Site HTML validation

25/10/2007, 14h53

Here's something that puzzled me for years. Google said, in not so many
words, that they could not care less if the site validates.

If a site renders its graphics as intented with major browsers, and is
search engine friendly, why would it benefit, in any way, from passing
validation checks?

i

25/10/2007, 16h05

In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>,
Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:

> Here's something that puzzled me for years. Google said, in not so many
> words, that they could not care less if the site validates.
>
> If a site renders its graphics as intented with major browsers, and is
> search engine friendly, why would it benefit, in any way, from passing
> validation checks?

If I'm understood by the majority of the population, why would I benefit
in any way from using proper spelling and grammar?

--
Philip
http://NikitaTheSpider.com/
Whole-site HTML validation, link checking and more

25/10/2007, 16h07

> Here's something that puzzled me for years. Google said, in not so many
> words, that they could not care less if the site validates.

True.

But unless you at least pass it through a validator, how do you know
the HTML doesn't have any serious errors in it?

25/10/2007, 16h08

On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote:
> In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>,
> Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>
>> Here's something that puzzled me for years. Google said, in not so many
>> words, that they could not care less if the site validates.
>>
>> If a site renders its graphics as intented with major browsers, and is
>> search engine friendly, why would it benefit, in any way, from passing
>> validation checks?
>
> If I'm understood by the majority of the population, why would I benefit
> in any way from using proper spelling and grammar?
>

Unlike spelling and grammar, HTML validation is not something that
visitors could see directly.

i

25/10/2007, 16h12

On 2007-10-25, Phil Payne <phil@isham-research.co.uk> wrote:
>> Here's something that puzzled me for years. Google said, in not so many
>> words, that they could not care less if the site validates.
>
> True.
>
> But unless you at least pass it through a validator, how do you know
> the HTML doesn't have any serious errors in it?
>

Maybe it does. For example, a <BODY> tag that never closes is a
"error". Maybe even a "serious error". But the visitors would never
see any detriment for it.

I think that it is quite clear that the website should be viewed with
a couple of major browsers to see if it renders correctly (as
intended). If it does, then what is the point of "validation"?

i

25/10/2007, 18h28

> I think that it is quite clear that the website should be viewed with
> a couple of major browsers to see if it renders correctly (as
> intended). If it does, then what is the point of "validation"?

Had one two or three days ago. It must still be around somewhere.

"Why does Google never use my description meta tag as my snippet?"

Answer - the preceding meta tag wasn't closed so the Googlebot was
flushing until it recovered and it missed the description meta tag.
Page indexed well, came up well in IE, Firefox and Opera (all I tried)
- just the snippet never worked.

Validation error.

I'm _NOT_ saying every page should be 100% clean, though I try to do
that myself. At the end of the day, if the validator can recover its
parsing well enough to get to the end, so can a bot. I'm just saying
the validator will often spot things the Mk I eyeball will miss.

And you never know what the bot is missing. I had a case of the
notorious broken Google verificaton met tag a few months back. A
handful of validation errors. Fixed the tag and the validator
immediately threw up two errors between the tag and </head> - it had
obviously been flushing.

25/10/2007, 19h40

On 2007-10-25, Ignoramus20839 wrote:
>
> On 2007-10-25, Phil Payne <phil@isham-research.co.uk> wrote:
>>> Here's something that puzzled me for years. Google said, in not so many
>>> words, that they could not care less if the site validates.
>>
>> True.
>>
>> But unless you at least pass it through a validator, how do you know
>> the HTML doesn't have any serious errors in it?
>
> Maybe it does. For example, a <BODY> tag that never closes is a
> "error". Maybe even a "serious error". But the visitors would never
> see any detriment for it.

No, it's not an error; the opening and closing BODY tags are
optional.

> I think that it is quite clear that the website should be viewed with
> a couple of major browsers to see if it renders correctly (as
> intended). If it does, then what is the point of "validation"?

If you know how to write conforming HTML, there's no need for it.

--
Chris F.A. Johnson, webmaster <http://Woodbine-Gerrard.com>
================================================== =================
Author:
Shell Scripting Recipes: A Problem-Solution Approach (2005, Apress)

25/10/2007, 20h30

On Thu, 25 Oct 2007 10:12:37 -0500, Ignoramus20839
<ignoramus20839@NOSPAM.20839.invalid> wrote:

>On 2007-10-25, Phil Payne <phil@isham-research.co.uk> wrote:
>>> Here's something that puzzled me for years. Google said, in not so many
>>> words, that they could not care less if the site validates.
>>
>> True.
>>
>> But unless you at least pass it through a validator, how do you know
>> the HTML doesn't have any serious errors in it?
>>
>
>Maybe it does. For example, a <BODY> tag that never closes is a
>"error". Maybe even a "serious error". But the visitors would never
>see any detriment for it.

Except that the number of visitors may itself be diminished because
Google can't properly have confidence in the content of the site and
thus might be reluctant to return it against queries. You make the
wrong argument.

>I think that it is quite clear that the website should be viewed with
>a couple of major browsers to see if it renders correctly (as
>intended).

They bend over backwards to allow for mistakes. I suspect Google would
prefer not to.

> If it does, then what is the point of "validation"?

I was recently trying to explain a technical point to some of Brad
Callan's deskers. Articulate they weren't, which made it a waste
of time trying to communicate with them. You need agreed grammatical
standards, protocols if you will, in order to communicate the finer
points of, er, some kinds of stuff. You know?

