From my point of view it is worth doing and once you have it done you dont really have to bother with it again, especially if your site uses templates.
There is nothing to say that search eninges like Google wont use it as a part of their ranking algo. Chances are they wont do that until they get their own pages to validate :)
There is a chart somewhere in WebmasterWorld that shows that on the DMOZ validate, or thats how it was.
Some of the fixes are easy as well for example your java one
simply add the following
Worth it in my opinion
It is also good for forward compatability with the next generation of browsers that may be less tolerant to invalid HTML.
If you do a search on this site there will be a few good threads on the subject I know doubt :)
In the short term it' probably not very important for you if your site works across browsers/platforms and has no SE trouble.
Looking to the future is different though. Forward compatibility will become a bigger issue of the next couple of years as XHTML becomes more widespread, and validation will ensure that as new versioins of browsers are realeased you can relax and not worry about it.
I suggest you validate, it's not really that tough. Also, get hold of the proper DOCTYPE for your page (html 4.01 Transitional / xhtml Transitional)
I don't feel so bad after reading around some of the threads. If Google has HTML errors, then I won't worry. Plenty of time for me to read a fat book and actually learn how to do it. I ran the validator on a couple of competitors homepages and they were way worse than ours. I have only a couple of errors, while they had hundreds!
there is no attribute "NAME" ... 'images/accommodation-rollover.jpg',1)"><img name="Image30" border="0"
document type does not allow element "BODY" here ... il-to-us-rollover.jpg','images/faq-rollover.jpg')">
Hey, and this page was written by a couple of amateurs! Not too bad!
Also see [webmasterworld.com...] where I made my longest ever post on a forum anywhere about this subject [no point repeating all it again here].
g1smd, if you want to reference a single message in a thread use a fragment identifier which is made up from the string "msg" and the msg # as displayed for each message. This will get you an URI like this:
There was a thread about this not too long ago where someone stated that you do not need the language attribute for the script tag (its deprecated). You can shorten it to...
after checking my site w/http://www.htmlvalidator.com/lite/?google=free+HTML+Validator, I checked some big boys
TONS of errors on all! on other checkers too.
so guess they dont care?
>> so guess they don't care? <<
Most probably, so if you do care then you will clean up when Internet Explorer 10 arrives and gives a "page cannot be displayed" message for any page that uses code that is not well-formed. Validation and correcting errors does not take very long to do. If you use proprietory Netscape or IE code extensions in your page then these will be flagged as errors. You can ignore these "errors", but you should not ignore any errors caused by wrongly nested, or missing tags. These should be fixed.
GoogleGuy stated a while back that they are lenient about coding errors. However, I think that the forward compatibility argument is a good argument. That and pride in your workmanship.
One of my clients was hyper about cross os compatibility and was mighty pleased to see the site function perfectly with NN 4.7, on all os's, even using a unix browser.
Funky code can lead to unpredictable display problems. I used to be afraid of validating but now I'm totally ok with it.
Using htmlKit helps to cleaning up the code, as it lists line by line the alt tags that need fixing, and one click takes you to the appropriate line in the code. A real timesaver. And it's free!