JavaScript vs. Server-Side Includes

[Christopher Locke]

At 11:52 AM 6/26/97 -0700, "John R. Little" <jrl at acpub.duke.edu> wrote:
>Recently Sam Khosh-khui posted a question about using 
>JavaScript to automatically generate the last-modified date 
>for inclusion in web pages.  Christopher Lock[e] responded 
>with a few tips and mentioned that he wanted to use 
>JavaScript to replace the similar SSI technique.  Chris 
>mentioned that the SSI technique "entails a nontrivial 
>performance hit on the server."
>
>I don't know how to measure that performance hit but wanted 
>to ask if anybody can verify or deny Chris's statement.  

as indicated in my previous msg in this thread, there's a very
useful chapter on SSI in O'Reilly's CGI book -- the full text is
online at http://www.ora.com/books/webref/cgi/ch05_01.htm

here are a couple quotes from that source:

"SSI sounds like a great feature, but it does have its
disadvantages. First, it can be quite costly for a server to
continually parse documents before sending them to the client."

"Alternatively, you can set the configuration so that the server
parses all HTML documents: 

   AddType text/x-server-parsed-html .html

However, this is not a good idea! It will severely degrade
system performance because the server has to parse all the HTML
documents that it returns."

as the latter is exactly what I do to enable SSI datestamping to
work on ALL pages, I would have liked to use the Javascript date
mechanism to get better performance.  and this is why i was
disappointed to find this JS function unsupported in MSIE.

>...does anyone have suggestions on how 
>I would measure performance and processor demand based on 
>SSI vs. JavaScript?

sorry but I haven't Clue One...

after thinking I had posted this to the list (I'd only sent it
to John) he replied in private mail:

>The whole thing is interesting to me because Shishir 
>Gundavaram, author of the O'Reilly book you mention, says 
>performance degrades.  Musciano, author of the article I 
>read on SSI and of another O'Reilly book (HTML: definitive 
>guide -- <http://www.ora.com/books/webref/html/index.htm>) 
>says another. I just wonder which is really right. I'm 
>inclined to go with the CGI book author, Gundavaram.  But, 
>since you say that you already have SSI enabled to parse 
>all HTML documents I'd like to get your reaction.  Did you 
>notice "serevely degraded performance"?  And if so, did you 
>dis-enable SSI for all HTML documents?  Or, were you 
>willing to live with it (which might suggest that 
>"severely" is too strong a word)?

indeed, this whole discussion has got me curious again as to
which is correct.  I do in fact enable SSI for all pages -- this
is at http://www.displaytech.com -- so if the server hit is
minor, I'd feel better.  I do suspect though that the Javascript
would be faster -- if only I could count on it working in most
resonably up-to-date browsers, and I know it fails in MSIE 3.02

best

chris