I’ve written an in depth article about nonlinear least squares fitting from a Bayesian perspective. In there, I derive the best fit parameters, their covariance, and credible bands around the best fit model from scratch. I am using only elementary linear algebra and calculus. I also try to shine a light on the influence of priors.
None of this is rocket science or novel, but I still feel there’s value to this, since it helps to understand what the method of least squares means from a Bayesian POV.
I've only read A Fire Upon The Deep, which I found super interesting in terms of setting, but the storytelling felt a bit clunky to me at the time. I was still intrigued by the author though and I read A Deepness In The Sky and was blown away.
Storytelling, setting, everything just clicked. I know there is a third book in the trilogy but it seems a bit back to the roots in terms of setting and I haven't touched just out of the fear that I wouldn't love it as much as ADITS... Should I?
Deepness and Marooned in Realtime are his favorites of mine. Both sequels to less impressive novels imo. I like his writing in general but those two were really brilliant.
If you haven't already read "The Ungoverned" I recommend it as well - although much shorter it's effectively the second part in a trilogy, bridging the two novels and featuring the same main character as "Marooned in Realtime".
I just read it a couple of years ago, missed the memo that there was a third book. It feels like book 3/4 but he has not finished the story and I worry if he ever will.
> It feels like book 3/4 but he has not finished the story
Worse: Children felt like the author stalling for time on the greater plot. Chekov's gun isn't just left unfired; the author takes it down, polishes it a bit, and puts it back on the mantle without firing a shot.
Assuming that it's still canon, The Blabber tells us how it ends, but there's some very interesting stuff in between we're still waiting for.
Probably just me, but it first book was such a schizophrenic experience. Loved every single bit of space part but couldn't care less about medieval cluster dog minds part, eventually grew bored of it and it felt disturbing. Especially when author thinks some minor medieval squabbles are equally important as the galactic space part, giving it same page count.
Suffered till the end, and lost any motivation to continue since following books proudly advertised that the form is there to stay.
There are vastly more diverse stories that still link well together, ie Hyperion cantos was an amazing experience for me. Literally 7 different stories coming together, from cyberpunk to high religious politics, but always connected to same technological universe.
> Especially when author thinks some minor medieval squabbles are equally important as the galactic space part, giving it same page count.
That was a major plot design point to me.
That pre-spacefaring civilizations have things that are incredibly important to them, but questionably important relative to order-of-magnitudes different power galactic civilizations.
One of the reasons I like Banks fiction as well -- in all likelihood, given evolution and technological maturity timescales, advanced civilizations will be wildly displaced in time as well as space.
Yes, but Banks version of the "extremely minor event in the grand scheme of things but terribly important to those participating in it" (i.e. Consider Phlebas) is much better.
Are you perchance channelling that space-Usenet poster who was utterly dismissive of the main plot until the very end? ;-)
Parent comment should be marked SPOILERS, this one also.
Srsly tho, I appreciate your perspective. I found the dog part a little slow at times but still super interesting due to their collective identity which was explored quite a lot. As such it wasn't just a medieval squabble, it was entirely in keeping with the nature-of-consciousness theme that permeated the space part of the story.
I do get it, no idea why everybody now starts explaining the story to me like I misunderstood it :) Its fine that person A likes object X, but person B doesn't, there is no rationalization needed since it won't change the opinion.
I went for that sci-fi book for the space and sci-fi parts, and found the other side not interesting. That's it. I think I generally like sci-fi happening in space or generally futuristic. For medieval I prefer reading fantasy (which I think would help story tremendously if included for folks like me, since there is already another 'magic' in that universe).
I think you misunderstand my reply. I'm not trying to change your mind, I'm trying to express disagreement without reaching for the downvote button, is all :-)
Deepness has the same "split brain". Half the story is set in the future, but the other chapters are told as a picaresque fantasy in what appears to be a kind of fictional alternative reality circa World War I, that is eventually revealed to be a "humanized" version of the alien civilization's history (similar to how the aliens are described in The Three-Body Problem).
