When I took my AI class, my professor would assign one paper a week and we had to turn in a "critique" of that paper. These papers were, of course, the seminal classics in the field, because what else would you want your intro class to be reading? I complained a bit to the professor that it was a bit silly to critique the seminal papers in the field and if you want to be sure we read it, we could just turn in a summary (which our critiques typically started with anyhow), but we were supposed to learn how to critique papers.
In hindsight, I suppose I should have appealed to the fact that critiquing the seminal papers in the field is a serious data set bias and that trying to "learn to critique" on the best papers ever written in a field was less likely to produce a useful "critiquing" skill and more likely to produce some overfitted garbage skill, but, hey, I hadn't taken AI yet! I didn't know how to express that.
(It did produce a garbage skill, too. I tried writing "real" critiques using my brain, but after getting Cs and Ds for the first couple, I learned my lesson, and mechanically spit out "Needs more data", "should have studied more", and as appropriate, "sample sizes were too small". Except for that last one, regardless of the study. Bam. A series of easy As. Sigh. I liked college over all, but there were some places I could certainly quibble.)
Anyhow, this is the paper that needs to be assigned towards the end of the semester, and students asked to "critique" it. It's a much better member of the training data set for this sort of skill.
Yeah, any "critique" or "argument" assignment seems to have weird grading biases.
In my freshman Composition class we needed to pick a controversial topic and argue one side. Highest grade on that assignment went to someone who argued that smoking causes lung cancer (this was 2003). The instructor explained that it was the most convincing paper. Those who picked an actually controversial topics got the lowest grades because their arguments were less of a slam-dunk.
I had to do something similar in a Physics class of mine. We had to write a paper critiquing and arguing against a well known physics theory. I was really into Special Relativity at the time, so picked that one.
My professor was a Nobel laureate, so already pretty intimidating. After we each picked a theory and wrote an outline he asked that we review with him. Well, he said, “I don’t know why you’d pick that. That’s going to be too hard.” So I asked if I could change, but he said that it’s too late for that and wished me luck… I was more than happy with the C+ I got.
Had a very similar experience in philosophy classes. The good thing about critiquing seminal works is that there's often a rich body of criticism already available. I found the only way to do well in those classes, since I lacked the mental horsepower to produce my own seminal critiques, was to parrot the literature that was out there already.
I think in philosophy it's a bit more valid. Still not maybe the best case, since these are still the papers that stood the test of time, critiquing some marginal paper would probably be a better practice of "critiquing", but philosophy always has a critique. The form may be impeccable, the writing may be spectacular, etc., but the philosophy itself always has room for rumination, discussion, etc.
For a science paper, if it's something people are still reading 30-50 years later, it was apparently good enough. I can always critique the paper for failing to solve String Theory and then draw out from String Theory a mathematical demonstration of how their solution for getting robots to navigate around boxes is very good, but that's more a reflection of me than the paper.
Love it! The sordid truths of data science laid bare! "We prefer subtitles over machine-learning-based detection because the presence of a cough in subtitles means it was prominent enough for a person to write it down. Also our postdoctoral fellow was the only one who completed the Tensorflow tutorial."
When their initial theory of cough count produced a trend line that didn't really work well, with some major outliers, they introduced the implausible, ad-hoc mechanisms like the Thriller Tripler effect and the Batman effect to make it fit. This is the part that isn't just parodical silliness, but has actual applicability. This is something you might see in a bad research paper.
> Editor’s Note: The claims in this paper have not been verified because the researchers have refused to yield their full dataset and methodology, citing “intellectual property rights,” “the sanctity of the First Amendment,” and “the Wright brothers never had to show their work.” We are publishing their paper here because they won’t stop mailing hard copies in triplicate to our homes and offices.
> We prefer subtitles over machine-learning-based detection because the presence of a cough in subtitles means it was prominent enough for a person to write it down. Also our postdoctoral fellow was the only one who completed the Tensorflow tutorial.
In my experience Closed Captions almost always describe sounds, specially if they happen offscreen or things like the type of background music; Subtitles don't, I assume they expect you can hear the sounds.
Subtitles frequently describe entire background conversations I can't hear when I'm watching at home. Maybe I should turn up the sound to a ridiculous level.
