Become a fan of Slashdot on Facebook


Forgot your password?
Sci-Fi Books

Ask Slashdot: How Could You Statistically Identify The Best Sci-Fi Books? 180

jimharris writes: Over at SF Signal I wrote a piece "How Well-Read Are You in Science Fiction?" There are three databases that collect lists of popular science fiction books that try to statistically identify the best books of the genre, [offering] combined list that shows which books were cited the most. They use different sets of best-of lists, but their results are often similar. The final lists are, Classics of Science Fiction, Worlds Without End Top Listed, and Premiosylista Comparativas: Comparativas: Ciencia ficcion (Spain).
Interestingly, each list has a different book in its #1 position (though both "Dune" and "Frankenstein" make the top four on at least two of the three lists). But is this really a good methodology for determining the classic canon? What would be the best way to statistically identify the greatest sci-fi books? (And have you read any good science fiction novels lately?)
This discussion has been archived. No new comments can be posted.

Ask Slashdot: How Could You Statistically Identify The Best Sci-Fi Books?

Comments Filter:
  • by Anonymous Coward on Sunday May 01, 2016 @11:39AM (#52022743)

    You need to define best, greatest, and classic before you can go further in your quest.

    • by Anonymous Coward on Sunday May 01, 2016 @12:00PM (#52022825)

      And not in the least decide what the definition of the term science fiction is (it's not as easy as it sounds).

  • You can't (Score:5, Insightful)

    by 110010001000 ( 697113 ) on Sunday May 01, 2016 @11:44AM (#52022753) Homepage Journal
    Art/literature/music is subjective. You can't rank them, except personally. Next question.
    • Do "works ignorance of whose details is most likely to produce a 'turn in your geek card' reaction []" have a rankable metric?

      • Your comment makes no sense. I started writing this post without visiting the link but stopped myself and your comment still was nonsense, like a bunch of English words randomly strung together.
        • by tepples ( 727027 )

          I'll rephrase with less embedding.

          Alice makes an inside joke alluding to a particular work of fiction. The joke goes over Bob's head because Bob has never read it. Alice notices this and tells Bob to turn in his geek card. With which works should one be familiar in order to get jokes that geeks are expected to get?

          • Depends entirely on the geeks.

            Dr. Who, Star Drek, Star Wars, PKD, Heinlein, Clark, Ellison, Asimov, LOTR, Comic Books, Forbidden Planet, Godzilla, all have constituencies that would call them 'essential'.

            Just reading all the PKD and Asimov would be a huge challenge. 'All the comic books' is a silly concept.

            Go meta one. If you don't get the South Park SciFi refs, 'turn in your geek card'. If funnybot starting to exterminate was a surprise, you are no geek.

      • My god, that totally trolled half of Slashdot with that one.

    • Exactly. If you want to know how well statistics fares at this, turn on any top-40 radio station.

    • by mwvdlee ( 775178 )


      You can most certainly rank literature by the number of pages, art by the size of the canvas or weight of the sculpture and music by the duration.
      Whether these rankings are useful is debatable, but they can definitely be ranked.

      • Bullshit.

        You can most certainly rank literature by the number of pages, art by the size of the canvas or weight of the sculpture and music by the duration. Whether these rankings are useful is debatable, but they can definitely be ranked.

        Exactly. Their criteria, according to TFS, is "combined list that shows which books were cited the most." I don't know if that's relevant, but going with it I ask, "cited where and by whom?" If they had polled /. the #1 slot might go to HHGTTG [] - sorry, no citation available.

    • by Bruce66423 ( 1678196 ) on Sunday May 01, 2016 @01:32PM (#52023329)

      If a lot of people are wowed by product A and bored silly by product B, it is irrational to argue that the two can't be ranked as to which is the one most worth investing time in to read, given that we have a finite amount of time to spend doing so. Therefore to get people to vote for their 'favourite' seems a rational way forward, despite its subjective foundation.

      Otherwise I know a wall with some paint drying that you can watch this evening...

