Organising complex webs of complex articles in large websites

Question

My question relates to how large websites, edited by several hundred people who work in a large organisation (such as a News website) organise their web pages. What form of mechanism works in the background?

• Does each article represent a row in a table on a database. If so, how are images and embedded videos incorporated.
• Or is each page an individual web file with "include" tags for sidebars, headers and styling. Certainly on the BBC website, each news page looks like it is an individual ".stm" file, but this doesn't strike me as a very efficient method when lots of people are collaborating.
• Or is each page one of many files, that are organised by databases but which are not stored in them?

I'm sure there are a myriad of methods? What do the largest organisations use?

Answer 1

I would think there could be a couple different approaches to such sites:

1. Content Management System based. In this case, the article may actually be a collection of fields that are multiple rows in a database so that it isn't quite that simple as you describe. The fields are likely a mix of Rich Text/HTML so they can contain some tags but the images and other things may be coming from something else. Note that in this case there could be XSLT renderings that know how to combine the fields to form a page.

2. Platform-based. Something like Facebook meanwhile may have a different way to handle editing and updating content and thus is outside of a CMS situation I'd think. I'm not sure how well this complements the other approach but I would think this is somewhat different in some ways. In this case, the idea here is that the company behind the big site has their own proprietary method for storing the content and its structure.

---

Just to add a little more: Future Media Standards & Guidelines from BBC's site may have some more details about how their site is structured including what the "STM" ones may be as don't forget that you may want to use something like Fiddler or Firebug to note all the requests that are actually made to request a web page as you may find it surprising how many files are downloaded from one URL initially.