Thursday, September 30, 2004
- In XHTML, attributes must always have a value. As opposed to old HTML, where one could create short-cuts using only the value of the attribute-value-pair. Thus in XHTML, you correctly need to expand the attribute selected to selected="selected” in select-boxes.
- In XHTML (and XML) you can quote attributes with a single, or double quote. Using <a href=’http://www.example.com’>Example</a> is therefore perfectly legal. You may not use no quotes at all, something whin in HTML was sometimes allowed (and depended on the type of characters in the attribute-value).
- In ASP’s VBScript, you can start a comment in C-style syntax using the double-slash: "//”.
- Today’s most popular browsers have two rendering modes: quirks and strict (or “compliant”). Strict-mode uses a leaner, more standardized interpretation of the page and can be triggered by using valid doctypes.
- CSS can be used to zoom in and out elements on the page. It can also be used to rotate elements. Most of what works today in this area however is proprietary. Internet Explorer introduced its own set of feature-rich CSS-enhancements called “filters”. They are rarely used on the web today.
- A long time ago, in the great browser wars, Netscape introduced the “blink” tag. Microsoft hit back by introducing the “marquee” tag. Both tags could be used to make things nervously flicker and shake, and many found them appalling. Consequently, the latest Netscape-versions do not render blink-tags anymore. Marquee on the other hand is still <marquee>around</marquee> and living in Internet Explorer.
- The most asked question in HTML newsgroups must be “How do I hide my HTML source?”
The answer is: you can’t, because HTML is interpreted by the client (not the server, like ASP, Python, PHP and so on).
The most effective way however to seemingly hide HTML can be found in a simple trick: add hundreds of empty lines on top of the page, crowned by a comment of this form:
<!– HTMLProtector 3000 –>
- The W3C (the World Wide Web Consortium) in 1996 introduced special entities like &audio; and &folder; which were intended to display small icons without sending any images to the client. The idea never took off.
- You can actually use CSS to draw a house.
- A web server can use any extension to send out any file format (the HTTP header contains the actual, standardized “mime type” or “content type”, as different systems use different extensions). So when the URL reads “...something.pdf”, this may actually be an HTML page, or a zip file, or anything else. For example, it’s normal to send out dynamic pages with an “.html” extension to be very search engine friendly. (And sometimes, the extension is “.htm”, which points at someone using old Windows-style – there is certainly no necessity to do so, though.)
- Before HTML2.0, this markup language was simply referred to as “HTML”, not “HTML1.0”. The first highly commercial dialect was HTML3.2; a mixture between HTML3.0, which the W3C built with cascaded stylesheets in mind, and proprietary extensions Netscape and Microsoft created for their browsers. In short, HTML3.2 was a big mess. HTML4 (and later on, XHTML1) tried to set the record straight.
- In theory, stylesheets come in two different flavors. There are user stylesheets, and author stylesheets. They are both balanced out to create one single design. A user stylesheet is what comes with the browser, or what the visitor of a web site deems to be his favorite layout setting (like font sizes, colors, link underlines). The author stylesheet on the other hand is what the webmaster of a site intends his page to look like.
On this situation, Jukka “Yucca” Korpela remarks:
“It would be like letting one person select a suit for you, someone else choose the shirt, defaulting shoes to whatever you happen to be wearing, picking up socks yourself, carefully but ignorant of what the other selections are.”
- Since stylesheet rendering in Netscape 4 was so bad, it was often better to shut it off completely. To do so authors could include media “print” in their CSS-link, alongside with media “screen” (and “projection”). Doing this would make Netscape 4 completely ignore all stylesheets, which is a much safer thing for the visitor.
- HTML wasn’t always intended to be read-only. Tim Berners-Lee, who invented it, had something writable in mind too. Today we take it for granted we can just type into documents of Word Format, but we build specific tools (like Wikis) to make the same possible for HTML. But compared to “normal” web pages, their wikified brothers are still extremely rare. And even when you can hit the “Edit"-button, things are anything but WYSIWYG.
- In theory, the object tag could have replaced the image (img) tag in 1998, when HTML4.0 was published. Using the object tag authors could easily tell the browser: if you can’t render this 3d-file, render this movie; if you can’t render this movie, render this image; and if you can’t even do that, here’s the text fallback. The object tag never worked in popular browsers.
- Not too long ago, WAP1 entered the scene and promised to bring mobile content to hand phones via its native WML (the Wireless Markup Language). However, hand phones could in theory cope with HTML, which can deliver platform independent content. Now in WAP2, people realized it was time to make use of existing documents on the web, and switched to a somewhat reduced form of XHTML (the “mobile content profile”, with support from Nokia and others).
- SOAP by many developers is considered to be overly complex for most uses. SOAP is intended as a language for different servers to talk to each other. (The Google Web API, for example, is based on SOAP.) Many recent Web Services make use of simple, parameter-based URL-requests, which in return delivers simple XML. That’s keepin’ it simple.

As the results in the latest SEO competition are moving up and down fast, Raymond Angel started to advertise his Seraphim Proudleduck site on Google – certainly the only somewhat stable position (as long as he can afford to pay for it). On a related note, while this page has fallen to the fourth position in Google web search, it somehow climbed to the top in Froogle...

BoingBoing ponders if Google is on fire.
Hellacious Riders promises to give out free 10GB email accounts. The registration form looks funny, not only because of the devilish logo on top: you are asked to provide your first name, your last name, and your full name. Then again, the site is “hosted with 100% solar energy.” Rumor has it the first new user filling the account with 10GB will get 1 Terabyte for free. But guess what: after I registered, I never received the confirmation mail containing my password. I think this might be the start of a not-so-beautiful and rather short friendship.
>> More posts
Advertisement
This site unofficially covers Google™ and more with some rights reserved. Join our forum!