Re: SEO theories

Gazing into my crystal ball I observed John Dunlop <usenet+2004
@john.dunlop.name> writing in news:MPG.1d34d5ca42feed829896b4
@news.ntlworld.com:

I heard somebody say most of the 'SEO theories' are a 'big
hoax'.

Comments?

Here's my take on SEO:

1. Realise that search engine bots are machines. They do not have eyes,
so they can't see a great Flash presentation. They don't have ears, so
great music is not going to sway them. They cannot submit forms, so form
based drop down navigation is lost on them. They do not have javascript
or any other kind of client side plug in available. All they see is the
plain text.

2. Write semantically correct markup without any presentational markup.
Save the presentation for CSS. Use external javascripts to enhance the
site, not for anything crutial like navigation.

3. Try not to use tables for layout. That's more markup for the bot to
parse, and more chance of an error. Tables should be used for tabular
data.

4. Validate to W3 standards, and fix the errors the validator comes up
with. Use a Strict doctype, which will force you to not use
presenational markup.

5. Try do have a domain name that has something do to with what you are
trying to promote. If not for the bots, at least for people who might
remember it.

6. For dynamic pages, don't use ?id= because the bot will think it's
going to get into an endless loop (it might), and it will go away.

--
Adrienne Boswell
http://www.cavalcade-of-coding.info
Please respond to the group so others can share

Adrienne <arbpen2003 (AT) sbcglobal (DOT) net> wrote:

Here's my take on SEO:

1. Realise that search engine bots are machines. They do not have
eyes, so they can't see a great Flash presentation.

But they should be able to extract readable text from a Flash
presentation. And links.

They don't have
ears, so great music is not going to sway them. They cannot submit
forms, so form based drop down navigation is lost on them.

Technically it's possible :-)

They do
not have javascript or any other kind of client side plug in
available.

Yet there shouldn't be a problem to extract the URL from:

.... onClick="window.open( 'http://example.com/' )" ...

Technically it's possible but interpreting JavaScript takes too much
resources.

All they see is the plain text.

And images ;-)

3. Try not to use tables for layout. That's more markup for the bot to
parse, and more chance of an error. Tables should be used for tabular
data.

Hence validate. But I agree that tables should be used as little as
possible (if at all) for layout.

4. Validate to W3 standards, and fix the errors the validator comes up
with. Use a Strict doctype, which will force you to not use
presenational markup.

Hello SEO Dave :-D.

6. For dynamic pages, don't use ?id= because the bot will think it's
going to get into an endless loop (it might), and it will go away.

Google has no problem with ?id=. Moreover, if you cause endless loops by
parameters, you are not restricted to id. And hence Google is probably
able to detect such loops.

Again: google can't see if a page is dynamic or static, and hence uses
heuristics that use the look of an URL. If google might consider to
spider dynamic pages slower in order to lessen the load on "dynamic"
sites (database queries etc.)

--
John Perl SEO tools: http://johnbokma.com/perl/
Experienced (web) developer: http://castleamber.com/
Get a SEO report of your site for just 100 USD:
http://johnbokma.com/websitedesign/seo-expert-help.html

"Adrienne" <arbpen2003 (AT) sbcglobal (DOT) net> wrote

Gazing into my crystal ball I observed John Dunlop <usenet+2004
@john.dunlop.name> writing in news:MPG.1d34d5ca42feed829896b4
@news.ntlworld.com:

I heard somebody say most of the 'SEO theories' are a 'big
hoax'.

Comments?


Here's my take on SEO:

1. Realise that search engine bots are machines. They do not have eyes,
so they can't see a great Flash presentation. They don't have ears, so
great music is not going to sway them. They cannot submit forms, so form
based drop down navigation is lost on them. They do not have javascript
or any other kind of client side plug in available. All they see is the
plain text.

2. Write semantically correct markup without any presentational markup.
Save the presentation for CSS. Use external javascripts to enhance the
site, not for anything crutial like navigation.

3. Try not to use tables for layout. That's more markup for the bot to
parse, and more chance of an error. Tables should be used for tabular
data.

