We’re just preparing beta 4 for our tireless beta testers with many minor changes including the new icon set. Time to direct your attention to some of the substancial changes we have made to DEVONthink 2.0.
What you will be able to see is DEVONthink 2.0’s ability to open multiple databases at the same time, a much smaller memory footprint, and much higher performance especially when using large databases. What you cannot see is that we have rewritten large parts of the underlying AI-based database kernel as well as the way DEVONthink stores its data. The kernel is responsible for the database management as well as for all AI functions such as classify, see also, topics and the AI-assisted search.
The modified database package format now also stores all document always as files and no longer keeps plain text, rich text, and HTML files only in the index structures. Consequently, Word files and email messages will no longer get converted to RTF but be stored in their original file formats and be displayed using QuickLook or alternative methods. Using ‘Open with’ you’ll be able to open them in any application that can open or edit files of these types and upon saving them in the external editor DEVONthink picks up the changes you have made.
In addition these improvements to the database structure make the database format much more robust — the index can simply be rebuilt when it should become corrupted of whatever reason — and enables DEVONthink to store meta data with every record. Version 2.0 (or 2.x) will allow you to add your own meta data fields to the database and use them in searches.
Well, so much for today. Back to editing the online help for 2.0 …