Use of meta tags

[Robert]

I have some questions about these meta tags:

<meta http-equiv="Content-Type" content="text/html;
charset=iso-8859-1">

What charset should I use? Is this latin-1? I have MacOS 10.2.6.
Where should I have looked for this information.

<meta http-equiv="Content-Script-Type" content="text/javascript">

If I include this meta tag, do I need to specify content type on the
script tag?

Robert

 

[David Dorward]

I have some questions about these meta tags:

<meta http-equiv="Content-Type" content="text/html;
charset=iso-8859-1">

This may be useful for pages designed to be used offline. Real HTTP headers
are preferable for pages designed for the WWW.

What charset should I use?

That depends on which characters you use. UTF-8 is good for most purposes
these days. Of course you need to make sure the file really is encoded in
that format. Claiming that a file is encoded in a format other then that
which is really is is likely to cause more problems then not specifying any
at all.

Is this latin-1?

I believe iso-8859-1 is Latin 1.

I have MacOS 10.2.6.
Where should I have looked for this information.

Somewhere other then comp.lang.javascript. Nothing you have raised so
touches upon being on-topic.

<meta http-equiv="Content-Script-Type" content="text/javascript">

If I include this meta tag, do I need to specify content type on the
script tag?

Yes you do, the above only applies to script placed inside intrinsic event
attributes (e.g. onload and onclick).

 

[Michael Winter]

[snip]

What charset should I use?

It depends on your content. If you're writing in English, ISO-8859-1 will
be fine. You can always specify characters outside that set with character
entities.

Is this [ISO-8859-1] latin-1?
Yes.

I have MacOS 10.2.6.

That doesn't matter.

Where should I have looked for this information.

I don't know of any specific reference. Googling for iso-8859 produces a
lot of results. It would be a good place to start.

<meta http-equiv="Content-Script-Type" content="text/javascript">

If I include this meta tag, do I need to specify content type on the
script tag?

Yes, for two reasons:

1) The default script type, as specified by that META element, applies to
intrinsic events.
2) The type attribute for SCRIPT elements is required.

By the way, it's best to provide this information in the server response
headers. Your server may already transmit the character encoding. It would
also be better to have sent this to alt.html or
comp.infosystems.www.authoring.html where it's on-topic.

Mike

 

[Lars Eighner]

In our last episode,

I have some questions about these meta tags:

<meta http-equiv="Content-Type" content="text/html;
charset=iso-8859-1">

What charset should I use? Is this latin-1? I have MacOS 10.2.6.
Where should I have looked for this information.

First http-equiv is to supply information usually supplied in http
headers by the server. Unless you know your server is wakko, you
know you are doing something very unusual, or you are catering to a
particular browser that is very broken or likely to have some
unreasonable or unusual defaults, you probably should leave
http-equiv alone. Don't include such meta elements unless you know
they are needed and you know what you are trying to accomplish with
them.

This particular one is next to useless. It says the document is
html. But if the browser doesn't already know (or assume) this
already, it won't know what a meta element is. It declares
iso-8859-1 as the character set, which is what most browsers in
English-speaking countries and much of Europe will assume if they are
told nothing.

Yes, it is more or less what is known as "latin 1" in some systems.
If you are using character entities (such as &eacute for accented
characters, and the q element for quotations, you probably won't
notice any problems even if you are not actually using iso-8859-1.

<meta http-equiv="Content-Script-Type" content="text/javascript">

If I include this meta tag, do I need to specify content type on the
script tag?

Yes.

 

[Thomas 'PointedEars' Lahn]

This may be useful for pages designed to be used offline. Real
HTTP headers are preferable for pages designed for the WWW.

*Any* page (better: document) may be used offline, that's why the HTML 4.01
specification recommends to include that element in any HTML document that
requires it according to the characters used therein.


PointedEars

 

[David Dorward]

Thomas 'PointedEars' Lahn wrote:

Why are you resurrecting three week old threads?

*Any* page (better: document) may be used offline,

They may also be transcoded to a different character encoding by a proxy
server or http user agent - which would make the meta declaration wrong.

that's why the HTML
4.01 specification recommends to include that element in any HTML document
that requires it according to the characters used therein.

Where does it recommend this?

 

[Thomas 'PointedEars' Lahn]

Thomas 'PointedEars' Lahn wrote:

Why are you resurrecting three week old threads?

Why don't you understand that NetNews is not an Internet chat?

They may also be transcoded to a different character encoding by a proxy
server or http user agent - which would make the meta declaration wrong.

That's not the concern of the author of the document but of the person who
misconfigured the proxy or user agent. The "meta" element should only be
used as alternative to HTTP headers.

Where does it recommend this?

<http://www.w3.org/TR/html4/charset.html#h-5.2.2>


PointedEars

 

[Randy Webb]

Thomas 'PointedEars' Lahn wrote:

Why are you resurrecting three week old threads?

Three things come to mind:

1) He is too stupid to wipe his ass so he replies to three week old
threads in the hopes that noone will bother to correct his stupid self.

2) He lacks the intelligence to answer in a timely manner and keep up
with the conversation.

3) Both of the above.

--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq
Answer:It destroys the order of the conversation
Question: Why?
Answer: Top-Posting.
Question: Whats the most annoying thing on Usenet?

 

[Jim Ley]

That's not the concern of the author of the document but of the person who
misconfigured the proxy or user agent. The "meta" element should only be
used as alternative to HTTP headers.

It's not a misconfigured proxy, charset transcoding is a legitimate
job of a proxy. You'll notice user agents like IE also do transcoding
when you save the files.

<http://www.w3.org/TR/html4/charset.html#h-5.2.2>

More precisely? It only says CAN in that section that I can see, not
even a SHOULD, I see no recommendation.