BB
--

http://www.fat-odin.com/
http://www.kruse.co.uk/close-ended-questions.htm
http://www.kruse.co.uk/seo-software-review.htm

25/10/2007, 20h30

On Thu, 25 Oct 2007 10:28:08 -0700, Phil Payne
<phil@isham-research.co.uk> wrote:

>> I think that it is quite clear that the website should be viewed with
>> a couple of major browsers to see if it renders correctly (as
>> intended). If it does, then what is the point of "validation"?
>
>Had one two or three days ago. It must still be around somewhere.
>
>"Why does Google never use my description meta tag as my snippet?"

Or it didn't have the key phrase in it.

>Answer - the preceding meta tag wasn't closed so the Googlebot was
>flushing until it recovered and it missed the description meta tag.
>Page indexed well, came up well in IE, Firefox and Opera (all I tried)
>- just the snippet never worked.
>
>Validation error.
>
>I'm _NOT_ saying every page should be 100% clean, though I try to do
>that myself. At the end of the day, if the validator can recover its
>parsing well enough to get to the end, so can a bot. I'm just saying
>the validator will often spot things the Mk I eyeball will miss.

Also my Mk11s.

BB
--

http://www.fat-odin.com/
http://www.kruse.co.uk/close-ended-questions.htm
http://www.kruse.co.uk/seo-software-review.htm

25/10/2007, 20h30

On Thu, 25 Oct 2007 08:07:00 -0700, Phil Payne
<phil@isham-research.co.uk> wrote:

>> Here's something that puzzled me for years. Google said, in not so many
>> words, that they could not care less if the site validates.
>
>True.
>
>But unless you at least pass it through a validator, how do you know
>the HTML doesn't have any serious errors in it?

Some that can affect navigation, understanding of the site's content,
etc. etc.

BB
--

http://www.fat-odin.com/
http://www.kruse.co.uk/close-ended-questions.htm
http://www.kruse.co.uk/seo-software-review.htm

26/10/2007, 09h24

> Here's something that puzzled me for years. Google said, in not so many
> words, that they could not care less if the site validates.

Because it's your site, not theirs, and you lose if it looks stupid in some
browsers, they don't.

Google also point out that if their spiders cannot get around the site, it
may not get indexed at all.

There's one reason for validation that matters; it maximises the chances of
the best possible experience for your visitors.

If that doesn't matter to you, then perhaps a career in public relations
would be a better bet?
--

Andrew
http://www.seo2seo.com/
http://www.sick-site-syndrome.com/
First things first - but not necessarily in that order.

The video that makes YouTube worthwhile:
Charlie The Unicorn - http://www.youtube.com/watch?v=Q5im0Ssyyus

26/10/2007, 10h24

> If a site renders its graphics as intented with major browsers, and is
> search engine friendly, why would it benefit, in any way, from passing
> validation checks?

Because you might want to target minor browsers as well. The main reason is
that you don't know which browsers that will visit your web site. If you
think you know, or if you do know (e.g. by reading your logs), then you'll
do fine supporting only the major browsers.

But if you want to support all spiders and all browsers for mobile devices,
then you end up with a large amount of browsers which you don't have a clue
about, whether your site works with them or not.

But if you only want to target major browsers (e.g. IE, Safari and Mozilla)
then it won't matter. So it's your choice and that might depend on how much
resources you have and if it'll be worthwhile to support the minority.

Mike

MHC Synthesizers and Effects
www.mhc.se

26/10/2007, 12h48

"Mike" <mike@nospam.com> wrote ...
> But if you only want to target major browsers (e.g. IE, Safari and
> Mozilla)
> then it won't matter. So it's your choice and that might depend on how
> much
> resources you have and if it'll be worthwhile to support the minority.

Not quite.
You better future-proof your site by passing validation checks.
You never know whats around the corner.

Personally, I never worry about the minutiae of validation; life is way too
short, and the validation sites waaaaaay too picky.

But the process is still useful, as it does show up your errors, and - more
important - your bad habits.

But obsessing about it is best left to people who do not need to make a
living.
--

Andrew
http://www.seo2seo.com/
http://www.sick-site-syndrome.com/
First things first - but not necessarily in that order.

The video that makes YouTube worthwhile:
Charlie The Unicorn - http://www.youtube.com/watch?v=Q5im0Ssyyus

26/10/2007, 18h32

In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>,
Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:

> On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote:
> > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>,
> > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
> >
> >> Here's something that puzzled me for years. Google said, in not so many
> >> words, that they could not care less if the site validates.
> >>
> >> If a site renders its graphics as intented with major browsers, and is
> >> search engine friendly, why would it benefit, in any way, from passing
> >> validation checks?
> >
> > If I'm understood by the majority of the population, why would I benefit
> > in any way from using proper spelling and grammar?
> >
>
> Unlike spelling and grammar, HTML validation is not something that
> visitors could see directly.

Without testing one's pages in every user agent that exists now and ever
will exist, one can't know if that assertion is true.

It's true that some invalid pages will render fine on an acceptably high
percentage of user agents (we all have a personal definition of
"acceptable"), and it is also true that a page can render poorly in a
browser even though the page is valid. But just as expressing myself
with good spelling and grammar give me the best chance at being
understood by those who read what I write, writing valid HTML gives me
the best chance at being understood by the user agents that read my
pages.

Personally, when I encounter someone who can't be bothered to use his or
her language properly (be it HTML or English or...), I get the
impression that they're not terribly concerned about being understood.
That's not the impression I want to give.

Hmm, I should also point you to this:
http://diveintomark.org/archives/200..._wont__you

Cheers

--
Philip
http://NikitaTheSpider.com/
Whole-site HTML validation, link checking and more

28/10/2007, 16h39

____/ Ignoramus20839 on Thursday 25 October 2007 14:53 : \____

> Here's something that puzzled me for years. Google said, in not so many
> words, that they could not care less if the site validates.
>
> If a site renders its graphics as intented with major browsers, and is
> search engine friendly, why would it benefit, in any way, from passing
> validation checks?

