SQL, or GoogleSQL, that is the question
Ignatius Fernandez
Exceptional database engineer specializing in database tuning, SQL tuning, automation, software engineering, and technical writing
a large class of users who, while they are not computer specialists, would be willing to learn to interact with a computer in a reasonably high-level, non-procedural query language. ... It is for this class of users that [SQL] is intended. For this reason, [SQL] emphasizes simple data structures and operations.
Has Google successfully solved a problem that does not exist? The problem stated in the paper introducing “GoogleSQL” is: “Fundamentally, the SQL language is difficult for users. For beginners, SQL is hard to learn. Even for expert users, SQL is awkward and annoying – it’s hard to write, hard to read and hard to edit.” (https://research.google/pubs/sql-has-problems-we-can-fix-them-pipe-syntax-in-sql/)
That opinion would have perplexed the creators of SQL who said that there is “a large class of users who, while they are not computer specialists, would be willing to learn to interact with a computer in a reasonably high-level, non-procedural query language. ... It is for this class of users that [SQL] is intended. For this reason, [SQL] emphasizes simple data structures and operations.” (https://dl.acm.org/doi/pdf/10.1145/800296.811515)
If you find that SQL (or any other language) is awkward, annoying, and hard to write, read, and edit, then let AI handle it for you. :-) Check out this sample transcript: https://chatgpt.com/share/6c2e7f63-25e9-4f66-b673-6e1581f1cb10.
Data Engineer freelance developer
1 个月ChatGPT missed one. Use a cross join with an assist from a junction table. ??
Sr. Architect IT @ Tractor Supply
1 个月I admit, I lean on ChatGPT more and more often lately. I predict a day when we can simply talk to our databases. Have you seen this: https://blogs.oracle.com/cloud-infrastructure/post/first-principles-oracle-sql-dialogs-generative-ai
Data Engineering & Architecture @Corporate Technology | Sr. Director of Software Engineering
1 个月That's an excellent point! Building on your observation, SQL and relational databases, which were somewhat overshadowed with the rise of big data and the push for denormalized schemas and the CAP theorem, are now making a strong comeback. We're seeing an increased emphasis on ACID compliance even in big data and distributed platforms like Databricks and Snowflake, which now offer robust support for these features.