Live data from Hacker News

Is Mars Habitable? (1907)

people.wku.edu

21–26 of 26 posts

Re: Is Mars Habitable? (1907)

#21
post #16

Earlier quoted context omitted.

There's a slew of bot-army template comments from new-ish accounts, getting comment credit to work up to voting rights for a spruiking up-vote spree. See: https://news.ycombinator.com/user?id=xzxa https://news.ycombinator.com/user?id=hsag https://news.ycombinator.com/user?id=kidfkj https://news.ycombinator.com/user?id=linkjp https://news.ycombinator.com/user?id=lndsh https://news.ycombinator.com/user?id=mnjksd https:…

I guess we have to enjoy the last year or so before better models are indistinguishable from us? With some better/clever prompting I’m not sure they could be as easily detected now.

I wouldnt mind a bot writing a comment on HN if it's thoughtful, which is HNs trademark. These comments are so damn void of any thought. You can flag them for being thoughtless alone.

Re: Is Mars Habitable? (1907)

#23
There was an interesting more recent article on life on Mars from idelwords. https://idlewords.com/2023/1/why_not_mars.htm

Not proposing little green men who are unlikely but suggesting there may be bacteria like organisms underground which seems quite possible. A reason I think Musk shouldn't be allowed to build a human settlement there in a hurry. It should be more like a national park with no biological humans. We can always have robots and FPV goggles for the visitors.

(1000+ comments on HN https://news.ycombinator.com/item?id=34213549)

Re: Is Mars Habitable? (1907)

#24

Earlier quoted context omitted.

I guess we have to enjoy the last year or so before better models are indistinguishable from us? With some better/clever prompting I’m not sure they could be as easily detected now.

I wouldnt mind a bot writing a comment on HN if it's thoughtful, which is HNs trademark. These comments are so damn void of any thought. You can flag them for being thoughtless alone.

That's the premise of xkcd 810, "Constructive": https://xkcd.com/810/>

It's one of the more spectacular cases in which I think Munroe utterly missed the plot. It's not that it's possible to create bots which at a superficial assessment are making constructive / helpful comments. It's that bots which do this often have covert goals or purposes, "alignments" in the LLM/ML context. These might or might not be known or evident to either the bots' creators or users, and might even be exploited by still other parties.

But bots are capable of a scale of operations which fundamentally changes the community dynamics of sites such as HN (or of Reddit, Twitter, Facebook, TikTok, etc., etc.). It also fundamentally violates the prime directive of HN, "[T]o make thoughtful comments. Thoughtful in both senses: civil and substantial":

https://news.ycombinator.com/newswelcome.html>

Or as dang so often notes, optimising for intellectual curiosity: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...>

Bots, so far as we're aware, are neither intellectual nor curious.

Re: Is Mars Habitable? (1907)

#25

It's really interesting how he constructs a self-consistent narrative with half true and half false premises. For example, he totally whiffs on the geology due to bad starting premises. At this time, the Earth was believed to be relatively young (on the order of millions of years) because simple calculations would show that even a very high initial temperature would have cooled to at most the surface temperature with…

Self-consistent narratives generated with half-truths and false premises describes one heck of a lot of philosophy and theology, something I note being current with Peter Adamnson's History of Philosophy podcasts, inclusive of Western and other traditions (Indian and Africana being completed, Chinese just having launched).

That's not to denigrate all philosophy (much of it is useful, and even the wrongheaded stuff often contains useful methods and constructs in testing arguments and beliefs), or even the wrongheaded bits (useful not only as bad examples but also to explain much of current belief and ideology, and to show origins of contemporary arguments and topics). But it does provide a strong caution about the dangers of leaping off into rational thought with parlous thin empirical foundation, or of false premises. Wrong belief is often far more harmful than no belief.

As to Mars's characteristics ... I'm sufficiently old to have seen an immense evolution of understanding about the Solar System and Earth itself. In the mid-1960s, the theory of plate tectonics was just being formally accepted, the far-side of the Moon mapped for the first time (the story of National Geographic's map project published in 1969 is a wonderful account of this: part 1 https://web.archive.org/web/20090129220141/https://kelsocart...> part 2 https://web.archive.org/web/20090130150935/https://kelsocart...>) The Luna, Mariner, Zond, and Surveyor missions provided first close glimpses of the Moon, Venus, and Mars from the late 1950s to early 1970s. The Viking Mars landers (1975) and Voyager outer-planetary missions (1979--1988) absolutely revolutionised understanding, and that based on what was often just a very narrow and short glimpse. But the books I'd read as a child, published from the 1950s to early 1970s are now woefully out of date.

Re: Is Mars Habitable? (1907)

#26
post #7

It's really interesting how he constructs a self-consistent narrative with half true and half false premises. For example, he totally whiffs on the geology due to bad starting premises. At this time, the Earth was believed to be relatively young (on the order of millions of years) because simple calculations would show that even a very high initial temperature would have cooled to at most the surface temperature with…

> There's a lesson here in the ability to form consistent, empirical, wrong theories given even slightly wrong inputs. Is this really a lesson? Every single theory evolved this way, starting with an almost certainly naive hypothesis, then with incomplete evidence leading to bad conclusions, then as better data came in, adjustments to the theory until we had something that fit the evidence to a high degree (though it'…

I think it is a lesson, yes. It's not a lesson to avoid trying to form theories. It's a lesson in the importance of out-of-model error and in assigning low confidence to arguments with many premises and logical steps along the way, because each premise and each step creates new ways to introduce small propagating errors.
Post reply on HN