4. Validate to W3 standards, and fix the errors the validator comes up
with. Use a Strict doctype, which will force you to not use
presenational markup.

5. Try do have a domain name that has something do to with what you are
trying to promote. If not for the bots, at least for people who might
remember it.

6. For dynamic pages, don't use ?id= because the bot will think it's
going to get into an endless loop (it might), and it will go away.

Not one of these will get you to the #1 spot in the SERP. A site can have
all these and be missing the part that is needed to be above the rest.
Links, keyword density, proximity, the right keywords to gain the best
SERPs, headings, titles, etc.

Stacey

On Tue, 05 Jul 2005 21:37:16 GMT, Adrienne <arbpen2003 (AT) sbcglobal (DOT) net>
wrote:

Gazing into my crystal ball I observed John Dunlop <usenet+2004
@john.dunlop.name> writing in news:MPG.1d34d5ca42feed829896b4
@news.ntlworld.com:

I heard somebody say most of the 'SEO theories' are a 'big
hoax'.

Comments?


Here's my take on SEO:

1. Realise that search engine bots are machines. They do not have eyes,
so they can't see a great Flash presentation. They don't have ears, so
great music is not going to sway them. They cannot submit forms, so form
based drop down navigation is lost on them. They do not have javascript
or any other kind of client side plug in available. All they see is the
plain text.

Generally true, there are exceptions like alt attributes, but for the
most part it's just the text that counts.

2. Write semantically correct markup without any presentational markup.
Save the presentation for CSS. Use external javascripts to enhance the
site, not for anything crutial like navigation.

I've seen no evidence or anything suggestive that presentational
markup has a negative effect on a pages SERPs. Been meaning to do some
tests, but not got around to it.

3. Try not to use tables for layout. That's more markup for the bot to
parse, and more chance of an error. Tables should be used for tabular
data.

For non SEO reasons I agree, but again tables are used extensively for
layout, I see no reason why this could be treated negatively by the
search engines. Why would they?

BTW I don't use tables for layout, everything I create is tables free
(unless tables are appropriate, or I'm feeling lazy :-)). It's good
advice to move away from tables for layout.

4. Validate to W3 standards, and fix the errors the validator comes up
with. Use a Strict doctype, which will force you to not use
presenational markup.

There's no evidence this helps again.

Interesting how this is becoming standard SEO advice on this NG
though.

It's like blind sheep following other blind sheep down the wrong path,
fortunately there's no big bad wolf on this path, just a bit longer
than the right path, so all you are doing is wasting time taking the
scenic route with the other sheep :-)

5. Try do have a domain name that has something do to with what you are
trying to promote. If not for the bots, at least for people who might
remember it.

Yes for the bots, hyphenate words, it helps a little.

6. For dynamic pages, don't use ?id= because the bot will think it's
going to get into an endless loop (it might), and it will go away.

Not true, the problem lies in the way sites with session IDs
constantly change their URLs as the bots return not the actual URL
structure. Every time a bot visits it gets a new set of links, for
this reason pages are indexed, but quickly dropped as new pages are
indexed and quickly dropped. The bots never get the same link twice!!

So don't use session IDs or at least disable it for bots.

David
--
Free Search Engine Optimization Tutorial
http://www.seo-gold.com/tutorial/

SEO Dave <seodave (AT) search-engine-optimization-services (DOT) co.uk> wrote:

On Tue, 05 Jul 2005 21:37:16 GMT, Adrienne <arbpen2003 (AT) sbcglobal (DOT) net
wrote:

2. Write semantically correct markup without any presentational
markup. Save the presentation for CSS. Use external javascripts to
enhance the site, not for anything crutial like navigation.

I've seen no evidence or anything suggestive that presentational
markup has a negative effect on a pages SERPs. Been meaning to do some
tests, but not got around to it.

Replace all your <h1>...</h1> (semantically correct mark up) by
<font size="4"><b>...</b></font> (visual mark up).

3. Try not to use tables for layout. That's more markup for the bot to
parse, and more chance of an error. Tables should be used for tabular
data.

For non SEO reasons I agree, but again tables are used extensively for
layout, I see no reason why this could be treated negatively by the
search engines. Why would they?

