Malazan Empire: Algorithms and automation - Malazan Empire

Jump to content

  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

Algorithms and automation

#1 User is offline   Azath Vitr (D'ivers 

  • Ascendant
  • Group: Malaz Regular
  • Posts: 3,610
  • Joined: 07-February 16

Posted 08 July 2019 - 11:25 PM

View PostGrief, on 08 July 2019 - 06:27 PM, said:

View PostAzath Vitr (D, on 08 July 2019 - 05:51 PM, said:

Tax the whole professional sports enterprise, or at least the most exorbitantly and unnecessarily overpriced aspects of it (with tax incentives for making meaningful improvements). Taxes (or charities) should effectively be returning that value to the people. It then becomes a selective tax on people who spend money on professional sports. (Amateur, participatory sports are a social good, promoting exercise, teamwork, sociability, etc. (though exercise is counterbalanced by injuries, sociability by aggression and hostility, etc.); but highly exclusionary and absurdly expensive pro sports tickets are not a social good. Pro sports encourage people to sit around watching other people play sports instead of exercising; and while they may reinforce community bonds, they do so by encouraging hostility towards other communities.)


Sure, you could selectively tax people for spending on that particular hobby. Though personally I would tend towards simply taxing the higher-wealth brackets far more. Professional athletes are far from the least deserving of their wealthy peers.

View PostAzath Vitr (D, on 08 July 2019 - 05:51 PM, said:

Technology has made this much easier. (While it also makes hacking or sophisticated schemes a possibility, they're almost certainly only minor issues that wouldn't affect most of the volume.) In a lottery system, a simple option would be to require the name on the ticket to match the name on government-issued ID or a passport (could that get conservatives on board? realistically though, none of this will happen in the United States in the near future).


It has. Then you get into a debate over what the "correct" value of a ticket should be since it's not being set by a market but rather by a central authority. In the event where this money is taken from the athletes, you're essentially removing wages from the worker to give to the consumer. Which feels somewhat debatable in terms of morality. Watching professional sports isn't particularly more productive than playing them, after all.

View PostAzath Vitr (D, on 08 July 2019 - 05:51 PM, said:

The main advantages of capitalism---quickly and reliably matching supply to demand, providing motivation, etc.---will plausibly be superseded by ubiquitous real-time personal data collection, AI, the internet of things, and the administrative surveillance state. It's perhaps too bad that this will probably first be achieved in China, given their severe restrictions on intellectual freedom (among other things...).


If anything this makes the problem worse. Rather than superseding capitalism, the system you're describing is essentially hyper-capitalist. Algorithmic pricing discrimination works to more accurately gauge exactly what a consumer is willing to pay on the market (i.e, where on the demand curve they fall). What you're describing seems more like optimizing market-economics than superseding it, to me. It's a technological way to ensure that companies maximise their surplus while giving the minimum they can to the customer (i.e, selling at exactly the price point where consumers will buy, rather than customers being able to find bargains where the product costs less than they would pay).

Indeed, the potential for social harm and anti-competitive behaviour that is facilitated by algorithmic pricing has been a pretty hot topic in policy circles for the last few years (for one example see the OECD, Algorithms and Collusion: Competition Policy in the Digital Age).

View Postworry, on 08 July 2019 - 05:53 PM, said:

Grief is right (until we abolish capitalism).


Oh hello new sig quote it's been a while.



I was referring to the traditional advantages of capitalism over managed economies (for example, actually existing forms of communism, which relied on a top-down approach to data science that generally proved less effective, even though they tried to collect and process vast amounts of information with the tools then available; but advances in data collection and AI are now making a more bottom-up, real-time approach, with vastly more data and much better predictive models, seem feasible in the near future), and to using these tools in the context of a (primarily or fully) managed economy. In a way it would still maximize a sort of abstract 'capital' in the form of measures of efficiency, adapting the main positive aspects of capitalism. But the equivalent of 'corporate profit' would go to the general welfare, and motivation for human performance (to whatever extent human or transhuman labor it hasn't yet been automated away) would be based on fine-grained, data-driven scientific approaches to what actually motivates the individuals in question most effectively (that can also be ethically implemented, perhaps with virtual proxies---if people want power, for example, or prestige, or whatever 'wealth' in a post-money world might be abstracted into---human brains almost certainly didn't evolve to want 'money', or even to hoard gold, but what money provides, symbolizes, or allows).

