homepage Welcome to WebmasterWorld Guest from 107.21.163.227
register, free tools, login, search, subscribe, help, library, announcements, recent posts, open posts,
Pubcon Platinum Sponsor
Home / Forums Index / Google / Google News Archive
Forum Library, Charter, Moderator: open

Google News Archive Forum

This 96 message thread spans 4 pages: 96 ( [1] 2 3 4 > >     
heading 1 Tags - Still Relevant?
bekyed




msg:156110
 8:39 am on Jul 6, 2003 (gmt 0)

Are the heading 1 tags still relevant with google or shall we remove them from our websites completely
What are your feelings on this?

 

Marcia




msg:156111
 9:10 am on Jul 6, 2003 (gmt 0)

bekyed, the H1 tags are relevant to web page construction and are a standard element. There's no reason to stop using them as long as they're used the way they're intended to be.

SEO practioner




msg:156112
 2:05 pm on Jul 6, 2003 (gmt 0)

Bekeyd

H1's are some of the best tools you can use to make a strong site in the serps.

Good luck

Nick_W




msg:156113
 2:11 pm on Jul 6, 2003 (gmt 0)

We had a very long and very interesting discussion on this just a few weeks ago. Check it out here [webmasterworld.com]. ;)

Nick

g1smd




msg:156114
 2:38 pm on Jul 6, 2003 (gmt 0)

I have sites at #1 with Headings constructed just the way that W3C describes. Follow their rules, you can't go wrong.

annej




msg:156115
 4:15 pm on Jul 6, 2003 (gmt 0)

Some people here seemed to be worried about using CSS to adjust the size of the H1 tags there is no evidence that Google counts down for CSS resizing. I figure as long as your H1 is larger than your H2 and that is larger than your text you are just fine.

Having said this I would LOVE to get rid of my H1 tags as I have a lovely graphic with the same words that I would much rather use. I took it off when we had discussions on this several months ago. If there ever is proof that H1 tags no longer count I would return to the graphic heading.

SEO practioner




msg:156116
 4:44 pm on Jul 6, 2003 (gmt 0)

"If there ever is proof that H1 tags no longer count I would return to the graphic heading"

Annej, H1's are tops

g1smd




msg:156117
 4:48 pm on Jul 6, 2003 (gmt 0)

Umm, if Google has some sort of heading penalty then I doubt that it is triggered by the size that you render the text in. It would more likely be triggered by the positioning of the tag, and the way that you are using it, not how you style it.

.

If you have code like:

<h1>A Heading at the top of the page</h1>

<p>A load of words about some stuff which is at least vaguely relevant to whatever the heading says it is supposed to be about.</p>

<h2>Yet another heading</h2>

<p>A load more words about random stuff but hopefully somewhat in tune with the words in the heading directly above, as well as the one at the top of the page</p>

then you are using the tag as intended, and should fear nothing.

.

If you are doing this:

<p>A load of words about some stuff
<h1>keyword keyword keyword</h1> which is vaguely relevant to something or other but with
<h2>keyword keyword keyword</h2> keywords highlighted as being headings dumped in the
<h2>keyword keyword keyword</h2> text when they are not actually proper headings at all.</p>

then you might be on to something.


Like I said before; if you use heading tags simply for marking headings and you should be alright.

Martin Dunst




msg:156118
 5:24 pm on Jul 6, 2003 (gmt 0)

annej,

[...] I would LOVE to get rid of my H1 tags as I have a lovely graphic with the same words that I would much rather use.

you can of course have h1 _and_ graphics.

<h1><img src="level-1-headline.png" alt="This is a level 1 headline"></h1>

regards
Martin

heini




msg:156119
 5:31 pm on Jul 6, 2003 (gmt 0)

Martin, I have never used that, but I seriously doubt the alt text gets any added importance by the header tag. Apart from that being incorrect html.

Martin Dunst




msg:156120
 5:44 pm on Jul 6, 2003 (gmt 0)

heini,

[...] I seriously doubt the alt text gets any added importance by the header tag

according to the w3c, the alt attribute is for user agents that can't display images.
googlebot is just one of those user agents.
the above example makes perfect sense, especially when it comes to accessibility.