      • by Dutch Gun ( 899105 ) on Sunday May 01, 2016 @02:38PM (#52023607)

        I think the trick in any sort of subjective ranking is to find a person or group of persons that tends you match your own personal tastes. Subjective ranking certainly isn't worthless when I recommend a book, movie, or TV show to my parents. That's because I can more or less accurately judge how well they'll like something since I know their tastes.

        I think this is where most algorithmic approaches tend to fail (like on Amazon), at least from what I've been able to see. I think they tend to find general correlation - that is, "those who like book x would also like book y", but I think a much more effective approach would instead be to search for other customers who's general ranking patterns tend to match your own most closely, and build a personalized recommendation group from which to mine predictive data. In this way, the predictions would be more or less tailored for each individual customers based on similar likes and dislikes, rather than being based on general popularity trends.

        For better accuracy, the algorithm should pay even more attention to statistical outliers. For instance, I generally love space-opera-y science fiction (Honorverse, Lost Fleet), as well as "harder" works, like The Martian. But I didn't care for the Heinlein I've read, so would wish to avoid more of him, and more important, more books in that general style. So, a clever algorithm would notice that trend of mine and find sci-fi fans who *also* didn't care for Heinlein, and give their recommendations a slightly higher bias based on that data point.

        Generally speaking, any sort of non-personalized "ranking" is going to simply be a popularity contest within the target audience you select. There's nothing wrong with that, so long as people understand that's all it can ever really be. Which, of course, they won't.

        • You're right of course; personal preferences make the process problematic. OTOH my experience of using Hugo and Nebula award lists to choose what to read has almost always worked out well, so I suspect the best of the best has a wider appeal than just to its obvious target audience.

          As far as Heinlein is concerned, 'Stranger in a Strange Land' and 'The Moon is a Harsh Mistress' are worth the effort. You either love or hate 'Time Enough for Love' - either it scratches where you itch or it doesn't...

          • As far as Heinlein is concerned, 'Stranger in a Strange Land' and 'The Moon is a Harsh Mistress' are worth the effort. You either love or hate 'Time Enough for Love' - either it scratches where you itch or it doesn't...

            Unfortunately, my introduction to Heinlein was "Number of the Beast", which was so unbelievably awful, it made me seriously regret that I couldn't get those hours of my life back and purge that literary vomit from my brain. I've since heard that was about the worst introduction to his work possible, but it's hard to get past that bad of an experience. "Starship Troopers" was not terrible, but I can do a whole lot better than 'meh'.

            That being said, I've heard so many people recommend "Moon" that I may have

            • Instead of asking what are the best works in all of sci fi, better to ask what are the best works of sci fi authors.

              Heinlein: Starship Troopers, Moon is a Harsh Mistress (although for younger reader I would recommend The Rolling Stones or Orphans of the Sky)
              Brin: Startide Rising, The Postman, Uplift Wars
              Asimov: Foundation trilogy, I Robot
              Cherryh: Cuckoo's Egg, Cyteen
              Banks: Excession, Player of Games, Feersum Enjin
              Niven/Pournelle: The Mote in God's Eye, Legacy of Heorot

              It's still subjective, but you'll get a

              • Also, the same list had 'The Color of Magic' as the only Pratchett book. I definitely wouldn't recommend that as the first book to introduce someone to the Disk World series.

                You shouldn't forget the metric the metric they used was "Number of times included in a "Best of" list on our site". So they're not saying it's his best work, they're saying it's the one most often included in lists, and that's probably only because it's the first Discworld novel.

                It's pretty cool that you can actually click on the entry for the book and see all of the lists that the book was included in.

              • I'm surprised you didn't include any Ringworld books in the Niven list. Also, Stephenson might be good if you like that style, (Snow Crash, SeveneveS, Cryptonomicon), Gibson (everything...) would be another good author. Though it looks like you mostly like hard sci fi, and these two authors are more in the cyberpunk style. I would also add the other robot books to Asimov, though again, this is a taste thing.


