XPost: comp.lang.forth

On 2025-06-10, Paul Rubin wrote:
anthk <[email protected]> writes:

I'm admittedly conflicted on responding to an off-topic thread,
and two months after the discussion has concluded at that.
Still, it mentions Lynx, and... Well, I'm going to cross-post
to (little-used) comp.infosystems.www.misc and set Followup-To:
on the odd chance someone might be interesting in discussing
this further there.

Most of these "web sites" are irrelevant to me.

By the by, I'd like to note that the lifestyle argument works
both ways. I've started using web c. 1998, and within a few
years, settled on Lynx as my primary browser. (I have a Lynx
"bookmarks" file dated August 2001, for example.) I doubt indoor
plumbing is a suitable comparison, but driving a car perhaps is.

And indeed, switching to, say, Chromium now feels like a big
lifestyle change to me. Not unlike starting to drive a car.
Sure, it has its benefits, but it also has its costs, both in
terms of responsibility, and in terms of buying gas (for a car)
or new hardware (for Chromium.)

Being somewhat of a retrocomputing enthusiast (from whence
interest in Forth), I'd say relying on Lynx fits my lifestyle
better anyway.

Can you read sfgate.com? That's a major news site near here.

I can read it via Wayback Machine [1] at the least; e. g.:

After November flop, California Forever launches new city concept

An aerial rendering of where the original planned community by
California Forever would fit into Solano County.

A California city tried to triple in size. Then came the rebellion.

[1] http://web.archive.org/web/20250730/http://sfgate.com/

(FWIW, I have this cheap China-made radio that I listen to news
broadcast locally on UHF/FM on. And every once in a while, I can
catch shortwave CRI broadcasts on it, too.)

Interestingly, I'm able to read apnews.com with lynx. With firefox,
I'm impeded by Cloudflare Turnstile which is basically a JS-dependent captcha.

The "solve-to-read" captchas generally are JS-based, IME.
(Unlike those for posting comments or registering an account.)

I haven't noticed sites skipping a captcha for non-JS browsers
myself, TBH, though I have noticed sites skipping JS-based ads
for Lynx. Can't say I feel disadvantaged by it.

I get a 403 from this with lynx: https://www.fanfiction.net/s/13768935/1/Harry-Potter-and-A-Galaxy-Far-Far-Away

Years ago, I've got an impression that Fanfiction (and some
other servers) reacted badly to "libwww" in Lynx' User-Agent:.
I'd venture to guess it might be related to an unrelated Perl
library (libwww-perl AKA LWP), presumably at one point popular
among bot writers, also having "libwww" in User-Agent:.

I /think/ Fanfiction acquired a bunch of restrictions on top
of that over the years, though. Generally, I'd suggest using
Wayback Machine here as well, but that particular story doesn't
seem to be archived.

FWIW, I've been able to read most of "Darth Vader: Hero of Naboo"
that way last year; see (URI split for readability):

https://web.archive.org/web/20240914132402/ https://www.fanfiction.net/s/11730208/1/Darth-Vader-Hero-of-Naboo

That site also uses Turnstile. Turnstile is becoming extremely
widespread across the net, to push away AI scrapers.

FSF has recently commented, if tangentially, on that in [2].
(They've pointed out that Anubis captcha might be free, but
it's still essentially malware.)

I believe I understand, to a degree, the issues involved
in running a website this day and age, but this particular
solution gets no sympathy from me. If anything, it seems
like a web counterpart to hostile architecture [3].

[2] http://fsf.org/blogs/sysadmin/our-small-team-vs-millions-of-bots
[3] http://en.wikipedia.org/wiki/Hostile_architecture

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

On 2025-06-10, Paul Rubin wrote:
anthk <[email protected]> writes:

Most of these "web sites" are irrelevant to me.

By the by, I'd like to note that the lifestyle argument works
both ways. I've started using web c. 1998, and within a few
years, settled on Lynx as my primary browser. (I have a Lynx
"bookmarks" file dated August 2001, for example.) I doubt indoor
plumbing is a suitable comparison, but driving a car perhaps is.

Well, by that analogy, what do you think would happen if you tried to take
a Ford Model T on to the Autobahn? The modern web is a “kitchen sink”
mess of technologies (well beyond JavaScript) that is going to be heavy
lift for any browser to support:

<https://www.w3.org/TR/>

For example, I have sites that make extensive use of server sent events
(SSE). While JavaScript is the main way to pull the data, the format *is*
just text that any browser could display. But give `lynx` that URL and it
just *sits* on the result, displaying *nothing* until the connection is
closed.

Interestingly, I'm able to read apnews.com with lynx. With firefox,
I'm impeded by Cloudflare Turnstile which is basically a JS-dependent captcha.

The "solve-to-read" captchas generally are JS-based, IME.
(Unlike those for posting comments or registering an account.)

I haven't noticed sites skipping a captcha for non-JS browsers
myself, TBH, though I have noticed sites skipping JS-based ads
for Lynx. Can't say I feel disadvantaged by it.

The thing to shoot for (i.e., what a modern “text” browser should target) is *accessibility*. That’s what standards are geared towards these days,
and that’s what sites are *supposed* to support. Often time the weight
of law and/or public opinion can be brought to bear against large
organizations that do not accommodate disabled people. Try those sites
with something like a screen reader and complain if they still don’t work *that* way.

I believe I understand, to a degree, the issues involved
in running a website this day and age, but this particular
solution gets no sympathy from me. If anything, it seems
like a web counterpart to hostile architecture [3].

The web itself is hostile. Much of what *was* The Internet has gotten
locked up by it, including Usenet. Instead of complaining that you can’t
get modern sites to work on some old HTML browser, maybe question whether
or not it was wise to have tried jamming everything into HTML in the
first place.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-09, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

I took the liberty of removing comp.lang.forth from Newsgroups:
as this discussion would be off-topic there.

By the by, I’d like to note that the lifestyle argument works both
ways. I’ve started using web c. 1998, and within a few years,
settled on Lynx as my primary browser. (I have a Lynx “bookmarks”
file dated August 2001, for example.) I doubt indoor plumbing is
a suitable comparison, but driving a car perhaps is.

Well, by that analogy, what do you think would happen if you tried
to take a Ford Model T on to the Autobahn?

Either I do not understand your point, or you’ve misunderstood
the one I’ve tried to make. The thing is: for a variety of
reasons, whether they want it or not, different people live
different lives.

Think of, say, people living in different parts of the world.

In this case, why should I be concerned with my inability to
take my Ford Model T (let’s suppose, for the sake of this
argument, that I do have one) to the Autobahn, when we do not
have the Autobahn here in the first place? (Why, /billions/
of people in the world don’t have the Autobahn anywhere near.)

We do have decent mass transit, though, and that’s what I’m
happily using. Yet I understand it well enough that in some
parts of the world people do not have such amenities and have
to resort to having a car to get to their workplace and back.

Kind of ‘to each their own’, if not quite that.

The modern web is a “kitchen sink” mess of technologies (well beyond JavaScript) that is going to be heavy lift for any browser to support:

Analoguously, the modern Earth is a “kitchen sink” mess of
human languages that is going to be heavy lift for any human
to learn. Настоящее Вавилонское столпотворение! Должно ли,
однако, /меня/ волновать то, что я не владею, к примеру,
албанским, испанским и (или) китайским? Я пишу на двух, и, как
мне кажется, владею еще одним на уровне read-only. Хотелось
бы освоить еще три-четыре, но, увы, с возрастом такого рода
подвиги appear to become harder to achieve. So there.

<https://www.w3.org/TR/>

I do not consider W3C to be a relevant authority when it comes
to /interoperability,/ which is to say, having more than two
independent implementations of a technology or stack.

FWIW, my go-to sources about web, or, rather, the parts of web
I’m interested in, are:

http://html.spec.whatwg.org/
http://rfc-editor.org/ (particularly IETF STD 99 / RFC 9112) http://262.ecma-international.org/ (particularly 5.1 and 6.0)

About the only thing I refer to W3C for is CSS specifications.

For example, I have sites that make extensive use of server sent
events (SSE). While JavaScript is the main way to pull the data,
the format *is* just text that any browser could display. But give
`lynx` that URL and it just *sits* on the result, displaying
*nothing* until the connection is closed.

Indeed, Lynx does not implement HTTP/1.1 in full, which is
apparently what SSE requires [1].

[1] http://en.wikipedia.org/wiki/Server_sent_events

More importantly, am /I/ part of your target audience? If so,
I fully expect for your website(s) to be accessible to a user
agent implementing HTTP/1.1 and HTML Living Standard. It’s my
responsibility to ensure that my UA supports them, and work
around whatever issues it might have with that. (Say, when
Lynx stumbles on chunked encoding, I might use Wget to download
the resource first, and pass it as a local file to Lynx then.)

If your website requires anything besides that (CSS, JS,
WebRTC, – whatever) to be usable, I might complain.

If I’m /not/ part of your target audience, why bother?

By the same merit, if my grandmother is part of your target
audience, I fully expect for your website(s) to be written in
German or (and) Russian. I’m afraid that regardless of how
good your website might be, she’s not at an age when learning
English, or any other new language, is still a viable option.

Otherwise, we might complain. Sometimes, it /does/ help.

For an example, an IRC pal of mine once complained to website
operators that the higher TLS version that they’ve started to
require is not supported by their old tablet /and/ that they
cannot upgrade because newer tablets are more power-hungry and
with only solar and no grid power, they hardly can use them.

The website operators complied and lowered the TLS version
requirement.

The thing to shoot for (i. e., what a modern “text” browser should target) is *accessibility*. That’s what standards are geared
towards these days, and that’s what sites are *supposed* to support.
Often time the weight of law and/or public opinion can be brought to
bear against large organizations that do not accommodate disabled
people. Try those sites with something like a screen reader and
complain if they still don’t work *that* way.

Last I’ve checked, Karl Dahlke was trying them regularly with a
braille display, and /did/ complain that JS APIs evolve faster than
he can implement them in Edbrowse. As we put it back in 2021 [2]:

Nevertheless, support for modern Web technology, and first and
foremost for the latest advances in Javascript, is largely a
moving target, and likely requires more effort than the current
team of developers can provide.

[2] http://am-1.org/~ivan/misc-2021/005.edbrowse.en.xhtml

Accessibility is an important point, true, but one I find more
important on a personal level, is that I /can/ patch Lynx (due
to its relative simplicity), and I very much /cannot/ patch
Chromium (due to its relative complexity.)

That’s one of the chief advantages of Lynx to me, and that’s
why I largely dismiss complaints by website operators that I’m
using ‘too old’ (or whatever) a browser that’s too inconvenient
and troublesome for them to add support for.

