Hi Googlers,

My question is this ... Whenever I ask questions on this forum, about
Google and listings, I get people telling me to validate my site with
W3C. I have always used Microsoft Frontpage for page building and the
code has never been able to be totally validated in the eyes of these
puritans. However, my sites have always managed to be listed on all of
the major search engines and acheive some heady heights in some cases.
If the W3C validity of my sites was that important to Google wouldn't
they have a tool which would show me the errors on their Sitemap
interface ?

Many thanks and a Happy Christmas to all.

Colin Hall ( www.badcol.com )

The tool is there for you to use: the w3 validator.

You need to understand why a robot needs valid code. It' simply because
it cannot parse a page properly and pick up text from code if the code
is incorrect and broken.

Googlebot is trying to be smart and not just pick up words and
sentences, but also give them weight if they are used in title, header
tags, or list items or anchor tags for links.

If the code is broken things just don't' work. Block level errors are
the worst. An invalid head section means bye-bye.

There are degrees in invalid code.You'll not know how invalid is valid
enough until you run the validator and study the results and see what's
what.

Don't fool yourself. Things have been ok until now for yuo because
you've been a) lucky b) not yet subjected to the latest robots and
their logic.

Can you afford to sit on a timebomb? You can if your site hobby only.
As soon as you want and need to be indexed because you're running a
business or for any other economic reasons, then you need to put all
the chips on your side, to give your site the maximum chances.

Frontpage may be creating atrocious code like all WYSIWYG editors - but
it can all be fixed in the html view.
Unfortunately you cannot rely 100% of these editors and decide you
don't want to know html and you don't need to know html.

Hi Webado,

I understand that you are sold on the idea of W3C validation, that's
nice. But I was hoping to hear from others who could tell me Google's
thoughts on this subject. We have had our run ins in the past and I
would rather keep that in the past and move on a bit.

web design tools. The examples that I see of W3C websites I think are a
bit bland, boxy and robotic structures and the sites that I am
impressed by are usually really creative, flowing and artistic. Surely
there is a place in Google's heart for art ?

Cheers

Colin

No one from google is really responding to any specific questions in
this group, so good luck with that.

No  is the simple answer,
if your site opens in explorer,firefox,opera and a few others then it
is fine,
About the robots breaking off because of  text or  W3C validation
errors in your site is,mmmmmmmmm bull i think,if someone can come in
who has worked out if  a site who has corrected the  W3C validation
errors and got a better result in serps just by doing that then let us
know?
I think it is only googlers who can say if is worth doing or not,But
dont hold your breath.

Hmmm... now that you mentioned it maybe I valguely remember your Colin
having a difference of opinion with my views. I have poor memory though
;)

The Webmaster Guidelines rather recommend valid code. I'm not sure why
people tend to glide over that. Maybe because it requires extra effort?

Is  www.yourhtmlsource.com bland and not artistic?

How about most if not all templates at  www.oswd.org ?

WYSIWYG editors left to their own devices produce invalid code as you
have no doubt noticed. Generally that can be rectified but sometimes it
requires the user to really know their editor thoroughly, which
sometimes is  even harder than knowing how to code good valid html and
css.

My point is that any and all designs CAN be made valid code and there's
no doubt in my mind about it.

The only bits  that can be thought of as iffy are some css constructs
which may be only valid in IE and which may be picked up as invalid CSS
 (colored scrollbars, opacity filters are a couple of examples) by the
CSS validator. These do not affect html, and html is what needs to be
valid first and foremost. The kind of quasi-valid invalid css that I'm
referring to has the nasty habit of not working in a browser like
Firefox - so it's a choice to make.

Does producing valid html code  require more work? yes and no. No if
you develop good habits and start out paying close attention to the
state of validation from the first page you make for a site. Yes if
you've put up a 1000 page site riddled with errors.

I'll add the big problem lots of people face: that cursed validation
meta tag. If used rigth out of the box as given by Google it's invalid
for all doctypes.
Whats' worse is that fro non xhtml doctypes it actually cuts the head
and prevents robtos from parsing the page altogether even if everything
else is 100% valid on the page.From there on in it all goes to hell in
a handbasket - no indexing, no crawling.

Yes, such a small thing and causing so much grief. But robtos are just
programs. They need the interpret the tags they find in a web page in
order to know what is relevant content and what is code. You don't want
to see html and javascript code appear in the index. That kind of
broken code signals the end of the page.

