As a matter of fact you’ve chosen the worst way for realization of a non-existing file or catalogue for this article. By this I mean things that you have written down in .htaccess There are some nicer and more elegant solutions. At this they don’t have unforeseen results (like a non-existing image). And if the think over properly, $REQUEST_URI is not a question to be discussed. It would be better to suggest some writing variants of .htaccess which would be interesting and provide some stuff for discussion.

Search engines treat such addresses as the normal ones – they divide everything into 200 Ok and 404 Not Found. All the rest doesn’t matter for them. mod_rewrite is tested by me at the time.

Keeper, could you show your solutions to everyone? Maybe they are not as nice as you believe…

Sure.
It is a rather strange method. There are Apache and mod_rewrite for it so why do you keep inventing something new?
However this method is often used and even commented…

It is the same Errordocument. The main problem is  —. The runner is to compile too much if everything is collected in one place.

How should I deal relative paths?
When I make hrc.com/consulting everything works and downloads properly, pictures are searched in hrc.com/pictures. But when I type hrc.com/consulting/ (it occurs if I cross via link) everything is performed in a similar way but paths of pictures and all the rest are built relatively to consulting directory which I don’t have as far as this virtual directories’ mechanism isn’t used by me 
And I don’t know how to avoid it …

Has anyone investigated how to transmit $QUERY_STRING within method "404"? I cannot manage it but I need it for search, for instance...

I’ve made redirection to the PHP-file which had something like header ("HTTP/1.0 200 Ok") in the beginning. But the problem is that this feature doesn’t function under windows and the same httpd.conf tried in Linux works properly   
I cannot realize what the matter is…

I’ve tried this all in the Netscape and it told me that beside Not Found I have 500 error as well connected with Errordocument. Then I switched off header which sends OK 
For some time it worked but Internet Explorer made redirection to the file which I place and Netscape makes everything nicely as I’ve entered it (http://127.0.0.1/go/to/news). I don’t like this bug. After that I’ve thought ‘If it works without header let’s place a .html-file’ and everything repeated    I return everything and it starts working in Netscape but not in Explorer    What is the matter and how to deal with it???