Instead of complaining that you can’t get modern sites to work on
some old HTML browser, maybe question whether or not it was wise
to have tried jamming everything into HTML in the first place.

First, why these two options have to be mutually exclusive?
I’d think that the very existence of this discussion here on
(non-HTML-based) Usenet is evidence enough that some of us
can do both. (For a given value of ‘complain’ in any case.)

Then again, my Lynx is 2.9.0dev.12 from January 2023. Is it
already considered “old”?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sun, 10 Aug 2025 08:15:07 +0000, Ivan Shmakov wrote:

I do not consider W3C to be a relevant authority when it comes to /interoperability,/ which is to say, having more than two
independent implementations of a technology or stack.

Which is exactly how standards like those endorsed by W3C are created.

For example, this is the reason why the WebSQL standard failed to pass:
because all the browser makers were using the same underlying DBMS engine (SQLite) for their implementations, which didn’t meet the threshold for
truly “independent” implementations.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
John Ames <[email protected]> wrote:

On Sat, 9 Aug 2025 19:53:24 -0000 (UTC)
Doc O'Leary , <[email protected]> wrote:

Well, by that analogy, what do you think would happen if you tried to
take a Ford Model T on to the Autobahn?

...you'd get passed by everyone going faster than 42 MPH but otherwise everything would work normally enough because the operating principle
of a roadway hasn't changed since the Neolithic, conventions for motor traffic have been broadly consistent since the '40s,

Then I should have used a better analogy; the lesser highways around where
I live have higher minimum speeds posted, and are full of people who would
not be polite to someone puttering around in traffic going 2x or 3x
faster. The “roadway” in the analogy would be maybe TCP/IP; that’s not the layer that’s causing the problems.

and fundamentally
people just want to get wherever it is they're going and aren't weirdly fixated on controlling what anyone else does...?

And yet we still come full circle back to people misunderstanding what the modern web is, which *is* about being fixated on controlling every aspect
of the browsing experience (or at least *trying* to). The only-static-
HTML web is about as lively as a non-binaries Usenet turned out to be.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

On 2025-08-09, Doc O'Leary wrote:

Well, by that analogy, what do you think would happen if you tried
to take a Ford Model T on to the Autobahn?

Either I do not understand your point, or you’ve misunderstood
the one I’ve tried to make. The thing is: for a variety of
reasons, whether they want it or not, different people live
different lives.

I’m contending that the problem is not that there are different cultures,
but that there are inevitable culture clashes. It just doesn’t make much sense to gripe about Lynx or any other browser not working well with the
modern web; it’s not the obligation of the web to “live the life” of Lynx.

More importantly, am /I/ part of your target audience?

I think that’s a false premise. Why should I be trying to limit my
audience to some “target” subset? If I have tabular data, for example, I want as many people as possible to use it however they wish. That
generally does *not* mean crapping it up with HTML formatting and calling
it good.

If your website requires anything besides that (CSS, JS,
WebRTC, – whatever) to be usable, I might complain.

How quickly the “different people” sentiment evaporates. :-)

If I’m /not/ part of your target audience, why bother?

If you can conclude you’re not part of someone’s “target audience”, why complain?

By the same merit, if my grandmother is part of your target
audience, I fully expect for your website(s) to be written in
German or (and) Russian. I’m afraid that regardless of how
good your website might be, she’s not at an age when learning
English, or any other new language, is still a viable option.

More of the false premise, but it’d be *really* off topic to fully unpack
why historical/cultural baggage is at odds with globalization.

Accessibility is an important point, true, but one I find more
important on a personal level, is that I /can/ patch Lynx (due
to its relative simplicity), and I very much /cannot/ patch
Chromium (due to its relative complexity.)

But, again, my argument is that a browser like Chromium aims to
implement the mess that is modern web standards, not just whatever
patchwork subset that Lynx supports. Some parts *could* be pulled out
for use with Lynx in some fashion:

<https://en.wikipedia.org/wiki/V8_(JavaScript_engine)>

But I don’t expect there’s enough of a user base for *any* text browsers
to justify that development effort. I still think the best angle to
approach non-GUI web browsers is some kind of accessibility argument.

That’s one of the chief advantages of Lynx to me, and that’s
why I largely dismiss complaints by website operators that I’m
using ‘too old’ (or whatever) a browser that’s too inconvenient
and troublesome for them to add support for.

In the abstract, I’m right there with you! I’ve gone so far as to try to make my sites *static* HTML when possible. But the fact remains that
there is usually some underlying data in play that is *not* HTML, so *my* rendering it in HTML for *you* to see becomes of questionable value.

Instead of complaining that you can’t get modern sites to work on
some old HTML browser, maybe question whether or not it was wise
to have tried jamming everything into HTML in the first place.

First, why these two options have to be mutually exclusive?

Ask the people who make web browsers. I’m quite happy to serve up a YAML file from my server that contains all the data you need about a product,
but there isn’t a web browser I know of that’ll render it to *your*
liking, so I’m forced to also render it in HTML to *my* liking. Again, it’s the web that’s being hostile here.

I’d think that the very existence of this discussion here on
(non-HTML-based) Usenet is evidence enough that some of us
can do both. (For a given value of ‘complain’ in any case.)

Then again, my Lynx is 2.9.0dev.12 from January 2023. Is it
already considered “old”?

In every way that matters, yes. Just like a new Usenet client could be released tomorrow and instantly be outdated, because the underlying
technology is not fit for purpose. Which is to say that, while usable
for *a* purpose (simple text discussions), Usenet is no longer anybody’s
idea of a *modern* messaging platform.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Doc O'Leary , <[email protected]> wrote:

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

On 2025-08-09, Doc O'Leary wrote:

Well, by that analogy, what do you think would happen if you tried
to take a Ford Model T on to the Autobahn?

Either I do not understand your point, or you've misunderstood
the one I've tried to make. The thing is: for a variety of
reasons, whether they want it or not, different people live
different lives.

I'm contending that the problem is not that there are different cultures,
but that there are inevitable culture clashes. It just doesn't make much sense to gripe about Lynx or any other browser not working well with the modern web; it's not the obligation of the web to "live the life" of Lynx.

Of course nobody's obliged. Equally nobody's obliged to live the
life of other browsers, hence the resurgence in Gopher use by
people fed up with modern Web design. Gopher works in Lynx but not
in Chrome or Firefox. But isn't it better to appeal to both crowds
by writing HTML that is usable in all the browsers your audience
chooses to use?

More importantly, am /I/ part of your target audience?

I think that's a false premise. Why should I be trying to limit my
audience to some "target" subset? If I have tabular data, for example, I want as many people as possible to use it however they wish. That
generally does *not* mean crapping it up with HTML formatting and calling
it good.

Regardless of whether you're "crapping up" the data in HTML or I
guess using other technologies like JS to format it client-side,
the user isn't going to see a way to access that raw data unless
you show a link to download it separately. Otherwise the data is
getting crapped up either server-side or client-side, at least
from a user's point of view. As a user who avoids running random
JS (even when I'm using browsers that support it), I obviously
prefer the server-generated crap.

Even if you want all the functionality of a website to be exposed
as a public API, you can add a method in API requests to
enable/disable the HTML rendering. Catering to Lynx users doesn't
shrink your audience, it offers everyone more options!

--
__ __
#_ < |\| |< _#

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
John Ames <[email protected]> wrote:

On Mon, 11 Aug 2025 21:51:09 -0000 (UTC)
Doc O'Leary , <[email protected]> wrote:

And yet we still come full circle back to people misunderstanding
what the modern web is, which *is* about being fixated on controlling
every aspect of the browsing experience (or at least *trying* to).

It's not misunderstanding, it's rejection.

Well, then I’d say they need to *reject* it if they’re going to reject it, not try to use some browser that hasn’t added any new feature support
since the 1990s and act shocked that things don’t work like they used to.
The problem remains that everyone gets “the web” pushed as the one-stop shop for all their online needs (even for things like writing mobile
apps), and that has resulted in the kitchen sink that is the modern web browser. Support other solutions if you don’t like the current state of affairs; I certainly do.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

John Ames <[email protected]> writes:

No, I think I'll stick with active scorn and spite towards web
designers who can't be bothered to do their job properly. The attitude
that it should be considered acceptable for web designers to dictate
people's choice of browser was contemptible in the '90s-'00s and it's contemptible now.

There's a web standard (HTML5) and it includes all those features that
you (and I) dislike. We could agree that it's a BAD standard. Some
people feel the same about ANS Forth. But it's there, and the big
browsers implement it, and web developers for the most part follow it.

I write C++ code sometimes. C++11 introduced a lot of new features that weren't in earlier versions. They were refined further in C++14 and
later. Am I irresponsible or not doing my job if I use those features,
instead of writing C++98 code in 2025?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
[email protected]d (Computer Nerd Kev) wrote:

But isn't it better to appeal to both crowds
by writing HTML that is usable in all the browsers your audience
chooses to use?

Again, I say “no”, and that it’s a mistake to be jamming things into HTML that don’t need to be HTML. I’d generally say that XML-based markup had its chance and has proven itself to be far too inflexible given our pace
of technological innovation. It’s an entrenched system that nobody likes, evidenced by *countless* layers that have been built on top of the web to
make it usable (e.g., I’ve done a lot of React development).

Regardless of whether you're "crapping up" the data in HTML or I
guess using other technologies like JS to format it client-side,
the user isn't going to see a way to access that raw data unless
you show a link to download it separately.

And I do, when I can. For my static sites that render Markdown, I provide
a link to the non-HTML source. And, like I said, while I *could* provide direct links to SSE endpoints, I don’t know any browser that is going to
do a good job handling it if it doesn’t already have SSE support.

As a user who avoids running random
JS (even when I'm using browsers that support it), I obviously
prefer the server-generated crap.

Well, then you need to be running a browser that supports *modern* server generated crap! 💩 :-)

Even if you want all the functionality of a website to be exposed
as a public API, you can add a method in API requests to
enable/disable the HTML rendering. Catering to Lynx users doesn't
shrink your audience, it offers everyone more options!

I mean, that *is* how most JavaScript web sites work these days. The
problem generally is that the API endpoints are buried in the JS code. I don’t see much payback in overhauling the rendering pipeline to `noscript` everything with links to those API calls on the off chance that someone is going to be combing through the XML or JSON responses and mentally reconstructing the behavior of what the dynamic page would be doing.
That’s an “option” request that would need to come with a pile of cash.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-11, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

This is going to be somewhat tangentially related to web proper
and is posted here merely for convenience & completeness. Perhaps
we'll veer back to topic in the downthread discussion (if any.)

Summary: I /am/ old-fashioned. I appreciate finer things in
life, and those finer things IMO tend to be either old-fashioned,
or just plain old. Hopefully sharing the URIs below will help
others see the light and better sense, and start appreciating
such finer things as well. (Or something.)

