Clarifications of "Viewpoints Research Trip"

Computations and data are dual; a computation is a function from data to data, and a datum is a function from computations to more computations. (In particular, a data type is described completely by its "fold" and "unfold" operations) This duality is at the heart of programming language design. The sole difference between call-by-value and call-by-name lambda calculi is whether they treat functions as computations or as data, yet it leads to an enormous difference in programming style.

The main sticking point is that, due to what remains of the von Neumann architecture, there is something called "code"; the stored representation of a program. Code is not computation; it is data. This is why one is able to write compilers, reverse-engineer DRM, etc. - if code was actually a computation, instead of merely data describing a computation, it would be impenetrable. One doesn't read compiled binaries for pleasure, but the fact remains that they are still legible - they can be disassembled, studied, and as the Boomerang project showed, even turned back into high-level code. Actual computations are completely ephemeral; however hard one looks to find a computation, one will only see data. But they are an incredibly convenient abstraction for understanding the universe, so they can't be ignored.

Code is mobile; it moves around whenever one updates or installs an application. Computation is not, because there's nothing that can be moved.

Computation should be avoided if possible; it is slow, error-prone, and generally a nasty business. Data should reign supreme.

The web of today is in fact mobile code; particularly, 99% of it is written in Javascript. They do "download code and run it within a “page” with constrained resources". (One gets an "This page is taking too long to run" message after some time) I'm not certain what you mean by "mini operating systems", but Google and Mozilla are experimenting with web-centric operating systems that boot straight to a browser, and the amount of code that needs to be added is apparently quite small.

There is no difference between data and code; trying to create a "“document format” that represents documents as data, not as code" is futile. What one can do, on the other hand, is write it in a more or less declarative language; in other words, shift the burden of implementation from the programmer to the compiler writer.

The difference between Word, Excel, and PowerPoint documents is artificial; HTML is better than all of them. (Although the default presentation of the data is Word-like, and thus some javascript and css are required to allow Excel / PowerPoint formats; but see http://www.w3.org/Talks/Tools/Slidy2/ for how easy it is; see Popcorn.js and other things for how the web is evolving beyond these tools - the main issue of course is that a browser is a document consumption tool, not a document creation tool, so most people do not have the requisite knowledge)

I, of course, am indeed trying to solve "the whole problem of data representation, interpretation, transmission, and re-interpretation." There have been some quite promising results lately; extensible grammars, compilers, backends... the problem I wish to solve is that of creating a language without being tied down to any design decisions; in other words, a completely modular system; it seems plausible today. The design decisions would be offloaded into libraries, and developed by their users; but the underlying language would be the same, so that one could borrow constructs / algorithms / optimizations from other people without difficulty.

My desires

• I want a distributed peer-to-peer scalable indexable spam-resistant low-latency high-bandwidth version-controlled real-time hardware-accelerated virtualized customizable Unicode-compliant multiplatform internationalized extensible fast embeddable scoped sandboxed self-verifying anonymously-developed profitable secure popular standardized open free memoizing multi-core multi-machine concurrent parallel programming language.
• I want a distributed peer-to-peer scalable indexable spam-resistant low-latency high-bandwidth version-controlled real-time hardware-accelerated virtualized customizable Unicode-compliant multiplatform internationalized extensible fast embeddable scoped sandboxed self-verifying anonymously-developed profitable secure popular standardized open free memoizing multi-core multi-machine concurrent parallel editor.
• I want a distributed peer-to-peer scalable indexable spam-resistant low-latency high-bandwidth version-controlled real-time hardware-accelerated virtualized customizable Unicode-compliant multiplatform internationalized extensible fast embeddable scoped sandboxed self-verifying anonymously-developed profitable secure popular standardized open free memoizing multi-core multi-machine concurrent parallel platform.
• I want them all NOW!

Can a portable language provide C-like unions?

Not unless it has a portable memory model.

My interests

• Ideas that I haven't encountered before
• Things to make achieving goals more efficient
• Humans
• Other stuff
  

List of things I don't care about

• Trolling
• Firmly held beliefs
• Names
• Sunk costs
• Unobserved events
• Definitions

UX story #1

It was Saturday morning, and Joe Programmer hopped out of his bed and
got onto his computer. It booted up instantly, as usual, greeting him
with a smiley face (as configured) and the latest interesting tidbits.
"What?" Joe exclaimed. "Bill made a trillion dollars off his latest
program? I need to get to work!"

Joe took a sheet from his "list of programs to be reimplemented in
Mathnerd314's programming language", and went to work.

"Hmm", he said, "this uses one of those Java/C# hybrids. I really hate
that crap. Good thing it's guaranteed bug-free. I don't get paid to fix
bugs."

Whistling, he went to work, mostly just typing back in the code, until
he came to a place with a base class and a few derived classes. "This
one's a puzzler," he said. "The derived classes don't override any of
the getters of the original class, but there's an overriden method that
doesn't call its parent! This won't work at all for Mathnerd314's
language, where both the base class method *and* the derived method are
called at once."

He IRC'd Mathnerd314, and explained the problem. "Oh, your universe is
impossible," said Mathnerd314.

Joe, and his problem with Mathnerd314's language, promptly ceased to exist.

The End

Slogans

"the ultimate programming language"
"the superset of your language"
"reverse-engineering compilers so you don't have to"
"replacing syntax with semantics since <date>"
"solving the halting problem, every day before breakfast"
"porting the world to run on thin air"
"more abstract than abstraction itself"

Wiki design

The Structure

• Dict: (Dictionary / Default namespace)

  • code / pseudocode for various important algorithms and data structures

  • working definition of each term

  • snippets of various PL documentation (grammars, semantics, ...)

  • authoritative names for the various PL's

  • relatively difficult to edit - protected to some (small) group

  • new terms (pages) should be easy to make

  • copying of portions wikipedia and lesswrong might be good

  • nothing should ever be deleted here unless "OMG THE SERVER'S RUNNING OUT OF SPACE!!!"

  • large #'s of redirects likely

  • disambiguations should be resolved in DictionaryChange (terms have only 1 meaning)

  • should be a relatively small area, mostly for people who want to know what they're talking about

  • creme da la creme of the site

• DictionaryChange (Talk: ?)

  • RFCT's (request for changes to terms)

  • clarification on whatever some random git posts as a new term

  • garbage-collected once the RFCT has been discussed

• Meta:

  • overview of the site

  • links to related sites

  • IRC channel policies

  • ... other meta stuff ...

• Dect: (Deconstruction)

  • detailed picking apart of PLs

  • analysis of which features work and don't work

  • contain research, summaries/links to research

  • stated opinions of some authorities / random gits

  • performance data / charts

  • cognitive behavioral therapy

  • /really intense/ discussion of the lambda calculus and its relatives

  • trolling for n00bs

  • ... other interesting stuff ...

  • stuff deleted if the author agrees (possibly via protracted silence)

  • goal is to find stuff worthy of a term / page in Dict:

• Perhaps some other namespace(s) for IRC logs, debates, flamewars, images, etc. so we don't have to use Dect: all the time

• User pages limited somehow and garbage-collected based on content

End goal: random user posts some code snippet in Dect:, we say use [[Dict:X]] [[Dict:Y]] and [[Dict:Z]] in [[Dict:AwesomePL]] to make it maintainable. (Maybe make Dict: the main namespace, so it's just [[X]] [[Y]] and [[Z]], and have autolinking enabled - wiki software to shorten it to [x] might be better)