I have fixed the code on a number of sites and they have all come
around.

Try it, you'll like it ;)

<<< Is  www.yourhtmlsource.com bland and not artistic?
How about most if not all templates at  www.oswd.org ? >>>

I've got to say it but they are a bit boxy and dull. That doesn't mean
that they aren't good, just a bit like a Volkwagon, rather than
characterful VW Beetle or a Mini. Just lacking a bit of style (IMHO).
Sometimes boxy is good, you know ... when you're building a box or
something like that, but now and then it's really good to get outside
of that box (Oh no, I'm starting to sound really Millennium)

Webado - I love your persistance though ...

I do tend to agree with Daz-in_Oz when he says that if a browser can
read it ok, then presumably there isn't much to worry about.

All the Best

Colin

A robot works differently from a browser. Search engines run robots not
browsers.

And taste obviously is a personal thing.

Try this one: www.designbyatfb.com - if anybody truly thinks outside of
a box AND uses valid code (not just valid by XHTML STRICT valid) that's
the one.

Nope, sorry ... I still think it looks very 1990's. I don't mind the
subject matter and I can see that the designer has made some effort in
brightening things up ... but IMHO it's boxy in the extreme ;-)

Plus, from my side of the pond, I can't see the site even getting into
the top 10 for much at all. I get much better results and I don't
really try much past putting Title / Description & Keyword tags into my
pages.

Hi There,
I too was unsure about this validation stuff until last week when
webado pointed me once again in that direction.
I found that all my pages at http://informationcyprus.com  (I use one
template) were riddled with code errors many of them put there by my
hosting/web building companies wysiwyg editor and front page which I
use to edit the templates.

The great thing is that a lot of silly mistakes I made when building
the pages were picked up by the w3c validator. Mistakes I would not
otherwise have known about.

I now run every new page through the validator, it only takes a couple
of minutes and any mistakes (not just with the coding) are picked up
straight away.

Kev

Top 10 what?

Assuming one sells things, maybe but otherwise what does it matter? I
sent you to a personal website. No sales pitch. Zip.
The layout of that site changes frequently incidentally. It's now a
sedate one LOL

Can you give me an example of what YOU consider updated design,
wonderful by your standards, and which is not valid and cannot be made
valid?

I personally don't have imagination and cannot use Photoshop either.
But having or not having such a skill has nothing to do with writing or
not writing valid code.

1990's, lol well I suppose that's better then 1950's, although Retro is
in. Hey, I'd say MY SITE may appear "boxy", but it validates. Despite
your supposition that my site doesn't get much traffic, let me assure
you it does rather well in it's niche.

I've several hits on my site from Google in the top #10 out of several
thousand and even a few million. And let me assure you google doesn't
like my site for it's content. It likes my site as it is valid code and
very easy to crawl.

I used to be  dedicated Front Page user myself, Adobe Image Ready
sliced and diced pages, loaded with graphics were my absolute fav's.
Imagine my horror, much like yours when I was first approached with,
egads 100's of warnings and errors, no doc type, no page titles, no alt
tags and on and on it went. Those folks who loved my work graphically
then, loathed my coding.

They gently and patiently taught me the right way, css, validation etc.

The site Webado gave as an example is my first real attempt at complete
css, html design outside FrontPage and I think it has done well by all
my efforts.

It may never win any awards, but I take great pride in the fact that
even though it may still work with errors and warnings, font tags and
all the other "stuff", I did manage without all that garbage.

And by the way, none of the sites I've ever built using valid css and
html have ever suffered the google sandbox. All have been crawled even
before I managed to finish a site long enough to build a sitemap.

Really, no one could have argued the "doesn't need to be valid"
agrument better then I could. I pitched my fits, swore I'd never change
the way I preferred to do things. But I did indeed reverse my postion
and I've never regretted it.

In fact just today, I used FrontPage to throw together a tabled very
simple page to display a few school event photos. But you know what, I
immediately opened the same page in my now favorite program (Topstyle
Pro) threw in a doc type, fixed up the no no's that FrontPage adds in
and validated it. Only then would I upload it. Today it's a matter of
pride.............if I can, then why not. It don't hurt.

It's a lovely world when you can take complete and utter pride in your
work because you took the time to do the best you could.

I really think that I tried to disuade Webado from participating in
this thread when I stated "We have had our run ins in the past and I
would rather keep that in the past and move on a bit". You are
disrespectful of anyone elses opinion and I'm bored of it. If you would
like a list of well designed sites that I find cool here you go.