But hypercapitalism of the sort you're referring to is already helping to create most of the information processing infrastructure necessary for a managed economy to replace capitalism.

Similarly, labor is increasingly going to be replaced by automation, including intellectual and even creative labor, rendering labor for capital superfluous. (The idea of labor as an intrinsic good is contrary to what major thinkers and authors in the western tradition believed until just the last few centuries. Aside from the potential for meditation (as an accidental consequence of interacting with human physiology, not logical necessity in the transhuman context), labor is odious. The idea that it's otherwise (not counting activities like self-perfection or cultivation), that work in itself is a positive, is an odious ideological tool for manipulating people to labor.
0

#2 User is offline   Obdigore 

  • ThunderBear
  • Group: High House Mafia
  • Posts: 6,165
  • Joined: 22-June 06

Posted 09 July 2019 - 01:02 AM

View PostGrief, on 09 July 2019 - 12:06 AM, said:

Automation may encroach on large parts of the workforce, but it's not clear that we're going to see the 'end of work' in the near-term. Humanity is also very good at creating new jobs and AI still has major limits for all the evangelism around it. If we automate everything to such a post-scarcity and post-work extent I agree then we're looking at a totally different society which needs a totally different social contract. But this is far from a near-term trend.


I will say I fear you disregard the amount of people doing 'do this thing right 2000 times a day' jobs that our economies have. For instance, long haul truck drivers make up the largest proportion of single income households in many US states, of any job. When that gets automated, all those people are out of work. Many of them, and their wives/kids, don't have any other skills, and you can be sure things like stocking supermarkets using automation is right around the corner.

Remember that during the Industrial Revolution, hundreds of thousands of people died in Europe because they didn't have skills that translated to the new workforce, and thus couldn't feed/house/whatever themselves. While that is tragic, I think people are dismissing the effect of this massive economic shift in our current consumer economies. If people dont have money to buy things, and they don't have an income and so aren't being taxed, neither the government nor corporations will be bringing in money. No one is preparing for it, but it is happening very fast.
Monster Hunter World Iceborne: It's like hunting monsters, but on crack, but the monsters are also on crack.
1

#3 User is offline   Macros 

  • D'ivers Fuckwits
  • Group: High House Mafia
  • Posts: 9,147
  • Joined: 28-January 08
  • Location:Ulster, disputed zone, British Empire.

Posted 09 July 2019 - 05:45 AM

Thread split required.

We already have automated check out systems, shelf stacking is next
0

#4 User is offline   Aptorian 

  • How 'bout a hug?
  • Group: The Wheelchairs of War
  • Posts: 24,785
  • Joined: 22-May 06

Posted 09 July 2019 - 06:40 AM

Libraries are also increasingly seeing more and more automation and streamlining.

Many of our tasks are centralized. Manual work is reduced by more intelligent systems. Even finding books and restocking them is currently being planned for replacement by a robot on wheels.

Coupled with Copenhagen's administration's hard on for unmanned hours, one day soon there just might not be any librarians in public libraries. You'll have to talk to a screen instead.

That they'll probably outsource to India.
1

#5 User is offline   Macros 

  • D'ivers Fuckwits
  • Group: High House Mafia
  • Posts: 9,147
  • Joined: 28-January 08
  • Location:Ulster, disputed zone, British Empire.

Posted 09 July 2019 - 07:02 AM

thats cute, you think there will be libraries in the future.
0

#6 User is offline   D'rek 

  • Consort of High House Mafia
  • Group: Super Moderators
  • Posts: 14,619
  • Joined: 08-August 07
  • Location::

Posted 09 July 2019 - 05:23 PM

An oldie but a goodie


View Postworrywort, on 14 September 2012 - 08:07 PM, said:

I kinda love it when D'rek unleashes her nerd wrath, as I knew she would here. Sorry innocent bystanders, but someone's gotta be the kindling.
1

#7 User is offline   worry 

  • Master of the Deck
  • Group: Malaz Regular
  • Posts: 14,823
  • Joined: 24-February 10
  • Location:the buried west

Posted 12 August 2019 - 07:53 PM




They came with white hands and left with red hands.
0

#8 User is online   Tiste Simeon 

  • Faith, Heavy Metal & Bacon
  • Group: Malaz Regular
  • Posts: 12,442
  • Joined: 08-October 04
  • Location:T'North

Posted 12 August 2019 - 08:06 PM

Why? What do Google/YT have to gain?
A Haunting Poem
I Scream
You Scream
We all Scream
For I Scream.
0

#9 User is offline   Aptorian 

  • How 'bout a hug?
  • Group: The Wheelchairs of War
  • Posts: 24,785
  • Joined: 22-May 06

Posted 12 August 2019 - 08:10 PM

I worry that it's not so much what they have to gain, than how much effort it would cost to fix.

