Most of the 9/11 questions raised by sights like reopen911.org, flight77.info, 911research, and 911truth.org appear to the layman, like me, to raise valid questions about the answers supplied by the 9/11 commission. Are these, in fact, just your run-of-the-mill consipracy theories? Is there coverage that rebuts these claims?

I wish everyone would stop using it OR that firefox always put new windows in a tab unless I explicitly chose new window.

I posted this as a story, but as usual, they are rejected so I put them here instead.

Basically about how XMLHttpRequest can't be stopped by the browsers stop button. It's a serious problem when google's services cause the browser to consume 100% CPU. Imagine the problems when less proficient developers write crappy applications.

Why Ajax Sucks (Most of the Time)

I want a means to generate a file's contents whenever it's referenced (and have it cached unless something has changed). I think this can be accompilshed with devices (at least on Linux) but I'm not sure that that's the right approach. Ideally I'd like to be able to write a script in, say, Bash, python or perl, that produced output that, for all intents and purposes, was the contents of a readonly file.

I want a way to bookmark my email messages. What would be very cool is if mozilla mail tagged email messages and threads with unique ids that the browser could access via something like mailbox://localhost/?mail=12345 and mailbox://localhost/?thread=12345.

For the time being, I suppose I'll just use perl to publish my mail folders and produce anchors that I can bookmark.

1. Dynamic, temporary/expiring email addresses. If the address includes a plain-text date of expiration, it will allow spammers to know that the address is no longer valid so don't bother sending it.
      http://www.mailexpire.com/
      http://www.spamgourmet.com/
      e.g.
          myID-20040113@somedomain.com will be good until January 13th, 2004
          myID-200401@somedomain.com will be good for the entire month of January 2004.

2. Single-sender (user/domain) addresses. They allow only one sender or domain. This is similar to point 1 except that the address is a hashed string that can be used to verify that the sender is the allowed sender. This is useful for registering with mail lists.
      e.g.
          XDuquXQMcZ/nK+4zING8Ew@somedomain.com will only allow malda@slashdot.org emails to pass.
          0jRDzvvEN4SZLpLoQu533w@somedomain.com will allow any @slashdot.org email.
      Optional:
          - Can be prepended with a string to indicate that the address is this type of address so don't waste bandwidth sending to it unless you know it will get through.
          - The Hash can be a hash of an encrypted value to add extra protection.

3. Everyone digitally sign ALL personal/business emails and include the recipient and sender addresses within the signature . If an email doesn't contain a valid, verified signature, it doesn't pass.

These measures don't imply that you can't retain open addresses that allow all email to pass.

There is a technological solution to spam that is 100% effective. Dynamically
generated, encoded and encrypted email address that, when decryted and decoded,
contain who is allowed to send the email and/or for how the long address is valid.
For example, I can encode/encrypt "org.slashdot" to indicate that the address is
only valid from slashdot.org. The address can become
b3JnLnNsYXNoZG90-20030801-myuser@mydomain.com.

Mail clients can autogenerate these address for the replyto field and decode and
decrypt them to autofilter incoming email. Utilities can be written as plugins for
browsers to generate these addresses registering with websites too.

Of course, MTAs would have to be aware of the format of the address but that's
already being handled by the likes of spamgourmet.com though, personally, I don't think
spamgourmet.com goes far enough.