      • comparing a good vs a bad/boring book is vastly different to ranking books in a genre, especially at the top of it. while the former is easy the later is impossible as it is subjective as the differences are personal. I hated Frankenstein but loved Dune, doesn't mean Dune is better (except for me), both a arguably exceptional pieces of literature in the genre.
      • by tlhIngan ( 30335 )

        If a lot of people are wowed by product A and bored silly by product B, it is irrational to argue that the two can't be ranked as to which is the one most worth investing time in to read, given that we have a finite amount of time to spend doing so. Therefore to get people to vote for their 'favourite' seems a rational way forward, despite its subjective foundation.

        That depends on the book - some authors are universally bad (L Ron Hubbard) and some are universally good. But those are the exceptions - a book

      • by hoggoth ( 414195 )

        I think it is useful to have a list ranked by recommendations, even if the list is biased, arbitrary and subjective. If that list leads me to read 'Exhalation' instead of 'Animorphs'; 'Shoggoths in Bloom' instead of 'Battlefield Earth' then it has served a useful purpose. As the parent says we have a finite amount of time.

    • Even with the greats, there is a lot of subjective views on it. Sometimes such flicks are popular due to lowbrow reasons. Frankenstein popularity while may be a classic story, its success carries on from the movie adaptation. Also some genius Science Fiction has passed away threw the years, because modern sensibilities had made such stories obsolete. While other unpopular stories of the time, my get renewed attention as such stories may be ahead of its time.

      There are a lot of factors that can become a cla

      • by Sique ( 173459 )
        Frankenstein was famous for 100 years without any movie adaption. I would say, it's exactly the other way around. It was adapted early and often for movies because it was famous already. And it is definitely quoted everywhere. Frankenstein's monster is an iconic figure by itself, and Frankenstein's laboratory is how the workspace of a crazy genius is portrayed since then.

        On the other hand, many of those lists are very U.S. centric. If it wasn't for Jules Verne, you would think that there wasn't any scienc

        • If it wasn't for Jules Verne, you would think that there wasn't any science fiction outside the anglophonic world, and if you substract H.G.Wells, science fiction seems to be an U.S. only phenomenon.

          Hmm, Peter F. Hamilton [], Robert J. Sawyer [], and Alastair Reynolds [] are a few non-Americans I can name off the top of my head. But I can see your point about non-English works, but I'm not sure it's terribly valid, a list of the best literary works is going to focus on works in the language the list was written in, because very few people are actually multi-lingual (as in can easily read books in more than 2 different languages).

          Also, Arthur C. Clarke [] was British and you pretty much can't have a list of "Best

          • by Sique ( 173459 )
            I don't doubt that there are very good american SF writers. I just never understood the case for Dune for instance. Ursupator dethrones king. King's son disappears, meets new people, learns a new art/gets an important magical item/tames a mighty beast, returns to topple ursupator and becomes the rightfull king. This is a plot that has been told so often, and with so many costumes, that just another costume variant does not add very much. The story of Dune could have been told without ever mentioning anythin
    • by bidule ( 173941 )

      Subjectivity is an error bar. You can still get a partial ordering and know that Dune is better than Lion's Pride.

  • Maybe something like a points race? Assign a point value for 1st thru 10th place, ask the Teeming Millions (sorry Cecil!) to make a list of their top 10 or top 20, perhaps out of a pre-defined list of 100 titles (made from the 3 lists/databases you mention), then calculate point totals for each title. This way in theory a title that consistently gets 2nd or 3rd place on a majority of lists is "better" than a title that has some 1st place mentions but doesn't even make it onto some peoples lists. See the

    • by tepples ( 727027 )

      You appear to suggest something similar to a Borda count [] to turn a ranking into a range vote. Doesn't a Condorcet count [] produce a less gameable result than a Borda count?

      • Yes, the Condorcet method is less gameable than Borda count. I don't think gaming the vote is going to be too much of an issue in ratings of favourite works, however, where most voters probably don't have too much riding on the outcome. I think there may be a different point in Condorcet's favour in this case, but there's also a practical difficulty.

        Borda count is going to be biased towards more well-known works, since these will be rated more often, so receive more points. The Condorcet method can be ca

    • Snoutcounting?