This was taken to a ridiculous extreme in the marvellous The Spanish Prisoner, where the whole point at the end is that you can't hear a certain conversation—but it's still subtitled!
Also, Closed Captions always describe sounds and lines of background dialog (like TV) that I definitely can’t hear in the sound.
As a non-native English speaker who can hear I often see movies and shows with English audio and English closed captions, the redundancy helps me. It happens all the time that the CC tell things that you just can’t hear in the movie.
> Subtitles don't, I assume they expect you can hear the sounds.
Many—though not as many as it should be—(English-language) movies explicitly include two English-language subtitles, "English" and "English SDH", standing for "English (subtitled for the deaf and hard-of-hearing)". The latter includes subtitling for sounds.
Which makes sense, "Closed Captioning" (CC) is different than "Subtitles", in that the purpose of CC is to provide on-screen text (captions) for deaf/near deaf viewers, while subtitles is captions for just dialogue.
During the apocalypse, I leaned on video for "company" while doing chores. I stumbled onto the Descriptive Video Service, which I guess is super duper Closed Captions.
I love it. I can half watch the show while fussing without missing anything. Even when the show has my complete attention, the added detail is kinda great. Not all the time, but enough to keep things interesting.
I accidentally turned this on at the beginning of Toy Story 4. The narration had me convinced that the movie was taking place in a future where Andy was in film school and we were watching him working through his first film based on his childhood toys.
Felt like a fool when I realized it was really super caption for the visually impaired.
There was a lot of skepticism that their planes were real, even years after they had first flown publicly. Information spread more slowly in those days.
They just took their time, and it became a joke because other people were already flying around when they came in public saying they were flying first.
Before anyone takes this too seriously -- and it is surely too late for that -- I should point out that they did nothing to disambiguate the length of the movie. It may not be the coughgeist but the ... er ... timegeist (they should have a german word for it).
Anyway, the point is, coughs per hour would help us know if audiences simply favor longer or shorter movies at different points in time.
That still requires some calculating before comparing movies, since it's much harder to compare a short film to a bladder buster. Using an average runtime as standard length F, then calculating coughs (C) per F would yield a value that can be more easily compared between films, regardless of length.
Would love to see more satire like this for different research areas. It's just subtle enough that the realization builds slowly, like reading one of those internet stories that ends with pulling up in bel-aire.
If you haven't already seen it, you may enjoy SIGBOVIK, The Association for Computational Heresy [0]. An annual conference mostly focused on computer science, but any field is welcome. I immediately thought of SIGBOVIK when reading this paper.
Like most films directed by Dennis Villeneuve I found Dune to be a work of art and visual masterpiece. His recount of how they prepared the set and the scene for the Gom Jabar ritual just shows how commited he is and how much passion he puts into his work. Absolute delight to watch.
It was really beautiful but the pacing was horrible for modern attention spams. I loved the movie, but only because I read way too many of the books and played all the games and watched every Dune before it. My partner had no idea what was going on and fell asleep 1/3 of the way through... even though she is normally a movie completionist.
That dude had a certain style, cinematic intellectualism with beautiful cinematography. It's definitely not for everyone. Dune is that unique blend of world building, political intrigue, religious exploration, coming of age, and a tiny bit of action. Hard to get all of that in one movie, and the trailers made it seem more Star Wars than Blade Runner.
It's the sort of movie to watch when you're seeking quiet contemplation, not popcorn pulp...
I can't stand the Avengers/Marvel style of pacing and switch off immediately with the constant CGI and fighting scene.
I was completely gripped with the new Dune but God help her if she tries to watch Lawrence of Arabia, Tarkovsky (e.g. Stalker), or 2001. For me it works, I appreciate others might not like it but I thought the style was great.
If somebody is going in expecting a space western like Star Wars or Star Trek, then I agree with you about the pacing. I think the movie moved along beautifully.
I didn't expect to like it because I didn't really like the book. For me, this is one of the few times where I think the movie is better than the book.
I hear this criticism a fair amount. I felt the polar opposite. I felt like Dune was made for me. I really enjoy that engrossing feeling of scale(both visually and aurally) that I am constantly bombarded by.
I am glad that a film like that can even be made these days given that I think you're right about modern attention spans.
> Hard to get all of that in one movie, and the trailers made it seem more Star Wars than Blade Runner.
