[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-gnuzilla] GNU LibreJS won't be removed from GNU IceCat
From: |
Stephan Kreutzer |
Subject: |
Re: [Bug-gnuzilla] GNU LibreJS won't be removed from GNU IceCat |
Date: |
Wed, 7 Mar 2018 20:14:50 +0100 |
Hi guys,
I don't know to which extend each of you is involved in web standards -- I
myself only observe it passively, so feel free to object to any of my following
statements:
To me, it looks like that there can't be a HTML6 any more than there was a
HTML4 and HTML5, because in HTML5, the crazy web guys decided that they neither
need a DOCTYPE any more, and the deprecated version attribute for the root
element was removed, so now, to my knowledge, there's no way to distinguish a
HTML5 document from a HTML6 document except reading the whole thing and
heuristically find out at the very end that it isn't the version/standard your
custom supports, or to find out according to which Schema that document wants
itself to be validated against. The idea here seems to be that older clients
are supposed to still be able to read/interpret documents that contain elements
of a newer standard, but only react to those elements they support and ignore
all others, at the expense of never being able again to introduce new
structures that conflict with the older standards, as there's no secure path
for safe identification and conversion any more.
Regarding the scripting and general conception, I arrived at the impression
that the web is more or less an application stack (forms and JS and media
elements and whatnot) and not for mostly static documents at all, because what
do you have for the latter? Headers, paragraphs, lists and only the most
primitive type of link, that's basically it? The initial concept and spec seems
to be focused on providing a mechanism to link together resources from
different systems in different formats in lists, so it's easier to navigate
them while the host system details are abstracted away by the URL [1], while
the CERN research data and publications themselves weren't (re)written in HTML
(and how could they, who would ask the world to convert all of their stuff for
this small Hypertext system that doesn't offer a lot for text?). Later, the
browser people abandoned the semantic web as there's more money in e-commerce,
online applications and centralized services like Google. What would happen if
ordinary people could run their own small clients/agents that can easily work
with data that's published on the web without the need to bring a big, bloated
browser that's able to parse whatever crap HTML might be out there and tries to
make something reasonable out of it? And now that effort is revived under the
new name "linked data", but in contrast to semantic web ideas, now humans are
supposed to invest their valuable lifetime to read API documentation and write
specific code for it.
I might be totally wrong with my perspective, but on 2018-12-09, it'll be the
50th anniversary of Doug Engelbart's Mother of All Demos, so there's a group
that tries to come up with something that's a better networked environment than
what we usually encounter today, along the lines of the early Internet pioneers
[2], Doug Engelbart [3], Ted Nelson [4], and I'll also add David Gelernter [5].
Things are all over the place as the new system isn't there yet, but
https://doug-50.info
is a place to learn more about those efforts and to get involved, as I would
doubt as of now that "HTML6" will help with some of the problems at hand.
Sincerely,
Stephan Kreutzer
[1] https://www.youtube.com/watch?v=x2GylLq59rI
[2] https://archive.org/details/ComputerNetworks_TheHeraldsOfResourceSharing
[3] https://www.youtube.com/watch?v=yJDv-zdhzMY,
https://archive.org/details/000127EngelbartColloquiumPart1 from 38:00 on,
https://archive.org/details/000127EngelbartColloquiumPart2 from 44:00 on, but
especially from 49:19
[4] https://www.youtube.com/watch?v=c3I54QXQPLA
[5]
https://www.youtube.com/watch?v=XrhM6uXMLZg&list=PLZQMfWBUelIge46VFOd53V1IhW-UzHScK
https://www.youtube.com/watch?v=8pTEmbeENF4
--
Sent with freely licensed software (GNU/Linux). See
https://www.gnu.org/distros/free-distros.en.html
For encrypted communication (GPG):
https://pgp.mit.edu/pks/lookup?op=get&search=0x25BE126289B321DC