Apart from that being incorrect html.

you're right here - i should have added that it's an html fragment.
here's the correct version:

--
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

<html>
<title>This is a level 1 headline</title>
<meta http-equiv="content-type" content="text/html; charset=iso-8859-1">
<h1><img src="level-1-headline.png" alt="This is a level 1 headline"></h1>
</html>
--

scnr :)

regards
martin

Marcia




msg:156121
 5:58 pm on Jul 6, 2003 (gmt 0)

Martin, that really would presuppose that alt text is being read and scored the same way as ordinary text within the H1.

Can you point us to where in the W3C specs it says that using an image as H1 rather than text is standard acceptable practice?

ruserious




msg:156122
 6:22 pm on Jul 6, 2003 (gmt 0)

@Marcia: I agree that Google won't give an image the added weight, that one would expect for regular text-headings.

However in css-circles there is a practice called the fahrner image replacement, which is debated over wether it really enhances accessability, but with it you can serve well structured text with heading to agents that won't respect css and have images served instead of headings for more modern browsers. Problems arise when you have modern browsers with css, but images disabled, because you can't have any alt-text for css-background-images.

I don't know which category google falls into with this, but since the fahrner image replacement involves the display:none directive, it might be considered as spammy...

martinibuster




msg:156123
 6:41 pm on Jul 6, 2003 (gmt 0)

heading 1 Tags - Still Relevant?

Questions like this betray a fundamental misunderstanding of HTML markup, and the role each Element plays within the document structure.

The H1 element is used for building a hierarchical structure for the ideas you are presenting in a web site. It is not a hacking tool for "tricking" Google.

Are the heading 1 tags still relevant with google...

This is precisely the approach that will backfire on you- it demonstrates a lack of understanding about what Google wants.

Everyday there are folks moaning about being on page 9 of the serps, and that they applied alt tags to their images, used H1, etc. I have little -perhaps no sympathy- for these folks for I strongly suspect that they did not comprehend what they were doing.

My suggestion is to do your research at the w3c and you will come back to us saying, "Wow...I know Kung Fu."

Either that or hire a professional.

Martin Dunst




msg:156124
 6:54 pm on Jul 6, 2003 (gmt 0)

marcia,

Martin, that really would presuppose that alt text is being read and scored the same way as ordinary text within the H1.

correct.
the alt attribute holds an _alternate text equivalent_ for user agents that, for example, cannot render images.

no, i can't point you to a w3c line saying "having images within h1 elements is okay, its alt-text will be treated as regular h1-text by user agents".
and to be honest: i doubt that such a document exists.
counterquestion: can you point us to a w3c document saying that img within h1 is forbidden?
it's okay for the w3c html validator, thus it's syntactically correct.

when images are turned off in a graphical user agent, the alt-contents are displayed as text. alt-text within h1-elements is displayed in the big h1 font style.
screenreaders read the alt-text as they would read regular text.
lynx displays the alt-text as regular text - and it displays alt-text within h1 in the centered h1-style.

as for googlebot, i don't know how it handles alt-text of images within h*-elements.
i think there's only one person here at webmasterworld who can give an authoritative answer to this question :)

regards
martin

martinibuster




msg:156125
 7:02 pm on Jul 6, 2003 (gmt 0)

If I were Google I'd take Martin's scheme and flag it as an attempt to game the algo-

I agree with Marcia. My statements about some folks misunderstanding HTML elements apply equally to Martin's scheme.

albert




msg:156126
 7:14 pm on Jul 6, 2003 (gmt 0)

I would strongly recommend to follow martinibuster.

Understand - by reading w3c - what's intended by structuring pages with tags.

If you set up pages as intended, H1 tags are really helpful. But w/o images.

My 2 c.

SuzyUK




msg:156127
 7:18 pm on Jul 6, 2003 (gmt 0)

fahrner image replacement

now it's getting interesting :).. I've been waiting for this one for nearly a year and have been using above mentioned technique, before it even had a name, with no problems so far..

it's just another way of using CSS to style an <hn> tag, yes it uses display: none; but why would Google think this property "spammy"? perhaps the W3C should remove it from the specs?

