SQLite founder says project is not open enough and needs to be modernized

SQLite

is an ACID-compliant relational database management system, contained in a relatively small library written in C

SQLite is a lightweight relational database engine accessible through the SQL language. Unlike traditional database servers such as MySQL or PostgreSQL, its distinguishing feature is that it doesn't replicate the typical client-server architecture , but rather integrates directly into programs.

SQLite is a complete single-file relational database management system . SQL, or Structured Query Language, is the industry-standard programming language for storing and retrieving data. Well-known SQL database management systems include Oracle, IBM's DB2, Microsoft SQL Server and Access, as well as the free software MySQL and PostgreSQL.

Recently, the project's founder complained that "SQLite is explicitly and unequivocally 'open source, not open contribution,'" mentioning that the project currently has a limited objective, which is to be a fast, small, and reliable integrated database.

The reason for this comment stems from two things. First, he and his team argue that SQLite should be enhanced with essential new features. These include, in particular:

  • Support for distributed databases available on multiple servers
  • Asynchronous I/O support via the new Linux io_uring API
  • eBPF to optimize SQLite by allowing certain operations to run in the kernel
  • Support for user-defined functions in Wasm (WebAssembly) to allow the use of other languages, which are compiled into Wasm, instead of C.
  • The libSQL project plans to use Rust in conjunction with C to implement these changes.

Secondly, a new fork of SQLite, called libSQL, has recently been created with the aim of modernizing the hugely popular embedded SQLite DBMS. Furthermore, it plans to use Rust along with C to implement the changes advocated by its founder.

As such, many might ask, why a fork and not make proposals to the SQLite project itself? And as already mentioned , the problem is that the SQLite project has a limited objective.

According to the SQLite development team, the DBMS is likely one of the five most widely deployed software modules of any description. Given that SQLite is used extensively in all smartphones, and there are over 4000 billion smartphones in use, each containing hundreds of SQLite database files, it's likely that there are over 4000 billion smartphones in use and one trillion SQLite databases in use.

Thanks to its extremely lightweight nature, SQLite is one of the most widely used database engines in the world. It is used in many consumer applications and is also very popular in embedded systems, including the latest smartphones.

Size isn't everything for the humble creator of SQLite, as he apparently intends to make the project profitable, since he doesn't receive royalties for the millions of copies used by commercial and non-commercial users. Regarding this, he mentions that he's not really interested in the monetary aspect, and his frustration stems from another issue.

This is an unusual approach, even in the world of free software. Most open-source code is licensed under an agreement like the GNU GPL (General Public License), which includes terms that guarantee the software remains free, although today several developers of major projects have reached a conclusion—not entirely disagreeing—that their products weren't receiving support from commercial users (something similar to what Qt reached).

“I went through all the licenses,” says Hipp, “and thought, why not just put it in the public domain? Why put these restrictions on it? I never expected to earn a dime. I just wanted to make it available to other people to solve their problem. »

"We're not trying to compete with those other engines," says Hipp. “Our goal is not to add all kinds of bells and whistles, but to keep SQLite small and fast. We set an arbitrary limit to keep the library space below 250KB.”

Finally, Glauber Costa mentions that " the success of SQLite combined with its relatively closed development makes a fork difficult to succeed ; another option is to wrap SQLite with additional functionality, but this has limitations," to which Costa alludes.

Source: https://devclass.com/


Add as preferred source in Google