  • Many science fiction lists look at the distant past. I think some of the best science fiction in recent years have been written by women, as they explore issues that most men writers ignore. The Wess'har Wars series by Karen Traviss, The Entire and The Rose series by Kay Kenyon, and Jenny Casey trilogy by Elizabeth Bear are my favorites.
    • by Z80a ( 971949 )

      The thing is much more complex than just "men and women" or even "person of ethnicity X or Y", at a point that if you try to boil down everything that might make a group better than others by such labels, you would take years, get insane and still get two or three people crumbling your system apart like a castle of cards near some pesky hyperactive kids.

      So, the best you can do is just judge people by their individual merits instead of worrying about group X or Y.

  • Google's Page Rank algorithm probably has as good a shot as anything else.
  • I can't even tell if my friend will like a book that I read for a certainty. Why would you think you could do so using statistics?

    • Tell me who you like and I can find people who also like the same list. Books that appear on lots of their lists are likely to be good, in your opinion. They are also likely to be books you have already read, but didn't list.

  • Best by what metric, criteria (or criterion)? I did a quick skim of the "classic" list and the newest was 1992 "Red Mars" (and another book). Why nothing newer? I also note that a lot if not most of these books are written as social commentary. Perhaps the list reflects back on the list maker?

  • Looking at the "Classics of Science Fiction", I have to question how many of these books have been actually read.

    Going through the first 100 list, I have read and attempted to read 90 of them (which is why I would put the challenge on the Subject line).

    "Frankenstein", while undoubtedly having a huge impact on modern society is basically unreadable by modern readers - the copy I have has a forward by Stephen King saying it took him four attempts over decades to get through the book. The same comments for "The Lost World", "Gulliver's Travels" and "Brave New World".

    I have never seen a copy of "R.U.R." (either as the original play, translated into English, or as a novelization), my copy of "Cities in Flight" is from 1966 and there are a number of books that are listed pre-1950 that I have heard about but never seen copies of. Now, as a counter-point, I just looked for many books I haven't seen on and did find them (the two cited in this paragraph are easily found and ordered) but I haven't seen them in new & used book stores even though I have looked for them.

    So, to get this list, how many were actually read and how many were checked off because the reviewer/person being surveyed has actually read them? So, going back to the original question how can you statistically identify the best science fiction books if not all of the "classics" haven been read?

    • I read Frankenstein a while ago, and yeah, it's pretty bad. It's still a "classic"; its influence on SF is unmistakable. So is R.U.R, even if few today (not me) have read it. Gulliver's Travels, on the other hand, is quite readable (though like Dante's Inferno, you miss all the contemporary references), as is Brave New World. Doyle's "The Lost World" is readable but I'm not sure it's all that influential; it seems to me it was pretty much forgotten until about 10 years ago.
    • by Ecuador ( 740021 )

      "Frankenstein", while undoubtedly having a huge impact on modern society is basically unreadable by modern readers - the copy I have has a forward by Stephen King saying it took him four attempts over decades to get through the book.

      That's strange, my wife recently finished Frankenstein and she told me the exact opposite, while urging me to read it. While she is a humanities PhD she actually puts down books that many people would say are "unreadable" like Ulysses (the James Joyce one), or Paradise Lost and actually English is not even her native language, so it is not like she reads unusually hard/unreadable etc books. Anyway, she told me it was an amazing book, that the narration method is so modern you'd think it was written recently

      • by Quirkz ( 1206400 )

        As a counter-anecdote, I really enjoyed Ulysses (and actually even read Finnegans Wake -- not that I claim to have understood it) but never got through Frankenstein. It's been too long to say for sure, but I've been pinning it on a distaste for novels written in "someone's old journal" format.

    • by werepants ( 1912634 ) on Sunday May 01, 2016 @02:42PM (#52023635)

      "Frankenstein", while undoubtedly having a huge impact on modern society is basically unreadable by modern readers

      What? Seriously? Frankenstein, along with many similarly classic works like 20,000 Leagues Under the Sea, War of the Worlds, are often placed in the children's section of book stores and libraries. I know that I read many classics growing up for just that reason - I saw some cool covers with aliens and monsters and submarines and got my start in sci-fi that way. If I was able to read that stuff as a kid, it's frankly absurd to call it "unreadable" for current adults.