I would sincerely hope there would be no problems with this as it is just part of the accessibility that CSS can add to a page, yes it has the capability to be misused but that's been the nature of the web so far and it's up to the individual to see if you can "trip" those filters if that's more important than building pages for the user

can an algo filter out when display: none; is appropriate or not? or does Google intend to take on the CSS specs too?

Yes it could throw up a flag for a manual review but that could be quite a lot of reviews... and as long as the text/image swap is legit then there should be no problems passing such a review anyway..

Suzy

soapystar




msg:156128
 7:28 pm on Jul 6, 2003 (gmt 0)

bekyed
what would make h1 irelevant is having

<h1>keyword1 keyword2 keyword3 keyword4 keyword5 keyword6 keyword7 keyword8 keyword9 keyword10 keyword11 keyword12 keyword13 keyword14</h1>

at best youve watered down any affect by having too many keywords, at worst you'll trip a filter!

tedster




msg:156129
 7:55 pm on Jul 6, 2003 (gmt 0)

We've had many discussions about this H1 issue -- and I think the root of many questions that come up is a lack of understanding of what HTML is and is not.

HTML is NOT a layout language, but we are so used to dealing with layout applications on a computer that we think HTML should be about how our information gets rendered. This is essentially a PRINT mindset, and not a WEB mindset.

HTML is about document structure, although you can still "force" HTML 4 to do some layout/rendering work. The separation of document structure and rendering information is what the movement to CSS and XHTML is all about. And Google is looking for the document structure in order to determine what a page is all about. Using text in one, clear H1 tag per page sends exactly that message.

It's been a while since I saw an reverse engineering analysis, but about a year ago, text in an H1 tag was 7 or 8 times more important than regular body text. And text in an alt attribute was about half as important as regular body text. That's 14 to 16 times different.

If you have some test domains you can play with you can experiment more, but the numbers have always been something like this.

Alt text for an image in an H1 tag is still alt text. Because of the potential for algorithm manipulation, it is much less heavily weighted than regular text -- and a WHOLE lot less important than a heading tag. If you think about HTML as markup that indicates document structure, then this makes plenty of sense. It's not some arbitrary decision by Google.

The actual weight for the elements in the Google algorithm has been tweaked since I saw those results (even with every update) -- and in fact, alt attribute text may be even less valuable than it was. But the principle that Google uses for their analysis has not changed, because it's welded tight to the nature of HTML itself.

tedster




msg:156130
 8:05 pm on Jul 6, 2003 (gmt 0)

why would Google think this property "spammy"?

Only if it becomes widely abused -- which it certainly can be if the hidden text doesn't match the visible graphic. You know what I mean -- the graphic says "you can look great" but the text say "product name, product name, product name"

Martin Dunst




msg:156131
 8:15 pm on Jul 6, 2003 (gmt 0)

martinibuster,

If I were Google I'd take Martin's scheme and flag it as an attempt to game the algo-

just to make sure i get it right:
you would flag an img inside an h1 as an attempt to cheat on google?
what about an img inside a p element then?
should this be considered cheating/spamming, even though the alt text holds important information e.g. for blind web users?
should user agents (and googlebot is just one amongst many) rely on alternate content at all?
or should it rather be silently ignored, for it might contain fake information?

I agree with Marcia. My statements about some folks misunderstanding HTML elements apply equally to Martin's scheme.

in your opinion, which element is misunderstood when using
<h1><img src="level-1-headline.png" alt="This is a level 1 headline"></h1>
(assume that the image has graphical text saying "This is a level 1 headline")

tedster,

Alt text for an image in an H1 tag is still alt text.

this suggests that alt-text is meant to be less important than "normal" text.
i can't agree here, because alt-text is meant to be a text equivalent, as stated in section 2.1 (Guideline 1. Provide equivalent alternatives to auditory and visual content. [w3.org]) of Techniques for Web Content Accessibility Guidelines 1.0 [w3.org]

regards
martin

Marcia




msg:156132
 8:15 pm on Jul 6, 2003 (gmt 0)

