Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon. Entire thread

S-expression HTML as Standard

Name: Anonymous 2013-09-17 13:25

Who the fuck thought it would be a good idea to use opening and closing tags as DA STANDARD for SHITML? Are the people at W3 dumb?

For those who don't know what I'm talking about,
look how ugly and unnecessarily long HTML is.

<html>
<head>MY WEBSITE</head>
<title>MY WEBSITE</title>
<body>
<p>THIS KIND OF MARKUP IS <b><i><u>SHIT QUALITY</u></i></b></p>
</body>
</html>

But compared to the S-expressions

(html (head "MY WEBSITE")
(title "MY WEBSITE")
(body (p "THIS MARKUP IS" (b.i.u "SATORI QUALITY"))))

It looks far cleaner and makes your eyes feel right at home. It's easier to parse and won't take up as a few kilobytes / megabytes (depending on the page) less Internet bandwidth. Sure it has a lot of parenthesis but it's more attractive than the repeating shitstorm of </tags>.

I know there are a lot of implementations to this but there should be one STANDARD to rule them all. The next version of HTML in the standard should be something like this.

Name: Anonymous 2013-09-17 13:28

HTML is not programming.

Name: Anonymous 2013-09-17 13:38

But >>2-san, HTML 5 + CSS 3 is Turing-complete!

Name: Anonymous 2013-09-17 13:47

Opening and closing tags aren't [i]that[/i] bad, SGML-likes lend themselves to markup well when the amount of markup isn't under a few percent of the entire data.

What's really horrible is that tags may be implicitly closed. Whoever thought of that should be shot, it defeats the entire purpose of using a SGML-like language.

Are the people at W3 dumb?
Of course they are, it's the same team that defends DRM hooks in HTML with the "argument" that it makes the web more open.

Name: Anonymous 2013-09-17 13:49

>>4
[i]that[/i]
H-how is that possible? I wasn't using [#] anywhere.

Name: Anonymous 2013-09-17 13:52

Parens hurt my eyes.

Name: Anonymous 2013-09-17 14:15

>>4
SGML-likes lend themselves to markup well when the amount of markup [b]isn't[/b] under a few percent of the entire data.
I really hope you meant to say is, not isn't.

Name: man(7) && mdoc(7) 2013-09-17 14:54

>>1
 Tim Berners-Lee,
 Yes, they can't do even proper \fIstyling\fR correct, and decide on things.
 We did, it was called [t]roff. Hipster Physicist didn't like it, so we are stuck in this mess.
.PP
>>2
 We know, it is \fBMarkup\fR
.PP
>>3
 Because Javascript was introduced. Should have kept it as Scheme.
>>4
 Meh. Other markups are better:
  org-mode, Markdown, Asciidoc, wikicreole, heck even wml.
.PP
>>5
 tablecat's broken regex.
.PP
>>6
 You an replace them with
.RS
 if <like == true>
    <statement_limiter = '<>' >
 else
    <statement_limiter = default >

Name: Anonymous 2013-09-17 18:14

>>8
 Because Javascript was introduced. Should have kept it as Scheme.
I agree about Scheme, but HTML5 + CSS3 really is Turing-complete on its own, you can encode rule 110 in it.

Name: Anonymous 2013-09-17 19:23

>>6

But angle brackets don't? You must have eye cancer.

Name: Anonymous 2013-09-17 19:24

>>9
 We know it is. But not enough to self reproduce.

 We should really make a petition, and make a UI support for those markups. It's sad the Chrome has 46.02% of the browser market. Its all Javascript, sigh. Why can't they do one thing correct.

Name: Anonymous 2013-09-17 21:47

>>1
I agree with you, but if you think beyond just that, you realize you don't even need an html or a body object. Using S-exprs is one step towards a saner web, the other one is redefining the shit standard that has stupid things like tag arguments, <p> | <br> | <blockquote> | <q> | <em> | <i>, other redundant tags and even the le whoops forgot my closing tags sorry xDDD leniency of most shitML parsers.

