How can I explain the potential ease of scaling and feature benefits of Segregated Witness to a non technical friend?
How can I explain the potential drawbacks of Segregated Witness in a non technical manner?
The name itself is a good hint as to what it actually does.
What we have today is a "witness" (i.e signature scripts) included in every Bitcoin transaction in non-segwit compliant transactions. This allows every full node within the Bitcoin network to verify the integrity of any transactions.
By removing the signatures in a Bitcoin transaction, you lose the ability to verify the integrity of the transaction. But in reality, not everyone needs to be able to independently verify the integrity of a transaction.
Thus, by segregating the signature from a transaction, we have a smaller transaction that describes the inputs and outputs exactly as it is. But if you want to know for sure that these are verified inputs, you'll look up the Segregated Witness set to find the corresponding scripts to validate the SegWit Transaction.
Some of the major benefits of SegWit are (in no particular order):
1) Effective block size limit increase
2) A solution to transaction malleability (no longer possible for a third party to alter the transaction ID)
3) Facilitate certain payment channel and lightning implementations by resolving the transaction malleability issue
4) Addition of a script version number to allow future script upgrades to be backward compatible
5) Reduced computational complexity of signature algorithm
An easy explainer from Aaron van Wirdum is here (with thanks to Pieter Wuille) https://bitcoinmagazine.com/articles/segregated-witness-part-why-you-should-care-about-a-nitty-gritty-technical-trick-1450827675/
Here is FAQ about SEGWIT benefits.
If you would like to know more technical details, probably segwits developer guide is very good start.