JFTR, the 38 resources referenced below, sans [4, 5] were
/at some point/ tested by me to be readable with Lynx, and
presumably should work with other "imaginary" browsers as well.
Aside of a single https: instance in [4] (which requires a
workaround), the list in [4, 5] is usable with mpg123(1).

Then again, my Lynx is 2.9.0dev.12 from January 2023. Is it
already considered "old"?

In every way that matters, yes. Just like a new Usenet client could
be released tomorrow and instantly be outdated, because the
underlying technology is not fit for purpose. Which is to say that,
while usable for *a* purpose (simple text discussions), Usenet is no
longer anybody's idea of a *modern* messaging platform.

Somehow, I'm reminded of Paul S Person's comment in
news:[email protected] :

Beethoven is a bit late for my taste. I prefer JS Bach and friends.
Also harpsichord music -- played on harpsichords, not on pianos.

I find that even though I'm no fan of harpsichords, I can
relate to the sentiment at large.

Web is as much part of human culture as music or computer
games. Why is it that I can appreciate centuries-old music
(or music written in styles that were popular centuries ago,
and no longer are), yet it's suddenly morally reprehensible
to enjoy "old-school," Lynx-compatible websites, such as
http://malvaceae.info/ and Wikipedia?

Once again, it fits rather well with the rest of my lifestyle.

Consider my choice of music; most of it is from Magnatune (see
[1]); a fair chunk - http://modarchive.org/ http://scene.org/
http://remix.kwed.org/ , a bit from http://opengameart.org/
(say, [2].) The rest is from local CDDA shops. No Youtube,
no Spotify, not even concert DVDs (because, seriously, "how music
changes through the years?" [3].) Though not a fan, I can relate
to people who get their cassettes from http://dataairlines.net/ .

No streaming, aside of a handful of Shoutcast / Icecast
radios [4, 5]. I particularly like publius shows at [5],
though my personal schedule is at odds with that of his shows
(e. g., [6]), so I have to resort to using station archives [7].

[1] http://users.am-1.org/~ivan/misc-2025/004.magnatune.en.xhtml
There's a .json version, too: http://pin13.net/mf2/?url=http://users.am-1.org/~ivan/misc-2025/004.magnatune.en.xhtml
[2] http://opengameart.org/content/arrival-1
[3] http://web.archive.org/web/2010/http://queenwords.com/lyrics/songs/sng13_01.shtml
[4] http://relay4.slayradio.org:8000/ http://listen.181fm.com/181-buzz_128k.mp3
http://radio.wanderingsheep.tv:8000/jazzcafe
https://iskatel.hostingradio.ru:8015/iskatel-128.mp3
[5] http://anonradio.net:8000/anonradio
[6] http://anonradio.net/schedule/
[7] http://archives.anonradio.net/


On that note, I've long had my sights on
http://archive.org/details/OTRR_Dragnet_Singles . Got to
actually listen to it recently. Yes, it's 1950s. Still good.

But perhaps 70 years isn't old enough for a cultural artifact
that's not a website? Well, I /do/ quite enjoy these authors:

http://en.wikisource.org/wiki/Author:Gilbert_Keith_Chesterton http://en.wikisource.org/wiki/Author:Howard_Phillips_Lovecraft http://en.wikisource.org/wiki/Author:Robert_Erwin_Howard

Or what about this one (one of my favorites since I've first
read it c. 1994; though not the Wikisource edition, of course):

http://en.wikisource.org/wiki/Kwaidan:_Stories_and_Studies_of_Strange_Things

If even that's not "old enough" for a book to be comparable to
a non-mainstream, Lynx-friendly website in 2025, then (1872
and, say, the "Sibylline Leaves" version from 1817):

http://en.wikisource.org/wiki/In_a_Glass_Darkly/The_Room_in_the_Dragon_Volant http://en.wikisource.org/wiki/The_Rime_of_the_Ancient_Mariner

If everything above fails, well, I do refer to the following
book at times, though mostly its "common era" (so to say) parts:

http://en.wikisource.org/wiki/Bible_(King_James)

Now, someone may object that "no one's ever interested in that
PD junk" and that "contemporary writers never release anything
in electronic form without Adobe DRM." And here my argument
gets weak: there're of course Baen Books, Tor Books, and a few
other publishers that make their books available in DRM-free form
(I personally have copies of "Princess Holy Aura" and "The Eye
of the World" in .epub that I can unzip(1) and read with Lynx.)
Still, I know of no Lynx-compatible website where you can
(legally) /obtain/ any of them: paywalls are typically JS-based.

There're several well-known websites hosting amateur writing, but
all of them seem to have been captch'd recently, presumably due
to bot abuse, and are no longer readable with Lynx (or archived.)

So, aside of using Wayback Machine to reach what of them has
been archived over the years, I can only suggest http://lib.ru/ .
In particular, http://fan.lib.ru/l/lokhard_d/ is one of my
favorite contemporary Russian (I mean language, not ethnicity;
though frankly, I've never asked) writers.


But let's get back to computers. As with many (I believe) of
my generation, my early interest in computers and programming
was stimulated by computer games, even though that interest
largely waned by c. 2005.

Take Wolfenstein 3D, for example, which I've first played in
1993 (if not 1992), and has been somewhat of a fan ever since.
I went as far as to write a fair bit of [8], and I still play
the game on occasion (under FreeDOS, too.)

That's an old game, sure. Of "new" ones, I play OpenTTD and
Crispy Hexen. Though of course the first one is a free
software clone of a game first iteration of which came out in
1994; and the second is a maintained fork of the engine for a
game originally released in 1995 (you'll need to obtain a .WAD
file separately; a demo is at [9], use Wayback Machine if
needed, and the complete one can be bought at GOG.)

"The Impossible Bottle" [10] is, so far as I can tell, a brand
new game released in 2020. It's not my usual fare, but I still
managed to complete it back in the day.

It's interactive fiction, though, so even being relatively
recent, and a competition winner at that, it's still pretty
old-fashioned.

But then again: so would be any computer game of interest
to me. Consider my requirements:

* no DRM;

* playability without connectivity (say, on an air-gapped PC);

* playability without GPU; (at least until affordable GPUs
start appearing at http://ryf.fsf.org/ .)

So, yes, either it's old-fashioned, or I won't play it.

[8] http://ru.wikibooks.org/wiki/Wolfenstein_3D
[9] http://classicdosgames.com/game/Hexen%3A_Beyond_Heretic.html
[10] http://linusakesson.net/games/the-impossible-bottle/


Then, of course, there're programming languages. ES 5.1 [11]
/is/ old. All the same, unlike newer versions, it has a bunch
of independent implementations: Duktape, MuJS, likely QuickJS,
and I'm not sure how many of independent ones are in the usual
list of Node.js, Gecko, Webkit and Safari. I find it
particularly convenient to run tests with $ duk testsuite.es
(as opposed to spawning a full-blown "real" browser.)

Ones I use daily, though, are Dash, Bash, Awk (both Busybox and
GNU versions), Perl 5. Occasionally I'd use Tcl (8.6 or Jim)
and NetBSD /bin/sh.

Recently, I've been looking for a Forth implementation.
GNU Forth? Well, actually, no: one that'd run, and compile
itself, on a relatively small, "old or old-fashioned" system,
such as 16-bit 8086/DOS, or a 8-bit AVR MCU.

Doubting that anyone still writes DOS software? Well, take
a look at the examples [12].

Frank Sergeant's Pygmy Forth [13] looks promising, even if
somewhat idiosynchratic. (Then again, what Forth isn't?)

[11] http://262.ecma-international.org/5.1/
[12] http://ibiblio.org/pub/micro/pc-stuff/freedos/files/dos/command/0.86/
http://ibiblio.org/pub/micro/pc-stuff/freedos/files/dos/fdisk/1.4.4
http://ibiblio.org/pub/micro/pc-stuff/freedos/files/kernel/2043/
http://ibiblio.org/pub/micro/pc-stuff/freedos/files/dos/emm386/jemm/5.85/
(Artistic license on this last one, though.)
[13] http://pygmy.utoh.org/pygmyforth.html , and also
http://ibiblio.org/pub/micro/pc-stuff/freedos/files/devel/forth/pygmy/


That'd be all for today. Conclusion per Summary: at the top.
HTH and TYC.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
John Ames <[email protected]> wrote:

No, I think I'll stick with active scorn and spite towards web
designers who can't be bothered to do their job properly. The attitude
that it should be considered acceptable for web designers to dictate
people's choice of browser was contemptible in the '90s-'00s and it's contemptible now.

Who are you to say how a job you don’t pay for is properly done? Who gave you the authority to dictate that the world use *your* pet browser? It doesn’t sound like you know how “choice” actually works.

I feel the opposite way. Web designers are nothing without the software developers that enable them to march forward as technology progresses.
The onus lies there. If some software isn’t being updated to handle the modern web, it stops being a “web browser”. It may be a perfectly fine HTML viewer, though. Enjoy *your* choice.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-11, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:
On 2025-08-09, Doc O'Leary wrote:

Accessibility is an important point, true, but one I find more
important on a personal level, is that I /can/ patch Lynx (due
to its relative simplicity), and I very much /cannot/ patch
Chromium (due to its relative complexity.)

But, again, my argument is that a browser like Chromium aims to
implement the mess that is modern web standards, not just whatever
patchwork subset that Lynx supports.

Unless you define "modern web standards" as "web standards
supported by modern browsers" (and, mutatis mutandis, the
other way around), Chromium supports a /selection/ of standards
in use on the web at large, just like Lynx. Sure, Chromium
supports a bigger chunk, but that's /exactly/ what makes
Chromium bigger and more complex and less appealing to me.

Is FTP a "modern web standard"? Unless I be mistaken, it was
supported by Chromium - until it wasn't. Who decided that FTP
is no longer "modern"? Google? Certainly not W3C.

Personally, I like Lynx' selection better. Even if I'm not
particularly interested in its support for once again popular
Gopher, and if I'd rather /not/ use it as my newsreader.

Some parts *could* be pulled out for use with Lynx in some fashion:

<https://en.wikipedia.org/wiki/V8_(JavaScript_engine)>

But I don't expect there's enough of a user base for *any* text
browsers to justify that development effort.

Conversely, if such an effort is undertaken, and, by some odd
chance, is fruitful, Lynx will quite possibly become too complex
for me to comfortably patch. As a result, I'll be forced to
maintain my own, JS-less fork of Lynx, or switch to another
browser altogether.

