32 replies to this topic

[reddevil]

I am in the middle of designing a new site layout and keep checking my W3C compliance which gives 40+ error messages. These are because I do not have any ALT tags for the .jpg bullets in my menus and titles and really don;t want to add all that extra html just for ALT tags.
Is this a reasonable choice to make?

Also, will 40+ W3C non-compliances affect my site in any way, especially the SERPS?? Therefore, why would the Validator say I am non-compliant with such a minor issue?

[qwerty]

It won't affect your SERPs, but you could just put alt="" in there if they're just bullets. In fact, you could set that image to be used in your <li>s, and then you'll have even less code, because you won't have to call the image, and even the W3C won't expect an alt attribute on it.

[reddevil]

Thanks for the No.1 Top Tip qwerty.

Wow, reduced code - that's just what the doctor ordered.

I'm a happy chappy.

[Matt B]

A Validator does nothing more than gauge the "valid use of code" according to the W3C. Alt attributes are a part of valid mark-up. Every image needs to have an alt attribute for user-agents who do not use graphical browsers to understand what is in place.

Even if the alt attribute is blank, it signifies that the image is not part of the information on the page, a spacer or sone other incedental design element. If you have graphics that are vital to explaning the purpose of the information, an image link or any other type of information (such as a graph, chart, etc) - it is critical to include that attribute information. The SE's don't account for compliance in their ranking algo's, but, it is a good goal to pursue. Best of luck to you.

[reddevil]

Cheers Matt, I will rest a little easier knowing that the SE's won't penalise me for non-compliant code, but it is always nice to get that successful validation message. We aim for perfection but have to settle for less (haha).

[Matt B]

There have been some testimonies of better rankings because of validated code, most of which seem anecdotal at best. My opinion is that the validation test finds mis-marked and open tags, when corrected, it can help a spider gather more content; thus, better rankigns.

Tim Mayer at Yahoo stated unequivically that validation is not a concern at Yahoo. Now, it's not to say that may not happenn in the future. However, if it ever would, i think you would see thousands of webmasters scrambling to learn CSS.

[mcanerin]

I agree that validation isn't the issue - an easy to spider page is. If validation was an issue then search engines would not index things like PDF's, word docs, text files, etc - none of which validate at all.

Further, let's say you make a perfectly valid HTML 4 strict page. If you changed the doctype and ran it through the validator as XHTML, it would fail miserably. Yet the code itself is completely fine and spiderable.

I do recommend validating, but mainly because it's an easy way to find real mistakes, and also gives you a warm feeling of accomplishment and professionalism.

Most of my own site validates, but there are a couple of pages that don't (typically ones with forms, or scripts on them). I'm not worried about it, and I've never seen any indication of a lack of a ranking ability for any of them. They all work fine in a text-only browser, and that's the only validator for a search engine that really matters.

Ian

[Michael Martinez]

If you contract with any organizations that require W3 compliance, this may become an issue. You'll just have to wait and see. I think it would depend on how much common sense the other side has. Compliance, however, is usually key with government entities that have to watch standards for accessibility.

[reddevil]

However, if it ever would, i think you would see thousands of webmasters scrambling to learn CSS.

Why to learn css? I presume that normal html would validate too?

[hooperman]

Hi,

this is my first post! Could all those .jpg bullets help you rank better if they were to be renamed something like:

"dog-trainer.jpg", and used with alt="dog trainer blah blah"

if you wanted to be found for the phrase "dog trainer"?

[qwerty]

Welcome, hooperman

No, that's the sort of thing that used to be done a lot, and I don't know if it ever worked, but it doesn't now. For one thing, alt attributes don't seem to be very important from an SEO perspective unless they're on images that anchor links. Now, if all of those images were links, they'd either all be pointing to the same page, and only the first one would matter to search engines, or they'd be pointing to different pages, in which case they'd just be competing with each other for relevance to the keyword phrase, so that wouldn't help either.

And all of that is in addition to the fact that it's just not what the attribute is for. If it's a picture of a dog trainer, then it makes sense for that to be its alt attribute. If it isn't, it's simply incorrect.

Why to learn css? I presume that normal html would validate too?

That depends. If you're in standards-compliant mode (that is, if the page has a proper document type declaration), then there isn't a lot of formatting you could do with the page without CSS and still have it validate. After all, you wouldn't even be able to use a <font> tag.


Oh... one other thing: in case anyone isn't familiar with how to use your own image for bulleted lists, as I suggested in my first post in this thread, it's really easy. Let's say the image you want to use is at /images/bullet.gif. Put the following in your CSS:

CODE

li {
list-style-image: url('/images/bullet.gif');
]

You may need to fiddle with margins and padding to get it positioned properly, but that's really all there is to it, and it means you'll be replacing

CODE

<span><img src="/images/bullet.gif" alt="" width="5" height="5">some text</span><br>

with

CODE

<li>some text</li>

Edited by qwerty, 09 September 2005 - 07:52 AM.

[Jill]

Hi,

this is my first post! Could all those .jpg bullets help you rank better if they were to be renamed something like:

"dog-trainer.jpg", and used with alt="dog trainer blah blah"

if you wanted to be found for the phrase "dog trainer"?

No, that would be a complete misuse of them, since your bullet isn't actually an image of a dog trainer. (Presumably!)

[Catz]

If you want your pages to validate, every image does need an alt tag and it should describe the graphic at hand rather than being used for alt tag stuffing, an old SEO trick that could now get you in trouble if abused.

For accessibility to those users with special equipment, such as readers for the blind, it helps them know what the image is, if you have blank spaces, alt="blank space" or even just blank would be fine.

If you're in standards-compliant mode (that is, if the page has a proper document type declaration), then there isn't a lot of formatting you could do with the page without CSS and still have it validate. After all, you wouldn't even be able to use a <font> tag.

If you use the DTD for HTML 4.01 Transitional you can use the font tag and still validate. One tag you can't use is the center tag, so learning a little CSS to do that is worthwhile.

Having just worked on some W3C validation for an e-commerce site with font tags throughout, it validated fine and the font tags were not an issue.

[JohneeMac]

Having just worked on some W3C validation for an e-commerce site with font tags throughout, it validated fine and the font tags were not an issue.

You shouldnt be using font tags,maybe this is because the ecommerce package has them built in, but personally i wouldnt touch a package like that.

It goes against everything that the web is trying to transform into.

[Catz]

There are many older sites on the web, especially in e-commerce. Most coded in HTML using font tags rather than XHTML using CSS. If you get the opportunity to work on a successful e-commerce site that has been around for many years, you will be working with font tags.

This is not due to any e-commerce package, this is due to the fact that people originally hand coded their pages when font tags were the norm, or created sites with programs such as Dreamweaver, many years ago, which also used font tags.

Just because the web is trying to transform does not mean that all these sites are no longer valid. They show up, they function fine, they can be viewed on common browsers and they bring in millions of dollars in sales each year.

Amazon and eBay use font tags in their code, Yahoo and Google use font tags in their code, even highrankings.com and searchenginewatch use the font tag on their home page. Just view source on any of these pages in the body and you will find the font tag being used.

Perhaps this is a good way to see how much experience people actually have in coding, how long they have been at this. Most (though not all) fighting the font tag are relatively new to the game. There is nothing wrong with change but you can't try to force your way on everyone.

The cost to transform these sites cannot be justified, which is why HTML 4.01 Transitional is still a valid document type. W3C DTD's are offered in both HTML and XHTML because there are so many businesses set up that have been online, coded in HTML, for many, many years.