Jim.

 

[Thomas 'PointedEars' Lahn]

It's not a misconfigured proxy, charset transcoding is a legitimate
job of a proxy. You'll notice user agents like IE also do transcoding
when you save the files.

And your point is ...?

More precisely? It only says CAN in that section that I can see, not
even a SHOULD,

My "should" was not a "SHOULD" in the IETF sense, more in common sense.

I see no recommendation.

The entire HTML 4.01 Specification is a Recommendation and includes:

| 5.2.2 Specifying the character encoding
| [...]
| The HTTP protocol ([RFC2616], section 3.7.1) mentions ISO-8859-1 as a
| default character encoding when the "charset" parameter is absent from
| the "Content-Type" header field. In practice, this recommendation has
| proved useless because some servers don't allow a "charset" parameter
| to be sent, and others may not be configured to send the parameter.
| Therefore, user agents must not assume any default value for the
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| "charset" parameter.
^^^^^^^^^^^^^^^^^^^^
|
| To address server or configuration limitations, HTML documents may include
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| explicit information about the document's character encoding; the META
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| element can be used to provide user agents with this information.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^


PointedEars

 

[Jim Ley]

And your point is ...?

That including of a META charset for html documents is unsafe, as it's
likely to be incorrect after transcoding (one of the points we've been
discussing in the thread)

My "should" was not a "SHOULD" in the IETF sense, more in common sense.

So you're now applying your common sense to the specification and
reporting it as fact, when it doesn't actually say any such thing.

Jim.

 

[George Hester]

Three things come to mind:

1) He is too stupid to wipe his ass so he replies to three week old
threads in the hopes that noone will bother to correct his stupid self.

2) He lacks the intelligence to answer in a timely manner and keep up
with the conversation.

3) Both of the above.

--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq
Answer:It destroys the order of the conversation
Question: Why?
Answer: Top-Posting.
Question: Whats the most annoying thing on Usenet?

You Randy are an Internet asshole. How old are you? Have you any loved ones? Cause
I can tell you, here, you are loved about as much as a can of droppings from a Dodo bird.

When you respond on point you do have useful information and I appreciate it.
But when it comes to public manners you have a lot to learn. You are always looking for a confrontation.

 

[Thomas 'PointedEars' Lahn]

[...] Thomas 'PointedEars' Lahn [...] wrote:
David Dorward wrote:
They may also be transcoded to a different character encoding by a
proxy server or http user agent - which would make the meta
declaration wrong.
That's not the concern of the author of the document but of the person
who misconfigured the proxy or user agent. The "meta" element should
only be used as alternative to HTTP headers.
It's not a misconfigured proxy, charset transcoding is a legitimate
job of a proxy. You'll notice user agents like IE also do transcoding
when you save the files.

And your point is ...?

That including of a META charset for html documents is unsafe, as it's
likely to be incorrect after transcoding (one of the points we've been
discussing in the thread)

A transcoding proxy or user agent must address issues arising from the
possibility that the character declaration technique described in the
Specification is used, and therefore remove or modify the character
declaration then; if it does not, it is broken.

So you're now applying your common sense to the specification and ^^^^^^^^^^^^^^^^^
reporting it as fact, when it doesn't actually say any such thing.

Don't be ridiculous. Again, among other things, the Specification says:

| Therefore, user agents must not assume any default value for the
| "charset" parameter.
|
| To address server or configuration limitations, HTML documents may include
| explicit information about the document's character encoding; the META
| element can be used to provide user agents with this information.  

Since reasonable authors are eager to provide useful content, they
would address the problem as recommended instead of ignoring it.


PointedEars

 

[Randy Webb]

George Hester wrote:

You Randy are an Internet asshole.

Yippeee!!!!!!!!! Let's have a party. I made it to the rank of "Internet
asshole".

How old are you?

Old enough to know how to quote, snip and use my intelligence.

Have you any loved ones?

Any WHAT? loved ones? Is that some Martian thing? Oh wait, I might have
some of those in the closet somewhere.

Cause I can tell you, here, you are loved about as much
as a can of droppings from a Dodo bird.

And you are mistakingly thinking that your opinion of me means more to
me than the molecular weight of whale shit on the bottom of the ocean.

When you respond on point you do have useful
information and I appreciate it.

And I am glad I could help you, and that you appreciate it.

But when it comes to public manners you have a lot to learn.

I have always said "When I stop learning, its time to die", so I
actually take that as a compliment and an assurance of a long life.

You are always looking for a confrontation.

ME? A Confrontation? Nah, only when I have a difference of opinion and I
try to discuss/debate that difference and it turns ugly. I won't back
down from being ugly, but I dont "look" for it. With people like
PointedHead, I have enough fun as it is, I don't have to look for more.

But the sad irony is that he actually has some good ideas sometimes. Its
when he starts babbling about three week old threads and he attempts to
change the meaning of words in English (which isn't his native language)
to fit his arguments that it gets too much to bear.

Besides, Boston is 1 game up, leading 2-0 as I type this, what do I care?

 

[David Dorward]

A transcoding proxy or user agent must address issues arising from the
possibility that the character declaration technique described in the
Specification is used, and therefore remove or modify the character
declaration then; if it does not, it is broken.

My MIME isn't fantastic, so I might be wrong here... but I suspect that
anything declared as text/something can be treated as text and thus
transcoded without the user agent needing to know anything about HTML in
specific.

Don't be ridiculous. Again, among other things, the Specification says:

| Therefore, user agents must not assume any default value for the
| "charset" parameter.

It wouldn't be assuming anything. The HTTP header trumps <meta> http-equiv
pseudo whatsits, so it would be following the information given in said
header.