Guidelines For Lem-Inspired Stories?

[Dragon of Improbability]

The Dragon has emerged from Nothingness again,

In the opinion of the most devoted Lem fans on Solaris Station: What are the guidelines to writing fiction (or science fiction) in the vein of Stanislaw Lem? More specifically: what are the steps required to write stories that are like Stanislaw Lem's?

[OlegY]

good enough scientific-related education =)

[Terminus]

Why would you want to write like Lem? Just don't occupy your mind with that... If you love his books it will show up in what you write.

It is true Lem was very well educated, very intelligent (his speculated IQ was around 200) and imaginative. But knowledge itself isn't most important - in cases of books like Solaris he, instead of knowing anything, had to invent a whole new branches of science connected with the planet. It showed that he had done his homework about how science works, developps, what conflicts it has within itself, and so on. I believe it gives a hint about at least one aspect of "Lem's method" - he got to know some mechanisms to great extent, and then let his fantasy do the rest, while keeping to the details with great care.  Also, he used Polish language such extraordinary ways... But this obviously could hardly be copied even if you were Polish, and you probably are not

By the way, let us know when you write anything

[OlegY]

"If you love his books it will show up in what you write".

Terminus, you are really getting the point! I bet you have your own experience of writing novels which are worth reading. Please give me the link. I cannot boast of something like that but my younger brother is training to compose stories and periodically asks my consultation. So I feel myself somehow adhere to story-writing activity 

[trx]

Composing stories (good ones) is not this kind of human activity that could be taught by any simple methods, resembling algorithms. If one thinks that he/she may achieve this skill by reading any guides ("10 steps to become a famous poet", etc.) or by writing for practice, he/she is deeply sunk in illusion. There are thousands of factors, making you a a great writer/mathematician/etc.: a set of your genes, your childhood, parents, access to books, libraries, culture. Dozens of them.

[Terminus]

Well, I'm not sure. Maybe a really good algorithm with large database could try its chance

[OlegY]

respect to trx. I agree. The art of story-writing is far from being practicing some algorithms. That is why  I like Lem and Strugatskyie. I think the crucial factor here is personal expirience of existance. Funny, it sounds somehow vague in English, but I'm sure my Polish friends can understand it alright. Plus something else of course. May be the ability of seeing things as they are and the courage of being able of sharing this knowledge with  others?
  Sorry, I'm getting too complicated.
I'm sure if you were brought up in an intellectual family having pure gene code and were graduated with high level academic education, but if you've never seen life as it is with its absurd stupidness, violence and psycosis - and from the other hand - with those men who can adequately stand it all showing good resistance and professionalism - then you hardly have a chance to be undersood by people. Leave alone composing stories which could impress us. 

[Terminus]

Alas, I still beg to differ.
Oleg gives an uplifting description of how wonderful person a good (sci-fi) writer must be: abnormally intelligent acute observer, magically tunelling his creativity into ideas never before seen, still keeping it all in agreement with contemporary (and, by projection/extrapolation) future science. All that is obviously true (though I dare say quite a few good writers fail to meet these criteria), but that's not against the point I was trying to make.
In short: creation of a story, even though Lem himself considered such task an ultimate test for AI, is still only as possible as creating AI. I too once objected the idea (of AI) but now I think that task is somehow overrated, not to say that programmers working on it are getting better - cause they still seem pretty far from it - but rather because the "I" in "AI" seems something different to me nowadays. More elusive and illusive  than ideal; an umbrella term encompassing few other, of which some recently became quite attainable, like e.g. problem solving or communication.
Besides, as proven by Lem's own 'Sci-Fi calculator', a "weak"(/"foolish") writer is not far from being a machine, merely choosing should Jack be eaten by Christmas Mechanic Gorillas or save Jean from the Aliens. So why couldn't another machine take it a little further?

[OlegY]

  that's great. I wish there could be another third man able to discuss this theme. To me the most interesting issue here is the difference between I and AI. I cannot boast of my high iQ (but I honestly can complain to my low one =) The question is that AI is the product of  I (could you imagine the contrary?) so it cannot be smarter that its producer. Is there someone to argue that?  So the perfect story-teller cannot exist

[tzok]

The question is that AI is the product of  I (could you imagine the contrary?) so it cannot be smarter that its producer. Is there someone to argue that?

I can imagine the contrary. I mean computer-designed bio-cells, made entirely by robots. That's sci-fi today, but possible in the future in my opinion. And as for the second question I also disagree with you. Look, humans can develop brilliant algorithms and learning techniques for computers. Then they can run the programs and wait, wait, wait until some super-AI develops. If you have some mathematical problem one person cannot solve due to its computational complexity, you can delegate the job to computers. Then the "only problem" human needs to solve is to program the algorithm appropriately. That's easier to do and with computer help - it yields a result. The same can happen with AI development. We will "only" have to program computers to do it.