Why go to a fine restaurant when you can eat on the sidewalk to save time and
money?

Validation leads to better and more maintainable. It s you spot markup
errors and sometimes eliminates the need to test your pages in more than a
single Web browser.

--
~~ Best of wishes

Roy S. Schestowitz | UNIX: Because a PC is a terrible thing to waste
http://Schestowitz.com | GNU is Not UNIX | PGP-Key: 0x74572E8E
http://iuron.com - proposing a non-profit search engine

28/10/2007, 16h45

____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____

> In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>,
> Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>
>> On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote:
>> > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>,
>> > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>> >
>> >> Here's something that puzzled me for years. Google said, in not so many
>> >> words, that they could not care less if the site validates.
>> >>
>> >> If a site renders its graphics as intented with major browsers, and is
>> >> search engine friendly, why would it benefit, in any way, from passing
>> >> validation checks?
>> >
>> > If I'm understood by the majority of the population, why would I benefit
>> > in any way from using proper spelling and grammar?
>> >
>>
>> Unlike spelling and grammar, HTML validation is not something that
>> visitors could see directly.
>
> Without testing one's pages in every user agent that exists now and ever
> will exist, one can't know if that assertion is true.
>
> It's true that some invalid pages will render fine on an acceptably high
> percentage of user agents (we all have a personal definition of
> "acceptable"), and it is also true that a page can render poorly in a
> browser even though the page is valid. But just as expressing myself
> with good spelling and grammar give me the best chance at being
> understood by those who read what I write, writing valid HTML gives me
> the best chance at being understood by the user agents that read my
> pages.
>
> Personally, when I encounter someone who can't be bothered to use his or
> her language properly (be it HTML or English or...), I get the
> impression that they're not terribly concerned about being understood.
> That's not the impression I want to give.
>
> Hmm, I should also point you to this:
> http://diveintomark.org/archives/200..._wont__you
>
> Cheers

I like this analogy very much, but I'll admit caring about cleanliness of code
while caring very little about proofreading what I write. In code, unlike
speech, subtle changes can make big difference, including major failures.

The presumption that what the user does not see has no effect on him/her can be
equated to the open source/proprietary software debate where binary blobs tend
to be buggier. The code behind them is messy (no visibility, no "code is
poetry") and thus harder to test. Predictability breeds maintainability.
Hacking code for a browser (or a set thereof) is a solution which is prone to
breakage.

--
~~ Best of wishes

Roy S. Schestowitz | "Signature pending approval"
http://Schestowitz.com | Open Prospects | PGP-Key: 0x74572E8E
Tasks: 113 total, 1 running, 111 sleeping, 0 stopped, 1 zombie
http://iuron.com - knowledge engine, not a search engine

28/10/2007, 21h11

In article <1483097.SThvHVRULv@schestowitz.com>,
Roy Schestowitz <newsgroups@schestowitz.com> wrote:

> ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____
> > It's true that some invalid pages will render fine on an acceptably high
> > percentage of user agents (we all have a personal definition of
> > "acceptable"), and it is also true that a page can render poorly in a
> > browser even though the page is valid. But just as expressing myself
> > with good spelling and grammar give me the best chance at being
> > understood by those who read what I write, writing valid HTML gives me
> > the best chance at being understood by the user agents that read my
> > pages.
> >
> I like this analogy very much, but I'll admit caring about cleanliness of
> code
> while caring very little about proofreading what I write. In code, unlike
> speech, subtle changes can make big difference, including major failures.

That's a good point. Computers can only dream of our (human) skills at
interpreting messy input. Someday they'll get there and the DWIMNWIS
command (Do What I Mean, Not What I Say) will finally work. For now,
they need all they we can give them.

--
Philip
http://NikitaTheSpider.com/
Whole-site HTML validation, link checking and more

28/10/2007, 23h09

On Sun, 28 Oct 2007 16:11:04 -0400, Nikita the Spider
<NikitaTheSpider@gmail.com> wrote:

>In article <1483097.SThvHVRULv@schestowitz.com>,
> Roy Schestowitz <newsgroups@schestowitz.com> wrote:
>
>> ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____
>> > It's true that some invalid pages will render fine on an acceptably high
>> > percentage of user agents (we all have a personal definition of
>> > "acceptable"), and it is also true that a page can render poorly in a
>> > browser even though the page is valid. But just as expressing myself
>> > with good spelling and grammar give me the best chance at being
>> > understood by those who read what I write, writing valid HTML gives me
>> > the best chance at being understood by the user agents that read my
>> > pages.
>> >
>> I like this analogy very much, but I'll admit caring about cleanliness of
>> code
>> while caring very little about proofreading what I write. In code, unlike
>> speech, subtle changes can make big difference, including major failures.
>
>That's a good point. Computers can only dream of our (human) skills at
>interpreting messy input. Someday they'll get there and the DWIMNWIS
>command (Do What I Mean, Not What I Say) will finally work. For now,
>they need all they we can give them.

Absolutely. Otherwise, they can have no confidence in their
interpretation of what they traverse, hence Google's reliance on
everything but its own algo.

