if your page does NOT use an xhtml doctype you must not close any tags
with /> at all.
In particualr those meta and link tags in the head of the page. This
results in a premature end of the ehad and of the page, with content
not being processed further.
I recommend you validate all the pages on yrou site, starting with the
homepage.
You may be able to fix things easily if you are using any kind of
templates or if you are abel to use search and replace for all pages
you have on your pc.
On Mar 5, 11:10 am, tapirgal <she...@tapirback.com> wrote:
Thanks, Webado. It would have been awhile before I found that problem.
Actually, many, many of my 1300+ pages do validate, such as the sea
lion page itself validates, but I wouldn't have realized that one or
several upstream had cut off the search. At least it validates on CSE
v 8.04.
The gift shop validates - it's the recreational (fun for me, less time
to play with) areas and my old tapir pages I didn't follow up on. :/
Well, it seems that my main mistake on these pages was using a meta
tag that close with " /> which should be reserved for XHTML and I
should have only used the closing bracket without the slash and space
in my HTML 4.01 Transitional doc.
What I don't understand is why these pages validated cleanly with CSE
Validator Pro v8.04 with the latest updates. I thought that was a good
piece of software. Any comments on this software? I've been using it
on my whole site. I found it easier to work with than any of the
online pages I found (such as the W3C page) and to understand the
corrections when I had so many pages with so many mistakes a couple of
years ago, and I kept using the program, which obviously didn't
consider the current mitake a mistake. <br /> are also not seen in CSE
as a mistake in HTML 4.01 Transitional.
> Well, it seems that my main mistake on these pages was using a meta
> tag that close with " /> which should be reserved for XHTML and I
> should have only used the closing bracket without the slash and space
> in my HTML 4.01 Transitional doc.
> What I don't understand is why these pages validated cleanly with CSE
> Validator Pro v8.04 with the latest updates. I thought that was a good
> piece of software. Any comments on this software? I've been using it
> on my whole site. I found it easier to work with than any of the
> online pages I found (such as the W3C page) and to understand the
> corrections when I had so many pages with so many mistakes a couple of
> years ago, and I kept using the program, which obviously didn't
> consider the current mitake a mistake. <br /> are also not seen in CSE
> as a mistake in HTML 4.01 Transitional.
CSE HTML Validator is good software. Those issues you mention are
extremely unlikely to affect search engine rankings. Also, remember
that CSE HTML Validator is designed for practical use, not simply to
check for strict compliance like the W3C validator. This makes CSE
HTML Validator more useful for most people.
If you want CSE HTML Validator to report errors for "/>" in HTML, then
you can uncheck the "XML compatibility" option in the Validator Engine
Options. This option is checked by default because it's not a problem
with today's browser/user agents.
> Well, it seems that my main mistake on these pages was using a meta
> tag that close with " /> which should be reserved for XHTML and I
> should have only used the closing bracket without the slash and space
> in my HTML 4.01 Transitional doc.
> What I don't understand is why these pages validated cleanly withCSEValidatorPro v8.04 with the latest updates. I thought that was a good
> piece of software. Any comments on this software? I've been using it
> on my whole site. I found it easier to work with than any of the
> online pages I found (such as the W3C page) and to understand the
> corrections when I had so many pages with so many mistakes a couple of
> years ago, and I kept using the program, which obviously didn't
> consider the current mitake a mistake. <br /> are also not seen in CSE
> as a mistake in HTML 4.01 Transitional.
Albert, it is VERY important the head of a page NOT be broken. Using
XHTML syntax in a non-xhtml document (e.g. no xhtml doctype) breaks
the head in the worst way, resulting in the page content not being
well parsed if at all.
On 6 mar, 14:47, Tidal_Wave_One <goo...@wiersch.com> wrote:
> CSE HTML Validator is good software. Those issues you mention are
> extremely unlikely to affect search engine rankings. Also, remember
> that CSE HTML Validator is designed for practical use, not simply to
> check for strict compliance like the W3C validator. This makes CSE
> HTML Validator more useful for most people.
> If you want CSE HTML Validator to report errors for "/>" in HTML, then
> you can uncheck the "XML compatibility" option in the Validator Engine
> Options. This option is checked by default because it's not a problem
> with today's browser/user agents.
> On Mar 6, 1:40 am, tapirgal <she...@tapirback.com> wrote:
> > Well, it seems that my main mistake on these pages was using a meta
> > tag that close with " /> which should be reserved for XHTML and I
> > should have only used the closing bracket without the slash and space
> > in my HTML 4.01 Transitional doc.