Name: Anonymous 2013-09-18 7:47

But it's also a repeating shitstorm of PARENS

Name: Anonymous 2013-09-18 15:35

>>13
I'd rather have a ))) than a </div></div></div?.

Name: Anonymous 2013-09-18 19:39

I agree that HTML's syntax isn't that great, but if you want to try bootstrapping a new ML onto the internet, why constrain it to be ``HTML, but with parens instead of brackets and slashes''? Why not make a self-sustaining system like Gopher or the original WWW? There's plenty wrong with the web these days that could be fixed by completely changing the underlying ML, so try and grab as much of it as you can. It's not like you're going to do anything anyway.

Name: Anonymous 2013-09-18 20:07

>>15
That's what I said in >>12, except for the "it's not like you even know how to do it" part.

Name: Anonymous 2013-09-19 0:17

>>1,15, etc
So given the impossibility of replacing HTML, how about this stupid solution? Write ()ML then compile it to HTML. You could even do it using a .js library on demand, allowing for improvements to the size of the content to be propagated at the cost of an increase in client-side processing time.

Name: Anonymous 2013-09-19 1:37

>>17
It has been done already, but I refuse to use any ML that retains the cruft of the HTML standard. (p text) is only slightly better than <p>text</p>, but the fact that p is stupid crap remains.

Name: >>8 2013-09-19 1:46

>>17 welcome to the 1990's where documentation generators will be the way of the future. No more need for nasty old databases talking to a script in the Common Gateway Interface to deploy output. And we will definitely not go with the live output while code is changing plan as we go. We will have none of that.

-CMS [sub]Agile based

Name: Anonymous 2013-09-19 2:05

>>18
I should have known it had been done. I agree with you about the tags as well. Div vs span vs p seems odd, and my God, css. Inheriting float from the previous sibling etc. is just silly. It simply isn't easy to do something that should be. The whole thing needs to be thrown out. That's sort of my thought with compiling to HTML: you wouldn't have a collection of HTML tags any more than C is a one to one map of assembly. Closer to the subject at hand, one could do for HTML what jQuery did for JavaScript, as limited as that may be.

That said, >>19 has a point as well: if you look at the HTML CMSs generate it's pretty terrifying. One would need to write an optimizing compiler, so to speak.

Name: Anonymous 2013-09-19 12:15

Your assignment:
Write a S-expression -> XML parser in Haskell or any functional programming language of your choice.
Extra credit for included unit tests!!!

Name: Anonymous 2013-09-19 16:33

>>17
How about not? That way, you would have to deal with the quirks of HTML and the quirks of your replacement. That sort of thing is why we have Automake and its family (the Parcae of programming), as well as all those goddamn X-to-javascript projects.

Name: Anonymous 2013-09-19 21:23

>>22
Well I've thought about that. JavaScript is difficult to write a converter for, because it's genuine code. HTML on the other hand is, as has been exhaustively explored in this thread, not a real programming language. Its sole output is visual. So it's possible that in compiling your ()ML you might receive build errors if a rendering engine (webkit, gecko, etc.) put something in the wrong place, or hid it when you wanted it to display, etc. It's probably possible to automate that process. So if you set out clear specifications for your language, with no confusion as to where things are or what's on top, you could be automatically notified if something failed to land where it should.

Name: >>23 2013-09-19 22:57

>>22
Also they got doom running by compiling C to JS, so that's got to be worth something.

Name: Anonymous 2013-09-19 23:08

>>24
I made a ``boat'' out of wooden sticks and snot when I was 5.

Name: >>25 2013-09-19 23:47

Did anyone play with it?

Name: Anonymous 2013-09-20 0:08

>>26
My younger cousins did. They were like 2, 3 and 5 respectively.

Name: >>25 2013-09-20 2:18

>>27
You had a user base! You're on your way, son! Just think - if your little cousins played with sticks and snot, how many parents would pay to keep their little ones occupied? Mister, you're on to something!

