Part of the big buzz surrounding Web 2.0 has been pull alerts as opposed to push alerts.

Push alerts / messages –  I send you a message.  This is how email and IM work.  I choose when, where, and how the message is sent and largely control how you receive it.  I send an email, you get it in your inbox.

Pull alerts / messages – I send a message which may (or may not) be intended for you primarily.  You decide when, where, how, and IF you receive it.  This most common form of this is RSS/ATOM feeds.  I publish to my blog / Twitter / whatever and you subscribe to me if you want to.  You can receive alerts via email, IM, Xanga, Facebook, Google Reader, BoxtheWeb, Sage, or a myriad of other options.

Some have said that push alerts are dying.

This makes some sense.  When I post on a forum, I don’t want to have their system email me every time there is a reply (email/push).  What I really want is to have easy access to a list of posts replying to mine to look over (RSS/pull).

However, this can go a bit too far.  Pull IM does exist to some extent, but it defeats the purpose.  I want you to see something NOW, it’s URGENT, INSTANT.  Pull does not fit this.

Push alert systems, however, just refuse to die!  Facebook/Myspace messages/wall posts.  Blog comments.  New friend-group messaging systems like Pownce.  Push is extremely popular.

The masses are rarely right, but perhaps we shouldn’t brush off push alerting altogether at this point.

The microformats-discuss mailing list reached an impossible signal/noise ratio for me awhile back. My inbox was just too flooded with discussions I did not find interesting and I archived 90% of it without even opening it. Well, today I went browsing through the archives and here is some of the interesting stuff I found summarised for those of you with similar issues:

• Scope was discussed again, some arguing that rel-tag needs to be defined as to where it applies. The upshot (which I fully support) was documented on the wiki.
• There was some discussion surrounding microformats detection and false-positives. Discussion concerned using the profile attribute of HTML to get around this, but since parsers should check for validity of uF as well as root class existence, it shouldn’t really be a problem.
• There is a bit of ongoing discussion which I do not 100% grok. It could have to do with saying an hCard is for the owner of a page, but that is determined by whether or not it is in an
  
  element. It could have to do with two hCards being for the same person, in which case I agree that class=url, rel=me on both of them pointing to each other demonstrates that very well. Or it may have to do with seeing if an hCard is useful at all… I don’t get that.
• There was some discussion caused by the announcement of Videntity’s hCard and XFN integration (yay!) Great potential there.
• There was the age-old ‘my rel-tag URLs are crap’ issue brought on by a Blogger-FTP issue. The Blogger problem makes sense and probably will not be ‘resolved’ except by removing rel=tag from FTP blogs. Blogger-FTP is crap.
• Someone questioned the value of XOXO itself! They were kindly directed here 😉
• There is a presentation (S3!) about AHAH. Of course, AHAH is built on AJAX and thus subject to cross-domain issues. They need a JSONP version.
• Finally, another bit on proving two hCards are for the same person. In a more dangerous way.