of finding the right person. If someone offered you a soulmate if you gave them a couple hundred dollars, you'd take it in a second, right? Paying ahead actually aligns your incentives better, because the site is no longer incentivized to drag you along single month after month to keep you paying.

This is bordering on logic like the following: - Water is really important, why don't you buy this $100 bottle of water. - The site has an incentive to improve your dating outcomes. No, it's primary objective is to maximise revenue, everything else is a side effect. - Paying more for something means someone will commit/ follow through, somehow raise incentives. This is just a guess, not supported or disproven by reality.

I like to think defensively especially when it involves companies. What are they doing, and what do they stand to achieve?

These apps have not shown any value to their users, paywall their content and have an aggressive-long-term subscription model because they have optimised themselves straight into the garbage can, by thinking short term.

Re: We were promised Strong AI, but instead we got metadata analysis

#209

Since Dartmouth AI looks like a story of naive visions, exaggerated promises, massive disappointment, recurrent divisionary tactics, rebranding and snak oil sale. Until finally a light on the horizon became visible and academically camouflaged wishful thinking could be materialized into usable products. A groping in the dark, nothing more. There are probably good reasons why the blind watchmaker needed billions of ye…

The blind Watchmaker -> creator of the physical Universe The seeing Watchmaker -> human engineering "(We) wish the human some progress, obviously egotistical and delusional, on whatever the human makes next" I believe this slightly poetic word salad references certain theological problems about the capacity of man compared to a Creator. The conclusion is that the future is uncertain and the human is flawed, but an in…

No creator, no teleology, no theology, no poetry, just Richard Dawkins:

https://en.wikipedia.org/wiki/The_Blind_Watchmaker

I thought it was common knowledge, but maybe I'm getting old.

Re: We were promised Strong AI, but instead we got metadata analysis

#210
post #77
post #60

I really like the expression "metadata analysis". It's very succinct, describes very much what AI/ML often boils down to in 2 (rather) simple words. I will try to remember this. The marketing guys in cooperation with journalists won the battle for now but words will be replaced by others again and again and even change their meaning, so maybe next time the shiny new technology will get more adequate wording? (well, m…

The minute 'AI' solves a problem the problem is removed from that class of problems that require 'AI'. If the problem is difficult enough we might award the new solution its own name.

“Moving the goalposts” has long been identified as a problem by AI enthusiasts, but the opposite is also often true—an AI is built to solve a limited or highly restricted form of a problem, and then proponents claim that the goalpost has been moved when it’s observed that the general problem remains unsolved.

It feels to me that we started calling ML “AI” when deep learning became powerful enough to work on less clearly structured problems like vision and NLP—but “find patterns in complex data” does not seem powerful enough for what I would consider “intelligence” (artificial or otherwise). I don’t think that stateless/idempotent ML is capable of what most people would recognize as intelligence; in part because I suspect a history is required for a system to be self-correcting over time.

Post reply on HN