quite effective at deep packet inspection and other man in middle attacks.

1) a true swf to html5/JavaScript engine on the fly - believe that when I see it running

2) slurps out any rtmp url's it finds doing dynamic conversion to h.264 - won't work with sites using dynamic logic inside swf to assemble resource URL

3) virtual machine on server running flash which then streams result to iPhone - huge expense for app owner

Plus not sure sites using rtmpe like hulu would work

Is worst case for redundant secure two site storage for volumes over 2tbytes

I had 3 backups of home data of about 300gbytes each.

Each one was almost but not quite the same due to some rather poor backup policies on mypart.

I was able to dedup per backup to get them small enough to combine and dedup the combo.

Left with one pure 150gbytes combo. Rsync is amazing

Only problem is getting your changes into the trunk.

For that I am glad they are very picky.

Ensuring everyone in project is using same one is more important than which one is chosen.

If you can use the same one for a whole company even better

1) Install Wikipedia software locally and use this for any locally created articles

2) The web server running this simply proxies out to en.wikipedia.org for that request if not available in the local version. The easiest way to do this is with Apache + rewrite rules

This means that users can get to articles locally and on wikipedia from the same command

You then need to consider the following

1) The search request needs to go to the local version of wikipedia then the external one and concatinate the results together - a small proxy script should be more than capable of doing this

2) You may want to create a reference table which maps external wiki articles to related internal ones. Again a small script could insert these into the external wikipedia articles during rendering