SE's will move and move more towards XML, and hence use the
*semantically* meaning of elements, hence a table could be regarded as
different content compared to a column of text in a table cell.

4. Validate to W3 standards, and fix the errors the validator comes up
with. Use a Strict doctype, which will force you to not use
presenational markup.

There's no evidence this helps again.

You gave some time an example which according to you might harm the SEO
efforts on a page. Again (until you get it), it's possible to have on
page HTML errors that *harm* SEO.

A validated page doesn't rank better per se, but you can be quite sure
that the parser bots use it get it right. Since you seems to get shocked
if I compare this with using a condom when having sex (odd from someone
who sold sex toys and lingerie), probably you understand the following:
Yoda hard to follow is, and when Yoda speeks he does, misunderstanding
occur might.

Interesting how this is becoming standard SEO advice on this NG
though.

More interesting is that the people who say it ain't so are the same
people who have hard times at understanding basic HTML and fixing very
basic validation problems...

It's like blind sheep following other blind sheep down the wrong path,
fortunately there's no big bad wolf on this path,

There is Dave, as you even admitted some time ago: code that does not
validate, but might harm the on page SEO efforts. You were able to come
up with an example, and if you understand HTML a bit more, you might
come up with many more ;-)

--
John Perl SEO tools: http://johnbokma.com/perl/
Experienced (web) developer: http://castleamber.com/
Get a SEO report of your site for just 100 USD:
http://johnbokma.com/websitedesign/seo-expert-help.html

SEO Dave <seodave (AT) search-engine-optimization-services (DOT) co.uk> wrote:

[ validation ]

Interesting how this is becoming standard SEO advice on this NG
though.

Even more interesting is how you confirm my own ramblings of last weekend
time after time... Adrienne seems to be into the SEO business...

It's like blind sheep following other blind sheep down the wrong path,

Or like the adult webshop (failure) owner starting a lingerie site
(failure?), and does some blog spamming..., and uses the result to stuff as
much as easy to get content in Google as possible, and preaching it's the
best way to do things (*cough*).

Claiming to make loads of money, yet regularly peeing in
alt.internet.search-egines like a tomcat to keep others out of his
territory ( 70 GBP / month customers? ) :-D

--
John Perl SEO tools: http://johnbokma.com/perl/
Experienced (web) developer: http://castleamber.com/
Get a SEO report of your site for just 100 USD:
http://johnbokma.com/websitedesign/seo-expert-help.html

On Wed, 06 Jul 2005 01:27:17 GMT, SEO Dave
<seodave (AT) search-engine-optimization-services (DOT) co.uk> wrote:

On Tue, 05 Jul 2005 21:37:16 GMT, Adrienne <arbpen2003 (AT) sbcglobal (DOT) net
wrote:

Gazing into my crystal ball I observed John Dunlop <usenet+2004
@john.dunlop.name> writing in news:MPG.1d34d5ca42feed829896b4
@news.ntlworld.com:

I heard somebody say most of the 'SEO theories' are a 'big
hoax'.

Comments?


Here's my take on SEO:

1. Realise that search engine bots are machines. They do not have eyes,
so they can't see a great Flash presentation. They don't have ears, so
great music is not going to sway them. They cannot submit forms, so form
based drop down navigation is lost on them. They do not have javascript
or any other kind of client side plug in available. All they see is the
plain text.


Generally true, there are exceptions like alt attributes, but for the
most part it's just the text that counts.

2. Write semantically correct markup without any presentational markup.
Save the presentation for CSS. Use external javascripts to enhance the
site, not for anything crutial like navigation.

I've seen no evidence or anything suggestive that presentational
markup has a negative effect on a pages SERPs. Been meaning to do some
tests, but not got around to it.

3. Try not to use tables for layout. That's more markup for the bot to
parse, and more chance of an error. Tables should be used for tabular
data.

For non SEO reasons I agree, but again tables are used extensively for
layout, I see no reason why this could be treated negatively by the
search engines. Why would they?

