Hosted ecommerce platform contains 79 xhtml errors - is this acceptable?

[movietub]

Take a look at this report:

http://validator.w3.org/check?uri=http%3A%2F%2Fwww.brightpearl.co.uk&charset=%28detect+automatically%29&doctype=Inline&group=0&user-agent=W3C_Validator%2F1.1

This is error report for our eommerce platforms home page. As their homepage is built on the the same platform as all the ecommerce (and other type) sites hosted by them, we and all their customers inherit the same 74 errors on our sites.

This is very frustrating as we just went through our own markup to ensure we are 100% xhtml 1.0 strict, and may as well have not bothered!

We are trying to ensure valid mark-up as recent tests by w3s revealed that Google does indeed reward valid markup with higher rankings.

 

[Toni Anicic]

Take a look at this report:

http://validator.w3.org/check?uri=http%3A%2F%2Fwww.brightpearl.co.uk&charset=%28detect+automatically%29&doctype=Inline&group=0&user-agent=W3C_Validator%2F1.1

This is error report for our eommerce platforms home page. As their homepage is built on the the same platform as all the ecommerce (and other type) sites hosted by them, we and all their customers inherit the same 74 errors on our sites.

This is very frustrating as we just went through our own markup to ensure we are 100% xhtml 1.0 strict, and may as well have not bothered!

We are trying to ensure valid mark-up as recent tests by w3s revealed that Google does indeed reward valid markup with higher rankings.

No, Google does NOT award w3 valid websites, if anything they could penalize you for having a valid website (long story, watch official explanation: http://www.youtube.com/watch?v=FPBACTS-tyg )

Anyway, 74 errors is kind of concerning code quality :S

 

[NextPoint]

The HTML looks fine to me. W3C validator often says that there are more errors than actually exist, so I wouldn't worry about it - correcting one issue like putting a forward slash within <br> can correct multiple errors that the validator reports; the most anal thing about this is that a lot of these so called errors don't make any difference as to how browsers and devices show the website content. It's better to focus more of your time on making sure that your website shows correctly on all major browsers - something that making your website W3C compliant doesn't guarantee at all, which defies the point of having a website that is accessible.

 

[quikshop]

We are trying to ensure valid mark-up as recent tests by w3s revealed that Google does indeed reward valid markup with higher rankings.

Its a reasonable aspiration to have 100% standards-compliant code but it has no positive impact on SERPS.

 

[movietub]

No, Google does NOT award w3 valid websites, if anything they could penalize you for having a valid website (long story, watch official explanation: http://www.youtube.com/watch?v=FPBACTS-tyg )

Anyway, 74 errors is kind of concerning code quality :S

This test disagrees :

http://www.hobo-web.co.uk/seo-blog/official-google-prefers-valid-html-css/ and I would say their method is fairly 'real world'.

But anyway, as you say, not ideal is it!?

Seems many of the errors could be fixed simply by not declaring their markup as 'strict' in the first place.

Why would anyone declare a markup standard they know they won't pass :|

 

[Toni Anicic]

Are you seriously trying to tell me that I should trust the results of a SEO company that is trying to sell their services over the official word of Google that tells us the opposite?

 

[movietub]

The HTML looks fine to me. W3C validator often says that there are more errors than actually exist, so I wouldn't worry about it - correcting one issue like putting a forward slash within <br> can correct multiple errors that the validator reports; the most anal thing about this is that a lot of these so called errors don't make any difference as to how browsers and devices show the website content. It's better to focus more of your time on making sure that your website shows correctly on all major browsers - something that making your website W3C compliant doesn't guarantee at all, which defies the point of having a website that is accessible.

Most of the errors are seperate, although similar in nature. Although I do appreciate that 74 errors will amost certainly not require 74 fixes to solve.

Valid code does not ensure correct display accross browsers, but valid code does tend to make hacking older browsers more straight forward. Invalid code is handled wildly by some browsers, whearas valid code normally produces mostly 'in order and reasonable' results.

The same does not apply to CSS, which is almost impossible to validate if you want to support most browsers back to IE6. But then CSS is less important than valid xhtml (imo).

 

[movietub]

Are you seriously trying to tell me that I should trust the results of a SEO company that is trying to sell their services over the official word of Google that tells us the opposite?

Just saying that until the debate is ended by someone producing firm evidence, then valid markup is the safest option.

And as has been said, whats the excuse for invalid markup? surely a platform to build a website on should itself clean coded?

 

[Toni Anicic]

Just saying that until the debate is ended by someone producing firm evidence, then valid markup is the safest option.

And as has been said, whats the excuse for invalid markup? surely a platform to build a website on should itself clean coded?

