To BackTrack Scan or Not…?
So the public beta has been out for less than a week - and WOW. All I can say. Seems our little project here has struck quite a chord. Thank You - your feedback is awesome. Keep it coming.
2 areas that are getting a lot of attention - application support, and backtrack scanning. I will be blogging about what apps are in the hopper for support shortly - users are giving us a fantastic list of what to focus on.
The big one I want to point out in this post is scanning. Should you do it?
First - it can take a very long time. We do this intentionally. We don’t want to eat your system while we troll through all your files trying to find relationships for Flow to map.
That being said - the question remains - should you do it at all?
Backtrack scanning is disabled by default - you can turn it on in the prefs or using the wizard as part of install. It will read every file it can - looking inside it to find references to other files. This is why it takes so long. Unlike spotlight or other search - we peer into the file looking for fonts, layers, plugins, links etc. No other application in history has done this - and it takes time.
The key is - Flow will do the same thing to any file you open with Flow - even if NOT scanned. The same indexing will happen just by opening in Flow.
This means you don’t really have to scan unless you want a full understanding of all files on your system or other volumes. However, even after a scan - we can’t tell you that the PDF was exported from InDesign as an example - as there is no record of it in the PDF for us to scan.
This is why the real value of Flow comes from watching you create your files. Drag and Drop, Copy / Paste, Import / Export - all get tracked in real time as you do it - no scan required at all.
BackTrack is really good for some folks - to give that complete picture. We have heard some great feedback on how to improve the experience - and are already working a couple of cool things.
In the meantime - I just wanted to remind you that you actually don’t NEED to scan for Flow to work. Just hitting save in Word will cause Flow to understand that file in it’s entirety with no scan.
Last - if you feel you want to scan faster (using more system resources) or slower (using less) - there is a preference in “Preferences –>Volumes” that will let you decide what, when and how to scan or to actually do it at all. You can also set the types of alerts you want via the Dashboard in “Preferences –> General” so you don’t get that temptation to scan “everything”.
S.
Tags: BackTrack, Flow, Scan, Scanning

February 27th, 2009 at 3:16 pm
Good advice. I had over 2TB of data on my RAID, it was taking forever to Backtrack. However, I have an issue where even though I have specifically excluded Versioning and Scanning on a connected server, I still get Flow warnings when I rename files. What gives?
March 5th, 2009 at 11:23 am
When you exclude a directory - this is for scanning only. If you happen to save a file into a location that is currently excluded - it will still be tracked.
We are working a couple of tweaks in this area - will be posting more soon.