BB
--

http://www.fat-odin.com/
http://www.kruse.co.uk/close-ended-questions.htm
http://www.kruse.co.uk/seo-software-review.htm

28/10/2007, 23h09

On Sun, 28 Oct 2007 15:45:07 +0000, Roy Schestowitz
<newsgroups@schestowitz.com> wrote:

>____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____
>
>> In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>,
>> Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>>
>>> On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote:
>>> > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>,
>>> > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>>> >
>>> >> Here's something that puzzled me for years. Google said, in not so many
>>> >> words, that they could not care less if the site validates.
>>> >>
>>> >> If a site renders its graphics as intented with major browsers, and is
>>> >> search engine friendly, why would it benefit, in any way, from passing
>>> >> validation checks?
>>> >
>>> > If I'm understood by the majority of the population, why would I benefit
>>> > in any way from using proper spelling and grammar?
>>> >
>>>
>>> Unlike spelling and grammar, HTML validation is not something that
>>> visitors could see directly.
>>
>> Without testing one's pages in every user agent that exists now and ever
>> will exist, one can't know if that assertion is true.
>>
>> It's true that some invalid pages will render fine on an acceptably high
>> percentage of user agents (we all have a personal definition of
>> "acceptable"), and it is also true that a page can render poorly in a
>> browser even though the page is valid. But just as expressing myself
>> with good spelling and grammar give me the best chance at being
>> understood by those who read what I write, writing valid HTML gives me
>> the best chance at being understood by the user agents that read my
>> pages.
>>
>> Personally, when I encounter someone who can't be bothered to use his or
>> her language properly (be it HTML or English or...), I get the
>> impression that they're not terribly concerned about being understood.
>> That's not the impression I want to give.
>>
>> Hmm, I should also point you to this:
>> http://diveintomark.org/archives/200..._wont__you
>>
>> Cheers
>
>I like this analogy very much, but I'll admit caring about cleanliness of code
>while caring very little about proofreading what I write. In code, unlike
>speech, subtle changes can make big difference, including major failures.

You want to read that again? UNLIKE SPEECH? Gee...

BB
--

http://www.fat-odin.com/
http://www.kruse.co.uk/close-ended-questions.htm
http://www.kruse.co.uk/seo-software-review.htm

01/11/2007, 08h57

____/ Big Bill on Sunday 28 October 2007 22:09 : \____

> On Sun, 28 Oct 2007 15:45:07 +0000, Roy Schestowitz
> <newsgroups@schestowitz.com> wrote:
>
>>____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____
>>
>>> In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>,
>>> Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>>>
>>>> On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote:
>>>> > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>,
>>>> > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote:
>>>> >
>>>> >> Here's something that puzzled me for years. Google said, in not so many
>>>> >> words, that they could not care less if the site validates.
>>>> >>
>>>> >> If a site renders its graphics as intented with major browsers, and is
>>>> >> search engine friendly, why would it benefit, in any way, from passing
>>>> >> validation checks?
>>>> >
>>>> > If I'm understood by the majority of the population, why would I benefit
>>>> > in any way from using proper spelling and grammar?
>>>> >
>>>>
>>>> Unlike spelling and grammar, HTML validation is not something that
>>>> visitors could see directly.
>>>
>>> Without testing one's pages in every user agent that exists now and ever
>>> will exist, one can't know if that assertion is true.
>>>
>>> It's true that some invalid pages will render fine on an acceptably high
>>> percentage of user agents (we all have a personal definition of
>>> "acceptable"), and it is also true that a page can render poorly in a
>>> browser even though the page is valid. But just as expressing myself
>>> with good spelling and grammar give me the best chance at being
>>> understood by those who read what I write, writing valid HTML gives me
>>> the best chance at being understood by the user agents that read my
>>> pages.
>>>
>>> Personally, when I encounter someone who can't be bothered to use his or
>>> her language properly (be it HTML or English or...), I get the
>>> impression that they're not terribly concerned about being understood.
>>> That's not the impression I want to give.
>>>
>>> Hmm, I should also point you to this:
>>> http://diveintomark.org/archives/200..._wont__you
>>>
>>> Cheers
>>
>>I like this analogy very much, but I'll admit caring about cleanliness of
>>code while caring very little about proofreading what I write. In code,
>>unlike speech, subtle changes can make big difference, including major
>>failures.
>
> You want to read that again? UNLIKE SPEECH? Gee...

Yes. Small mistakes and the code won't run (or be rendered properly).

--
~~ Best of wishes

..oÊÊ‡ sÉ buÄ±É¥Ê‡ É¥É”ns ou s,ÇÉ¹ÇÉ¥Ê‡ 'É¹ÇpuÇq 'ÊŽÉ¹É¹oÊ Ê‡,uop :ÊŽÉ¹ÉŸ
..oÊÊ‡ É ÊÉs Ä± Ê‡É¥bnoÉ¥Ê‡ Ä± puÉ ...ÇÉ¹ÇÉ¥ÊÊŽÉ¹ÇÊŒÇ soÉ¹Çz puÉ sÇuo .É¯ÉÇÉ¹p 1nÉŸÊÉ uÉ
Ê‡ÉÉ¥Ê 'É¥É¥É¥É :É¹ÇpuÇq
http://Schestowitz.com | Open Prospects | PGP-Key: 0x74572E8E
Tasks: 111 total, 1 running, 109 sleeping, 0 stopped, 1 zombie
http://iuron.com - knowledge engine, not a search engine

01/11/2007, 08h59

____/ Nikita the Spider on Sunday 28 October 2007 20:11 : \____

> In article <1483097.SThvHVRULv@schestowitz.com>,
> Roy Schestowitz <newsgroups@schestowitz.com> wrote:
>
>> ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____
>> > It's true that some invalid pages will render fine on an acceptably high
>> > percentage of user agents (we all have a personal definition of
>> > "acceptable"), and it is also true that a page can render poorly in a
>> > browser even though the page is valid. But just as expressing myself
>> > with good spelling and grammar give me the best chance at being
>> > understood by those who read what I write, writing valid HTML gives me
>> > the best chance at being understood by the user agents that read my
>> > pages.
>> >
>> I like this analogy very much, but I'll admit caring about cleanliness of
>> code
>> while caring very little about proofreading what I write. In code, unlike
>> speech, subtle changes can make big difference, including major failures.
>
> That's a good point. Computers can only dream of our (human) skills at
> interpreting messy input. Someday they'll get there and the DWIMNWIS
> command (Do What I Mean, Not What I Say) will finally work. For now,
> they need all they we can give them.
>
No, that's dangerous. They should at least generate warnings (if not errors),
like a compiler. The worst small mistakes are the ones you are not aware of.