Not to mention it'd break my mental model of Lynx, which,
HTML-wise, is merely a tool to convert HTML to text:

$ printf %s\\n "<ol ><li >Foo</li><li >Bar</li>" | lynx --dump --stdin
1. Foo
2. Bar
$ lynx --dump -- http://example.com/
Example Domain

This domain is for use in illustrative examples in documents. You may
use this domain in literature without prior coordination or asking for
permission.
...
$

I still think the best angle to approach non-GUI web browsers
is some kind of accessibility argument.

Such an argument certainly has its merit, but I doubt it'd be
all that surefire. The better outcome I can think of is that
Big Money pulls some "accessibility specification" out of
somewhere, makes one or two of their more popular websites
conform to it, and then claim it's up to the gov't to provide
the people affected with the tools that actually implement the
requisite accessibility features. At public expense, of course.

Meanwhile, there'll be dozens upon dozens of websites that either
it would be too much pain to actually sue; or whose operators
simply wouldn't have the kind of budget it takes to make them
conformant to that specification. So at best, these websites
would be closed down and no longer accessible to /anyone./

I seriously doubt that such an "accessible web" would be readily
readable with Lynx. Rather, I'd bet on something that requires
an "accessibility browser" that costs 3e3 USD per copy and
that would be distributed by some gov't agency or another in
select few countries as part of "disability benefits."

Not to say that braille displays are cheap, mind you.

That's one of the chief advantages of Lynx to me, and that's
why I largely dismiss complaints by website operators that I'm
using 'too old' (or whatever) a browser that's too inconvenient
and troublesome for them to add support for.

In the abstract, I'm right there with you! I've gone so far as to
try to make my sites *static* HTML when possible. But the fact
remains that there is usually some underlying data in play that is
*not* HTML, so *my* rendering it in HTML for *you* to see becomes
of questionable value.

It's not one of my favorite pastimes, but I've found that
unless their author is being particularly creative, I /can/
extract data from HTMLs when needed.

Instead of complaining that you can't get modern sites to work on
some old HTML browser, maybe question whether or not it was wise
to have tried jamming everything into HTML in the first place.

First, why these two options have to be mutually exclusive?

Ask the people who make web browsers.

This is Usenet. People who make web browsers are of course
welcome to join this discussion and tell us why a person who
complains can't question and (or) vice versa.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-11, Lawrence D'Oliveiro wrote:
On Sun, 10 Aug 2025 08:15:07 +0000, Ivan Shmakov wrote:

I do not consider W3C to be a relevant authority when it comes
to /interoperability,/ which is to say, having more than two
independent implementations of a technology or stack.

Which is exactly how standards like those endorsed by W3C are created.

For example, this is the reason why the WebSQL standard failed to
pass: because all the browser makers were using the same underlying
DBMS engine (SQLite) for their implementations, which didn't meet
the threshold for truly "independent" implementations.

I would've appreciated positive examples better, TBH. Anyway,
a bit of digging revealed [1], so I guess I was wrong; thanks.

The current specification features are implemented across major
browsers like Chrome, Firefox, and Safari.

[1] http://w3c.github.io/geolocation/reports/implementation.html

On the one hand, now I'm curious how many other browser-related
TRs they have that only mention the Big 3 above.

On the other [2] http://w3.org/TR/encrypted-media-1/ :

This specification does not define a content protection or Digital
Rights Management system. Rather, it defines a common API that
may be used to discover, select and interact with such systems as
well as with simpler content encryption systems.

Which is to say, they provide a way to arbitrarily restrict
interoperability in an entirely standards-compliant and
interoperable manner.

W3C's role in making the Recommendation is to draw attention to
the specification and to promote its widespread deployment. This
enhances the functionality and interoperability of the Web.

I'm afraid at this point, I can't take their alleged focus on
interoperability seriously.

It'd be akin to admitting that a politician is honest as they
only ever took a bribe once or twice.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-11, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

I'm contending that the problem is not that there are different
cultures, but that there are inevitable culture clashes.

Whatever little I can do about culture clashes ("explain your
point, try to understand the other party, do not escalate"),
I believe I do. Other than that, I find it unproductive to be
concerned with things I can do nothing about.

It just doesn't make much sense to gripe about Lynx or any
other browser not working well with the modern web; it's not
the obligation of the web to "live the life" of Lynx.

Web is not a person, and cannot have obligations as such.

Moreover, almost all web pages that I visit (whether pointed
from web search, from Wikipedia articles, or from "old" IRC and
netnews), are readable with Lynx. (Examples below.)

Of course there's a certain selection bias in that; and it was
never my intent in the first place to ascertain the fraction
of /all/ websites in existence that can or cannot be read with
Lynx, but so far as my experience goes, the observation holds.

From whence, while I /do/ agree that switching from, say,
Chromium to Lynx /is/ a big lifestyle change, I'm going to argue
it's /less/ because there're millions of websites that one
cannot read with Lynx, and /more/ because there's a /handful/
of such websites that are used by millions.

Other than the aforementioned selection bias (Wikipedia doesn't
like to reference resources that Wayback Machine cannot archive,
for example; and if Wayback Machine /can/ archive it, chances
are, Lynx will be able to read it, too), I believe there're two
major factors at play.

* The long-term maintenance cost for "simple" sites is less than
that for "complex" ones. A major PHP (WordPress, MathJax,
- you name it) upgrade can put quite a few extra workhours
on your plate; while a "just bunch of files" website can be
left untouched for a decade, and served with pretty much any
HTTP server (case in point: http://am-1.org/dimath-2016/ .)

Hence: there's a number of unmaintained /but still useful/
"simple" websites; while "complex" ones, once no longer
actively maintained, tend to die out.

* The way search engines interact with site's JavaScript (or
CSS / WebSockets / whatever, for that matter) will necessarily
be different than the way a person with a browser does. Too
much emphasis on interactivity and (or) "modern standards"
can hinder search engines just as well as it hinders Lynx.

My only "problem" with "modern" web is my indifference to it.
Someone starts talking about Facebook this, or Youtube that,
and, frankly, I'm at a loss how to feel and what to suggest.
I tend to respond along the lines of "if it hurts you to use it,
then don't?" but I can't say that's received all that well.

Some of the webpages I've read recently (i. e., from Lynx'
history) follow.

http://en.wikipedia.org/wiki/Flat_Earth http://taiga-experiment.info/taiga-hiscore http://supercheats.com/pc/warcraftorcsandhumanscheats.htm http://analog.com/en/resources/app-notes/an-740.html http://bradrodriguez.com/papers/moving8.htm http://spectrumcomputing.co.uk/entry/26880/ http://malvaceae.info/Genera/Lavatera/herbaceous.html http://w3c.github.io/geolocation/reports/implementation.html http://ru.wikibooks.org/wiki/?curid=14978 http://ibiblio.org/pub/micro/pc-stuff/freedos/files/repositories/1.4/ http://otvet.mail.ru/question/6603461
(a copy of a chunk from http://ru.wikipedia.org/wiki/?curid=50515 ) http://nanochess.org/ http://t3x.org/t3x/ http://pygmy.utoh.org/

More importantly, am /I/ part of your target audience?

I think that's a false premise. Why should I be trying to limit my
audience to some "target" subset? If I have tabular data, for example,
I want as many people as possible to use it however they wish.

I apologize if the term is misleading, but it's the best I was
able to come up with. My point is, intentionally or not, you
/do/ limit your audience. Can someone who can only read
Armenian make use of your websites? How about a five year old?

(Can /you/ read http://sw.wikipedia.org/ ? If so, please kindly
share pointers to your reference materials, as I'd like to be
part of their target audience, too.)

The way I understand it, "target audience" is about making sure
certain people /will be able/ to make use of your resources.
(Rather than making sure some won't be.)

It may take effort to "limit" your audience, but it sure takes
some to "unlimit" it as well.

Which is perhaps why you seem to be opposed to Lynx: including
Lynx users in your target audience means extra work, no?

If your website requires anything besides that (CSS, JS, WebRTC, -
whatever) to be usable, I might complain.

How quickly the "different people" sentiment evaporates. :-)

Does it, though? Wouldn't the notion of "different people"
include those who can't walk, and those who can't talk? those
who only read Danish, and those who only read Spanish? those
who read web with Chromium, /and/ those who read web with Lynx?

By the by, I'd like to stress it out that /not requiring/ some
feature provided by the Big 3 browsers is not the same as
/not using/ that same feature. I do not mind reading webpages
that use JS (or whatever, except for actual, honest to goodness
DRM / W3C EME) so long as they are still readable /without/ it.

If I'm /not/ part of your target audience, why bother?

If you can conclude you're not part of someone's "target audience",
why complain?

Exactly! When I have an issue with a website, I try to
complain to its operator directly. If I haven't complained
to you, it likely means one of the following:

* I'm not aware of your website;

* I do not find your website useful right now;

* I've found that for my uses, it works adequately as is;

* I've figured out how to fetch the data I need from it with
curl(1), without needing /any/ browser.

Most of the time, my problem is not with the site proper, but
rather with a certain widely used service, that, as part of its
"DDoS protection," requires solving a (JS-based) captcha before
letting you obtain "protected" content. Reading a copy of the
resource at Wayback Machine usually suffices as a workaround.

I'm quite happy to serve up a YAML file from my server that contains
all the data you need about a product, but there isn't a web browser
I know of that'll render it to *your* liking, so I'm forced to also
render it in HTML to *my* liking.

How would you know the capabilities of the software I use?

Sure, I /do/ use Lynx; but I use a plenty of programs besides.
I get my weather forecasts from OPeNDAP endpoints under
http://nomads.ncep.noaa.gov/dods/gfs_0p25_1hr/ , for example.
The end result looks like:

## sqrt (vgrd10m * vgrd10m + ugrd10m * ugrd10m)
0d00 7.2 6.5 5.1 4.6 4.7 4.8 3.5 3.6
0d08 3.9 2.7 2.0 1.2 1.0 0.3 0.8 1.1
0d16 1.8 2.2 1.5 2.1 3.0 2.2 1.8 1.9
1d00 2.1 2.4 2.0 3.2 3.9 4.5 4.6 3.9

I /think/ no HTML is involved anywhere in the process.

Granted, I do dislike languages (whether code or data) that
count whitespace as part of their syntax, so I'd certainly prefer
XML or JSON to YAML. And HTML /is/ data, so when needed, I
typically /can/ extract what I need even from a bunch of HTMLs.

So, I suppose what I'm trying to say is: my interest in your
data does /not/ necessarily imply interest in running your
application(s) on my computer(s). Javascript is an application
programming language; from where I stand, providing access to
your data from your application is no substitute to publicly
documenting your formats and data request protocols. Providing
readily readable HTML is appreciated, but not compulsory.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

Summary: I /am/ old-fashioned. I appreciate finer things in
life, and those finer things IMO tend to be either old-fashioned,
or just plain old.

