auto redirection

in some of my sites, i have the inital index page run for
a split second, just to preload some elements,
do flash detection, and then redirect the user to
either a flash or non-flash version (subdirectories).

i plan to add in a robots.txt and robots meta, but i was hoping
to just tell it to NOT FOLLOW, and simply place all the keyword
and description metas within the index.

any thoughts on how all this may be interpreted by the spiders?

Most spiders work by page content rather than meta-tagging, or by comparing meta-tags to page content and other fancy things... but as far as I understand it, if you don't have any content on the page, then they won't like it one bit, and as you then have then not follow the links, then all further content no matter how valuble and juicy will be ignored.

The redirect will also hurt you - search engines hate them.

thanks for the input.
i've changed the format, but its raises new concern.
now........in a single page i first run js flash detection,
and the outcome of the detection then defines a php
variable (which ultimately displays the real content).
there is NO auto-refresh.

my question is........even if i include content relating to the
main topic, or keywords, in any outcome,
will spiders be able retrieve the information properly ?
- considering that the information is first hidden in the php.

in googling this question, many people say that google, itself,
has no problem indexing php. any thoughts ?

Google cannot see the PHP. PHP is server-side, googlebot requests a page (say index.php) the server reads the php, and then the mark-up (e.g. html) is served to the bot (just like it is served to you or I).

You also realise that google can't read your flash site?

Why not have the page with content describing your site etc. which the bots can all have a good look at, and have two links, one to the html site and one to the flash site, then the spiders can follow into your html site, and any users who have flash can click on the flash site.

This would make the system more bot and people friendly IMO.

For example: Some users who have flash might prefer to view your html site.

as a designer and as someone whos just taking orders from
the boss, its not just my preference, but my order to
display the flash if the user has it installed.

also.....
im sorry i can not remember the links just yet, but i have read
many articles that argues that google does now read flash
content. afterall, it is fairly easy to disect an swf.
i have seen no proof of either side.

none the less, i've required the marketing buffs to write up
straight text, that will remain static throughout the site, at
the bottom of every page - including keywords and good
information. my real concern is still the php side. i understand
that the spider will view the page once its generated by the
server, but its hard to tell what outcome the spider will see.
does the spider truly have flash installed? if so, what version?

its difficult for me to realize how a spider would view because
im setting up the site as, essentially, one single document.
depending on the users (or spiders) viewing preferences,
determines the content displayed. super-dynamic.

i completely agree that root index would be much more
search engine friendly, if it were only html with links thru
to seperate versions of the site, but i really dont have any
choice, so im trying to come up with the best possible
solution.

i appreciate your input and would be happy to read more.
thanks