      • 'Unreadable' doesn't necessarily mean 'written at a level that is too difficult to read'.

        It can mean 'This is such utter crap that I just can't read it.'

        I can think of a book or two I stopped reading because the main characters were so unlikeable, but I don't recall not finishing one because of sheer badness.

        • Perhaps - but even in that sense of "unreadable", would Frankenstein qualify? I'm no great enthusiast for literature - a good chunk of british classics are "unreadable" in that sense - I would put Jane Eyre in that category, for instance. There's no real plot or tension, nothing interesting happens, and it's basically just a wonderfully written pile of nothing at all.

          You can't say that about the early sci-fi classics, though - there are interesting ideas and motivated, proactive characters in all of them.

    • I think if something has been well written, it will hold up pretty well. I've not read Frankenstein, though I probably should. However I've read many HG Wells old science fiction, and while different was still quite good. You have to keep in mind the context of when the stuff was written however to really appreciate it. Same with HP Lovecraft.

      The only ones I had real difficulty were truly terrible writing. Two of those come to mind, and both were modern books. The end of the wheel of time series and the oth

  • Such gigantomania revenge-phanatisies as "Dune" - way overrated by a certain demographic of SF Fans - would totally scew the results.

  • Ask Slashdot: How Could You Statistically Identify The Best Flavor?

    Sure, you'll get a result if you're successful... That doesn't mean you're going to like it.

  • Looking at the Classics of Science Fiction list (is the clue in the name?) there is nothing in the top 100 that's less than 30 years old.

    The Worlds without End appears to have been Slashdotted - update: contains about a dozen from the last 3 decades
    and the Premios y Listas list names 22 titles from the 1950's and only 23 from the past 30 years

    Has the genre run out of ideas, do people prefer the simple, "space opera" style of times past or is modern science just too abstract and mind-boggling to base ent

  • If I were to define the best SciFi book, it would be the Analog magazine, at least when John Campbell was in charge. And since a lot of SciFi fans will doubtless be reading this, I'm going to take the opportunity to point out that Project Gutenberg (a project which has been active since the early days of the internet--in fact, the ArpaNet) has digitized a bunch of articles from Analog magazine, some dating back to the 1930s (but especially important for people like me, many from the 1960s). You can find t

    • Campbell fostered a certain subgenre, so if you follow Astounding/Analog only you're going to have a very limited view of the field. I wouldn't call what he encouraged above average as opposed to limited. If you like Campbell-type stuff, great, but there's lots of SF I'd consider just as good outside it.

  • Ask Slashdot: How Could You Statistically Identify The Best Sci-Fi Books?

    Read a lot of them and the one you like the best is the one you like the best.

    Anything else is just time wasted that you could be spending on reading more sci-fi books. Or anything else you enjoy.

  • My "best" is not the same as your "best".

    • by abies ( 607076 )

      My "best" is not the same as your "best".

      But it seems that your 'obvious' is same as everybody's else 'obvious'.

      • by JustNiz ( 692889 )

        Sigh yet another self-righteous smug asshole on Slashdot. its clearly not very fucking obvious to the OTA is it?

  • The Worlds Without End list is pretty terrible at the beginning (The Handmaiden's Tale, really? That theme was well-trodden in SF when Atwood wrote it. Hyperion and the Doomsday Book are odd choices as well). The Classics list is a better list, though it's quite influenced by the age of the work (not surprising for a 'Classics' list).

    A better list, IMO, would break the genre up into periods and subgenres and assign key works from each; for instance there's no need for both "Left Hand of Darkness" and "Th

    • by dgatwood ( 11270 )

      I think that the entire notion of creating a list using statistics about the number of people who read them, by ranking, etc. is an exercise in pointlessness. In the first case, you're duplicating the Amazon sales rank. In the second case, you're duplicating the Amazon review rank.