--
~~ Best of wishes

..oÊÊ‡ sÉ buÄ±É¥Ê‡ É¥É”ns ou s,ÇÉ¹ÇÉ¥Ê‡ 'É¹ÇpuÇq 'ÊŽÉ¹É¹oÊ Ê‡,uop :ÊŽÉ¹ÉŸ
..oÊÊ‡ É ÊÉs Ä± Ê‡É¥bnoÉ¥Ê‡ Ä± puÉ ...ÇÉ¹ÇÉ¥ÊÊŽÉ¹ÇÊŒÇ soÉ¹Çz puÉ sÇuo .É¯ÉÇÉ¹p 1nÉŸÊÉ uÉ
Ê‡ÉÉ¥Ê 'É¥É¥É¥É :É¹ÇpuÇq

25/10/2007, 14h53	#1
Ignoramus20839 Messages: n/a Hébergeur:	Site HTML validation Here's something that puzzled me for years. Google said, in not so many words, that they could not care less if the site validates. If a site renders its graphics as intented with major browsers, and is search engine friendly, why would it benefit, in any way, from passing validation checks? i

25/10/2007, 16h05	#2
Nikita the Spider Messages: n/a Hébergeur:	Re: Site HTML validation In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>, Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: > Here's something that puzzled me for years. Google said, in not so many > words, that they could not care less if the site validates. > > If a site renders its graphics as intented with major browsers, and is > search engine friendly, why would it benefit, in any way, from passing > validation checks? If I'm understood by the majority of the population, why would I benefit in any way from using proper spelling and grammar? -- Philip http://NikitaTheSpider.com/ Whole-site HTML validation, link checking and more

25/10/2007, 16h07	#3
Phil Payne Messages: n/a Hébergeur:	Re: Site HTML validation > Here's something that puzzled me for years. Google said, in not so many > words, that they could not care less if the site validates. True. But unless you at least pass it through a validator, how do you know the HTML doesn't have any serious errors in it?

25/10/2007, 16h08	#4
Ignoramus20839 Messages: n/a Hébergeur:	Re: Site HTML validation On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote: > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>, > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: > >> Here's something that puzzled me for years. Google said, in not so many >> words, that they could not care less if the site validates. >> >> If a site renders its graphics as intented with major browsers, and is >> search engine friendly, why would it benefit, in any way, from passing >> validation checks? > > If I'm understood by the majority of the population, why would I benefit > in any way from using proper spelling and grammar? > Unlike spelling and grammar, HTML validation is not something that visitors could see directly. i

25/10/2007, 16h12	#5
Ignoramus20839 Messages: n/a Hébergeur:	Re: Site HTML validation On 2007-10-25, Phil Payne <phil@isham-research.co.uk> wrote: >> Here's something that puzzled me for years. Google said, in not so many >> words, that they could not care less if the site validates. > > True. > > But unless you at least pass it through a validator, how do you know > the HTML doesn't have any serious errors in it? > Maybe it does. For example, a <BODY> tag that never closes is a "error". Maybe even a "serious error". But the visitors would never see any detriment for it. I think that it is quite clear that the website should be viewed with a couple of major browsers to see if it renders correctly (as intended). If it does, then what is the point of "validation"? i

25/10/2007, 18h28	#6
Phil Payne Messages: n/a Hébergeur:	Re: Site HTML validation > I think that it is quite clear that the website should be viewed with > a couple of major browsers to see if it renders correctly (as > intended). If it does, then what is the point of "validation"? Had one two or three days ago. It must still be around somewhere. "Why does Google never use my description meta tag as my snippet?" Answer - the preceding meta tag wasn't closed so the Googlebot was flushing until it recovered and it missed the description meta tag. Page indexed well, came up well in IE, Firefox and Opera (all I tried) - just the snippet never worked. Validation error. I'm _NOT_ saying every page should be 100% clean, though I try to do that myself. At the end of the day, if the validator can recover its parsing well enough to get to the end, so can a bot. I'm just saying the validator will often spot things the Mk I eyeball will miss. And you never know what the bot is missing. I had a case of the notorious broken Google verificaton met tag a few months back. A handful of validation errors. Fixed the tag and the validator immediately threw up two errors between the tag and </head> - it had obviously been flushing.