I imagine they're constantly tinkering with their algorithms, they're just not sure how to fix it with out either banning large swaths of content randomly or having to manually approve or disapprove mountains of content.
0

#10 User is offline   Cause 

  • Elder God
  • Group: Malaz Regular
  • Posts: 5,922
  • Joined: 25-December 03
  • Location:NYC

Posted 12 August 2019 - 09:29 PM

Yip, it may just be that the mindset of people speaking to spread proganda is to manipulate SOE as much as possible.
0

#11 User is offline   Whisperzzzzzzz 

  • Reaper's Fail
  • Group: Malaz Regular
  • Posts: 2,479
  • Joined: 10-May 10
  • Location:Westchester, NY

Posted 12 August 2019 - 10:19 PM

Two of the most reputable, well-researched, evidence-backed health and nutrition sites — SelfHacked and Examine — have seen their traffic eviscerated. Most of their traffic previously came through Google, and a recent Google search algorithm update heavily de-ranked them (along with much shadier, unsourced "health" sites, like Mercola and Dr. Axe).

On SelfHacked and Examine, almost every sentence contains a citation linking to a peer-reviewed study. Yet, somehow, that's not good enough for Google.

https://selfhacked.c...the-next-level/

https://examine.com/...date-july-2019/
0

#12 User is offline   Gorefest 

  • Witness
  • Group: Malaz Regular
  • Posts: 2,988
  • Joined: 29-May 14
  • Location:Sheffield

Posted 13 August 2019 - 08:53 AM

I guess that is the issue with algorithms, you always lose some of the good with the bad. And people who understand the algorithms can probably tweak their videos in such a way that they actually benefit from them. But as there are too many videos being put up on a daily basis for hands-on checking by staff, what is one to do? It's the bane of a social media age. Look at the amount of drivel you encounter on sites such as Reddit or Facebook, it is just becoming impossible to police effectively.
Yesterday, upon the stair, I saw a man who wasn't there. He wasn't there again today. Oh, how I wish he'd go away.
0

#13 User is offline   Cause 

  • Elder God
  • Group: Malaz Regular
  • Posts: 5,922
  • Joined: 25-December 03
  • Location:NYC

Posted 13 August 2019 - 09:00 AM

View PostWhisperzzzzzzz, on 12 August 2019 - 10:19 PM, said:

Two of the most reputable, well-researched, evidence-backed health and nutrition sites — SelfHacked and Examine — have seen their traffic eviscerated. Most of their traffic previously came through Google, and a recent Google search algorithm update heavily de-ranked them (along with much shadier, unsourced "health" sites, like Mercola and Dr. Axe).

On SelfHacked and Examine, almost every sentence contains a citation linking to a peer-reviewed study. Yet, somehow, that's not good enough for Google.

https://selfhacked.c...the-next-level/

https://examine.com/...date-july-2019/


I'm not so sure this is a bad thing. A quick look at self hacked leaves me feeling that its not doing any harm but I'm not sure its doing any good. Articles on Canabis oil for depression, hydrogen water and Glutathion supplements. All referenced, all clearly saying to take the article with a grain of salt and all utlimatly meaningless.

Canabis oil might be shown to do all sorts of things for rats or for cells in vitro but this means nothing if you take it orally. There is no proof. Is it harmful, no. Is it workin, who knows? If your depressed and know it, I would advize seeking medical help and not self treating your self diagnosed ailment.

Hydrogen water? even the article says the little research is biased and who knows what it does.

Glutathione. An important co-factor for perhaps the most abundant protein family of enzymes in nature. If your not sick your body is making enough for you. Gluthione taken orally gets destroyed anyway, your better off taking the amino acids that your body needs to make the tri-peptide. Though again you likely don't need it.

