Hi Roman,
> What is the difference (technical implementation "under the hood" from Oracle) between lock conversion and escalation?
Oracle itself does not use / need lock escalation, so the only technical implementation is lock conversion. A great detailed explanation of its implementation is published in book "Oracle Core: Essential Internals for DBAs and Developers" by Jonathan Lewis. You can find this chapter (lock implementation starts at page 77) here. Especially the graphic makes it very obvious and you can interpret my previous description of the "shrink space" DDL with it.
> Why you do such emphasis on this?
Because lock escalation can introduce a new level of issues (e.g. deadlocks), which you do not face on Oracle in that context. I am quoting for SQL server RDBMS (knowledge base article #Q323630) which uses lock escalation.
Lock escalation is the process of converting many fine-grained locks (such as row or page locks) into table locks. Microsoft SQL Server dynamically determines when to perform lock escalation. When making this decision, SQL Server takes into account the number of locks that are held on a particular scan, the number of locks that are held by the whole transaction, and the memory that is being used for locks in the system as a whole.
....
Large scans or large numbers of Bookmark Lookups may increase the chance of lock escalation; additionally, it increases the chance of deadlocks, and generally adversely affects concurrency and performance.
Maybe i am a little bit too picky, but this is mixed up a lot 
Regards
Stefan