I’m thinking that does say more about you than what things might be objectively “finer”. I often find old things to be garbage by some measures, just like I often find new things to be garbage by other
measures. Sturgeon’s Law has been widely followed all throughout
history.

it's suddenly morally reprehensible
to enjoy "old-school," Lynx-compatible websites, such as
http://malvaceae.info/ and Wikipedia?

Straw man argument. I don’t see anyone gatekeeping what you enjoy. All I’ve said is that the modern web does *more* than what Lynx sounds like
it supports, so don’t expect everyone else in the world to cater to your “finer” software experience.

And, like I said before, I’m more “old fashioned” in that I would prefer my tabular data in a simple CSV file, not jammed up with HTML formatting. Wikipedia is a site full of lists (and other data) that are made less
usable because of that.

Now, someone may object that "no one's ever interested in that
PD junk" and that "contemporary writers never release anything
in electronic form without Adobe DRM."

Why fabricate these straw man arguments? Nobody has objected to what
*you* like. The contention has entirely been the “stop liking what I
don’t like” meme. I *love* that archive.org has things from 100 years
ago like Buster Keaton. I do eventually hope to find some time to get
to listen to their early radio archives, too. I don’t give a good god
damn if some social media influencer thinks it’s stupid to spend my time
on that instead of trying this new face cream that’s on sale for a
limited time, but click like and subscribe before you go buy it!

Still, I know of no Lynx-compatible website where you can
(legally) /obtain/ any of them: paywalls are typically JS-based.

Fortunately, like web technology, what is legal is changing all the time!
It appears to be legal these days to download anything you want so long
as you’re using it to train an “AI”. Now, personally, I think Lynx is even *worse* as an AI than as a modern web browser, but that choice is
yours to make. :-)

There're several well-known websites hosting amateur writing, but
all of them seem to have been captch'd recently, presumably due
to bot abuse, and are no longer readable with Lynx (or archived.)

But, again, did they need to be web sites in the first place? Writing
and publishing existed before the Internet. Usenet is here, too, if
they can stomach being “old fashioned”.

It's interactive fiction, though, so even being relatively
recent, and a competition winner at that, it's still pretty
old-fashioned.

Genre’s are only “fashioned” when they are new. New styles of music don’t destroy old ones. 3D shooters don’t obsolete good interactive fiction games. New web technologies don’t change your HTML browser. All
for the better, or for the worse.

Doubting that anyone still writes DOS software?

Nope. I simply don’t care, because I don’t use DOS myself. If I *did* have that need for some strange reason, I’d *would* care. I just care
about getting things done in the most reasonable way possible. If that
somehow means using Forth or COBOL or some other “old fashioned” technology, I use it. But I don’t fetishize it as being anything “finer”, especially when it isn’t.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

2025-08-15, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

Summary: I /am/ old-fashioned. I appreciate finer things in life,
and those finer things IMO tend to be either old-fashioned, or just
plain old.

I'm thinking that does say more about you than what things might be objectively "finer".

But of course! I even carefully worded my claim to begin with
an explicit "I am."

it's suddenly morally reprehensible to enjoy "old-school," Lynx-
compatible websites, such as http://malvaceae.info/ and Wikipedia?

Straw man argument. I don't see anyone gatekeeping what you enjoy.

In news:[email protected] upthread, Paul Rubin
compared refraining from using modern websites (and modern
browsers) to refraining from using indoor plumbing:

Yes, it's possible to live your life without using those sites, just as it's possible to live without indoor plumbing or refrigeration at home.

Back in 2021, one of the irc://mbrserver.com/ participants has
argued that, basically, "everyone uses Spotify."

Elsewhere on the web (I don't have a pointer at hand; it's been
a while), there was a comment along the lines of "it's stupid
to use a text browser in 2015."

The sentiment /does/ exist.

All I've said is that the modern web does *more* than what Lynx
sounds like it supports,

I wasn't addressing you alone; in fact, my target audience (as
par the course) was at least this group at large.

Also, there's this one BBS I frequent; it has a dedicated, if
barely active, board for sharing pointers to websites that one
can read with Lynx. The idea was to post a list of such
pointers here on Usenet, then reference it in a BBS post.

I ended up choosing a different board for that latter post,
though, and won't doublepost now, but still.

so don't expect everyone else in the world to cater to your "finer"
software experience.

The thing is: it's not mine /alone./ It's a minority lifestyle,
granted, but 1. humanity consists of minorities; 2. about the
only thing that ever seems to "work" for minorities is: every
once in a while, politely remind the others that 'we exist.'

Which is one of the things I was aiming for with this thread.

And, like I said before, I'm more "old fashioned" in that I would
prefer my tabular data in a simple CSV file, not jammed up with HTML formatting. Wikipedia is a site full of lists (and other data) that
are made less usable because of that.

Can relate, though can't help but point out that: 1. there's
also Wikidata, which is used as the source for /some/ of
Wikipedia's HTML; 2. the ?action=raw markup isn't /that/
different from what you'd get exporting a formatted spreadsheet
into a .csv; 3. the data for these lists is ought to come from
/reliable sources,/ so it might make sense to use those sources
directly, rather than relying on Wikipedia's copy.

(FWIW, when I need to get data from Wikimedia wikis, I decide
between ?action=raw and HTML on a case-by-case basis.)

Now, someone may object that "no one's ever interested in that PD
junk" and that "contemporary writers never release anything in
electronic form without Adobe DRM."

Why fabricate these straw man arguments?

Why violate http://en.wikipedia.org/wiki/Principle_of_charity ?

It would've been a strawman argument were it made to advance a
point. In this case, the purpose is exposition.

Nobody has objected to what *you* like. The contention has entirely
been the "stop liking what I don't like" meme.

Curiously, do you mean a particular direction here? Or both?

Still, I know of no Lynx-compatible website where you can
(legally) /obtain/ any of them: paywalls are typically JS-based.

Fortunately, like web technology, what is legal is changing all the
time! It appears to be legal these days to download anything you
want so long as you're using it to train an "AI". Now, personally,
I think Lynx is even *worse* as an AI than as a modern web browser,
but that choice is yours to make. :-)

When it comes to all things legal, at the end of the day, the
choice is /court's/ to make.

There're several well-known websites hosting amateur writing, but
all of them seem to have been captch'd recently, presumably due
to bot abuse, and are no longer readable with Lynx (or archived.)

But, again, did they need to be web sites in the first place?

And why not? I, for one thing, can't readily think of a better
alternative myself.

Writing and publishing existed before the Internet.

Sure. But I'd argue that this day and age, distributing your
writing worldwide from an HTTP/1 server /would/ be cheaper
than, say, publishing a fanzine.

Usenet is here, too, if they can stomach being "old fashioned".

Usenet is oriented towards discussions, not publishing.

For instance, when you publish your work, you expect that it'd
get an identifier that it, and your future revisions of it, can
be referenced with. Unlike HTTP, NNTP doesn't provide such an
identifier. You have Message-Id:, and you can Supersede: (or
Control: cancel) your previous revision, but so far as I can
tell, there's no support for making new revision "take place"
of a prior one.

The practice I'd advocate for is cross-posting (an excerpt of)
your HTTP-published work to Usenet, then using the latter in
place of your webpage's "Comments" section.

I just care about getting things done in the most reasonable way
possible. If that somehow means using Forth or COBOL or some other
"old fashioned" technology, I use it. But I don't fetishize it as
being anything "finer", especially when it isn't.

The exchange from news:107dsmb$2u9o3$[email protected] upthread
(below) /can/ be read as if you were erring in the opposite
direction: instantly dismissing a technology (Lynx, Usenet)
/because/ it is old and no longer popular.

Instead of complaining that you can't get modern sites to work on
some old HTML browser, [...]

Then again, my Lynx is 2.9.0dev.12 from January 2023. Is it
already considered "old"?

In every way that matters, yes. Just like a new Usenet client
could be released tomorrow and instantly be outdated, because the underlying technology is not fit for purpose. Which is to say
that, while usable for *a* purpose (simple text discussions),
Usenet is no longer anybody's idea of a *modern* messaging platform.

In particular, you didn't specify to /whom/ it so matters, so
it can be read as "In every way that matters" to /you,/ as well
as, possibly, everyone (worthy of mention) else.

Conversely, if technology being old is irrelevant to you (even if
"so long as it /does/ get the job done"), why bring it up at all?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

On 2025-08-11, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

I'm contending that the problem is not that there are different
cultures, but that there are inevitable culture clashes.

Whatever little I can do about culture clashes ("explain your
point, try to understand the other party, do not escalate"),
I believe I do. Other than that, I find it unproductive to be
concerned with things I can do nothing about.

Yet that’s exactly what much of your position seems to be about. Just because the Internet has connected the world doesn’t mean that everyone
has the resources to cater to 8 billion people making 8 billion
different choices. The cost of choosing a “non-mainstream” browser is
on the people who make that choice.

My only "problem" with "modern" web is my indifference to it.
Someone starts talking about Facebook this, or Youtube that,
and, frankly, I'm at a loss how to feel and what to suggest.
I tend to respond along the lines of "if it hurts you to use it,
then don't?" but I can't say that's received all that well.

But *none* of that has to do with the technologies used and *everything*
to do with the cultures those sites promote. While it could be argued
that “doom scrolling” dynamic content is a large contributing factor,
that doesn’t make those modern web features *themselves* a problem.
Social media sites didn’t invent manufactured outrage, trolling, or any
other toxic behavior.

More importantly, am /I/ part of your target audience?

I think that's a false premise. Why should I be trying to limit my audience to some "target" subset? If I have tabular data, for example,
I want as many people as possible to use it however they wish.

I apologize if the term is misleading, but it's the best I was
able to come up with. My point is, intentionally or not, you
/do/ limit your audience. Can someone who can only read
Armenian make use of your websites? How about a five year old?

The onus is not on me to do so. It is on *them* to provide the resources needed to make use of what I offer. Stop trying to burden me with the
cost of 8 billion different choices.

(Can /you/ read http://sw.wikipedia.org/ ? If so, please kindly
share pointers to your reference materials, as I'd like to be
part of their target audience, too.)

Pointer: learn the language. Pointer: use language translation software. Pointer: *pay* someone who has done either to do it for you. Pointer:
don’t be an asshole by saying/implying that *they* are doing something
that wrongly limits you.

The way I understand it, "target audience" is about making sure
certain people /will be able/ to make use of your resources.
(Rather than making sure some won't be.)

Then, again, the argument should be *against* HTML and other XML-based
formats. My file-based resources are mainly text, sometimes spiced up
with a little Markdown. I have table data that might be in a CSV, or
in a database like SQLite. And, of course, *all* the big sites are
pulling from database server to generate pages. How you want to use
those resources should be for *you* to decide, not for *me* to figure
out how to convert it to HTML to *my* liking.

