1. 4

    I am in week 3 of the OCaml MOOC: https://bit.ly/2y2V989 Very interesting course, I currently find OCaml interesting because it influenced F# & Rust. I tried F# but find the CLR slow and the language tools are not a good fit when you are in a Unix-like system and and prefer a plain editor rather than IDEs. I still find Rust very interesting, but less so for business type software. Also with OCaml you can easily apply the patterns described in https://pragprog.com/book/swdddf/domain-modeling-made-functional

    1. 1

      Hello @eterps -

      very intersting statement: * “I still find Rust very interesting, but less so for business type software.”* I am curious why this is the case? Is it the state of the library ecosystem, or the fact that it is too much overkill to use Rust’s version of memory management for business software?

      1. 2

        The second one, for most business type software I doubt that memory management is the foremost concern. I do like Rust’s library ecosystem, its package management and its modern tooling in general. Rust’s type system is also flexible enough to describe business problems in a similar way as: https://fsharpforfunandprofit.com/posts/no-uml-diagrams/ , if there was something like Rust with garbage collection I would be very interested. For low-level programming Rust is perfect as it is. Maybe something like http://gluon-lang.org can fulfill that role in the future, however currently its type marshalling is a productivity killer.

        1. 1

          Thank you for clarifying your comment. Pity that these is not equivalent to garbage collected Rust that has all the bells and whistles you would like to have, and not the baggage of the JVM/Mono juggernaut.

    1. 2

      Over a decade ago one of my professors told me that the main barrier to implanted stimulation/recording devices was materials science. The brain tends to be very paranoid about foreign matter inserted in it and quickly encapsulates it. There have been numerous attempts to put neurotrophic factors on implanted electrodes (to encourage neurons to grow towards them) and to come up with conducting but more neutral materials (like diamond, or other funky carbon forms) and I don’t know if there has been a proper breakthrough.

      Neural interfaces are interesting because you start out feeling you need to understand how the brain codes information and it turns out the brain can adapt it’s code (to an extent) to what you can read/write just like it adapts to all other situations.

      1. 4

        There was an article not too long ago mentioned on the Skeptic’s Guide to the Universe podcast: http://www.bme.utexas.edu/news/1030-ultraflexible-probe

        Flexibility was key to reducing scar tissue formation.

        1. 2

          One of the many troubles of understanding the brain is the fact that it is a bio-electric mashine which introduces an incredible depth of biochemical, biological and biophysical aspects. We are speaking here about interdependent problem domains, from the key-keyhole principle and neurotransmitters, energy/matter flow, all the horrors of cell biology and fun protein circus. We have very little understanding at this stage due to the depth and breadth of the challenge. I guess it is one of the great puzzles of life - and looking at the state of our society, perhaps it is good we do not understand it yet, even if it means that we can not heal certain psychological diseases such as depressions, psychosis, et cetera.