>>>Guideline 1. Provide equivalent alternatives to auditory and visual

Exactly - but <h1>...</h1> is markup intended to provide a hierarchical size measure of TEXT to indicate relative importance as compared to <p> and <hn> that are lesser. The intention has never been stated as an enclosure for graphics. It can easily happen accidentally in WYSIWYG editors like Dreamweaver, BUT - it's the TEXT that would get a boost for relevance by being marked up with the <h1>. An alt is an alt is an alt - it is NOT text that's marked up by an HTML tag. Can Google tell the difference between text on a page and the alt attribute? Yes.

soapystar is right, all that keyword stuffing like that in <h> tags does is dilute.

My feeling is that unless someone is a seasoned optimizer who's been around the block more than a few times, can accurately assess relative risk and is willing to take calculated risks, they'd best stay with normal accepted practice and keep away from any--ummm... let's call them "creative optimization techniques."

Those folks would also be well advised to be very careful about advising others to stray from the norm as well. And for those of us who delight in hearing new things, as many of us do, we'd best be very careful about implementing them without considering all possibilities carefully.

For example, the alt attribute is perfectly legitimate. But how about if a 3px graphical horizontal rule has keywords in the alt attribute? How about having an all graphics page and stuffing alt attributes with paragraphs of text - and sticking an Hn or two in there for good measure?

>>can an algo filter out when display: none; is appropriate or not? or does Google intend to take on the CSS specs too?

We don't know Suzy, and there's nothing to stop them from doing it. Not that there's anything wrong with display: none; if it's used as it's intended, but even if Google doesn't filter it out algorithmically, if a site ranks real well in a category that's even near competitive, some poor soul someplace will get upset because someone else's better spam is beating out their less effective spam, come whining to us here, and trot off to the Google spam report to whine to GoogleGuy.

If it's reported and gets scrutinized by one of Google's staff people - let's say it's a woman who hasn't had her daily ration of chocolate - and they see something that isn't normal run of the mill, they're liable to go checking through that site with a fine tooth comb like they're hunting for cooties to see what else is under the hood there. Just like other webmasters and seo's do. Tell the truth now, don't we?

I was just thinking about these very elements and attributes earlier this morning and posted a question that's been sitting out there for a while.

[webmasterworld.com...]

Sometimes re-examining these basics can be refreshing when we've taken them for granted and haven't looked them through for a while.

tedster




msg:156133
 8:48 pm on Jul 6, 2003 (gmt 0)

this suggests that alt-text is meant to be less important than "normal" text.
i can't agree here, because alt-text is meant to be a text equivalent

Yes, that's what is intended in the W3C spec. But there has been so much abuse of the alt attribute (see Marcia's examples above) that search engines devalued that text in self defense. This is a place where search engine pragmatism over-ruled theoretical approaches.

Things "could" shift back -- but they won't as long as keyword stuffing is so prevalent.

Chris_R




msg:156134
 9:08 pm on Jul 6, 2003 (gmt 0)

this suggests that alt-text is meant to be less important than "normal" text.
i can't agree here, because alt-text is meant to be a text equivalent

There is a lot of talk about standards and structure - and all of that is meaningless as far as google is concerned.

Most users don't read the alt text and many web authors don't use it - or use it "incorrectly".

Google is designed for users. If the alt attribute included in their algos would lead to more relevant results - they'd be foolish not to use it.

<h1>, <h2> - whatever - google has used in the past - and I presume is using now. Back in the very early days - they weren't stupid and converted everything to relative sizes. I know H1 isn't a "size", but we all know it is in reality.

PUT:
BIG STUFF
Lots of little stuff.

Google will figure it out - if you put lots of big stuff - google will assume it is normal size - as that is what is running through your document. If they could figure that out when they were working with logos - there is no reason to assume they can't with dozens of PhDs running around.

As others have said - lay out your documents in a logical fashion. All else is secondary....

