Friday, November 2, 2018

Daily Blog #526: Where are we going?

Hello Reader,
         A long day of flying but luckily I have inflight wifi. Today I wanted to point you to a tweet from Pasquale Stirparo

https://twitter.com/pstirparo/status/1058301948145922048?s=09

I have a dream where, in our community, everyone would stop building his/her own "yet another new" #FOSS tool to fix the same problem as may other already there, instead of joining forces and build fewer but better and longer term solutions. #FridayThoughts #DFIR #InfoSec

I wanted to talk about this because while there are exceptions to any generalization (in this case things like plaso, volatiloty, autopsy, reg ripper and more) there are many of us who are putting out one off tools that we develop to prove a concept or learn an artifact that then becomes abandonware over time , left for some examiner to find in the future hoping it will solve their issue.

I'm not here to criticize this method, I think I've done many times as have others who work with me, but I can see Pasquale's point. If we could work together to bring our tools and research into something that could be integrated and maintained long term everyone would benefit. You see this in these larger projects but eachbis purpose built for the needs of the users.

Plaso -timeline creation
Volatility - memory forensics
Reg ripper - registry analysis

The big exception i see is:
Autopsy - forensic suite

But autopsy is written in Java and uses ipython which means it's outside of whereany of us feel comfortable.

Our commercial vendors are trying hard (for the most part) to create solutions for us but they can't keep up with a worldwide community that finds new artifacts almost every day. So we need to figure out how we can work together to take advantage of the brain trust that's out there who through good code review and sample projects can lead the next generation of examiners to learn how to solve their own problems and build and hopefully maintain their own future.

I don't have the answers yet, but I'm very interested in talking about what it could look like and how it could work together. The dfir future isn't windows, or Macs, or Linux, or Java, or elk, or python, or rust (sorry Matt) it's all of these things because we have to be where the data is no matter the shape, form or storage.

I know several academic grants are focusing on these macro problems like CASE (https://github.com/casework/case) so there are potential solutions out there we just need to start getting our industry to seriously talk about how we want to approach the problem. While I'm sure different answers will be reached , any profrprotowards a solution will eventually find a winner in the minds of the examiners eho use it.

No comments:

Post a Comment