Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: Best talks of 2018?

news.ycombinator.com

Post reply on HN
Print view View original

111–120 of 205 posts

Re: Ask HN: Best talks of 2018?

#111

Post by agentultra »

maxhallinan wrote: post #37
ttiurani wrote: post #20

"Maybe Not" by Rich Hickey: https://www.youtube.com/watch?v=YR5WdGrpoug Another excellent talk by the creator of Clojure, and like the previous ones, relevant for all programmers.

I thought this was one of the notably bad talks this year. The whole premise that a function of Maybe a should be a function of a without an API change is neither intuitive to me nor really justified by Hickey. Different things are different. It's sad to see someone build such a wall around himself when faced by something (type theory) that he doesn't understand.

I agree.

Further his exposition on `Either a b` was built on a lack of understanding of BiFunctors.

The icing on the cake was his description of his own planned type theory. What he described was, as I could decipher from his completely ignorant ravings, a row-based polymorphic type system. However he passes off his insights as novel rather than acknowledging (or leveraging) the decades of research that have gone into type theory to describe the very system he is trying to build.

Worse, he continued to implore his audience to spread FUD about type theory, claiming several times, "it is wrong!"

Re: Ask HN: Best talks of 2018?

#112

Post by agentultra »

chubot wrote: post #43
maxhallinan wrote: post #37

Earlier quoted context omitted.

I thought this was one of the notably bad talks this year. The whole premise that a function of Maybe a should be a function of a without an API change is neither intuitive to me nor really justified by Hickey. Different things are different. It's sad to see someone build such a wall around himself when faced by something (type theory) that he doesn't understand.

Are you sure that he doesn't understand it, or is it possible that you haven't worked with the same scale of systems he has? Here's my response to his talk, which I found insightful: https://lobste.rs/s/zdvg9y/maybe_not_rich_hickey#c_povjwe Also, I think your comment suffers from the problem here, where you invoke "type theory" without any elaboration: https://lobste.rs/s/zdvg9y/maybe_not_rich_hickey#c_ioeyob Rich ha…

I haven't seen any of Rich's contributions to any major open source Haskell project. I can't really speak to his experiences with Haskell in proprietary code bases. Has he been a prolific Haskell contributor/hacker/user?

From his talk he hasn't convinced me that he understands Haskell's type system. Not only does he misunderstand the contract given by `Maybe a` but he conflates `Either a b` with logical disjunction which is definitely a false pretense. He builds the rest of his specious talk on these ideas and declares, "it [type theory] is wrong!"

He goes on to describe, in a haphazard and ignorant way, half of a type theory. As I understood it these additions to "spec" are basically a row-based polymorphic type system. Why does he refuse to acknowledge or leverage the decades of research on these type systems? Is he a language designer or a hack?

I can't even tell to be honest. He has some good ideas but I think this was one of his worst talks.

Re: Ask HN: Best talks of 2018?

#114

Post by nickpsecurity »

sifoobar wrote: post #99
olodus wrote: post #47

Earlier quoted context omitted.

Was about to suggest this myself. I see a few that didn't like this and I can see why, but I thought it was interesting to listen too. Always interesting to hear someone knowledgeable argue for something you don't think you agree with. It is also interesting since Rich is in the quite unique position of handling a functional language with dynamic types (a dying breed in the functional space imo). I don't think Rich f…

Dying breed based on what? They're different compromises, there is nothing inherently superior about static types. It's more popular right now, because we're very much into rules at the moment; but I have a feeling that's peaking. Typically, you need to move back and forth a number of times before it clicks. So you have all these people on the net fighting over which end of the spectrum is the Right one, and next mon…

There is one advantage of formal specs in general that static types inherent: they can enforce a correctness policy on a design that works for all inputs with no runtime overhead. SPARK Ada eliminating entire classes of errors with mostly-automated analysis is one example. Rust blocking temporal and some concurrency errors without GC is another. I'll note they're both hard to eliminate with testing, too.

With that, comes the next benefit: less debugging and lower maintenance costs. It has to be balanced against annotation costs. Most projects claimed to come out ahead if it was just safety rather than full correctness. You can also generate tests and runtime checks from specs to get those benefits.

So, for high reliability with lower maintenance, there's definitely an advantage to knocking out entire classes of error. That leads to last advantage Ill mention: ability to warranty (market) and certify (regulators) your code free of those defects [if the checker worked]. The checkers are also tiny on some systems. Rarely fail. Inspires extra confidence.

Re: Ask HN: Best talks of 2018?

#116

Post by RickJWagner »

My favorite talk was called "Generations @ Work", presented at Red Hat Summit 2018.

It was about different generations in the workplace, what each grouping typically liked, disliked, was motivated by, and how they communicated differently.

