If you need to learn to build a website, you need to learn HTML — using those visual tools or WYSIWYG Editor may help a bit, but for the best result, you should learn HTML.

Before you jump into the HTML5, you can start with HTMl4.01. Why not XHTML?

There is a very good article at SitePoint titled “HTML vs XHTML”. It is really worth reading.

The author also point out that we should opt “for HTML for Page Weights”:

One possible reason for using HTML 4.01 over XHTML (of any kind or level of strictness) might be that page size is a very important consideration. For example, you may be creating a page that needs to be downloaded over a restricted connection, perhaps to a mobile device of some kind. By using HTML 4.01, you’re able to reduce the markup by not using quote marks and not using closing tags where the spec indicates that they’re optional.

HTML5 does have quite a few new stuff but if you master HTML 4.01, you will find that you have mastered most of the basics to construct a web page.

ScottyDM’s comment to this article speaks my mind:

It’d be great to get away from the half-baked structure of HTML and go to a pure XML solution, but until MIME type “application/xhtml+xml” works on the vast majority of browsers that’s not gonna happen for most websites. Meanwhile, pretending that your XHTML pages will render the same in “application/xhtml+xml” as they do in “text/html” isn’t useful. For this reason I’ve switched back from XHTML to HTML, but I use XHTML coding rules when they don’t violate HTML rules: such as lower case tags, using quotes around attribute values, closing all tags where possible (but no ” />”), etc.

…. CSS doesn’t always give the expected results when you don’t properly close paired tags such as

and

. Another reason to follow many of the XHTML rules when coding HTML.

That is exactly what I’ve been doing.

For further reading, you may want to check out: http://reference.sitepoint.com/html/html-vs-xhtml (HTML Versus XHTML)