Which is perhaps why you seem to be opposed to Lynx: including
Lynx users in your target audience means extra work, no?

Not like you imply. I’m *not* opposed to Lynx itself, but the *culture*
that you’re promoting by trying to champion it. It’s a fine HTML browser for text-centric pages. Most of what I do is *not* native HTML,
*especially* the text-centric content.

How quickly the "different people" sentiment evaporates. :-)

Does it, though? Wouldn't the notion of "different people"
include those who can't walk, and those who can't talk? those
who only read Danish, and those who only read Spanish? those
who read web with Chromium, /and/ those who read web with Lynx?

Not if their *culture* is to complain that *I* must take on the entire
cost of their 8 billion differences. Reasonable accommodations must be *reasonable*.

If I'm /not/ part of your target audience, why bother?

If you can conclude you're not part of someone's "target audience",
why complain?

Exactly! When I have an issue with a website, I try to
complain to its operator directly.

Whoosh!

I'm quite happy to serve up a YAML file from my server that contains
all the data you need about a product, but there isn't a web browser
I know of that'll render it to *your* liking, so I'm forced to also
render it in HTML to *my* liking.

How would you know the capabilities of the software I use?

That’s my point! That’s the inherent flaw with “web browsers” both old and new. Why the *hell* does the server dictate the page HTML that *you*
have to display? Why am I giving you the same <banner> and <nav> and
<footer> blocks over and over and over? It needs an overhaul all the way
back to it’s foundation.

So, I suppose what I'm trying to say is: my interest in your
data does /not/ necessarily imply interest in running your
application(s) on my computer(s). Javascript is an application
programming language; from where I stand, providing access to
your data from your application is no substitute to publicly
documenting your formats and data request protocols. Providing
readily readable HTML is appreciated, but not compulsory.

All web browsers think otherwise. Even the ones that rely heavily on client-side scripting do DOM manipulation in the end to give you an HTML
page. Don’t blame web sites when JavaScript is *the* client-side way to
do partial content updates. You’re welcome to propose a different
standard to do that which you approve of, and then patch Lynx to support
it.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

Is FTP a "modern web standard"? Unless I be mistaken, it was
supported by Chromium - until it wasn't. Who decided that FTP
is no longer "modern"? Google? Certainly not W3C.

FTP pre-dates the web. It’s deprecated for general use for the same
reasons that `telnet` is: security issues. Modern equivalents would be
SFTP and SSH. Neither have anything to do with the web, even if URLs
with those schemes may appear on web pages and clicking on them may work
(in some fashion).

Not to mention it'd break my mental model of Lynx, which,
HTML-wise, is merely a tool to convert HTML to text:

Then you’re supporting my points about HTML itself being a big part of
the problem. If text is where everyone is starting from *and* going to,
stop insisting on crapping it up with HTML in the middle! Champion other
open formats, most of which existed well before web browsers came onto
the scene and tried to shoehorn everything into some variant of XML.

I seriously doubt that such an "accessible web" would be readily
readable with Lynx.

Nobody’s fault but Lynx. I still don’t understand why you expect the
world to revolve around the limited development resources that go into non-mainstream browsers.

Not to say that braille displays are cheap, mind you.

Screen readers come for pretty much free on every mobile device. Again,
the onus is on Lynx to provide at least as good a text experience as a
blind user can get today. Nobody wants to hear you complain that you
choose to use Lynx and don’t like it.

Instead of complaining that you can't get modern sites to work on
some old HTML browser, maybe question whether or not it was wise
to have tried jamming everything into HTML in the first place.

First, why these two options have to be mutually exclusive?

Ask the people who make web browsers.

This is Usenet. People who make web browsers are of course
welcome to join this discussion and tell us why a person who
complains can't question and (or) vice versa.

Because it’s not rational; there is no “just asking questions”
necessary. The reality seems to be that the web has grown beyond what
Lynx developers are willing/able to support. Your complaints are not
tangible support to either improve Lynx or eliminate any web standards. Whatever path you choose should cost nobody but you.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
John Ames <[email protected]> wrote:

I'm not dictating what anyone else should use, or even what they should specifically work to support.

But you are. You just *think* you’re being sly about it by pretending
that it’s the oh-so-evil people running web sites that are making it hard
for the oh-so-good people making their 8 billion different choices.
Sorry, no, you’re just trying to reframe “the other” as the dictator so *you* can be the dictator of what a *true* web site should be.

As far as citing authority, I'll hand the mic over
to Tim Berners-Lee, a.k.a. The Guy Who Invented The Web:

"Anyone who slaps a 'this page is best viewed with Browser X' label on
a Web page appears to be yearning for the bad old days, before the Web,
when you had very little chance of reading a document written on
another computer, another word processor, or another network."

I sure hope that was something he naively said back in the 1990s, because it’s disingenuous bordering on signs of senility if it is more recent. I
ask you to think about that quote critically. It’s basically saying he
got everything *perfect* on Version 1.0 (technically, HTML 2.0/HTTP 1.0).
That nothing was interoperable before he came along with the one, true “document”.

Yeah, propriety data formats do suck, but there were plenty of open
formats that existed before the web tried to make them all vanish in a
puff of HTML. I’ll still take a common CSV file over trying to tease
some data out of a page with an embedded <table>, and countless other not-invented-here choices that got us to where the web is today.

You can not like change all you like, but rapid change is pretty much
the hallmark of our technological world. Like I said, the real
complaint to level against that change is not that the web isn’t just
HTML any longer, but that shoving everything into HTML was ever a good
idea in the first place. I take non-mainstream browsers seriously
when they approach the modern web from *that* perspective.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

2025-08-15, Doc O'Leary wrote:

Straw man argument. I don't see anyone gatekeeping what you enjoy.

In news:[email protected] upthread, Paul Rubin
compared refraining from using modern websites (and modern
browsers) to refraining from using indoor plumbing:

Yes, it's possible to live your life without using those sites, just as it's possible to live without indoor plumbing or refrigeration at home.

That’s not gatekeeping, that’s just an apt metaphor. Again, *you* can
do whatever you want. Just accept any burdens that come with that
choice. Don’t for a second think the world wants to hear how it should change because you chose poorly.

Back in 2021, one of the irc://mbrserver.com/ participants has
argued that, basically, "everyone uses Spotify."

Simply an incorrect assertion without any other context. Were you being
banned from IRC for not using Spotify? Are there musicians that are
*only* on Spotify? Those would be gatekeeping.

Elsewhere on the web (I don't have a pointer at hand; it's been
a while), there was a comment along the lines of "it's stupid
to use a text browser in 2015."

Again, simply expressing opinions, right or wrong, is not controlling
behavior. I mean, it’s a decade later and you’re still using a text browser, after all. Worst. Gatekeeping. Ever!

so don't expect everyone else in the world to cater to your "finer" software experience.

The thing is: it's not mine /alone./ It's a minority lifestyle,
granted, but 1. humanity consists of minorities; 2. about the
only thing that ever seems to "work" for minorities is: every
once in a while, politely remind the others that 'we exist.'

You have those going in the wrong direction. The reason accessibility
got traction is that the proponents smartly made it about *usability*.
They were inclusive rather than othering themselves as a minority.
There are people that like to watch things with closed captioning on,
because there is a compelling case for doing so in many circumstances,
even for people with perfectly good hearing. That’s why I encourage you
to focus on the gap between Lynx and a screen reader.

And, like I said before, I'm more "old fashioned" in that I would
prefer my tabular data in a simple CSV file, not jammed up with HTML formatting. Wikipedia is a site full of lists (and other data) that
are made less usable because of that.

Can relate, though can't help but point out that: 1. there's
also Wikidata, which is used as the source for /some/ of
Wikipedia's HTML; 2. the ?action=raw markup isn't /that/
different from what you'd get exporting a formatted spreadsheet
into a .csv; 3. the data for these lists is ought to come from
/reliable sources,/ so it might make sense to use those sources
directly, rather than relying on Wikipedia's copy.

Rubbish. I go to something like:

<https://www.wikidata.org/wiki/Q6629082>

and I see no easy way to get anything close to a CSV. The raw editing
markup is also “visual”; more akin to Markdown formatting than any proper data representation. And the big problem with “use those sources
directly” is that they are usually in a worse, less standardized HTML
layout than Wikipedia offers! In any case, there is absolutely *nothing* Wikipedia has to do because I would choose to do things differently; zero complaints coming their way from me about it.

Why fabricate these straw man arguments?

Why violate http://en.wikipedia.org/wiki/Principle_of_charity ?

Because often times people see generosity as weakness and abuse it.
When logical fallacies are repeated, it points to a fundamental flaw in
the underlying argument, not any mere “oopsie!”

Nobody has objected to what *you* like. The contention has entirely
been the "stop liking what I don't like" meme.

Curiously, do you mean a particular direction here? Or both?

I certainly am not going to exclude myself from the potential of being
on the wrong side of an argument. But, again, I’m not the one upset
that the modern web isn’t as locked in to server-generated HTML as it
once was, going so far as to complain to sites that are completely
standards compliant! It’s not a problem for me if *you* like HTML. But
it doesn’t sound like you actually do, given your usage of Lynx.

Still, I know of no Lynx-compatible website where you can
(legally) /obtain/ any of them: paywalls are typically JS-based.

Fortunately, like web technology, what is legal is changing all the
time! It appears to be legal these days to download anything you
want so long as you're using it to train an "AI". Now, personally,
I think Lynx is even *worse* as an AI than as a modern web browser,
but that choice is yours to make. :-)

When it comes to all things legal, at the end of the day, the
choice is /court's/ to make.

And they did. Fire up your Lynx AI and go crazy.

Usenet is here, too, if they can stomach being "old fashioned".

Usenet is oriented towards discussions, not publishing.

It’s a platform to be used however people want to use it. It certainly *used* to be a place where all kinds of one-way publishing took place. Especially the binary groups!

For instance, when you publish your work, you expect that it'd
get an identifier that it, and your future revisions of it, can
be referenced with. Unlike HTTP, NNTP doesn't provide such an
identifier.

I don’t know *what* you’re talking about. You mean a URI? Nothing is keeping anyone from using them on Usenet any more than anybody is
preventing them from being abused on the web (dead links, redirects, cookie-based personalization, etc.). Personally, I’m more in favor of intrinsic identifiers, like an SHA-256 hash of the content.

You have Message-Id:, and you can Supersede: (or
Control: cancel) your previous revision, but so far as I can
tell, there's no support for making new revision "take place"
of a prior one.

Neither can a publisher break into my house and swap out copies of a
book I have. You’re conflating “publishing” with a *lot* more than it inherently need be.

I just care about getting things done in the most reasonable way
possible. If that somehow means using Forth or COBOL or some other
"old fashioned" technology, I use it. But I don't fetishize it as
being anything "finer", especially when it isn't.

