>>6
Neither.
>>1,6
Differences:
1. One has this thing going for it that it has a humongous spec that brings homogeneity to all the implementations in a COBOL-like fashion. Unlike COBOL, this thing's spec is frozen in time and will have to carry the burden of its design flaws for millenia. If this thing ever perishes it will be from its near homogeneity.
2. The other was born from an academic exercise in computation, a 50 page spec documents it roughly every 10 years. Every implementation of this one is in fact, a completely different lisp that happens to have functions in common, this is good for speciating the language into niches and if its a permissively-licensed open source lisp, who cares about standards (the other scripting languages don't even have any). Other than the aforementioned 50 page spec there are bite-sized RFCs with code to maintain some new features across implementations, though these sometimes don't get consensus because someone points out the functionality won't be possible on his potatoe powered lisp machine, or "this is an american term, we should use this french term instead", etc. They are either compilers, scripting engines and a scant few are image-based, the spec doesn't put restrictions on what it can be really.
3. Another is an evident fork in philosophy that sparked off from the sixth revision of #2's spec. The seventh upcoming report will mean that #2 will now become #2 and #3. #3 is a superset of #2, it will perhaps be this generation's #1.
4. The other lisps tries to "improve" #2 and give it a completely new name to wow the reddits of their grand invention.
5. A few exists for no other reason than that portable scripting language libraries hadn't happened yet at the time they were implemented.
6. The rest existed without knowing about #2 and probably half assed anything resembling #1 as well.