Name: Anonymous 2013-09-20 8:52

>>28
That doesn't exclude the user base from being fucking retarded, though.

Name: Anonymous 2013-09-21 10:52

Lisp a shit.

Name: Anonymous 2013-09-21 11:57

>>29
A fool and his money...

Name: Anonymous 2013-10-04 21:47

It's Official, the web is going to hell:
http://boingboing.net/2013/10/02/w3c-green-lights-adding-drm-to.html

Time to work on Sexp-ML. Anyone got any good libraries to get started, than org-mode? Maybe I should extend a web browser to read/interpret org-mode/

Oh and https://bbs.progrider.org/prog/read/1380816422/1
You are welcome for the bump. Here is another thread we had about this:
https://bbs.progrider.org/prog/read/1378687067/22-

Good luck!

Name: Anonymous 2013-10-04 21:52

>>32
#20 By: Kimmoth, October 3rd, 2013 03:15

It's your 10th birthday, and your favourite uncle has just turned up.

Giddy with excitement, you run to the door to see what he's brought for you.

'Yay-' you start to say, as he unloads a roundhouse kick in your face.

Happy birthday, kid! Have fun eating through a straw the rest of your life.

...Fuck you, [Tim Berners-Lee].

Name: Anonymous 2013-10-04 21:58

>>32
I'm being cynical, but it was about time that the current web dies anyway. This is our chance to fix what is fundamentally wrong with its standards.

Name: Anonymous 2013-10-04 22:30

>>32,34
LOL, I love this guy's comment:
#9 By: TheGlitchEcliptic, October 3rd, 2013 01:52

We need an xkcd-style super villain to create an implausible scenario highlighting the potential abuses of this.

"The terrorists have planted thirteen bombs across the city and their locations are hidden in this website's source code!"

"I can't access the source code, sir!"

"Damnit! If we don't get those locations we might as well call off the bomb squads and call in the coroners!" The timers are already counting down!"

"Hang on, I just googled a web forum where an amateur programmer offers a Firefox plugin that exploits a workaround!"

"Thank god!"

"The link's no good anymore! It was hit with a DMCA takedown!"

"What?! Who filed the complaint?"

"....the terrorists, sir."

"Damnit! How are they always one step ahead?!"

Name: Anonymous 2013-10-04 22:39

>>32
I feel physically disgusted by reading this. What the fuck? This is way too much for me.

Name: Anonymous 2013-10-04 22:45

>>34
Your not being cynical, you are actually being an optimist, thinking we can actually fix it, and making an implementation in time. I hated the web anyways. USENET was and should have been the norm. I just hope the secushare guys can make an translators for must protocols out there, even a http2psyc gateway:
http://about.psyc.eu/Software

Name: Anonymous 2013-10-04 23:02

>>37
Your not being cynical
And you're not even being ironic

Name: Anonymous 2013-10-05 11:41

>>37
Usenet was never good for multimedia presentations and hyperlinked pages.

Name: Anonymous 2013-10-05 12:59

>>38 Why, [u]Fuck[/u] you⸮

>>39
D fock? Any message you send is encoded in proper MIME, which any client can use to interpret however specification required. And hyperlinking was as easy as saying:

More info can be found at: nntp://some.channel:some_user@someserver:/location/of/doc.

Since it is distributed in nature, and you can edit your message as to not have identifying information on the nodes. It is still the best federated P2P pseudonym network. high and low latency capabilities, and great caching. Even better when entire channels were a community hosted DHTs, of many different things.

Don't blame the protocol for your perceived lack of UI support for multimedia presentations and hyperlinked pages. Most common ones could and by default do:
wikipedia.org/wiki/List_of_Usenet_newsreaders#Free.2Fopen_source_software
wikipedia.org/wiki/Comparison_of_Usenet_newsreaders
Newer Posts
Don't change these.
Name: Email:
Entire Thread Thread List
All trademarks and copyrights on this site are owned by their respective parties. All comments and posts are the responsibility of their own posters.
- Tinychan + 2ch Mode + RSS -