This is because directors usually don't have anything to do with the trailers (in fact, it's typically outsourced to companies who specialize in making them). Trailers tend to reflect what the studio execs want people to think the movie is about, which trends towards "what gets the most people to buy tickets?"
Sadly, I don't think this will ever change.
I did enjoy the movie, though. Villeneuve really did the book justice, and I cannot wait for part two.
>It was really beautiful but the pacing was horrible for modern attention spams.
It's also pretty horrible from past audiences' point of view given that they'd expect to see more than a prologue in a 155m film. Villenueve makes truly beautiful movies but it comes at a cost.
Lynch-dune is mostly good, but Villeneuve's is so much better in a beautifully indescribable way that only cinema can really do.
Villeneuve has been dreaming of making his Dune since he was a boy and it shows. ignoring the visuals, the tone is just so much more ominous and alien than Lynch's. One thing that really dates old movies versus their modern counterparts (of sorts) is the sound design. Dune's sounds are absolutely fantastic.
Dennis Villeneuve has been watching David Lynch's Dune since he was a boy, and it shows. Villeneuve lifted dozens of shots directly from Lynch's version, and relied on the same key scenes as the 1984 Dune. I started out impressed by all the homages to Lynch's Dune, but beyond a certain threshold "homage" becomes "ripoff".
In Lynch's the Harkonen's were brutal. In Villeneuve's Dune Gurney had to explain to us that they were brutal, because no actual brutality made it to screen. Baron Harkonen was lower energy than Jeb Bush.
Villeneuve basically just took Lynch's Dune and drained the colour out of both the imagery and the performances.
The biggest criticism of Lynch's Dune us that it doesn't make any sense, and I still totally lost the plot 2/3 of the way through the new Dune as well.
Old Dune is kind of trippy view that happened once and I can't imagine it happening ever again, not with that budget and quality actors. Its still most approachable of all of Lynch's work. I like it a lot for sort of nostalgic feeling for the 80s vision of the future like from some pulp comics.
New Dune is completely different beast, can't wait till second part comes. That universe is rich for other stories, tv series etc. which seems to be the direction all major studios are moving to.
There are very different kinds of trippyness though. The Lynch version seems quite determined about avoiding that cheesy "bunch of swirling colors" brand of trippyness. Sequences like that exist, but they aren't the trippy bits. The trippy qualities are more between the lines. The TV serial... not so much, that one was more "oh, he's having a vision, switch on the mood light vfx!". Haven't seen the Villeneuve yet, which one would it be?
I think Villeneuve might go more in the trippy direction with the next one.
If you compare (say) 2049 and dune you can tell that the latter has been made to hold the audiences hand a lot (blade runner didn't make it's money back). Now he's sold the concept I imagine he'll have more confidence/freedom with part 2.
I think there's a better version of Dune somewhere between Lynch and Villeneuve.
Lynch didn't trust the audience enough. The wierding modules, the amount of voice-over, the limitations of the special effects of the time, etc. Also, the studio didn't fully trust the material, forcing Lynch to cram everything in one movie. The movie starts off well-ish enough. But after Paul and Jessica escape the Harkonnens, we just sort of yadda yadda years away and jump to Paul getting ready to attack Arrakeen. I think only the last 30 or so minutes is after the escape (excluding credits). Could be longer, but the jump is jarring. The movie moves at a decent clip otherwise. Set up with the emperor whinging about the Atreides, the Atreides getting ready to leave, the arrival, establishing the Atreides are not the Harkonnens, the coup, the escape, the time jump montage, the final attack.
Villeneuve's movie covers up to the escape and that's it. Villeneuve's film is 19 minutes longer. So Villeneuve basically had the opportunity to expand the beginning of the story by about an hour. He's going to be able to cover the same ground as the last 30 minutes of the Lynch movie with at least 2 hours.
The issue is that Villeneuve kind of wastes his time. We see Dr. Yueh. And that's really it. We never get even a mention of imperial conditioning or why it's shocking that it's Dr. Yueh that betrays the Atreides. Piter is seen, but I don't think ever named. We don't meet Feyd at all in the movie. We get way more backstory about a bull's head than we do any character.
If we had more of Lynch's development and Villeneuve's aesthetic, we'd have a near perfect version of the film.