25/10/2007, 19h40	#7
Chris F.A. Johnson Messages: n/a Hébergeur:	Re: Site HTML validation On 2007-10-25, Ignoramus20839 wrote: > > On 2007-10-25, Phil Payne <phil@isham-research.co.uk> wrote: >>> Here's something that puzzled me for years. Google said, in not so many >>> words, that they could not care less if the site validates. >> >> True. >> >> But unless you at least pass it through a validator, how do you know >> the HTML doesn't have any serious errors in it? > > Maybe it does. For example, a <BODY> tag that never closes is a > "error". Maybe even a "serious error". But the visitors would never > see any detriment for it. No, it's not an error; the opening and closing BODY tags are optional. > I think that it is quite clear that the website should be viewed with > a couple of major browsers to see if it renders correctly (as > intended). If it does, then what is the point of "validation"? If you know how to write conforming HTML, there's no need for it. -- Chris F.A. Johnson, webmaster <http://Woodbine-Gerrard.com> ================================================== ================= Author: Shell Scripting Recipes: A Problem-Solution Approach (2005, Apress)

25/10/2007, 20h30	#8
Big Bill Messages: n/a Hébergeur:	Re: Site HTML validation On Thu, 25 Oct 2007 10:12:37 -0500, Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: >On 2007-10-25, Phil Payne <phil@isham-research.co.uk> wrote: >>> Here's something that puzzled me for years. Google said, in not so many >>> words, that they could not care less if the site validates. >> >> True. >> >> But unless you at least pass it through a validator, how do you know >> the HTML doesn't have any serious errors in it? >> > >Maybe it does. For example, a <BODY> tag that never closes is a >"error". Maybe even a "serious error". But the visitors would never >see any detriment for it. Except that the number of visitors may itself be diminished because Google can't properly have confidence in the content of the site and thus might be reluctant to return it against queries. You make the wrong argument. >I think that it is quite clear that the website should be viewed with >a couple of major browsers to see if it renders correctly (as >intended). They bend over backwards to allow for mistakes. I suspect Google would prefer not to. > If it does, then what is the point of "validation"? I was recently trying to explain a technical point to some of Brad Callan's deskers. Articulate they weren't, which made it a waste of time trying to communicate with them. You need agreed grammatical standards, protocols if you will, in order to communicate the finer points of, er, some kinds of stuff. You know? BB -- http://www.fat-odin.com/ http://www.kruse.co.uk/close-ended-questions.htm http://www.kruse.co.uk/seo-software-review.htm

26/10/2007, 09h24	#11
Andrew Heenan Messages: n/a Hébergeur:	Re: Site HTML validation > Here's something that puzzled me for years. Google said, in not so many > words, that they could not care less if the site validates. Because it's your site, not theirs, and you lose if it looks stupid in some browsers, they don't. Google also point out that if their spiders cannot get around the site, it may not get indexed at all. There's one reason for validation that matters; it maximises the chances of the best possible experience for your visitors. If that doesn't matter to you, then perhaps a career in public relations would be a better bet? -- Andrew http://www.seo2seo.com/ http://www.sick-site-syndrome.com/ First things first - but not necessarily in that order. The video that makes YouTube worthwhile: Charlie The Unicorn - http://www.youtube.com/watch?v=Q5im0Ssyyus

26/10/2007, 10h24	#12
Mike Messages: n/a Hébergeur:	Re: Site HTML validation > If a site renders its graphics as intented with major browsers, and is > search engine friendly, why would it benefit, in any way, from passing > validation checks? Because you might want to target minor browsers as well. The main reason is that you don't know which browsers that will visit your web site. If you think you know, or if you do know (e.g. by reading your logs), then you'll do fine supporting only the major browsers. But if you want to support all spiders and all browsers for mobile devices, then you end up with a large amount of browsers which you don't have a clue about, whether your site works with them or not. But if you only want to target major browsers (e.g. IE, Safari and Mozilla) then it won't matter. So it's your choice and that might depend on how much resources you have and if it'll be worthwhile to support the minority. Mike MHC Synthesizers and Effects www.mhc.se

26/10/2007, 12h48	#13
Andrew Heenan Messages: n/a Hébergeur:	Re: Site HTML validation "Mike" <mike@nospam.com> wrote ... > But if you only want to target major browsers (e.g. IE, Safari and > Mozilla) > then it won't matter. So it's your choice and that might depend on how > much > resources you have and if it'll be worthwhile to support the minority. Not quite. You better future-proof your site by passing validation checks. You never know whats around the corner. Personally, I never worry about the minutiae of validation; life is way too short, and the validation sites waaaaaay too picky. But the process is still useful, as it does show up your errors, and - more important - your bad habits. But obsessing about it is best left to people who do not need to make a living. -- Andrew http://www.seo2seo.com/ http://www.sick-site-syndrome.com/ First things first - but not necessarily in that order. The video that makes YouTube worthwhile: Charlie The Unicorn - http://www.youtube.com/watch?v=Q5im0Ssyyus

26/10/2007, 18h32	#14
Nikita the Spider Messages: n/a Hébergeur:	Re: Site HTML validation In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>, Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: > On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote: > > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>, > > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: > > > >> Here's something that puzzled me for years. Google said, in not so many > >> words, that they could not care less if the site validates. > >> > >> If a site renders its graphics as intented with major browsers, and is > >> search engine friendly, why would it benefit, in any way, from passing > >> validation checks? > > > > If I'm understood by the majority of the population, why would I benefit > > in any way from using proper spelling and grammar? > > > > Unlike spelling and grammar, HTML validation is not something that > visitors could see directly. Without testing one's pages in every user agent that exists now and ever will exist, one can't know if that assertion is true. It's true that some invalid pages will render fine on an acceptably high percentage of user agents (we all have a personal definition of "acceptable"), and it is also true that a page can render poorly in a browser even though the page is valid. But just as expressing myself with good spelling and grammar give me the best chance at being understood by those who read what I write, writing valid HTML gives me the best chance at being understood by the user agents that read my pages. Personally, when I encounter someone who can't be bothered to use his or her language properly (be it HTML or English or...), I get the impression that they're not terribly concerned about being understood. That's not the impression I want to give. Hmm, I should also point you to this: http://diveintomark.org/archives/200..._wont__you Cheers -- Philip http://NikitaTheSpider.com/ Whole-site HTML validation, link checking and more