      What would be far more interesting, IMO, would be to use machine learning to actually analyze the text of books within a genre, and then recommend other books based on which books a given individual likes and dislikes. And by

  • by BarryHaworth ( 536145 ) on Sunday May 01, 2016 @10:15PM (#52025709) Homepage
    In answering this question for books it might be instructive to look at what happens in another artistic field, that of the movies. Although there are some major differences (Movies cost a lot more to make and therefore there aren't so many made each year for a start) the comparison might shed a little light.

    With rating movies statistically there are a number of methods:

    - Box office takings, such as Box Office Mojo []

    - DVD and Video sales

    - Movie audience figures (when broadcast on television or similar)

    - Industry awards, such as the Academy Awards [] or the Baftas []

    - Ratings from critics, such as Rotten Tomatoes [] or Metacritic [] - Ratings from general users, such as IMDB []

    - and finally, "Best of" listings voted on by critics or interest groups.

    I include the last not because it is really a very good statistical comparison as compared to any of the other methods, but because it is the only one analogous to the sorts of lists being considered in the Worlds Without End rankings.

    To get a good statistical ranking for books or movies we need to get a comprehensive set of data that covers all (or most) of the entries, and which applies the same rankings to each. None of the rankings for Movies which I have listed really does that, but some do better than others in some ways at least. For example, ticket and unit sales cover all movies, though they have the problem that the number of people going to movies, and the price they pay per ticket, have increased over time so that the ranking metric isn't the same for all movies. It also has the disadvantage that ticket sales are not necessarily related to how good a movie is. Industry awards can probably be assumed to cover all movies released in a given year and therefore cover the whole population, but have the problem that the award givers may not cover all entries equally, and may be subject to bias. Critical judgement, whether from professional critics or members of the public, also have the problem of coverage - I personally cannot expect to be able to see every movie made, and the ones I do see will be affected by by things like advertising budgets which are not necessarily related to how good the movie actually is.

    With books we do have some similar data sets. Figures for number of books printed, or sales on the likes of Amazon can be compiled, though these have the same problem of not being related to quality. I don't know of any compilation sites for professional book critics (anybody?), but there are sites such as Goodreads [] where members of the public can give their subjective rankings. Industry awards also exist, such as the Hugo [] or Nebula [] awards, but these have the disadvantage of being subject to politics (*cough* Puppies vs SJW anyone?). Finally, there are "Best of" lists, such as the ones cited by Worlds Without End.

    Books have a problem compared to movies in that far more books get published than movies get made. While a good critic can expect to see all the movies that come out in a year (at least all those released theatrically), reading every book that is published is impossible. This eats into the quality of critical rankings out there, or even into Industry awards. Any "Best of the Year" list can't really hope to be definitive, because a book - especially a ground breaking, iconoclastic new classic - will take time to find a wide audience and be widely recognised.

    For my money, I think the likes of Goodreads are probably the best bet as an objective, comprehensive and timely statistical source for

  • 1) "Blindsight" by Peter Watts is the best Science Fiction book.

    2) Popularity is not a measure of quality.

    3) The actual best way is to build a true machine intelligence that reads every book after succeeding humanity.

    • 1) "Blindsight" by Peter Watts is the best Science Fiction book.

      Haha! I don't know if I would go that far, but it certainly ranks up there on my list as well. That book changed the way I thought about reality, intelligence, and a hell of a lot of other things.

  • First, based on some of the comments I see, a large percentage of slashdotters these days have no clue what sf&f actually is: 99.999% of it is *WRITTEN*, and has *NO* *RELATIONSHIP* to any movie, tv show, or video. None. In my personal library - I'm an average active fan - I have between 3,000 and 4,000 books... and *maybe* 20 or 30 0f them related to anything ever filmed.

    Second, the whole concept of "what's your favorite book" is, IMO, asked by people who don't read. My "favorite"? In what subgenre? Cy

"It's my cookie file and if I come up with something that's lame and I like it, it goes in." -- karl (Karl Lehenbauer)