I didn't love A Fire Upon the Deep because, while the sci-fi parts were amazing, the medieval fantasy telepathic dog society was much less interesting, basically a political fable with some children thrown in; it seemed like a different novel altogether, and in fact the plots don't really ever intersect except when the adults finally land. It felt like two very different books cobbled together from two different authors. The dog half reminded me of Terry Pratchett for some reason, another author whose schtick involves making the story go faster and increasingly more breathless towards the end, like some Hollywood action movie, often involving a bunch of running or flying.
I liked Deepness even less, because once again you have fantasy politics, and while I understood the conceit, the storytelling just didn't work for me. Like before, the science-fiction stuff was fantastic and full of great ideas, but the other half was the exact opposite.
Just a note that the dogs/Tines aren’t telepathic but communicated via high frequency sound. The idea was exploring an entity that existed as a society of minds and how that would look from a practical perspective. There’s a story in the compilation discussed here that feature the Tines in another setting one may find interesting.
I had to read the book a couple of times before I completely understood this, but I'm pretty sure there's an implicit metaphor that ties the two plots together, and it's about the feedback between technological progress and intelligence, and how physical constraints (or accidents of evolution) limit progress. The society of the Tines is constrained by their peculiar path to sentience - they can't split up pack members and they can't co-mingle packs without losing coherence. From their perspective, early 20th century electronics are the equivalent of neural prosthetics, and might as well be gifts from the Transcend.
Ah, it's been a while. It doesn't change anything, though. They might as well have been telepathic, since it appeared they formed a sort of hive mind when close to each other.
I've seen others who felt similarly about the enjoyability of the space part vs the dog part. But for some reason I felt the opposite, the space part was alright but I didn't really connect with any of the characters or plot there much. But I loved the part on the dog planet, for me that world felt like he started with a question of "What might it look like if a pack animal achieved human-like level of intelligence?" and explored from there, and what he came up with made a lot of sense to me and was really interesting to think about.
I'm completely with you. I found the space parts a chore to read through to get back to the most interesting stuff. The whole "zones of thought" concept just seems an obvious device to allow Vinge's ideas to work.
By contrast, the thrill I got when I first realized that the dog hive mind was evolution's stab at an acoustic modem was fantastic, and I read through all of this aspect of the book with great wonder.
I think the dog part really suffered from a quite common syndrome, that the sci-fi author gets some really cool idea/concept, and has to explore it in some context / story, but doesn't really deliver on it. For me the idea/concept was great, the story just boring filler needed to explore the idea.
I disagree, they have some stuff to think through, even if they're not as shiny as the first one. Perhaps he did himself a disservice by showing all he's got at the very begining of the series, though.
That stuff has the Star Wars vibe where Ep 1-3 have much more grandiosity than the original series - luckily the weren't chronologically first.
The problem isn't the highlights of it - it's the low-lights, which make up most of those books. The main POV character is incredibly annoying to follow around - he's a dense, not-too-intelligent, but incredibly-long-winded everyman, who is never told to shut up by anyone around him.
I don't need an everyman perspective[1] on an epic science-fiction story! I am an everyman! I can form my own opinion of it!
They are also rather heavy on author-dictated philosophy (which is not particularly interesting, or insightful, and manages to fall flat in a pretty major self-contradictory way) - whereas the first two Hyperion books have much less of an author's voice in them, and are much more of a presentation of an interaction of different, competing value systems. That is interesting! Listening to the chosen one go on and on about choosing again (when none of their disciples actually do) is not!
The overall plot of them is fine. It's, unfortunately, all in the execution.
[1] Or at least, not one that's hammered into me for ~60% of the book.
Strongly agree. There are some enjoyable bits in 3&r4 ( I just enjoy episodic travel adventures), but for the most part they are pretty bad, especially compared to the masterpiece of the first one (the second was an adequate ending to the story).
I disagree. Yes #1 was best. I thought #2 was the weakest link, but with the redeeming feature of getting you to #3 and #4. Are they perfect, no, but they do have good moments aplenty and I thought the ending worth the journey.
Edit: thinking back, the ending of #2 was bloody spectacular as well. I think the sheer volume of character/world building must help as preparation for impactful endings.
None of this is rocket science or novel, but I still feel there’s value to this, since it helps to understand what the method of least squares means from a Bayesian POV.