28/10/2007, 16h39	#15
Roy Schestowitz Messages: n/a Hébergeur:	Re: Site HTML validation ____/ Ignoramus20839 on Thursday 25 October 2007 14:53 : \____ > Here's something that puzzled me for years. Google said, in not so many > words, that they could not care less if the site validates. > > If a site renders its graphics as intented with major browsers, and is > search engine friendly, why would it benefit, in any way, from passing > validation checks? Why go to a fine restaurant when you can eat on the sidewalk to save time and money? Validation leads to better and more maintainable. It s you spot markup errors and sometimes eliminates the need to test your pages in more than a single Web browser. -- ~~ Best of wishes Roy S. Schestowitz \| UNIX: Because a PC is a terrible thing to waste http://Schestowitz.com \| GNU is Not UNIX \| PGP-Key: 0x74572E8E http://iuron.com - proposing a non-profit search engine

28/10/2007, 16h45	#16
Roy Schestowitz Messages: n/a Hébergeur:	Re: Site HTML validation ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____ > In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>, > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: > >> On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote: >> > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>, >> > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: >> > >> >> Here's something that puzzled me for years. Google said, in not so many >> >> words, that they could not care less if the site validates. >> >> >> >> If a site renders its graphics as intented with major browsers, and is >> >> search engine friendly, why would it benefit, in any way, from passing >> >> validation checks? >> > >> > If I'm understood by the majority of the population, why would I benefit >> > in any way from using proper spelling and grammar? >> > >> >> Unlike spelling and grammar, HTML validation is not something that >> visitors could see directly. > > Without testing one's pages in every user agent that exists now and ever > will exist, one can't know if that assertion is true. > > It's true that some invalid pages will render fine on an acceptably high > percentage of user agents (we all have a personal definition of > "acceptable"), and it is also true that a page can render poorly in a > browser even though the page is valid. But just as expressing myself > with good spelling and grammar give me the best chance at being > understood by those who read what I write, writing valid HTML gives me > the best chance at being understood by the user agents that read my > pages. > > Personally, when I encounter someone who can't be bothered to use his or > her language properly (be it HTML or English or...), I get the > impression that they're not terribly concerned about being understood. > That's not the impression I want to give. > > Hmm, I should also point you to this: > http://diveintomark.org/archives/200..._wont__you > > Cheers I like this analogy very much, but I'll admit caring about cleanliness of code while caring very little about proofreading what I write. In code, unlike speech, subtle changes can make big difference, including major failures. The presumption that what the user does not see has no effect on him/her can be equated to the open source/proprietary software debate where binary blobs tend to be buggier. The code behind them is messy (no visibility, no "code is poetry") and thus harder to test. Predictability breeds maintainability. Hacking code for a browser (or a set thereof) is a solution which is prone to breakage. -- ~~ Best of wishes Roy S. Schestowitz \| "Signature pending approval" http://Schestowitz.com \| Open Prospects \| PGP-Key: 0x74572E8E Tasks: 113 total, 1 running, 111 sleeping, 0 stopped, 1 zombie http://iuron.com - knowledge engine, not a search engine

28/10/2007, 21h11	#17
Nikita the Spider Messages: n/a Hébergeur:	Re: Site HTML validation In article <1483097.SThvHVRULv@schestowitz.com>, Roy Schestowitz <newsgroups@schestowitz.com> wrote: > ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____ > > It's true that some invalid pages will render fine on an acceptably high > > percentage of user agents (we all have a personal definition of > > "acceptable"), and it is also true that a page can render poorly in a > > browser even though the page is valid. But just as expressing myself > > with good spelling and grammar give me the best chance at being > > understood by those who read what I write, writing valid HTML gives me > > the best chance at being understood by the user agents that read my > > pages. > > > I like this analogy very much, but I'll admit caring about cleanliness of > code > while caring very little about proofreading what I write. In code, unlike > speech, subtle changes can make big difference, including major failures. That's a good point. Computers can only dream of our (human) skills at interpreting messy input. Someday they'll get there and the DWIMNWIS command (Do What I Mean, Not What I Say) will finally work. For now, they need all they we can give them. -- Philip http://NikitaTheSpider.com/ Whole-site HTML validation, link checking and more

28/10/2007, 23h09	#18
Big Bill Messages: n/a Hébergeur:	Re: Site HTML validation On Sun, 28 Oct 2007 16:11:04 -0400, Nikita the Spider <NikitaTheSpider@gmail.com> wrote: >In article <1483097.SThvHVRULv@schestowitz.com>, > Roy Schestowitz <newsgroups@schestowitz.com> wrote: > >> ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____ >> > It's true that some invalid pages will render fine on an acceptably high >> > percentage of user agents (we all have a personal definition of >> > "acceptable"), and it is also true that a page can render poorly in a >> > browser even though the page is valid. But just as expressing myself >> > with good spelling and grammar give me the best chance at being >> > understood by those who read what I write, writing valid HTML gives me >> > the best chance at being understood by the user agents that read my >> > pages. >> > >> I like this analogy very much, but I'll admit caring about cleanliness of >> code >> while caring very little about proofreading what I write. In code, unlike >> speech, subtle changes can make big difference, including major failures. > >That's a good point. Computers can only dream of our (human) skills at >interpreting messy input. Someday they'll get there and the DWIMNWIS >command (Do What I Mean, Not What I Say) will finally work. For now, >they need all they we can give them. Absolutely. Otherwise, they can have no confidence in their interpretation of what they traverse, hence Google's reliance on everything but its own algo. BB -- http://www.fat-odin.com/ http://www.kruse.co.uk/close-ended-questions.htm http://www.kruse.co.uk/seo-software-review.htm