I really liked it because of the content, but also because it was given by a small panel, I was a member of it. :)

Why did I propose such a talk? I usually submit technical talks to conferences, sometimes they are chosen. This year I did NOT have a technical talk selected by the Summit committee, but on a whim I had submitted this 'softball/managerial' title. It was the one that got me a ticket to Summit! You never know.

Re: Ask HN: Best talks of 2018?

#117

Post by olodus »

sifoobar wrote: post #99
olodus wrote: post #47

Earlier quoted context omitted.

Was about to suggest this myself. I see a few that didn't like this and I can see why, but I thought it was interesting to listen too. Always interesting to hear someone knowledgeable argue for something you don't think you agree with. It is also interesting since Rich is in the quite unique position of handling a functional language with dynamic types (a dying breed in the functional space imo). I don't think Rich f…

Dying breed based on what? They're different compromises, there is nothing inherently superior about static types. It's more popular right now, because we're very much into rules at the moment; but I have a feeling that's peaking. Typically, you need to move back and forth a number of times before it clicks. So you have all these people on the net fighting over which end of the spectrum is the Right one, and next mon…

Wow wow chill out. It wasn't meant as a negative comment. I was just making the observation most new languages are staticly typed and that a lot of what you hear in the functional space is about types (type theory, dependant types, categories...) though this might be just what I've had infront of me recently. I retract my statement. I love the dynamic functional languages. Erlang and Clojure are amazing (now that I think about it elixir is a lot in the news, as I said my statement probably was wrong). And as I said in my previous comment, I think Rich talk made some really interesting points.

Re: Ask HN: Best talks of 2018?

#118

Post by olodus »

sifoobar wrote: post #99
olodus wrote: post #47

Earlier quoted context omitted.

Was about to suggest this myself. I see a few that didn't like this and I can see why, but I thought it was interesting to listen too. Always interesting to hear someone knowledgeable argue for something you don't think you agree with. It is also interesting since Rich is in the quite unique position of handling a functional language with dynamic types (a dying breed in the functional space imo). I don't think Rich f…

Dying breed based on what? They're different compromises, there is nothing inherently superior about static types. It's more popular right now, because we're very much into rules at the moment; but I have a feeling that's peaking. Typically, you need to move back and forth a number of times before it clicks. So you have all these people on the net fighting over which end of the spectrum is the Right one, and next mon…

And I don't consider clojure dying at all btw. In the field of data science especially it is thriving. I was tired ok... So... There...

Re: Ask HN: Best talks of 2018?

#119

Post by sifoobar »

nickpsecurity wrote: post #114
sifoobar wrote: post #99

Earlier quoted context omitted.

Dying breed based on what? They're different compromises, there is nothing inherently superior about static types. It's more popular right now, because we're very much into rules at the moment; but I have a feeling that's peaking. Typically, you need to move back and forth a number of times before it clicks. So you have all these people on the net fighting over which end of the spectrum is the Right one, and next mon…

There is one advantage of formal specs in general that static types inherent: they can enforce a correctness policy on a design that works for all inputs with no runtime overhead. SPARK Ada eliminating entire classes of errors with mostly-automated analysis is one example. Rust blocking temporal and some concurrency errors without GC is another. I'll note they're both hard to eliminate with testing, too. With that, c…

Sure, in return for bending code over backwards to fit rules you get guarantees; but pretending that is anything but a different compromise isn't helping. If Ada was all that, everyone would be using it; and that's not going to happen, because it's not a realistic approach to programming.

The academic approach doesn't look very constructive to me, never did. Once you have a language nailed down so hard that errors aren't possible, the complexity of dealing with it will be more or less the same as dealing with the errors in the first place.

I would much prefer a focus on more powerful tools that fit into the current "unsafe" ecosystem while offering a more gradual and flexible path to improved safety.

Re: Ask HN: Best talks of 2018?

#120

Post by zengid »

reikonomusha wrote: post #30
ttiurani wrote: post #20

"Maybe Not" by Rich Hickey: https://www.youtube.com/watch?v=YR5WdGrpoug Another excellent talk by the creator of Clojure, and like the previous ones, relevant for all programmers.

I don’t like Rich’s appeal to human intuition and the physical world as justification for not using—and even disparaging—mathematical structures. His quote from memory was something like “There’s no such thing as a Maybe Sheep,” and that was aggravating to listen to. He then gives the simplest possible examples, and fails to convince me that his approach scales to the real world. At some point he starts talking about…

He's a genius but I think he's getting bitter about how people are happily sticking with static strongly typed languages. At the end of the day he's trying to sell people on using Clojure, so of course he's going to rip on other languages.
Post reply on HN