Re: websites: public vs private information?

On Mon, 5 Jun 2000, Dodi Schultz <schultz[_at_]compuserve.com> wrote:
>
> I ask as not only not a techie but barely a *sub*techie (and a
> confused one at that): But surely the discrete elements existed
> somewhere, fixed on some hard drive? If I'm reading, say, a
> Slate article: someone wrote that article, and the words do
> exist in the author's computer, no? Similarly, the page layout?
> And the type specs? And didn't someone have to bring it all
> together, somewhere, before tossing it all out there for my
> browser to pick up? Is there not some employee of Slate who,
> prior to making all these elements available to me and the
> rest of the public, put it all together first, and doubtless
> filed -- *fixed* -- the result on a computer owned by the
> magazine?
>
> Please comment further, Barry?

OK, I will take another crack at it...

[Most of the following was written before reading the thread that has developed the last couple of weeks. I will update it only slightly after reading the whole thread, see the very end for the new twists]

The page you see when you visit a web site, especially a technically sophisticated one, actually consists of many elements that are brought together for the first time when you cause your browser to visit a url.

I am simplifying here, but these elements include the content itself (e.g the text of a Reuters article on a news site), as well as the layout information.

Decisions about how to actually combine and deliver the sum total of all the necessary information is made, in real time, at the server end. This may mean consultation with many other computers to decide, and possibly personalize the information.

As an example, even though you and I may visit the same URL on Yahoo News, we likely will see different banner ads. The decision about which banner ad to display is made based on our past browsing history at Yahoo (and possibly other places).

This is done in collaboration with the ad server company, and it is important to understand the ad itself does not come from Yahoo's computers. Your browser actually is told to make a request to a URL on the ad server's computer (e.g. doubleclick.com). You can verify this by checking out the source of HTML in your browser.

Let's take a look at that news article from Reuters now. Obviously, it was written by Reuters and made available to Yahoo. But also it is made available to countless other news sites simultaneously. In the interest of timeliness and accuracy, it may be to both Reuter's and the news site's benefit to gather the text of the article on demand.

For instance, last night I did a quick scan and many web sites had stories about the end of the Stanley cup playoffs, which occurred in sudden death fashion, within seconds of the end of the game. How did this happen?

Reuter's (and other similar services) write the article. Yahoo's web servers know to look for updates of articles before constructing their news headline page. This is similar to the way your browser is told to check for ads on the ad server, except Yahoo does it for you. The article actually resides on the Reuters server.

I am simplifying, because for performance purpose, Yahoo will cache the article after the first time it is retrieved, while continuing to check for updates regularly. This is called caching.

Yahoo, using several servers of its own, will wrap the 3rd party content (and some content of its own perhaps -- imagine a newspaper site that does all described above, plus provide its own stories) with the "look and feel'" of Yahoo.

It then sends on the HTML to your browser, which then displays the look and feel and Reuters story (in this case), but must fetch the ads itself. So that is how what you see in the browser never exists anywhere except *in the browser* even though it is clear where it was requested from.

This is a very simplified example that does not address additional technologies in common use at both the browser and the server ends that can confuse the picture even more. For example, the content that is merged may come from places under differing degrees and types of IP protection.

What I really wonder is, can it get so complex that authorship can not be determined or may even not exist at all?

Hope this helps...

Barry

Barry Caplan
<bcaplan[_at_]i18n.com> Received on Sun Jun 11 2000 - 17:51:07 GMT