How do you change the font size and face of a compressed html (.chm) file in Winblowz?

Create a css file and link it to ie:

Click Tools ==>> Internet Options ==>> Accessibility ==>> Format Documents using my style sheet

then browse to your sheet, restart ie and/or your chm file, and that's it!

Here's an example (I'm still tweaking mine):

/* Html Elements */
body
{
font-family: Segoe UI, Trebuchet MS, Candara, Verdana;
font-size: 15px;
}
pre
{
font-family: Consolas, Century Schoolbook Mono BT, Monaco, Bitstream Vera Sans Mono, Dejavu Sans Mono, Lucida Console;
font-size: 13pt;
}
a
{
font-family: Segoe UI, Trebuchet MS, Candara, Verdana;
font-size: 15px;
}
b
{
font-family: Segoe UI, Trebuchet MS, Candara, Verdana;
font-size: 15px;
}

I was thinking about what if I could build a peer-to-peer antispam system which might solve the problem of spam once and for all. For example, using a Kademlia-based distributed-hash-table might work. Your client-side outlook or Thunderbird plugin would do a hashing of your complete email message after the headers, and/or hashes the attachments and/or the fields contained in the headers, then quickly, using the genius of Kademlia (which I really don't understand too well) possibly combined with a local bayesian filtering system, you are able to quickly find out if a message is: 1. spam 2. mailing list 3. not spam. To prevent spamming just leave out 3 & 2, so that any message that gets reported at all is going to be spam; why would a spammer report their own messages as spam?

Of course they could try reporting known common email like mailing list messages, etc... that aren't spam as spam, but a local adaptive filter similar to gmail's spam filter, that adapts to each user would sort that out.

A further protection against spammers spamming the system would be a way to trust a certain node, like everyone has an option to generate a guid that identifies you as unique. Then over time, you build up a trust percentile with other people who marked the same messages as spam, and the more closer they do, the more they trust your choices, and you trust them.

This could even work for usenet, which is filled with so much spam that it makes it nearly unusable, especially since in the case of usenet generally everyone has the exact same message beyond the headers.

So after this genius brainstorm I searched for 'kademlia spam' and find this guys page:
SpamWatch - A Peer-to-peer Spam Filtering System
http://www.cs.berkeley.edu/~zf/spamwatch/
Also he wrote this interesting paper:
http://www.cs.berkeley.edu/~zf/papers/ata_middleware.pdf

Here's a similar project, but unrelated:
http://www.mailavenger.org/

Also other things could be implemented, like for example, quickly reporting email worms and the like.

For more in-depth Kademlia information:
http://en.wikipedia.org/wiki/Kademlia