[OlegY]

Thanks, tzok.
First. Who has produced the robots? Human.
Second. "We will "only" have to program computers to do it" - to develop AI. Sorry, pal. We are now still as far from that as we were when Lem wrote HMV. It's mere sci-fi so far.

I wish  you read his later works.
Lem was trying to ensure the audience that interstellar voyages, contacts with extraterrestrial mind, artificial intelligence, journey through time and other old-fashioned sci-fi stuff is not essential now.
Yes, we are facing technological breakdown. Brand new magical substances, medicine recycling life organs and rejuvenation, more astrophysics resuts, Large Hadron Collider and so on. But we are still alone, we are still human, we are still animals. Wild and stupid. And that is the problem to handle and work out. Not AI.   

[tzok]

First. Who has produced the robots? Human.

So? That doesn't exclude the other way round.

As for the rest of your response. I think you are mixing two things. First - is there EVER a possibility to create true artificial intelligence? Second - if yes, where are we now?

For the first question, I think the answer is true. And also in my opinion, Lem with all his pessimism from later works will agree. He always pointed out all dangers and odds, but still remained on position, that someday it will become possible.

For the second question, I am not even willing to doubt in what you say. It is true, that we are all still "wild and stupid animals".

I just thought we are concentrating on the first question

[Terminus]

Just as well, if we're 'wild & stupid', then it should be easier to construct AI that matches our intellect

[OlegY]

Just as well, if we're 'wild & stupid', then it should be easier to construct AI that matches our intellect

Terminus, no way 

Ok, let me be so stupid not to believe that a today's "perfect consumer" could invent and produce technology for improving himself. The first stage to that for him is to learn to become aware that he's nothing more than a "perfect consumer". As I've read in one of Eugine Lukin's books: To get started from zero you should first achive that zero point.
 

Tzok: "Then the "only problem" human needs to solve is to program the algorithm appropriately."
Well that's where we differ. It's a classical problem: to put the question the right way means to almost answer it, something like that. 

One more from you: "is there EVER a possibility to create true artificial intelligence? Second - if yes, where are we now?"

I'm an optimist, I believe yes. Where are we  now? The same place as ever before, may be even farther from the solution than we where in Rennaissance age for instance. At that time  naturalists constructed mechanisms and made an autopsy by their own hands. And now you have only be able to push the computer key to be a scientist.
I think the first thing on the way to create an AI is to establish what an "ordinary" intelligence is and how it can be reproduced or replicated.
Do you agree with that? 
Thanks for keeping the dialog alive!

[tzok]

OlegY, I think I understood what you meant. You are saying, that people are treating technology in a consumerist manner - to have better experiences, more fun, more joy, etc. but no one really looks for a good direction to choose in further development. And looking at all of this from some perspective makes us doubt in any real possibility of AI development, at least in the near future. If that's what you meant - I agree.

Tzok: "Then the "only problem" human needs to solve is to program the algorithm appropriately."
Well that's where we differ. It's a classical problem: to put the question the right way means to almost answer it, something like that.

Of course my words were a large oversimplification . I just wanted to highlight a mechanism, that it's easier to write:

Kod: [Zaznacz]

for i := 0 to 1000:
    output ithan to write numbers from 0 to 1000 yourself.

One more from you: "is there EVER a possibility to create true artificial intelligence? Second - if yes, where are we now?"

I'm an optimist, I believe yes. Where are we  now? The same place as ever before, may be even farther from the solution than we where in Rennaissance age for instance. At that time  naturalists constructed mechanisms and made an autopsy by their own hands. And now you have only be able to push the computer key to be a scientist.

I think that you're stereotype-thinking a bit here . Renaissance is often portrayed as a golden age of humanism, but remember that what we learn now on history are stories about individuals from that time. We all associate Renaissance with Leonardo da Vinci - a brilliant genius... but one person.

And look at modern civilization. Life expectancy has never been so long (it drastically increased in 20th century). We can cure many illnesses, which were fatal not so long time ago. We can travel all around very quickly and easily. We can communicate instantly with people on other the Earth. For the first time in history, the knowledge is free - if you have access to the Internet, it's no matter you're from western Europe or maybe some poor country in Africa.

Just to state it clearly - I am generally on the same side as you . I am a misanthrope myself - this civilization and this common consumerism are disgusting me. I am saying all of this above for the sake of objectivity. Because with all the greatness of Renaissance or maybe Enlightenment, these were cruel and savage times not less that ours. And also with all stupidity and dumbness of today, we are fortunately not in the worst times ever . Every time has its pros and cons.

I think the first thing on the way to create an AI is to establish what an "ordinary" intelligence is and how it can be reproduced or replicated.
Do you agree with that? 

Yes I do, and that's a good starting point. However that's also so much of an open question that I don't think we'll find an answer to