Absolutely! Villeneuve caught the Dune-universe as I pictured it in my head incredibly well, from visuals over story telling to sound. And it showed just how alien and dangerous / harsh Arrakis is.
I'm just a tad worried about Feyd-Rautha so, Sting was just brilliant in that role!
Arrakis is alien but it's depicted as relatively friendly in the dream-sequences. They used sounds of waves crashing on beaches, which makes a lot of sense as to the Fremen it is their home.
My only gripe is that Arrakis doesn't feel hot enough, but I also don't really care.
[sigh]. You could have just googled this, but from the Appendix:
> He found that in the wide belt contained by the 70-degree lines, north and south, temperatures for thousands of years hadn't gone outside the 254-332 degrees (absolute) range
That's -2 F to 137 F in the arctic circle.
> Kynes and his people turned their attention from these great relationships and focused now on micro-ecology. First, the climate: the sand surface often reached temperatures of 344° to 350° (absolute). A foot below ground it might be 55° cooler; a foot above ground, 25° cooler.
That's 114 to 125 degrees Fahrenheit for "a foot above ground".
I don't think it's been underrated, in critics' circles, for 15-20 years or so by now. Since the tales about its production, and the related art, have fully emerged, it has been widely re-evaluated - also because Lynch has gone on to become a bit of a sacred cow after Twin Peaks and his later work.
The original series set the bar for thriller tv series from then on - both on the good elements (always one more secret, one more cliffhanger; living on the edge of fantasy) and the bad (making up stuff as they go along; disappointing finale; unceremoniously cut).
I have only watched a couple of episodes of series 3 but by now I've seen enough Lynch to dread his output, narratively speaking.
* Why would soldiers sent to fight on a Desert-Planet wear black, vulcanised, full body rubber Hazmat suits?
* What on Earth is a "Wyrding Module" supposed to be?
* What exactly is achieved by charging into battle while holding a pug on ones arm?
* It says "Ornithopter" in the books, implying something vaguely animal-shaped, not a hovering metal box.
* Why do communication devices in the far future resemble telephones from the early 1900s?
* What exactly was the point of bringing the late-stage navigator to the meeting in a room-sized spice-tank, when his subordinate did all the talking anyway?
just speculating (usually set design isn't accounting for anything beyond "looks cool"):
- Hazmat suites can contain climate control, and the Tuareg wear dark clothing as well
- Wyrding Module: In Lynch's words: He didn't want Kung-Fu on sand dunes
- No idea what you mean, but charging into battle has since been proven to be viable tactic by the Avengers, so it has to work
- Ornithopter refers to the propulsion, like a bird, instead of a simple helicopter; I'd have to watch Lynch's Dune so to see how those actually look like in his film; Villeneuve nailed them pretty well
- Same reason why the first Enterprise under Archer used fancier screens than the ones under Kirk or Picard; IMHO Lynch borrowed a lot of his aesthetics from WW1, and the Dune universe is surprisingly low tech anyway
- The navigator: A wild guess, I always understood it as a way to show the importance of Spice and Arrakis to the Guild when they had to send one of their Navigators to talk directly to the Emperor instead of using proxies
Ok, I take Lynch pug and raise you by a guy carrying a shield but no sidearm then. The charging into battoe after breaking formation is similar in both cases.
If you're referring to something in the Villenueve Dune, I don't remember it in enough detail to know what you're talking about. And I only know the Lynch one from snippets and memes. :) I think I agree about breaking formation. Just once I'd like to see a movie that took battle formation seriously.
I referred to our beloved Cpt. America. For some reason we accept all kinds of absurdities from Marvel but call out other films dor much less. It's a film, a lot of stuff is just there to look cool. Which is a pitty, because those films that do shiw things like formations properly, e.g. Alatriste, still look cool if you ask me.
For Cpt. America this makes total sense...his shield is both a weapon and a defensive device, as demonstrated throughout the movies many times.
But carrying a small dog (a pug) while charging into battle? What exactly is the point? Where is the logic? Even if its the royal dog, when the compound is overrun by the archenemy of the duke, the royal family unaccounted for, the most effective weapons sabotaged, the last thing his most capable military commanders would think of doing, is saving the royal pet. And even if that was the intention, how does carrying it into the thick of the fight achieve that?
