Thread: Maths knowledge

[Mario's last post in entirety...]

O_o

Yes. Your point being?

Seriously, what are you trying to say with those comments? I realize the question at hand was rhetorical, but the pure "wrongness" of it needed to be addressed. Are you agreeing with me that offering the question in support of your position was silly? Are you trying to pass off the "wrongness" of it by being obtuse?

*shrug*

I'll have to ignore most of that for the sake of time, but I do have to address one thing.

It seems the perfectly adequate approach for a browser.

Allowing a random collection of optionally terminated, partially terminated, fully terminated, optionally nested, partially nested, fully nested, and interleaved declarations of almost every language element is completely inappropriate for any machine application.

If nothing else, convincing several hundred thousand developers to interpret such an abstract and underspecified language the same is impossible.

Soma

^_^

You could have simply said "yes" to "Are you trying to pass off the "wrongness" of it by being obtuse?" and saved yourself the typing.

The language "standard" (I use the term very loosely.) is underspecified. The "standard" specifically allows the garbage that is "tag soup". (The garbage of "tag soup" is a direct result of having optional closure and allowing partial and interleaved closure.) The various valid or invalid interpretations of "tag soup" is irrelevant.

What is relevant to the discussion is your offering that producing HTML must be programming because to produce HTML one "must follow a rigid set of syntactic and semantic rules". One need not "follow a rigid set of syntactic and semantic rules" to produce HTML because no "rigid set of syntactic and semantic rules" exist.

The HTML version in development, HTML5, actually does have some rigid lexing and parsing rules, but these new rules are all optional allowing "tag soup" to continue by nature of providing some level of standard semantics to the garbage. However, the required semantics are still underspecified.

Soma

I wasn't aware it would be made to look like that. It was in answer to one specific aspect. HTML, or anything for that matter, cannot be defined as a programming language solely on that. Neither I ever tried to imply it could.

Fine. You didn't mean to imply that.

Now would you like to address my other concerns? (Actually, I don't really care. I'm just saying that you didn't consider anything I said that argues why I say that HTML isn't a programming language.)

And the word "solely" hasn't entered into the discussion as far as I'm concerned, but that's really not important.

Simply because we don't have a sound and clear definition for what is a programming language.

Just because you, and most everyone else, would like to use different definitions doesn't mean that a "sound and clear" definition isn't available. Programmers especially will argue about the precise definition of anything.

The definition is Turing completeness, and all that requires is two properties: 1. The ability to conditionally branch to another part of code. 2. The ability to set and read from some form of memory.

That's interesting and all, but just so as you know you can artificially limit an otherwise "Turing Complete" language without changing the utility of the language in specific circumstances.

I have to think logically and in terms of the language syntax, its semantics, idioms and within the context of what I want to do.

I have to do this when writing a short story in English for consideration by my publisher who has very strict "steampunk" standards.

Any language not turing-complete cannot be a programming language.

YES THEY ........ING CAN!

They may not be of general use, but many domain specific declarative languages are not "Turing Complete".

A quick check... because I knew I had heard of these before... also reveals other programming languages that just like SQL show the implications of adopting that idea.

There are a lot of programming languages that aren't "Turing Complete" while still being of extraordinary use in their domain.

I think what can be said instead is that all general-purpose programming language are (should be) turing complete.

Truth.

This is the correct statement.

Soma

Yog Sothoth!

I tried to get back to work. I really did. It keeps nagging at me. The stupidity. The insanity.

I must give an example!

Any language not turing-complete cannot be a programming language.

Imagine a language very like C++. Image a language so very like C++ it is in fact C++.

Now imagine an embedded interpreter for our imagined language. Image that this embedded interpreter is itself targeting a very "low resource" embedded microcontroller.

Now let us transform our imagined language to be more fitting to what resources our imagined interpreter makes available. Let us restrict recursion so that no implementation may recourse to recursion. Let us remove `goto', `break', and `continue'. Let us restrict existing loop constructs to `MAXIMUM_ITERATIONS' iterations no matter the consequences. In short, let us force any executable to halt!

Our very C++ like language is now no longer "Turing Complete", yet the majority of semantically and syntactically valid C++ source written will still execute exactly as the programmer intended.

You may define `MAXIMUM_ITERATIONS' as high as you like. You can define it to be so large that it would take every molecule in the universe to build the ram necessary to store the massive integer. It doesn't matter. The restrictions on our imagined language makes it impossible to achieve a "Turing Complete" status despite the fact that it has only been barely limited.

Soma

shall we have a poll? ;-)

I take the side of "HTML is a scripting language"

There is an adverts section on this page which displays adverts for different sites. Any idea what that is written in? HTML or Javascript? Is it programmed or scripted?

To answer the original question, I'll echo the thought that it depends on what field of programming you find yourself in. Working heavily with databases, you'll find relational algebra helps (though you'd probably understand most of the concepts naturally without first learning the mathematical theory), and boolean logic always plays a role in programming. I recently started studying cryptography just for fun, and I really wish I was better / more familiar at linear algebra and discrete mathematics. You can get by with the basics, but occasionally I run across a statement I really have to think about before I can wrap my head around it.

HTML or Javascript?

My guess is this page contains JavaScript that decides when advertisements should be loaded (or it's decided when PHP is producing the page). That JavaScript then sends an HTTP request to a server for an advertisement, and the response is an animated GIF. There might be a bit of HTML wrapped in there somewhere as a wrapper / container.

Is it programmed or scripted?

Scripted, as it's either PHP or JS, both of which are scripting languages. However, I don't think it's fair to say that doesn't mean it was programmed. The "ML" in HTML means markup language. XML by itself is NOT a programming or scripting language by any means. It is, however, the basis for languages like Jelly, and a data expression format for systems like J2EE, etc... Just because it's used heavily in scripting or programming doesn't make it a programming language anymore than my text editor, even if it is a language with specific rules.