Electrum devs cannot change the fact Bitcoin uses Lamport sigs. As currently implemented, Lamport sig validation scales quadratically with tx size.
Nobody can fix that until we have segwit and may then change to Schnorr sigs.
Nobody can fix that until we have segwit and may then change to Schnorr sigs.
There is always more than one way to make something work, saying there is only 1 way shows a lack of imagination.
No offense to Lauda, There is always more than 1 way to Skin a Cat.
Until the Electrum dev actually come out and say they can't make it work, your assumptions are irrelevant.
There you go again, avoiding the concrete specifics of the O(n^2) attack and retreating into lazy hand-waving generalizations about "always" and "something" and useless slogans about cats.
It must really suck going through life encumbered by such sloppy, random thought processes. You must greatly resent those of us with the ability to function at much higher levels of focused attention to detail.
There is no known way to change the current Lamport sig implementation such that it avoids quadratic scaling.
That algorithm is not trivially parallelizable. We can't get there from here. If we want lightning fast multi-threaded validation, that requires segwit+Schnorr.
Gavin already explained why.
Here, I'll repeat it one more time in the hope that this spoonful of nutritious information somehow manages to make its way into your mental metabolism.
Here comes the plane! *Vrrrrooooom!* Open wide! Nummy nummy knowledge for sweet widdle baby kiklo!
The attack is caused by the way the signature hash is computed-- all n bytes of the transaction must be hashed for every signature operation.
Perhaps if you took the time to read his post and understand all of it, I wouldn't have to sit here spoon feeding you premasticated fact and wiping most of it off your chin/bib/high chair when you just spit it out rather than successfully starting the digestion process.