The exchange from news:107dsmb$2u9o3$[email protected] upthread
(below) /can/ be read as if you were erring in the opposite
direction: instantly dismissing a technology (Lynx, Usenet)
/because/ it is old and no longer popular.

I’m not sure how; I wouldn’t even be here (using a Usenet client that
I wrote from scratch, no less!) if that were the case. I clearly said
anything may still be suitable for *a* purpose. You want to use Lynx
to turn HTML into text? Seems reasonable. You bother someone running
a site that works just fine for web browsers following standards
published 10 years ago? Not so reasonable.

This is not complicated. Your charitable needs are exceeding my
generosity.

Conversely, if technology being old is irrelevant to you (even if
"so long as it /does/ get the job done"), why bring it up at all?

Because I am not the world. Your position is not in opposition to me.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sat, 16 Aug 2025 21:42:15 -0000 (UTC), Doc O'Leary , wrote:

That’s the inherent flaw with “web browsers” both old and new. Why
the *hell* does the server dictate the page HTML that *you* have to
display? Why am I giving you the same <banner> and <nav> and
<footer> blocks over and over and over? It needs an overhaul all the
way back to it’s foundation.

That has already happened. HTML just represents the document structure nowadays. All styling and layout is controlled via CSS. This way, it is possible to repurpose the same page content for many different uses.

And yes, there are means available for users to impose their own styles on someone else’s web page, too.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

On Sat, 16 Aug 2025 23:36:57 -0000 (UTC), Doc O'Leary , wrote:

For your reference, records indicate that John Ames
<[email protected]> wrote:

As far as citing authority, I'll hand the mic over to Tim Berners-Lee,
a.k.a. The Guy Who Invented The Web:

"Anyone who slaps a 'this page is best viewed with Browser X' label on
a Web page appears to be yearning for the bad old days, before the Web,
when you had very little chance of reading a document written on
another computer, another word processor, or another network."

I sure hope that was something he naively said back in the 1990s,
because it’s disingenuous bordering on signs of senility if it is more recent. I ask you to think about that quote critically. It’s basically saying he got everything *perfect* on Version 1.0 (technically, HTML
2.0/HTTP 1.0).

He was saying no such thing.

That nothing was interoperable before he came along with
the one, true “document”.

Considering he invented the WWW, yes it is fair to say nothing was “interoperable” because nothing *existed* along these lines before him.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-17, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

In news:[email protected] upthread, Paul Rubin compared
refraining from using modern websites (and modern browsers) to
refraining from using indoor plumbing:

Yes, it's possible to live your life without using those sites, just as it's possible to live without indoor plumbing or refrigeration at home.

That's not gatekeeping, that's just an apt metaphor.

The day I will believe that words are powerless to convince a
person, or at least influence their behavior, would be the day
I stop using Usenet. Perhaps stop talking altogether, even.

Again, *you* can do whatever you want.

Who-a, I can?! Surely there's a catch, there's always a catch...

Just accept any burdens that come with that choice. Don't for a
second think the world wants to hear how it should change because
you chose poorly.

Ah. Thought so.

Either way, my disagreement regarding the aptness of the
metaphor above prompted me to start this subthread with
news:[email protected] . And despite the
explanations given, I still disagree.

And of course I accept the consequences of my choices.
Do you accept those of yours?

Usenet is oriented towards discussions, not publishing.

It's a platform to be used however people want to use it.

It's a technology that's better suited for some uses than
others. A spoon works better as a spoon than as a spade,
and vice versa.

For instance, when you publish your work, you expect that it'd get
an identifier that it, and your future revisions of it, can be
referenced with. Unlike HTTP, NNTP doesn't provide such an identifier.

I don't know *what* you're talking about. You mean a URI?

Or ISBN, - anything that works. Any form of reference you can
give to someone else, when you're asked for your work, or when
you're providing a pointer to it on your own initiative.

In the HTTP/1 header, http://furry.de/jumpy/NotQuiteSilence.txt
has July 2009 in Last-Modified: , which I presume is when the
author has published the file. He can share that pointer, and
he can then edit the text on the server, and the pointer will
/still/ point to the latest revision of the work. That's
usually what authors want and need.

When I gave my students pointers to course materials, I did it
with http: URIs. I couldn't have used news: even if they were
supported by the browser(s) they use (say, if they all were
using Lynx instead of Chromium as their primary browser), as
such URIs refer to something that, standardly, cannot be edited.

Can you suggest that kind of a pointer for a work published on
Usenet? I deem myself fairly knowledgeable of the applicable
RFCs, but can't think of nothing suitable. news: URIs would
work fairly well for specific revisions or specific comments,
but hardly for the cases like this.

There certainly /is/ use case for private copies (like with
Wget) and public archives (like Wayback Machine.) It's somewhat
old-fashioned, though: modern web standards don't standardize
archival of websites in any way, and in fact provide a plenty
of facilities to make completely standards compliant websites
completely unarchivable.

Personally, I’m more in favor of intrinsic identifiers, like an
SHA-256 hash of the content.

You mean, like this?

http://users.am-1.org/~ivan/misc-2022/sfn.oIXvO0FgwKI2peJnKbe__sn-K4uXM97sK2lAyDNgcMs.png

The exchange from news:107dsmb$2u9o3$[email protected] upthread
(below) /can/ be read as if you were erring in the opposite
direction: instantly dismissing a technology (Lynx, Usenet)
/because/ it is old and no longer popular.

I'm not sure how; I wouldn't even be here (using a Usenet client
that I wrote from scratch, no less!) if that were the case.

And that's what made it all the weirder to read. Not unlike
finding a person commenting "command-lines are great, but it's
not the 1970s any more" in news:vpr689$3c4q9$[email protected] -
as if Usenet wasn't about the same age.

Conversely, if technology being old is irrelevant to you (even if
"so long as it /does/ get the job done"), why bring it up at all?

Because I am not the world. Your position is not in opposition to me.

You seem to be equating the world with your understanding of it.
It is my position in this thread that your understanding is not
without certain flaws.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-17, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

It's a minority lifestyle, granted, but 1. humanity consists of
minorities; 2. about the only thing that ever seems to "work" for
minorities is: every once in a while, politely remind the others
that 'we exist.'

You have those going in the wrong direction. The reason accessibility
got traction is that the proponents smartly made it about *usability*.
They were inclusive rather than othering themselves as a minority.

I'd think that "othering themselves as a minority" is more
applicable to Gopher and Gemini enthusiasts, rather than Lynx'.

If anything, unless its author actively works against it,
a "webpage" readable with Lynx /will/ be readable with Chromium
just as well. Not necessarily the other way around.

It's how I write my HTMLs, actually: my target audience is
mostly Chromium users, but I still feel safe testing with Lynx.

There are people that like to watch things with closed captioning
on, because there is a compelling case for doing so in many
circumstances, even for people with perfectly good hearing.

Thanks to your polite reminder, I am now aware that there's
also a minority who like to watch things with closed captioning
on, even though they have perfectly good hearing.

And of course it's only natural to search for a common ground
and find solutions that work for a greater fraction of people.
Kinda something we're doing in this very thread, in a way.

That's why I encourage you to focus on the gap between Lynx and
a screen reader.

I have next to zero experience with screen readers, and as such,
would be ill-equipped for such work, regardless of its merit.

Can relate, though can't help but point out that: 1. there's also
Wikidata, which is used as the source for /some/ of Wikipedia's HTML;

I go to something like: <https://www.wikidata.org/wiki/Q6629082>
and I see no easy way to get anything close to a CSV.

Whatever little experience with Wikidata I have is severly out
of date. My complaints on the WMF policy changes back around
the time Wikidata appeared were dismissed, so I largely ceased
my participation in most of their projects, including Wikidata.

(I still contribute to a select few WMF wikis on occasion.
Say, http://commons.wikimedia.org/w/?curid=76613333 .)

There's reportedly a SPARQL endpoint one can use to request
data in "data" formats, but I don't recall ever using it.
(I've used Dbpedia's back in the day, but it's been too long
and I can't seem to recall any useful examples of that, either.)

Conceptually, though: Wikipedia is chartered as an encyclopedia,
/not/ a database, hence you're just not being part of their
target audience in this specific case.

That said, there /is/ an ongoing trend of moving the "data"
portion of Wikipedia to Wikidata; simply because often enough,
people at, say, http://de.wikipedia.org/ want the same data in
their articles as, say, http://pl.wikipedia.org/ . When such
data is in a database, its target audience can be wider (and
the maintenance costs lower) than when it's on a wikipage.

The raw editing markup is also "visual"; more akin to Markdown
formatting than any proper data representation.

Yep. I've meant the table markup specifically, though, which
is like (to quote a Wikipedia article from my archives):

| Vary || Tells downstream proxies how to match future request headers
to decide whether the cached response can be used rather than requesting
a fresh one from the origin server. ||
* Example 1: <code>Vary: *</code>
* Example 2: <code>Vary: Accept-Language</code>
|| Permanent

No idea if, say, LibreOffice spreadsheet can have a bullet list
in a cell, but other than that, I can easily imagine the above
exported from there as CSV:

"Vary","Tells downstream [...]","Example 1: [...]","Permanent"

Similar enough, from where I stand.

In any case, there is absolutely *nothing* Wikipedia has to do
because I would choose to do things differently; zero complaints
coming their way from me about it.

I see nothing wrong in complaining per se. Self-regulating
systems, such as society, /require/ feedback loops, such as
complaints.

Say, back in the day, I've complained that "$ tree -- dir "
didn't work, and "$ tree --help 2>&1 | less " required "2>&1"
against my expectations. Guess what? Rather than telling me
all about "the resources needed to make use of what [they] offer",
or "the cost of 8 billion different choices", the developer
just changed the code.

In this case, though, complaining might've been unreasonable,
simply because so far as I can tell, Wikimedians understand
the benefits of the approach you'd be complaining in favor of
adopting, and already work towards it.

You're of course welcome to join their effort as an unpaid
volunteer, particularly if you're interested in the results
becoming usable to you sooner. Though not being a participant to
either Wikipedia(s) or Wikidata, I have no idea whether what you
want is within reach of the technology they employ right now,
or even if your use case will be something within their notion
of "target audience" even with the above in mind.

Curiously, do you mean a particular direction here? Or both?

I certainly am not going to exclude myself from the potential of
being on the wrong side of an argument. But, again, I'm not the one
upset that the modern web isn't as locked in to server-generated
HTML as it once was,

Neither am I. I /am/ upset, but for a different reason.

going so far as to complain to sites that are completely standards compliant!

Not only I complain to sites; just two days ago, I've
complained about one of these very standards they're compliant
to; see news:[email protected] .