Martin Dunst




msg:156135
 9:09 pm on Jul 6, 2003 (gmt 0)

marcia,

first of all, i'd like to say that i'm far from being what is called an seo. i'm an html author, no more.

being in a google forum, it's obvious that everybody is focused on a client called googlebot and how google will analyze the data it collects.
though in this very question, google shouldn't be the main focus.

google say that authors should ask themselves:
"Does this help my users? Would I do this if search engines didn't exist?" (Google Information for Webmasters [google.com]).
in this specific case, namely img within h1, it would be a clear yes for both questions.

1.) would it help my users?
yes, it can do so.
for example, i can use a font style in the image that does not exist on the client computer. still, this font style might help give the right visual mood to the page.
on the other hand, users with non-graphical user agents won't get hurt, for there is the alt text providing the text equivalent. non-graphical user agents (or those not displaying images because of user preference) won't even request the image, thus they won't have any extra traffic or loading time.

2.) would i do this if search engines didn't exist?
yes, why not?
content is accessible to all clients, looks "better" on graphical clients, markup is valid, presentation transforms smoothly to the client's settings.

i don't know what google thinks of all this.
that's why i say that i wouldn't hesitate to use this "technique".
i wouldn't even consider it a technique, i'd say it's valid html.

at this point, the google self-test get's turned around:
"should i do without this html construct, BECAUSE search engines exist and might consider it an attempt to cheat?"
the answer is up to each author.

tedster,

Things "could" shift back -- but they won't as long as keyword stuffing is so prevalent.

for the same reason, we could all make invalid table-layouts, just because old netscape4 produces better visual results this way than with strictly separated markup/css.

knowing what search engines can do, this might be a risky approach. that's why every author should make their own decision.
but it sure ain't a mis-interpretation of the markup concept or a "evil technique" per se.

regards
martin

Chris_R




msg:156136
 9:16 pm on Jul 6, 2003 (gmt 0)

Search Engines are designed to index text.

You can't EXPECT search engines to index stuff that the user can't easily cut and paste onto their clip board.

I don't think you are going to get banned for it, just not anything I would remotely consider. People are free to design webpages any way they want. Google can't be EXPECTED to index things a certain way, because someone's interpretation of a standard says so.

SuzyUK




msg:156137
 9:33 pm on Jul 6, 2003 (gmt 0)

Search Engines are designed to index text.

Which brings me to my "off my soapbox" point..

<h>..</h> elements probably will (imho) receive no greater (google) weighting than ordinary text. It's not that they're not relevant, they are structure. But the ways to "abuse" them will ultimately outweigh their "relevancy" and the only filter that will successfully remove the abuse is to remove the weighting.

Hey I don't think it's fair either!

Suzy

heini




msg:156138
 9:47 pm on Jul 6, 2003 (gmt 0)

I don't think it makes much sense to devaluate header tags. You might as well devaluate text.
Not that a certain well known search engine didn't think at one point that would be the way to go: put all weight on external factors, forget what's on the page, those pesky webmasters cheat on us anyway :)
There is a reason why in said SE lots of pages rank highly even though they essentially have no content at all.

No, SEs have to analyze docs and try to determine their relevancy to queries. It doesn't make sense to disregard the content and it's structure in such an analysis.
Disregarding headers would be like saying: okay, this page uses the word widgets 4 times, lets disregard that word in our analysis.

shep_man




msg:156139
 9:48 pm on Jul 6, 2003 (gmt 0)

I just had my first good luck with H1 tags. A few weeks back I finially figured out how to use CSS and went thru my site adding H1 and H2 titles. This morning I was checking my SERPs and was suprised to see my site jump from 4 to 2 on one of my main keywords. I checked the cache and sure enough my page with H1 and H2 tags showed up for the first time. It was especially sweet because I knew the webmaster that I had passed up. He knew I was catching up and I found out he was tweaking his site to stay ahead of me.

This 96 message thread spans 4 pages: 96 ( [1] 2 3 4 > >
Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google News Archive
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About
© Webmaster World 1996-2014 all rights reserved