Yes, we accept absudities in movies, if they make sense in the setting. If Cpt. Rogers were to carry a pet hamster around in his pocket for seemingly no reason, while fighting Hydra, it bet the acceptance of this would be kinda low, Marvel or no.
Any story has an implicit assumption about the kinds and levels of absurdity you can expect. People get weirded out when you violate those expectations. Marvel, by virtue of their source material if nothing else, has set a much higher baseline of goofiness for their movies than most others. Whereas we all sort of know Dune was meant to be Serious Scifi. And I guess people who decide they like the Lynch adaptation have re-calibrated their expectations, and found the movie internally consistent on some level. It's all about audience expectations (and sometimes Hollywood ignorance/laziness, I guess).
There is a lot of truth to that. I just realized that I have a hard time to discuss fictional work with people, that base their criticism on a highly subjective view of "logic". Especially when that "logic" fails to be satisfied by actual, historical events and actions.
Heck, there was one guy that stormed the beaches of Normandy with a sword, internet meme culture considers that be cool. Reality had stranger things happening then a lot of fiction. And still people complain about fictional people in a fictional story not acting logical. As long as the fictional logic is in itself consistent, and that includes visuals and style of the fictional world, I'm fine with it.
The Lynch version deserved all the drubbing it received. Weirding modules, heart plugs, and rain on Arrakis: trampling all over the novel. The exposition by way of overdubbing was horrible too.
Dune is not a good movie though, unfortunately. It's like 1/3 of a good movie. By any conception of storytelling, it doesn't constitute a story. As its own piece of media, without knowledge of the book, it doesn't stand as a coherent work. Compare it to something like Fellowship of the Ring which could be watched alone.
It got progressively bad and by the end the director's touch was nowhere to be seen, just rushing to finish the story and blasting that overbearing Hans Zimmer score. Then the final line "This is only the beginning" came. Meanwhile, “Blade Runner”, “Arrival”, “Prisoners”, “Sicario” are all great.
It has always weirded me out that Zimmer keeps getting jobs. His overly pathetic drone has always been like a parody soundtrack. But people plays it straight as if he is actually making something that adds to the movies. Dune was peak Zimmer so if you like him I guess it makes sense to call it his best. But to me it was him making fun of himself with a straightface and nobody calls him out on it.
The score was actually quite thoughtful, attempting to use rhythm and harmony and vocal style that was fitting for the universe, ie. very far removed from our current styles.
I promise it's worth your time. Being honest, I fell in love with the score before I even knew it was Zimmer's, and once I found out, my first thought was "of course."
To each their own, but if you want an idea of what went into the score, that video above is insightful.
I have seen it. Appreciate you sharing it though.
Point is I don't buy his angle. To me it wasn't a fresh take on sounds of the future. It meshed bad with the movie and was a bunch of trite tropes like detached arabic-like wails and dry buzzers. Soundcloud is full of amateur electronic artists doing it better. But in the end its all taste, and I wouldn't call the movie (and soundtrack) bad. Just not anything near good. I'm kind of envious of you who like it.
While I loved the overall production, for me the choice of actors for several of the side characters actually made it feel like a more generic Hollywood action movie.
For me it had been better to pick less well-used actors than Jason Momoa, Josh Brolin, Javier Bardem etc. Just robbed it of some of the mystique I think.
Mid-price movies disappeared when streaming became a thing.
We either get low-budget stuff (10-20 million), mostly horror. Or movies that are huge gigantic blockbusters that MUST succeed.
The mid-price stuff is gone. The ones that have enough budget to make a director's vision come true, but not so much that it brings in people from The Company suggesting their pet things to be added.
I'm sure they're out there, but it just boggles my mind that someone would decide to go watch Dune or not based on say Jason Momoa having a few minutes on screen.
Although that's really only recognizable in hindsight. In 1984 Stewart was still a fairly obscure character actor who had minor parts in BBC shows like I, Claudius.
The first half was about world building. It was atmospheric and interesting, however elusive. Right in the middle when the s** hit the fan, the film started sinking in the quicksand of plotting and by the end the director was MIA. The second half was so Hollywood I felt it was written by Kathleen Kennedy.
Dune is very very pretty, and there are some good performances in it, but over all I found it a terrible movie and a bad rendition of the source material. I think it's so visually and aurally overwhelming that folks don't notice how lousy it is.