BTW I don't use tables for layout, everything I create is tables free
(unless tables are appropriate, or I'm feeling lazy :-)). It's good
advice to move away from tables for layout.

4. Validate to W3 standards, and fix the errors the validator comes up
with. Use a Strict doctype, which will force you to not use
presenational markup.

There's no evidence this helps again.

Interesting how this is becoming standard SEO advice on this NG
though.

and this only how long after, beginning with the Yahoo Directory,
Yahoo, Google and MSN have all stated in their guidelines that they
prefer sites to be validated? Amazing how quick some folk catch on. Or
not.

Mind, we might open debate here about why the engines haven't said
anything either way (to my knowledge - anybody know different?) about
preference given to either CSS for layout or tables. My guess would be
that, so long as a site validates, they can equally happily read the
content in both.

BB
--
www.kruse.co.uk/ seo (AT) kruse (DOT) demon.co.uk
seo that watches the river flow...
--

John Dunlop wrote:

Would you agree those six pieces of advice (with the possible
exception of the last) are a subset of the superset of
authoring guidelines?

There is no either / or Jock. Think about it, what does a Search Engine
Robot have to go on? The content of the page, the structure of the site,
the links coming to a page, their anchor text and certain bits of meta
data such as the age of the domain, link creation rates etc.

If we take the content then an indexer has some W3C standard that it can
use to try and understand the content of a page - I'm assuming that
semantic analysis is still way beyond the computational power of search
engines. So it uses a mixture of headings, title, bold text, citations
etc etc. Now if a website owner choses to ignore standards - authoring
guidelines if you will they will also create a page that is potentially
(depending on the skill of the algorithm designer) less optimum. At this
level SEO is saying "this is what a robot expects to find on a page,
this is how it expects a site to be linked".

Unfortunately a great many webmasters still ignore how HTML should be
used (I'm not even talking about standards compliance here). They still
persist in using H4 for H1 level headings because "it looks nice" or
links that say "click here" or pages that are 100k long etc. Their sites
suffer for it because they provide a poor user experience, no matter how
good the content.

Your original claim that "most SEO theories are a hoax" doesn't make a
lot of sense. A theorem is an assumption based on a set of limited
information or knowledge. For example anchor text and inbound links is a
theory for ranking high in Google for the keywords in the anchor text,
this is based on the observation that websites with a lot of inbound
links with certain anchor text rank highly for those keywords. Another
theory is that the Title element is very significant in non-competitive
SERPS. If you take the trouble to look back over the archives of this
group rather than throwing 8th grade philosophy you will find a number
of examples of SEO theories that are not hoaxes. What your 'friend'
heard elsewhere isn't really of much concern here unless you can provide
some examples.

----------------------
http://www.abcseo.com/

On Wed, 06 Jul 2005 12:49:56 +0200, davidof
<david.george (AT) g-dumpthisbit-mail (DOT) com> wrote:

John Dunlop wrote:
Would you agree those six pieces of advice (with the possible
exception of the last) are a subset of the superset of
authoring guidelines?

There is no either / or Jock. Think about it, what does a Search Engine
Robot have to go on? The content of the page, the structure of the site,
the links coming to a page, their anchor text and certain bits of meta
data such as the age of the domain, link creation rates etc.

Hey, I could have written that.
Why didn't I?
Well, no matter.
I'll write it some day!

BB

--
www.kruse.co.uk/ seo (AT) kruse (DOT) demon.co.uk
seo that watches the river flow...
--

"John Bokma" <john (AT) castleamber (DOT) com> wrote

Adrienne <arbpen2003 (AT) sbcglobal (DOT) net> wrote:

snip

6. For dynamic pages, don't use ?id= because the bot will think it's
going to get into an endless loop (it might), and it will go away.

Google has no problem with ?id=. Moreover, if you cause endless loops by
parameters, you are not restricted to id. And hence Google is probably
able to detect such loops.

Somebody should read the Google guidelines more often.
"Don't use "&id=" as a parameter in your URLs, as we don't include these
pages in our index. "
http://www.google.com/intl/en/webmas...uidelines.html

I wouldn't use them as Google states they will not include them in the
index.

Stacey