[eeps] New EEP draft: -discontiguous declaration

Raimo Niskanen <>
Wed Jun 8 15:45:44 CEST 2011


On Wed, Jun 08, 2011 at 02:01:04PM +1200, Richard O'Keefe wrote:
> 
> On 7/06/2011, at 10:00 PM, Raimo Niskanen wrote:
> > 
> > A Markdown document looks as plain text when you view it; plain text email.
> > That is another possible view of "plain text".
> 
> I cannot understand you here.
> 
> Markdown does NOT look like plain text.
> It is full of special characters which do not bear their usual meanings.
> It is at ALL times, however viewed, a complex markup language.
> 
> If by "plain text" you mean only that it is something that can be
> usefully viewed using TextEdit or Emacs or even vi, where "usefully"
> means "you can figure out what some of the text is, even if there is
> weirdness you do not fully understand", and mean to contrast this
> with the strange semi-binary formats used by some word processors,
> then yes, agreed, but in that sense, so are LaTeX and XHTML "plain
> text".

Both LaTeX and certainly XHTML has **MUCH** more obscuring
markup than Markdown. And that is basically what I am trying to say.

> 
> We are agreed that Markdown (like LaTeX, XHTML, and to a large
> degree RTF) is a "text" format rather than a binary format.
> But to me, something counts as PLAIN text only if every character
> that you see represents itself only.

Then I that is your definition.

But I still think "looks like plain text" describes
Markdown much better than XHTML.

> 
> >> 
> >>> Or, check out the EEP repository; or from Github you can
> >>> download it as a .tar.gz file.
> >>> Edit eeps/eep-9999.md, ./build.pl, Browse eeps/eep-9999.html
> >> 
> >> Well, ever cooperative, I did just that.
> >> Downloaded the .tar.gz (a little tricky because there is no
> >> 'git' on this machine: rcs cvs svn hg even darcs but no git).
> >> Unpacked it, and are my eyes deceiving me?
> >> 
> >> There is no eeps/eep-9999.md
> > 
> > ./build.pl builds any file named eeps/eep-[0-9][0-9][0-9][0-9].md,
> > so if you copy the template eeps/eep-0033.md to e.g eeps/eep-9999.md
> > and edit it it will be built into eeps/eep-9999.html. You can of course
> > choose any number that is not occupied, but I thought 9999 was a nice
> > temporary number to use.
> 
> Well, if someone says to me "edit frobnitz.zappo", I expect there to be
> a "frobnitz.zappo" file already there that I can edit.
> 
> > 
> >> 
> >> Markdown is supposed to produce XHTML, we are told.
> >> However,
> >> 
> >> 	perl md/Markdown.pl eeps/eep-0033.md > 33.htm
> >> 
> >> does not produce well-formed XML (the rock bottom entry level
> >> requirement for something to count as XHTML), let alone valid
> >> XHTML.
> >> 
> >> m% tidy 33.htm >/dev/null
> >> line 1 column 1 - Warning: missing <!DOCTYPE> declaration
> >> line 1 column 1 - Warning: inserting implicit <body>
> >> line 1 column 1 - Warning: inserting missing 'title' element
> >> Info: Document content looks like HTML 3.2
> > 
> > That it produces body content that can become valid XHTML when
> > included in an XHTML file did not surprise me that much when
> > thinking about how it normally is used, that is as a part
> > of a containing page.
> 
> Producing an XHTML FRAGMENT is not the same thing as producing
> XHTML.  It may not have surprised someone who is familiar with
> the tool chain, but it most certainly surprised ME.
> 
> There is no reason why the process that includes the output in
> "a containing page" cannot strip off the outer elements it has
> no use for.

It is much easier to add surrounding tags to a known well
formed body than to parse and strip off XHTML tags from
a complete XHTML document since no tools are needed
except concatenation.

> 
> >> 
> >> The headings are used wrongly in eep-0033.md,
> > Is text before first <h1> allowed? I.e the <pre><code>...
> > </code></pre><hr /> section?
> > 
> 
> >From ISO/IEC 15445:2000(E) :
> 
> <!ELEMENT HTML        - -  (HEAD, BODY) >
> 
> 	The <HTML> ... </HTML> tags are required.
> 
> <!ELEMENT HEAD        - O  (TITLE) +(LINK | META | STYLE) >
> 
> 	The <HEAD> tag is required and so is a <TITLE>.
> 
> <!ELEMENT BODY        - O  ((%block;)*,(H1,DIV1)* ) +(DEL|INS) >
> 
> 	The <BODY> tag is required.  There may be material
> 	before the first <H1>, but it may not contain any <Hx>
> 	tags.
> 
> <!ELEMENT H1          - -  (%text;)+ >
> 	
> 	The <H1> and </H1> tags are required.
> 
> <!ELEMENT DIV1        O O  ((%block;)*, (H2,DIV2)* ) >
> 
> 	The <DIV1> and </DIV1> tags are not required, and indeed,
> 	should not appear.  They have no attributes.
> 
> 	ISO HTML element and attribute names are not case sensitive.
> 
> 
> 

Thank you for that explanation.


-- 

/ Raimo Niskanen, Erlang/OTP, Ericsson AB


More information about the eeps mailing list