01/11/2007, 08h57	#20
Roy Schestowitz Messages: n/a Hébergeur:	Re: Site HTML validation ____/ Big Bill on Sunday 28 October 2007 22:09 : \____ > On Sun, 28 Oct 2007 15:45:07 +0000, Roy Schestowitz > <newsgroups@schestowitz.com> wrote: > >>____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____ >> >>> In article <wqmdnSg_D_QZLL3anZ2dnUVZ_gudnZ2d@giganews.com>, >>> Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: >>> >>>> On 2007-10-25, Nikita the Spider <NikitaTheSpider@gmail.com> wrote: >>>> > In article <rLGdnaRdbvBBAr3anZ2dnUVZ_uzinZ2d@giganews.com>, >>>> > Ignoramus20839 <ignoramus20839@NOSPAM.20839.invalid> wrote: >>>> > >>>> >> Here's something that puzzled me for years. Google said, in not so many >>>> >> words, that they could not care less if the site validates. >>>> >> >>>> >> If a site renders its graphics as intented with major browsers, and is >>>> >> search engine friendly, why would it benefit, in any way, from passing >>>> >> validation checks? >>>> > >>>> > If I'm understood by the majority of the population, why would I benefit >>>> > in any way from using proper spelling and grammar? >>>> > >>>> >>>> Unlike spelling and grammar, HTML validation is not something that >>>> visitors could see directly. >>> >>> Without testing one's pages in every user agent that exists now and ever >>> will exist, one can't know if that assertion is true. >>> >>> It's true that some invalid pages will render fine on an acceptably high >>> percentage of user agents (we all have a personal definition of >>> "acceptable"), and it is also true that a page can render poorly in a >>> browser even though the page is valid. But just as expressing myself >>> with good spelling and grammar give me the best chance at being >>> understood by those who read what I write, writing valid HTML gives me >>> the best chance at being understood by the user agents that read my >>> pages. >>> >>> Personally, when I encounter someone who can't be bothered to use his or >>> her language properly (be it HTML or English or...), I get the >>> impression that they're not terribly concerned about being understood. >>> That's not the impression I want to give. >>> >>> Hmm, I should also point you to this: >>> http://diveintomark.org/archives/200..._wont__you >>> >>> Cheers >> >>I like this analogy very much, but I'll admit caring about cleanliness of >>code while caring very little about proofreading what I write. In code, >>unlike speech, subtle changes can make big difference, including major >>failures. > > You want to read that again? UNLIKE SPEECH? Gee... Yes. Small mistakes and the code won't run (or be rendered properly). -- ~~ Best of wishes ..oÊÊ‡ sÉ buÄ±É¥Ê‡ É¥É”ns ou s,ÇÉ¹ÇÉ¥Ê‡ 'É¹ÇpuÇq 'ÊŽÉ¹É¹oÊ Ê‡,uop :ÊŽÉ¹ÉŸ ..oÊÊ‡ É ÊÉs Ä± Ê‡É¥bnoÉ¥Ê‡ Ä± puÉ ...ÇÉ¹ÇÉ¥ÊÊŽÉ¹ÇÊŒÇ soÉ¹Çz puÉ sÇuo .É¯ÉÇÉ¹p 1nÉŸÊÉ uÉ Ê‡ÉÉ¥Ê 'É¥É¥É¥É :É¹ÇpuÇq http://Schestowitz.com \| Open Prospects \| PGP-Key: 0x74572E8E Tasks: 111 total, 1 running, 109 sleeping, 0 stopped, 1 zombie http://iuron.com - knowledge engine, not a search engine

01/11/2007, 08h59	#21
Roy Schestowitz Messages: n/a Hébergeur:	Re: Site HTML validation ____/ Nikita the Spider on Sunday 28 October 2007 20:11 : \____ > In article <1483097.SThvHVRULv@schestowitz.com>, > Roy Schestowitz <newsgroups@schestowitz.com> wrote: > >> ____/ Nikita the Spider on Friday 26 October 2007 18:32 : \____ >> > It's true that some invalid pages will render fine on an acceptably high >> > percentage of user agents (we all have a personal definition of >> > "acceptable"), and it is also true that a page can render poorly in a >> > browser even though the page is valid. But just as expressing myself >> > with good spelling and grammar give me the best chance at being >> > understood by those who read what I write, writing valid HTML gives me >> > the best chance at being understood by the user agents that read my >> > pages. >> > >> I like this analogy very much, but I'll admit caring about cleanliness of >> code >> while caring very little about proofreading what I write. In code, unlike >> speech, subtle changes can make big difference, including major failures. > > That's a good point. Computers can only dream of our (human) skills at > interpreting messy input. Someday they'll get there and the DWIMNWIS > command (Do What I Mean, Not What I Say) will finally work. For now, > they need all they we can give them. > No, that's dangerous. They should at least generate warnings (if not errors), like a compiler. The worst small mistakes are the ones you are not aware of. -- ~~ Best of wishes ..oÊÊ‡ sÉ buÄ±É¥Ê‡ É¥É”ns ou s,ÇÉ¹ÇÉ¥Ê‡ 'É¹ÇpuÇq 'ÊŽÉ¹É¹oÊ Ê‡,uop :ÊŽÉ¹ÉŸ ..oÊÊ‡ É ÊÉs Ä± Ê‡É¥bnoÉ¥Ê‡ Ä± puÉ ...ÇÉ¹ÇÉ¥ÊÊŽÉ¹ÇÊŒÇ soÉ¹Çz puÉ sÇuo .É¯ÉÇÉ¹p 1nÉŸÊÉ uÉ Ê‡ÉÉ¥Ê 'É¥É¥É¥É :É¹ÇpuÇq

Outils de la discussion
Afficher une version imprimable Envoyer un lien vers cette page par email