Suzanne — Sat, 28 Jun 2003 14:26:39 +0000

sorry, no -- it doesn't follow the include links because they don't exist to it.

Say index.php has this include (SSI works the same):

<? include "whatever.php"; ?>

A spider won't go get whatever.php on its own. It won't know it exists. HOWEVER, it will know whatever is inside whatever.php because the server will have taken that information and written it into the page that IS crawled -- index.php.

So the page comes to the spider as you see it when you browse, not as it's written on the server. It sees one file, not two (in this example), and that one file is the combination of both individual files residing on the server.

The Webmistress — Sat, 28 Jun 2003 09:58:02 +0000

Quote: Originally posted by Suzanne
It doesn't follow include links, just the page. SSIs (or includes) do not affect SEO at all.

So if the spider wont follow the links then it does affect SEO as the site will not get fully crawled and indexed!

Suzanne — Sat, 28 Jun 2003 00:18:37 +0000

Just supporting what's been said -- the spider calls the page, then the server puts it together for the spider. It doesn't follow include links, just the page. SSIs (or includes) do not affect SEO at all.

HOWEVER, they can affect server performance. If you have many pages that are mostly static, you may want to do something like many blog software applications do, and generate those pages into static files and only update them as needed so the server is only issuing text files, not parsing documents.

Megan — Fri, 27 Jun 2003 19:00:43 +0000

I'd hever though of that one but I'm assuming that it wouldn't be a problem. I've been trying to do some research on the topic but there isn't a whole lot out there on SSI's and SE's. One message board post I found stated that since the spider sees the rendered HTML page it will therefore see the SSI content just as the user sees it in the browser.

Quote: When the server is asked to serve the parent page (the template) it assembles the ssi (body text) and the template into a complete HTML page and serves it. All the spider sees is the complete HTML page.

So, you needn't worry about using ssi's. The entire page is created "server side" before being served to the spider or any other inquisitive individual.

(this is from Webmaster World forums )

The Webmistress — Fri, 27 Jun 2003 16:02:53 +0000

How do SEs deal with SSIs? If the navigation is called from the SSI do the SEs see it?

Megan — Fri, 27 Jun 2003 15:51:07 +0000

Have you looked into using Server Side Includes onstead of frames? I haven't seen your design, but this is usually a useful way of applying your template to every page without worrying about frames and associated search engine problems.

Suzanne — Fri, 27 Jun 2003 14:59:22 +0000

Julia's advice is the best option.

Second best, because sometimes you need to use frames (usually in applications only):

For the pages IN the frames (not the frame itself) you can try the below. This is JUST A SUGGESTION, not tested, and not inclusive of all browsers:

   if ((top == self) {
            top.location.replace(newURL);
   }

You'll need to read up more on how to specify what page is in the frame, or you'll lose people and annoy them.

Welcome to TWF

The Webmistress — Fri, 27 Jun 2003 07:37:18 +0000

Personally I wouldn't use frames, you have the problem of the search engines actually crawling the site properly and also pages breaking out of the frameset from a results page. This can be fixed by putting a framestuffer code onto each page - search google for one.

As for making the site load quickly, you really need to just make sure that each page (you're not going to be able to preload the whole site) is a reasonable size and that any graphics are well compressed.

Even if you have a huge site consisting of thousands of pages you don't have to use frames and the site will load as quickly as your page size allows it to.