This post originated from an RSS feed registered with Agile Buzz
by James Robertson.
Original Post: Browser Laments
Feed Title: Richard Demers Blog
Feed URL: http://www.cincomsmalltalk.com/rssBlog/rademers-rss.xml
Feed Description: Richard Demers on Smalltalk
I've been remodeling my wife's website to give it a better look and feel, but also to convert it from plain HTML to a combination of CSS and XHTML. I've only been concerned about getting it working on IE, but
David Buck
tells me I've probably been a bit short sighted about ignoring other browsers.
Actually, though, getting it to work on IE alone has been quite a challenge. I referred to several books and websites about CSS, I followed the examples as closely as possible, and then I got bizarre effects from IE. It's really annoying how much effort it takes to get something to work. (Un)fortunately, stubbornness is one of my personality traits.
So where's the fault here?
Let me confess that I don't use adequate tools for web building. I got hooked early on with the use of markup languages. In the 1980's, I used IBM's Script/Bookmaster to dynamically generate a document of over 1000 consistently formatted pages, complete with a dynamically generated table of contents and index. I've tried several so-called WYSIWYG word processors, and they are only good for small documents. (MS's Word is plain awful, but Lotus WordPro is pretty nice -- if it still exists!) So now, I crank up an editor and key in whatever HTML markup I want.
But hey, this is exactly the tendency I was
denouncing,
about programmers who won't use an Interactive Development Environment, who insist on using text editors, files, bulk compilers, and other such obsolete tools in their programming work. A good web IDE would undoubtedly handle most of the markup problems I've been facing.
So... can anyone recommend a decent, inexpensive IDE for website work?
When HTML was first designed it was intentionally kept simple and focused on the generation of web pages, but it didn't stay that way. A variety of enhancements to accommodate ever more complex page layouts and presentation effects was coupled with incompatible implementations by a variety of vendors, leading to -- what else? -- chaos. Then the XML crowd got involved and made the point that content should be kept separate from presentation, and that led to CSS and XHTML. The thing is, (X)HTML is still about the generation of web pages, and the separation of form and content isn't that easy to do.
If you are creating an entire website, you want the pages to have a consistent look and feel. It's best to put certain aspects in a separate stylesheet that can be applied to all pages. This works great for presentation parameters, such as borders, colors and fonts, and layout details, such as margins and padding. But each page must individually express the structure that uses the presentation parameters of the stylesheet.
For example, if all pages are to have a header, an index column, a content column, and a footer, then classified <div> tags must be repeated in each page for that structure. It is not possible to express the structure in a stylesheet, have each page link to the stylesheet, and pass in appropriate contents to each element of a structural template. But then, I'm a programmer, and that's what a programmer would expect to be able to do.
All this led to the inventions of JavaScript, and web pages generated by server code from a variety of different kinds of templates, all of them specified according to their own syntax and semantics. Like I said, chaos! What we don't have is the equivalent of what I was using in the 1980's, a single, comprehensive tag language for document generation; one that includes markup tags for content, structure and presentation within a programming framework; and one that operates within client browsers without any special server support.
If I've missed anything obvious here, I do hope someone will enlighten me!
And now we come to the browsers. It's been a long time since anything meaningful has been done to improve IE. Even fixing its many reported security holes has been hit-or-miss hacking. I was glad to hear that Firefox has put enough marketing pressure on Microsoft that they are finally working on a major update. But are they just going to "polish the turd?" or do they plan to do something meaningful about the chaotic mess I've been talking about? There's a real opportunity here for Microsoft to be innovative, work closely with other browser vendors, provide some leadership, and clean up this mess.
Microsoft could start by publishing their IE plans, and being open to comments from the Internet community. They could even follow an
architectural process such as the one I described. I won't hold my breath, though.