I like www.erwinpenland.com - Very smooth, dark and extremely quick
download times - W3C says it is not a valid document.

I like www.flyingpuppet.com - A very talented and dark artist - Again
not valid by W3C standards

and just in case you think I just have a thing about square format
sites there is the superb www.leosantos.com - a very talented animator
who the W3C nerds say is not valid.

Well, in the words of Rod Stewart (Para-phrasing somewhat) "If loving
them is wrong ... I don't wanna be right".

These 3 sites have more content and creativity than any W3C validated
sites that I've seen, so IMHO it's about time the industry standard
stepped up a gear, before it gets forgotten altogether.

Personally, I think that W3C is invalid. Web Browsers have moved on
since they declared themselves in charge and you know what? I prefer a
web with Flash and with CSS, I enjoy the web where a browser can read
an W3C 'invalid' page. I especially like a web where people can work
without rules holding them back and even use a WYSIWYG editor if it
makes things easier for them to express themselves.

I'm guessing, but I think that you're a little scared that there are
websites out there that are beyond your reach and when you offer your
web design services to people you have to have a prepared speech as to
why you build such boxy sites. I think you would be better served
climbing down from your soap box and getting creative with the rest of
the world, rather than telling everyone that they are 'Invalid'.

If you believe that, it's your own loss. I am not scared of anyhting or
anybody. I know my limits -  so I have no reason to worry. You want
fancy, get a fancy graphic artist. You want good, get a fancy graphic
artist who knows what he's doing web-wise.

I am not on a soap box, I rather think you are.

I think you're right Webado, knowing your limits is a good thing. I
just don't understand why you hijack every post with this pro W3C
stuff. Unless of course you are one of these paid bloggers who is paid
to promote websites on this kind of forum. It annoys me intensly that
every time that I ask a respected forum like this a question you decide
to sing the same song and dilute my initial question. It has happened
enough now to call it a trend and either it's just me you pick on or
you do it to everyone.

As for involving someone that you knew would take offense to my
comments, that's just childish. To msass, I appologise and offer the
hand of friendship.

Anyway, I think I've come to the conclusion that having a site that is
'Valid' in its coding is less important to me than the artistic freedom
that I get from using software. To my mind the html coding purity is
something incidental and secondary to content, and I believe that
Google is now advanced enough to tell good content when they see it.

All the Best for Christmas & New Years to everyone. Once again sorry
for upsetting you msass, it was not my intention and I think a bit of
planned Jerrymandering on Webado's part.

Colin :-)

I don't see how webado could be hijacking the thread for pro-W3C stuff
when the topic is "Is W3C validation really essential for Google to
list my site ?" :-)

As someone who has made his share of website crawlers and indexers I
believe there is a difference that needs to be made between
1. Block level parsability and semantic use of HTML tags
2. Full W3C validation

The later one is the one people like to discuss, but which IMHO has
little direct influence on search engine crawling and indexing. It does
matter, however, and if you can do it, then it makes a lot of sense to
get full validation. If you're a professional web-designer and you
can't create pages which validate then to me that's like being a
copy-writer and not being able to spell...

Getting a page to parse on a block level does make a big difference in
terms of crawling and indexing. If a page cannot be parset, the search
engine will not be able to extract the full value of your content --
it'll resort to a plain-text extractor and miss the context which you
might have used. If you don't use the semantic HTML tags you will also
lose out on additional value from context.

Example:
- a text line with a long "font"-tag
vs
- a text line with a "h1"or "h2" tag

The first one will be just like any other text line in the document,
the second one will be treated as a header and the content following it
can be seen as belonging to the header. It makes a big difference when
extracting information and it does not take any additional work to use
the tags correctly - with CSS you can even style it to look like
whatever you want.

If your page validates properly, then it will generally also be valid
on a block level -- meaning you have this part covered as well. Of
course it can't test to see if you're using markup correctly, that's
still your job :-).

Comparing a tiny, "insignificant" (on a global scale) web-site to say
Google or ebay and saying that "validation doesn't matter" doesn't
really make sense. Where small tweaks can make a big difference on a
small site, they generally don't matter on a site the size of Google.
Google could put up a pure flash website and still rank #1 for "google"
:-). After a certain size, the off-site factors are much more important
than the on-page factors -- but you have to get there somehow, so until
then, make sure you do the homework and iron out those details. Every
little bit counts.

