XML

Date: December 17, 2004

      First there was a need to have datasheets and specs look the same in appearance so people could look at one and another and not be confused by different placement, order, configurations. This is needed when you are dealing with hundreds of datasheets like big parts houses (think TI, Moto, etc).
     
      So SGML was created and in the 1990s was evaluated by even small companies. It was expensive ($5,000 per seat) and hard to learn (over 180 commands). So it had a high learning curve and small companies could not justify committing to it. It could allow rigid definition of document data that would control how it looked and was placed on a page.
     
      HTML came along (1993-95 depending on who you are talking to) and it had only 30 commands and people saw that it could hold tables and lists, and figures, and stuff. It was cheap (free) and easily learned, and they began to generate things. You could put tables and lists and sections of manuals in files and then cross reference them and use them to assemble training classes or refer to them with press releases, and any number of section-table-image reuse. People put up with different colors looking different on different monitors (hence the 230 "safe" colors). And they told others what default font settings to use to properly view their stuff. And then they told them the resolution they needed on their monitor to view their stuff.
     
      But as people got proficient, they wanted more control (rather than telling the user "best if viewed at ---". So HTML was added to, and added to, until HTML 4 with a whole lot more commands (some of which most people never see or use) and a reasonable amount of control on the page, using frames and layers and style sheets.
     
      But it was obvious that if only PC users read a file created on a PC, things were cool. If they switched to a UNIX or a Mac, it was different. A file done on a high-res Mac blows up on a low-res PC monitor.
     
      This bothers people so they wanted more control of appearances and so along came PDF.
     
      Now people could control what a document looked even if they used silly fonts that no one else had because Acrobat would "emulate" the font and anyone could see the file.
     
      But this negates "reuse" in its pure form, where you want a database to supply changing information to a document. A PDF document is fairly rigid.
     
      So various schemes were proposed and database applications grew and we got Oracle. And HTML got forms.
     
      But this was still not enough, so further changes were necessary for web pages and documents to share information from a database, and this is called XML, which is a subset of SGML (more than HTML is) and passes along types of information (title, author, track, movie, actor, book) instead of just format (header1, list, table element, address, paragraph).
     
      And thus was grow closer and closer to SGML every day. It only took 10 years.