> > What I don't understand is why these pages validated cleanly withCSEValidatorPro v8.04 with the latest updates. I thought that was a good
> > piece of software. Any comments on this software? I've been using it
> > on my whole site. I found it easier to work with than any of the
> > online pages I found (such as the W3C page) and to understand the
> > corrections when I had so many pages with so many mistakes a couple of
> > years ago, and I kept using the program, which obviously didn't
> > consider the current mitake a mistake. <br /> are also not seen in CSE
> > as a mistake in HTML 4.01 Transitional.
> > Sheryl- Masquer le texte des messages précédents -
It sounds like you're talking about strict standards compliance, but
what modern search engine or web browser do you know of that will not
be able to parse or handle "/>" at the end of a meta tag? I'm not
aware of any but would like to know if there is indeed one.
Thanks,
Albert
On Mar 6, 3:11 pm, webado <web...@gmail.com> wrote:
> Albert, it is VERY important the head of a page NOT be broken. Using
> XHTML syntax in a non-xhtml document (e.g. no xhtml doctype) breaks
> the head in the worst way, resulting in the page content not being
> well parsed if at all.
It won't complain. But the robot simply appears to skip over everything. Seen lots of sites doing poorly, pages dropping out of the index, or gone suppplemental (when that still existed officially). They often have this in common - broken head.
On Thu, Mar 6, 2008 at 4:26 PM, Tidal_Wave_One <goo...@wiersch.com> wrote:
> It sounds like you're talking about strict standards compliance, but > what modern search engine or web browser do you know of that will not > be able to parse or handle "/>" at the end of a meta tag? I'm not > aware of any but would like to know if there is indeed one.
> Thanks, > Albert
> On Mar 6, 3:11 pm, webado <web...@gmail.com> wrote: > > Albert, it is VERY important the head of a page NOT be broken. Using > > XHTML syntax in a non-xhtml document (e.g. no xhtml doctype) breaks > > the head in the worst way, resulting in the page content not being > > well parsed if at all.
But a "broken head" to googlebot is probably something other
(something much more serious) than simply using "/>" to end a meta
tag. Do you have any supporting links that say that "/>" is
responsible for googlebot's skipping of info? I seriously doubt that
is the problem. I think it is far too speculative to simply blame it
on a "broken head" because of "/>". It is much more likely that it is
caused by more serious issues or just changes in how or why pages are
indexed.
Also, I think many pages have "broken heads" since many pages are
poorly written. But many of these "broken pages" still do well in the
search engines.
I'm not saying that validation is not important of course. It is
important for many reasons, but I think to mess up a search engine
ranking due to an "invalid" document requires a serious validation or
document structure issue and not simply using "/>".
On Mar 6, 3:56 pm, webado <web...@gmail.com> wrote:
> It won't complain. But the robot simply appears to skip over everything.
> Seen lots of sites doing poorly, pages dropping out of the index, or gone
> suppplemental (when that still existed officially). They often have this in
> common - broken head.
For a robot to parse pages semantically, all markup has to be good,
unbroken.
Modern Googelbots parse semantically.
There there's ancecdotal evidence of the broken mata tag causing pages
to go suppelemental (back when that still existed). Hence the urban
legend that "submitting to sitemaps makes the site tank".
On 6 mar, 17:52, Tidal_Wave_One <goo...@wiersch.com> wrote:
> But a "broken head" to googlebot is probably something other
> (something much more serious) than simply using "/>" to end a meta
> tag. Do you have any supporting links that say that "/>" is
> responsible for googlebot's skipping of info? I seriously doubt that
> is the problem. I think it is far too speculative to simply blame it
> on a "broken head" because of "/>". It is much more likely that it is
> caused by more serious issues or just changes in how or why pages are
> indexed.
> Also, I think many pages have "broken heads" since many pages are
> poorly written. But many of these "broken pages" still do well in the
> search engines.
> I'm not saying that validation is not important of course. It is
> important for many reasons, but I think to mess up a search engine
> ranking due to an "invalid" document requires a serious validation or
> document structure issue and not simply using "/>".
> On Mar 6, 3:56 pm, webado <web...@gmail.com> wrote:
> > Googlebot.
> > It won't complain. But the robot simply appears to skip over everything.
> > Seen lots of sites doing poorly, pages dropping out of the index, or gone
> > suppplemental (when that still existed officially). They often have this in
> > common - broken head.- Masquer le texte des messages précédents -