John

I'm fairly new to all this, having just done my first site
(http://www.landmarkplc.com) in dreamweaver. It all looks OK when you
start it but, again, when I checked the W3C validator there were around
10-20 errors per page. If you use a basic template in a wysywig program
it is just as simple to create a valid layout and check it on W3C
before you carry on building your site tree from there. Once you've
done that and have a decent error free base it would make no difference
to the amount of time it takes, check the site again and I'm pretty
sure you'd only get the odd error. Validating is hassle, as I found out
after about three hours for only ten or so pages, but it's probably
worth it, as knowing rediculous companies like MS and all that,
unvalidated websites or websites with errors upon parsing the code will
probably be "barred" with IE 7-style "red address bars" eventually
which is ok, ie fixable, if you are maintaining your own site, but if
you're a web designer, like yourself colin, it would be a nightmare.

 I think the way to go, and a good way to prove your "competence" at
design (not exactly how competent you are just your attention to detail
on all parts of your site, not just that visible to the customer), is
just to jump through the hoops to make sure your clients get the best
available design and coding.

Also I would recommend scrapping frontpage, as it really, truly sucks
in comparison to proper products :p.

Thanks to ABerry & Softplus for your constructive words ( also to
webado, who I admire for his stickability ), at last we're back on
track and able to discuss this sensibly.

A question that is getting to me is has anyone seen any proof data
online that can positively state that this validation makes a
difference to search engine positions. I can find a lot of opinions,
but when I go actually looking for test data I can never find anyone
who has run two sites off against each other to see if one climbs
better than the other. I'd certainly get validating if I knew that it
made a concrete difference to anything other than my piece of mind.

Also, is there a validation tool that will automatically validate pages
? I'd like to run it and see if my structures would be blitzed.

Cheers

Col ( I got banned from webmasterworld for fighting too !!! )

I'm not sure there is, as, like you say, it could be fairly
catastrophic, or even just a small change could make a hidden
functionality malfunction or simply die, I think you're probably better
off doing it by hand. Page by page on the same site you get some of the
same errors so it gets to a point where you can use "find and replace"
to fix it, which only takes about 10 seconds site-wide if you're using
a decent wysiwyg program, but, depending on how much your pages vary
from one another, you can get a variety of different errors popping up,
so it involves a more hands on approach. Don't take my word as gospel,
though, some genius somewhere or other may have created a program to do
it :p Have a look, google it!

Colin, I was'nt offended, no not at all. I can certainly indentify with
your position. Been there done that.

Webado, this lady is a gem. Someone well worth knowing. Webado was one
of the patient folks who encourage me in the "beginning" and she
continues to do so. She her self has gone through the same issues. Lol,
I can remember many a time her gnashing her teeth and crying the
travails of Validation not even to mention the differences between
browsers. She didn't use me in the sence you might think. I was quite
pleased when I followed the link to this thread I found in my stats.
Gee she thinks enough of my journey to promote it as an example.

Happy happy......And may you all have the happiest of holidays!

ABerry5555 said "Have a look, google it!"

lol, I'm never using Google again, they've got 63 errors in their front
page alone ;-0

All the Best

Col :-)

Shame, because you're flat wrong.

No browser cares tuppence for the content of many of the head metatags.
 But some things ( http://www.isham-research.co.uk/broken_metatag.html
) _DO_ break head parsing and cost you the effect of the meta tags that
follow.

I would be the last to say you need to be 100% clean in validation to
get a good SERPs position.  Manifestly not true - do any search you
like and validate the top three - usually all three (and never less
than two) will have errors.

But there are errors, and there are errors.  Gross syntax problems,
serious head errors, and then endless whining about "alt" keywords.

Every page should be put through a validator.  Period.  Most of the
reported problems can probably be ignored - but it's like having a
second opinion from a doctor.  And once a month or so the damn things
spot something I didn't see.  That makes it worth it.

Your HTML looks very clean - I think if there were blatant HTML errors
on your code that the robots would avoid it.  Overall very nice - think
about using more CSS in your site - it will simplify code and reduce
load time. IMHO most people who use Frontpage are web beginners, and
you obviously have a handle on code and web page design - switch to
Dreamweaver for your WYSIWYG editor and you will like it better once
you get to know it.  Oh, and I ran your site in the validator and it
came up with 16 errors.  Most of them are ALT tag missing errors - can
if really hurt to run it?