[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Replace HTML4 doctype declaration

From: Jean-Christophe Helary
Subject: Re: Replace HTML4 doctype declaration
Date: Sun, 19 Dec 2021 11:17:43 +0900

> On Dec 19, 2021, at 11:01, Jacob Bachmeyer <jcb62281@gmail.com> wrote:
> Gavin Smith wrote:
>> On Fri, Dec 17, 2021 at 09:08:24AM +0900, Jean-Christophe Helary wrote:
>>> Is there a place *now* where HTML4 is a requirement ? HTML5 is the only 
>>> *current* standard and does not specifically target web applications.
>> We output HTML4 to get some flexibility in the output, but I am not sure
>> how useful the HTML4 doctype declaration is any more and perhaps we should
>> switch to the simpler HTML5 "<!DOCTYPE html>" header.  It looks like we are
>> trying to conform to a standard that nobody cares about anymore.
>> Does anybody object if I go and change texi2any to output this instead of
>> the HTML4 Transitional doctype?
>> As far as I remember the main difference is the use of the <tt> tag which
>> has been removed from the standard, but I am sure there are other usages
>> too.  It will have been discussed on this list before.
>> The GNU project's attitude to standards applies here:
>> https://www.gnu.org/prep/standards/html_node/Non_002dGNU-Standards.html#Non_002dGNU-Standards
>> "The GNU Project regards standards published by other organizations
>> as suggestions, not orders. We consider those standards, but we do
>> not “obey” them. "
> I would suggest staying with HTML4; it is upwards compatible in that an HTML5 
> parser should generally be able to read it, and the HTML4 DOCTYPE actually 
> declares a version, instead of the "eternal now" HTML5 uses.  That may be 
> suitable for shiny Web apps that change day-by-day and are always loaded from 
> the cloud, but an actual fixed version is much better for documentation, 
> which may be read from an archive years from now.  There will be no question 
> as to the correct interpretation of an HTML4 document, ever, while HTML5's 
> versionless declaration leaves open future questions of "now which version of 
> HTML5 is this?".
> I cite the GNU project's official attitude towards standards in arguing to 
> reject the deprecation of HTML4.  At most, we might want to move towards 
> emitting valid "strict" HTML4.01, which is more of a semantic markup than the 
> "transitional" form.

I don't understand the worry about using <!DOCTYPE html> but sticking to valid 
"strict" HTML4.01 output. We can "declare" in the documentation that the output 
conforms to strict HTML 4.01 and be done with the header.

Jean-Christophe Helary @brandelune

reply via email to

[Prev in Thread] Current Thread [Next in Thread]