For one thing, most of the characters are elided, they appear but they have so few lines and so little consequence that they might as well have been left out. Thufir Hawat especially was woefully neglected. Piter De Vries? No one even says his name! If you haven't read the book you wouldn't know who David Dastmalchian is supposed to be.
Incendies is his first movie that got him a nomination at the Academy Awards. The budget is smaller than Dune and it's in French and Arabic, so the mass appeal isn't there, but it's well worth the watch. To quote Denis Villeneuve: "[it's] a modern story with a sort of Greek tragedy element".
The movie is filmed in Jordan and Montreal, which have a similar feel to Dune, and I would say that he probably took inspiration from his Incendies days to make Dune.
The movie is based on a play by Wajdi Mouawad, who now works at the very prestigious théâtre national de la Colline as the director.
Bladerunner 2049 is so slow with such a bland uneventful story and I don't think that cinematography or visuals are enough to compensate for a weak story. I just found the whole thing incredibly boring.
To make a parallel between cinematography and special effects - I think at this point people are kind of dismissive of movies that emphasize special effects and there's this old George Lucas quote that "a special effect without a story is a pretty boring thing". But swap "special effect" for "cinematography/visuals" and suddenly movies like Bladerunner 2049 are treated like they are masterpieces even though they hardly have any story to tell.
> Bladerunner 2049 is so slow with such a bland uneventful story
Well, it follows a similarly slow and relatively uneventful story in the original, so that's not really a critique. That's not really the point of BladeRunner, anyway - the point is the ambience, the worldbuilding, and the philosophical questions it raises, which 2049 provides in good amount. It will never have the same cultural impact that the original had, but it is an extremely respectful sequel with incredibly beautiful cinematography.
Not everything is a space opera (Lucas' is terrific!). 2048 has strong emotions stretched over a weak story. 1984's somehow had more story, but also quite slow paced.
Some of my favorite movies almost don't have a story at all, neither sfx.
This is one of the few movies of the last two decades that brought me to tears at the reveal towards the end. Once it hits you, you'll look back at the entire movie in a different mindset and it's emotionally jarring. It also helps not having read the source story.
Don't know if you have but may I recommend reading the original novel from Ted Chiang named "Story of your life"? It's a bit different from the movie but still incredibly mind bending.
Also, any novel from the collection "Story of your life and others"[1] is worth reading and thinking about. Ted Chiang is an exceptional writer.
> Susanne Schötz, Robert Eklund, and Joost van de Weijer, for analyzing variations in purring, chirping, chattering, trilling, tweedling, murmuring, meowing, moaning, squeaking, hissing, yowling, howling, growling, and other modes of cat–human communication.
sure the music was fine, it might win best achievement on music score, or music edition... but best picture? you cannot evaluate a picture from the music lol
This sounds like an hour of throat singing, which whilst technically interesting, I don’t quite see (or rather, hear) how it guarantees an Oscar. No-one coughs at all
How about this (/s):
During a pandemic, people are hyper aware of coughing in their environment. It becomes a subconscious trigger (like a baby's cries). Every time a cough occurs in a film, it causes that same shift and elevated processing -- and has the side effect of making a viewer better pay attention/enjoy a scene.
So instead of desensitization, coughing becomes comfortable, reassuring, or perhaps even attractive due to it's being such a common part of life...hence the coughgeist may be reduced due to the appreciation of each cough.
In hindsight, I suppose I should have appealed to the fact that critiquing the seminal papers in the field is a serious data set bias and that trying to "learn to critique" on the best papers ever written in a field was less likely to produce a useful "critiquing" skill and more likely to produce some overfitted garbage skill, but, hey, I hadn't taken AI yet! I didn't know how to express that.
(It did produce a garbage skill, too. I tried writing "real" critiques using my brain, but after getting Cs and Ds for the first couple, I learned my lesson, and mechanically spit out "Needs more data", "should have studied more", and as appropriate, "sample sizes were too small". Except for that last one, regardless of the study. Bam. A series of easy As. Sigh. I liked college over all, but there were some places I could certainly quibble.)
Anyhow, this is the paper that needs to be assigned towards the end of the semester, and students asked to "critique" it. It's a much better member of the training data set for this sort of skill.