Also the simple truth for the bio-hacking movement is that its all anecdotal. Even proper medical studies done by scientific professionals have problems with reproducibility. Its unethical to perform the kinds of controls and experiments needed to really determine with accuracy what a humans diet is doing to the body. I once read a published article that showed that circumcision prevents heart attacks. The evidence was clear as day but cutting off some skin on the penis is very unlikely to do anything heart related, its likely that people whoa re circumcised being jewish, muslim, maybe wealthier (its popular in North America) is the contributing factor but that's not what they saw. Science used to say an aspirin a day was good for you, now they advise against it. They used to say eat only unsaturated fats, now they are considered carcinogenic. This is the greyest area of science imaginable.
0

#14 User is offline   Cause 

  • Elder God
  • Group: Malaz Regular
  • Posts: 5,922
  • Joined: 25-December 03
  • Location:NYC

Posted 13 August 2019 - 09:09 AM

View PostGorefest, on 13 August 2019 - 08:53 AM, said:

I guess that is the issue with algorithms, you always lose some of the good with the bad. And people who understand the algorithms can probably tweak their videos in such a way that they actually benefit from them. But as there are too many videos being put up on a daily basis for hands-on checking by staff, what is one to do? It's the bane of a social media age. Look at the amount of drivel you encounter on sites such as Reddit or Facebook, it is just becoming impossible to police effectively.


I have always liked the idea of a small, controlled portion of the internet. If its possible. If the internet as we know it is the average and the dark web is too horrible for my eyes to see Id be happy for a small content controlled section of the internet (open to abuse as all things are obviously) where content is hand checked. The obvious question is who filters and checks the content but Id be happy to pay a fee to access this part of the net, and websites may need to pay a fee too but instead of www. you type wwwww. and you can be comfortable knowing that theirs no porn and your child can surf this safe zone. The only news articles will be from reputable sources, the medical adise was ready by three doctors and none raised any objections to what was written and so it was approved.

Who watches the watchmen and all that, and these days you have to read more than one news source to get the truth perhaps but a corner of the internet where posts which tell you the new iPhone can be charged in the microwave are absent would be nice.
0

#15 User is offline   Kanese S's 

  • TMI Frigate Bird of Low House PEN
  • Group: Mott Irregulars
  • Posts: 1,947
  • Joined: 26-April 11

Posted 30 January 2020 - 09:20 PM

View PostD, on 09 July 2019 - 05:23 PM, said:

An oldie but a goodie




Having worked in tech, CGP Grey is spot on with this one.

Remember that to your employer, you are essentially just a biological machine, or even just a biological component to a larger machine.
If automation saves a business any money at all, they will absolutely invest in doing that to save labor costs, because even if they know that throwing lots of people out of work is bad for the economy overall, no business can afford to be left behind by its competitors. If I run a business and everyone else in my industry automates, and I don't, I'm going to fail. It's essentially a prisoner's dilemma.

People think that some tasks are too complex to automate, but most complex tasks are just a series of simpler tasks.

People think that oh, we'll just have more programming jobs, but there is no reason why computer programs themselves can't take over programming.


This is a big part of why I became a socialist.

Laseen did nothing wrong.

I demand Telorast & Curdle plushies.
0

#16 User is offline   Macros 

  • D'ivers Fuckwits
  • Group: High House Mafia
  • Posts: 9,147
  • Joined: 28-January 08
  • Location:Ulster, disputed zone, British Empire.

Posted 26 June 2025 - 04:45 AM

Not sure where to put this.
Only a preliminary study, and needs further research but a recent MIT study shows signs that chatgpt is rotting our brains, significantly.

Huzzah for the AI revolution!
0

#17 User is offline   Cause 

  • Elder God
  • Group: Malaz Regular
  • Posts: 5,922
  • Joined: 25-December 03
  • Location:NYC

Posted 26 June 2025 - 03:03 PM

View PostMacros, on 26 June 2025 - 04:45 AM, said:

Not sure where to put this.
Only a preliminary study, and needs further research but a recent MIT study shows signs that chatgpt is rotting our brains, significantly.

Huzzah for the AI revolution!


I saw this. It’s not that scary or suprising a conclusion. People who use ChatGPT to cheat learning and work don’t learn anything. Meaning if my high school teacher asked me write and essay on the American revolution and I ask ChatGPT to do it instead and copy paste it, then yes 1 hour later I won’t recall any of the facts or have any context of what my own essay says or means.

Conversely if I’m a proficient nuclear physicist and use ChatGPT to help generate some graphs or check my writings tone I’m still a proficient nuclear physicist who saved hours of work.

The paper is basically warning against the first case. That ChatGPT can lead to people either willfully, or accidentally misleading themselves into thinking they are learning when they are not.
1

#18 User is offline   QuickTidal 

  • Lord of the Kicks
  • Group: Team Quick Ben
  • Posts: 22,111
  • Joined: 05-November 05
  • Location:Victoria Peak
  • Interests:DoubleStamping. Movies. Reading.

