Speaking at BlockShow Asia in 2019, which we have already dealt with on several occasions, Xinxi Wang, director of the Litecoin Foundation, returned to the age-old question known as the "scalability trilemma"; without going too far we explain that a trilemma is nothing but a problem with three variables of which only two can be satisfied simultaneously. In the case of a blockchain the variables are security, decentralization and scalability (ie the number of transactions that can be processed in the unit of time); of these three fundamental characteristics only two can be satisfied at the same time, so that a currency can be decentralized and secure, but it will be little scalable (as in the case of bitcoin), or it can be secure and scalable, but it will not be decentralized (as in the case of XRP), while the third case is not desirable, a decentralized and scalable currency that was not even safe would not interest anyone. During his speech at BlockShow Xinxi then declared that:

Inevitably there will always be a trade-off between scalability, security and decentralization. For example, for Bitcoin and Litecoin we always choose decentralization and security instead of scalability. That's why Bitcoin and Litecoin are very slow 

To solve the scalability problem, as is well known, decentralized coin developers have come up with the Lightning Network solution which, in fact, does nothing but scale off the chain (ie outside the main blockchain) and is therefore a substantially centralized solution that is grafted onto the main blockchain that in any case remains decentralized. Just talking about Lightning Network Xinxi then stated that:

This scalability solution is revolutionary because it can handle millions of transactions per second or even more. So the two levels work together to mitigate mutual deficiencies, one is not scalable, the other is, but it is also centralized

Then there are projects that claim to be all three together, that is, scalable, safe and decentralized; these types of networks are based on consent protocols and new hashing algorithms, whose security has only been postulated but never tested in the field; because a consent protocol and a hashing algorithm can be considered safe it takes several years of testing, at least a dozen, and in fact in these projects new bugs are constantly emerging which, in some cases, are corrected in time without causing damage but in many others they are corrected precisely as a result of attacks that have damaged users. This is why, as argued by the director of the Litecoin Foundation, a compromise will always be needed, trying to build the infrastructure on several levels, a decentralized and secure first and then a second, higher level, which will instead be scalable and, therefore, inevitably centralized.