It's not a problem for me if *you* like HTML. But it doesn't sound
like you actually do, given your usage of Lynx.

Whether I do or do not like HTML is largely irrelevant to
the point I'm trying to make, which is: I do /not/ like being
locked to any single application, be it .js, .exe, .tex, .sh,
Microsoft Word, Widevine, Adobe DRM, - absolutely whatever.

And like I've mentioned before, Lynx is not the only program
I use. Think of Wget, for instance: not a "browser," yet a
tool for acquiring copies of webpages, such as for mirroring
or archival. Does /it/ support "partial content updates"?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

On Mon, 18 Aug 2025 10:45:55 -0700, John Ames wrote:

I will clarify for the sake of being clear ...

Gee, I wonder what other reason there might be for wanting to “clarify”, given that it means “make clear” ...

I'm talking about basic, *basic* stuff here - things like not depending
on Javascript to load and display static page content ...

Funny, I did an example of that just the other day. A friend had put
together a formatted table of data in a web page that was close to a
megabyte in size. I knocked it down to a small fraction of that -- a bit
over 100K -- by using JavaScript to generate the table layout from the raw
data (which I included in the page).

I also added functions to sort the display of the data on selected
columns. That only added about 3K to the page size.

... not hiding all your site navigation behind a hamburger button and
CSS pop-over ...

Bear in mind the point of CSS is precisely to separate document structure
from layout. If the semantics of the page can be gleaned from an
examination of the HTML structure without regard to the styling, then what
are you complaining about?

It's not dictatorial to expect that of Web designers ...

Feel free to show us examples of your way of designing the Web.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

For your reference, records indicate that
Ivan Shmakov <[email protected]d> wrote:

On 2025-08-17, Doc O'Leary wrote:

For instance, when you publish your work, you expect that it'd get
an identifier that it, and your future revisions of it, can be
referenced with. Unlike HTTP, NNTP doesn't provide such an identifier.

I don't know *what* you're talking about. You mean a URI?

Or ISBN, - anything that works.

Ten seconds on the ISBN Wikipedia page is all it takes for *anyone* to
see that “different ISBN is assigned to each separate edition and
variation of a publication”. You clearly are not interested in
presenting a sincere viewpoint that comports with reality. I’m done
with you.

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
John Ames <[email protected]> wrote:

I will clarify for the sake of being clear, though I doubt it'll keep
you from firing back with another "no u" - I'm *not* demanding that
anybody specifically work to support Browser XYZ. What I *do* expect out
of Web designers is some bare minimum of thought put into designing
with an eye towards graceful degradation, which (while never perfect)
has been possible since the beginning and remains so today.

NO U! :-)

Seriously, you’re laying blame on the *wrong* people. Web pages work/
look like they do because someone in management (and/or marketing)
*told* the designer to make it that way. It’s fundamentally the “culture” argument I’ve been making.

I'm talking about basic, *basic* stuff here - things like not depending
on Javascript to load and display static page content, not hiding all
your site navigation behind a hamburger button and CSS pop-over, and
for the love of all that is good and holy *not* redirecting unfamiliar
user agents to a screw-you-for-not-using-an-Approved-Browser page.

All culture. At least to a point; there is the technical angle that
I’ve brought up: there is no “graceful” way to degrade what JavaScript does. There is no alternative in the standard to replace just part
of a page, no support for a “reference implementation” of CSS that
would universally give you site navigation how *you* want it, no
ethical rules of publishing that intrinsically require a request to
get a uniform response. Ironically, though, someone *could* build a
browser that tried to “sandbox” the whole web through a user-centric interface with support for things like that (which is *kinda* what
screen readers aim to do), and *that* would get complaints for being
“best viewed in” dictatorial!

It's not dictatorial to expect that of Web designers; it is (or ought
to be) a basic qualification of the profession, in the same way that,
if you build a chair that falls apart the moment someone sits a little
too far to the left in it or clunks the occupant with a clown hammer
because they didn't do a little dance first, you're objectively a bad furniture designer.

The modern web is not designed for you, but the dictator that pays the
team to put the site together. If *they* use Lynx, yeah, the site
would work well in Lynx. If they give a damn about accessibility,
that’s what the site will be. Most *want* you smacked by the clown
hammer, though . . .

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

On Tue, 19 Aug 2025 21:58:38 -0000 (UTC), Doc O'Leary , wrote:

There is no alternative in the standard to replace just part of a
page ...

Sure there is. The DOM lets you do that.

... no support for a “reference implementation” of CSS that would universally give you site navigation how *you* want it ...

Browsers let you define custom overrides for site CSS, don’t they?

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 2025-08-19, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:
On 2025-08-17, Doc O'Leary wrote:
For your reference, records indicate that Ivan Shmakov wrote:

For instance, when you publish your work, you expect that it'd get
an identifier that it, and your future revisions of it, can be
referenced with. Unlike HTTP, NNTP doesn't provide such an identifier.

I don't know *what* you're talking about. You mean a URI?

Or ISBN, - anything that works.

Ten seconds on the ISBN Wikipedia page is all it takes for *anyone*
to see that "different ISBN is assigned to each separate edition and variation of a publication".

So it won't work, indeed. (Kinda reminds me that one case
years ago when I've made a remark to the effect that it makes
sense to run RTP over SCTP. Another participant also found it
out of touch with reality.)

That said, the point of inadequacy of Usenet to the task at
hand still stands.

You clearly are not interested in presenting a sincere viewpoint
that comports with reality. I'm done with you.

Thanks for participating in this thread anyway; it went better than
I've expected (i. e., news:[email protected] ),
all things considered.

I'm going to post a few followups with a handful of leftover
points for the benefit of anyone who might be reading this
thread (now or later.) JFTR, the primary point I've been
trying to defend is along the lines of: using "modern web
standards" (i. e.: Javascript) to 'gateway' access to the data
you publish /is evil./

One of the secondary points is that using a browser that
has no Javascript support, or allows it to be disabled, is
advisable for finding out if such gatewaying takes place,
and for complaining if it is the case. (Unless the website
operator is known to be unresponsive to complaints.)

Examples follow.

* http://devuan.org/ : not evil. To quote: "This site is free
of cookies and javascript".

* Wikipedia(s), Wikimedia Commons: not evil; the Javascript
used is often (always?) not essential, and even in the cases
it might be, the APIs the JS uses are well-documented (see,
e. g., http://commons.wikimedia.org/w/api.php ) and are
usable and useful outside of browsers (say, with curl(1).)

* http://earth.nullschool.net/ : JS is essential, but the GFS
data this webapp visualizes is available from elsewhere anyway
(e. g., http://nomads.ncep.noaa.gov/dods/ ; the underlying
GFS GRIB files used to be available as well, though I haven't
used that option in years, so no idea if they still are.)

Using application's Javascript doesn't interfere with one's
ability to use other applications to process that same data,
including applications that have nothing to do with "modern web."

* Youtube : evil. Employs Javascript to interfere with user's
attempts to access data outside of authorized applications.
(Not particularly successful, as work-arounds exist.)

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

John Ames wrote:

These are not hard things - in fact, it usually takes more work to do
the Bad Behavior than to not do it. They don't require designers to
spend hours fiddling with their site design to work around Browser XYZ's esoteric CSS support or tendency to choke on emoji glyphs or whatever;
they just require designers to not do things that they shouldn't be
doing anyway.

Web designers continually change web sites that don't need to
be changed. They continually add "features" that don't help
me, "features" that make it harder for me to use the web
sites.

Web designers don't make these changes because of user demand.
Web designers don't make these changes because the changes
help users.

Web designers make these changes because they help the
web designers.

If the policy of web sites was "If it ain't broke, don't fix it",
if the policy was not to make changes for the sake of change,
then most of these web designers would become unemployed.
Most of these web designers aren't productive; they are
actually destructive. Most of them aren't needed.

However, they try to make themselves seem needed by
continually howling:

"Keep paying us! Keep paying us! Keep paying us to make
changes so that fewer and fewer people can use your web site!
Keep paying us! Keep paying us! Keep paying us so that fewer
and fewer people can use your web site!"

Most web designers are destructive parasites that ought to be
fired. Those that remain ought to be chained down.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

In article <108f8c9$2s3o1$[email protected]>,
B. Pym <[email protected]> wrote:

John Ames wrote:

These are not hard things - in fact, it usually takes more work to do
the Bad Behavior than to not do it. They don't require designers to
spend hours fiddling with their site design to work around Browser XYZ's
esoteric CSS support or tendency to choke on emoji glyphs or whatever;
they just require designers to not do things that they shouldn't be
doing anyway.

Web designers continually change web sites that don't need to
be changed. They continually add "features" that don't help
me, "features" that make it harder for me to use the web
sites.

Web designers don't make these changes because of user demand.
Web designers don't make these changes because the changes
help users.

No. The web was designed to access scientific articles, valuable
peer reviewed and reliable. Now they want you to look every day
to see what has changed.

Web designers make these changes because they help the
web designers.

Web designers are the lowest in the hierarchy. Content creators
are the people who should be in charge.
<SNIP>

Most web designers are destructive parasites that ought to be
fired. Those that remain ought to be chained down.

Agreed.

Groetjes Albert


--
The Chinese government is satisfied with its military superiority over USA.
The next 5 year plan has as primary goal to advance life expectancy
over 80 years, like Western Europe.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

XPost: comp.lang.forth

For your reference, records indicate that
Lawrence =?iso-8859-13?q?D=FFOliveiro?= <[email protected]d> wrote:

On Tue, 19 Aug 2025 21:58:38 -0000 (UTC), Doc O'Leary , wrote:

There is no alternative in the standard to replace just part of a
page ...

Sure there is. The DOM lets you do that.

I’m not sure how “The DOM” is meaningful in the context of a browser without JavaScript. Please reference where the standard mentions the alternative you’re suggesting. If possible, give a URL to a reference implementation that works with Lynx or, if not, then some mainstream
browser with JavaScript disabled. Thanks.

... no support for a “reference implementation” of CSS that would universally give you site navigation how *you* want it ...

Browsers let you define custom overrides for site CSS, don’t they?

Only in theory; I think they’ve been twisted from the very beginning by
the people who want “pixel perfect” pages. I mean, take The CSS Zen
Garden for example. It’s a beautiful example of how a page can be
radically re-rendered with different CSS. But the CSS is *not universal*, because it can’t be applied to any other HTML page to get the same nice layout(s).

The shortcoming is especially obvious when you consider styles that are
tied to the element’s `id`. Different sites will have completely
different or, worse, conflicting `id`s. Nevermind that they will also
have completely different HTML structures, especially if they’re software generated pages. It’s no solution at all if I have to write custom CSS
for *every damn site* I visit!

--
"Also . . . I can kill you with my brain."
River Tam, Trash, Firefly

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)