Posted 26 June 2025 - 03:11 PM

View PostMacros, on 26 June 2025 - 04:45 AM, said:

Not sure where to put this.
Only a preliminary study, and needs further research but a recent MIT study shows signs that chatgpt is rotting our brains, significantly.

Huzzah for the AI revolution!


My brother in law has it on his phone and uses it for everything...and yeah can confirm, he's not that bright.
"When the last tree has fallen, and the rivers are poisoned, you cannot eat money, oh no." ~Aurora

"Someone will always try to sell you despair, just so they don't feel alone." ~Ursula Vernon
0

#19 User is offline   HoosierDaddy 

  • Believer
  • Group: Malaz Regular
  • Posts: 8,049
  • Joined: 30-June 08
  • Location:Indianapolis
  • Interests:Football

Posted 26 June 2025 - 04:01 PM

View PostCause, on 26 June 2025 - 03:03 PM, said:

View PostMacros, on 26 June 2025 - 04:45 AM, said:

Not sure where to put this.
Only a preliminary study, and needs further research but a recent MIT study shows signs that chatgpt is rotting our brains, significantly.

Huzzah for the AI revolution!


I saw this. It's not that scary or suprising a conclusion. People who use ChatGPT to cheat learning and work don't learn anything. Meaning if my high school teacher asked me write and essay on the American revolution and I ask ChatGPT to do it instead and copy paste it, then yes 1 hour later I won't recall any of the facts or have any context of what my own essay says or means.

Conversely if I'm a proficient nuclear physicist and use ChatGPT to help generate some graphs or check my writings tone I'm still a proficient nuclear physicist who saved hours of work.

The paper is basically warning against the first case. That ChatGPT can lead to people either willfully, or accidentally misleading themselves into thinking they are learning when they are not.


Right. It's a tool. How are you using it? If you are using it to think for you... you aren't learning how to think. Unfortunately there is already a graveyard of the mind in a lot of places because education has been so diminished and we are just digging graves deeper with it.

Also, it's a TOOL. How is it being forged? Unfortunately, GIGO has been a thing from the beginning and we are now training wholesale "AI" on lots of GI. What do we expect?

This post has been edited by HoosierDaddy: 26 June 2025 - 04:02 PM

Trouble arrives when the opponents to such a system institute its extreme opposite, where individualism becomes godlike and sacrosanct, and no greater service to any other ideal (including community) is possible. In such a system rapacious greed thrives behind the guise of freedom, and the worst aspects of human nature come to the fore....
0

#20 User is offline   amphibian 

  • Ribbit
  • Group: Malaz Regular
  • Posts: 8,074
  • Joined: 28-September 06
  • Location:Upstate NY
  • Interests:Hopping around

Posted 26 June 2025 - 07:19 PM

View PostCause, on 26 June 2025 - 03:03 PM, said:

Conversely if I’m a proficient nuclear physicist and use ChatGPT to help generate some graphs or check my writings tone I’m still a proficient nuclear physicist who saved hours of work.

The paper is basically warning against the first case. That ChatGPT can lead to people either willfully, or accidentally misleading themselves into thinking they are learning when they are not.

I disagree majorly with this and it is because using AI to do any of this kind of work - whether it is for neophyte or expert - is functionally saying "I don't want to do this work and I may or may not check for validity or veracity".

Anything that I have to certify or stand on my career + reputation and deliver, I will not use AI for. It's too risky and the lawyers etc who are doing this are dumb as heck.

I do recognize the corner case where using AI can improve some things for an expert, but it is in a form that is akin to niche, bespoke AI usage with a very vigilant person right there who knows how to use it plus has the considerable capacity to check what is happening. That's a rarer situation than most people will encounter.

This article below mostly encapsulates my feelings right now about usage. I do add that my experiences with people using AI in ways that cross over into my work often feature shitty copy edited paragraphs that eliminate First Nations history, meaningful context about diversity, and flatten longer sections into something like pablum. This is happening because the AI is putting together a statistically likely thing, not an actual and valid thing. First Nations people aren't statistically prevalent, so their names and history get chopped right out as "too hard to understand".

https://www.nytimes....innovation.html

This post has been edited by amphibian: 26 June 2025 - 07:26 PM

I survived the Permian and all I got was this t-shirt.
0

Share this topic:


  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users