Yes, the relational model is the key, that is my point. The SQL language is an entirely arbitrary syntax applied to it. You don't need it to work a relational database, just like you don't have to program in C to write a program for a typical processor.
I don't care about how many applications have a performance problem with SQL generation. *My* application has a problem with it. I also don't care about Richard's query. *My* queries are much easier to form without contorting them into a SQL query. This is the entire point of my changes: avoid arbitrary bottlenecks and conventions that get in the way of performance and ease of use. It's not useful for your average person or typical application, but it's extremely useful for people who want to get the benefit of the database engine encapsulated within it. On Fri, Jun 5, 2015 at 2:10 PM, James K. Lowden <jklowden at schemamania.org> wrote: > On Thu, 4 Jun 2015 15:11:55 -0700 > Darko Volaric <lists at darko.org> wrote: > > > Are you seriously saying that that SQL syntax is friendly? How can you > > defend SQL syntax other than on grounds of history or > > standardization? > > The first and best defense of SQL is that it has at least some > basis in the relational model. It expresses relational project, > select, join, union, and intersection directly, and with contortions > relational division. Like Algol-60, it's an improvement on its > predecessors and on many of its successors. > > > If you're more comfortable and familiar with JSON > > the yes it is easier and you can avoid an unnecessary conversion step. > > I wonder how many applications you've profiled for which SQL generation > and parsing were a significant share of the run time. Usually once the > data are of any appreciable size I/O becomes the most important > component. That's why modern DBMSs have so many features to minimize > I/O. > > While you're working on your new syntax, I hope you'll keep Richard's > query in mind. SQL, verbose as it is, is pretty clean compared to > most ORM syntaxes I've seen. A syntax that can express his query more > succinctly and is also "better" along the lines you describe would be > an achievement. > > --jkl > _______________________________________________ > sqlite-users mailing list > sqlite-users at mailinglists.sqlite.org > http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users >