Whats the excuse? Did you watch the video I linked? It's explained pretty well. If the website is valid, it's likely it's not displayed well in all browsers = bad user experience = not good for Google to display your website to users.

 

[NextPoint]

Are you seriously trying to tell me that I should trust the results of a SEO company that is trying to sell their services over the official word of Google that tells us the opposite?

I totally agree. It's always important to read between the lines when read information provided by a source that is likely to have their own agenda - this quote from my article proves my point:

It’s clear at least Google seems to prefer valid HTML & CSS. We will keep an eye on the results and see if Google decides to include the other pages, or anything changes.

Google seem to have favoured their test pages, but they haven't taken into many other accounts like whether Google simply hadn't got around to checking the other pages, the placement of the other pages within the website, what value the links were that they used to point at the pages - even the position of links on pages could gain different point values from Google - nobody knows for sure how Google operate and they frequently change the rules.

What you can be assured of is that Google want to provide the best content to their customers - making sure that your content markup is easy for Google to read and loads fast will provide a lot of benefit, but going overboard with W3C validation will not provide any extra benefits to a page that is just well constructed.

 

[edmondscommerce]

It does seem like Google don't care too much about validation

http://validator.w3.org/check?verbose=1&uri=http://www.google.co.uk/

 

[sysops]

Most of the errors result from the fact that they have declared a Strict document type. If they just change the document type to XHTML 1.0 Transitional, it goes down to 6 errors.

There are errors and there are errors. The ones on that site are not a big problem. They should however change the doctype.

 

[movietub]

Whats the excuse? Did you watch the video I linked? It's explained pretty well. If the website is valid, it's likely it's not displayed well in all browsers = bad user experience = not good for Google to display your website to users.

Sure I watched the video. And I do understand that Google don't reward perfect markup, but good markup (in keeping with the doc type decleration) may well assist bots viewing and recording the page content.

Also, they don't specify if it the HTML or CSS that they are sacrificing in the name of accesibility. I suspect it's more CSS.

I can't think of an occasion when using valid markup has stopped me getting something to work on IE6 for example, but there has NEVER been an occasion I've managed to keep CSS valid and still have IE6 display everything. At least not on any but the most basic layouts imaginable.

I can't think of any downsides to valid html, and the upsides are still under debate. The fact that google state they don't 'actively reward' valid markup, does not mean that certain markup errors do not throw bots and affect rankings.

 

[movietub]

Most of the errors result from the fact that they have declared a Strict document type. If they just change the document type to XHTML 1.0 Transitional, it goes down to 6 errors.

There are errors and there are errors. The ones on that site are not a big problem. They should however change the doctype.

That was my comment to them, but they emailed back saying they pushed forwards to XHTML 1.0 Strict ages ago in order to keep current!

They seem to have missed the point somewhat. You have to actually 'be' current and accurate or not bother at all surely!?

 

[movietub]

I've looked in to this a little deeper now. Their system is also not generating 'alt' tags for product images.

Apart from this being a requirement to validate, I understood that there was a clear SEO benefit for using descriptive alt tags (I'm aware that renaming images with friendly titles themselves also goes along way).

Would it not be more normal for a cart to simply autogenerate an alt tag using the product name? It would at least connect all images to text and title contextually in the eyes of a bot.

 

[edmondscommerce]

I suppose this is one of the big drawbacks of a hosted system.

This would be really easy to do, but as its not your code all you can do is request it and hope they are willing to implement it.

 

[movietub]

I suppose this is one of the big drawbacks of a hosted system.

This would be really easy to do, but as its not your code all you can do is request it and hope they are willing to implement it.

I understand that of course. There are drawbacks, but also many positives for us choosing a hosted system.

The question is, does ignoring the requirement to use alt tags, when it is an easy fix, count as unreasonable? I wouldn't pay a web designer if he delivered a site with basic html errors and no alt tags on images.

 

[edmondscommerce]

yes get in touch with them and request it, its easy enough so hopefully they will agree

 

[FreelanceSoftwareDeveloper]

Certainly some of the errors are basic such as image alt tags and even use of uppercase instead of lowercase.

If this was the other way around you just know the hosted solutions would be spouting the 'you get what you pay for' line, so ask them why what you pay isn't enough to fix those basic errors at least.

The product name as the image alt tag is so simple to code you can even get it for 'free' with open source carts

Seriously though ask the question, I don't have any issue with hosted carts and do understand how the business model suits many companies I just don't like how they sell themselves and put down open source solutions (yet run on LAMP servers?), especially when many open source solutions would never let simple stuff like this slip through the net.