Re:In Praise of PostgreSQL
by gregjor on Saturday August 07, 2021 @03:28AM
(#28096559)
What a bizarre love letter to a piece of software. Some of this looks cribbed from the postgres.org site.
"After 25 years of persistence, and a better logo design, Postgres stands today as one of the most significant pillars of profound achievement in free software, alongside the likes of Linux and Firefox." The better logo really does it. All three of the "pillars" the author names are clones or derivatives of commercial products (Ingres, Unix, Netscape Navigator). Postgres development was partly funded by DARPA at Berkeley, and then through licensing to IBM, before it was open-sourced. Only Linux can be called a "profound achievement" of free software if the author means the cathedral and bazaar model, and then just barely.
"PostgreSQL has taken a complex problem and solved it to such an effective degree that all of its competitors are essentially obsolete, perhaps with the exception of SQLite." Uh, no. The whole notion of "competitors" in the world of open source software is flawed. PostgreSQL has not made Oracle, SQL Server, and DB/2 obsolete, nor has it made MySQL obsolete, or any of the other database managers we have to choose from. PostgreSQL is about half as "popular" as MySQL (https://db-engines.com/en/ranking), and both are below Oracle.
"... providing the best implementation of SQL" and in the footnote "No qualifiers. Itβs straight-up the best implementation of SQL." I can't tell what the author means by "best," it's like writing that Oreos are the "best" cookies. In terms of standards compliance DB/2 is better. In terms of performance and scalability Oracle, SQL Server, and DB/2 are all objectively "better." In terms of compared to the main open source alternative, MySQL, the two are nearly neck-and-neck with features, compliance, and performance.
"SQL is usually the #1 bottleneck in web applications, and Postgres does an excellent job of providing you with the tools necessary to manage that bottleneck." Not in my experience. Sometimes excessive database traffic causes performance issues in web apps, but many other problems show up, like Apache/nginx contending for resources with MySQL or PostgreSQL (hint: don't run those on the same server/VPS). If the code makes too many database queries, or badly-written queries (WordPress and pretty much every ORM) that's a fault in the application code, not the database, and PostgreSQL doesn't magically fix it.
"After 25 years of persistence, and a better logo design, Postgres stands today as one of the most significant pillars of profound achievement in free software, alongside the likes of Linux and Firefox." The better logo really does it. All three of the "pillars" the author names are clones or derivatives of commercial products (Ingres, Unix, Netscape Navigator). Postgres development was partly funded by DARPA at Berkeley, and then through licensing to IBM, before it was open-sourced. Only Linux can be called a "profound achievement" of free software if the author means the cathedral and bazaar model, and then just barely.
"PostgreSQL has taken a complex problem and solved it to such an effective degree that all of its competitors are essentially obsolete, perhaps with the exception of SQLite." Uh, no. The whole notion of "competitors" in the world of open source software is flawed. PostgreSQL has not made Oracle, SQL Server, and DB/2 obsolete, nor has it made MySQL obsolete, or any of the other database managers we have to choose from. PostgreSQL is about half as "popular" as MySQL (https://db-engines.com/en/ranking), and both are below Oracle.
"... providing the best implementation of SQL" and in the footnote "No qualifiers. Itβs straight-up the best implementation of SQL." I can't tell what the author means by "best," it's like writing that Oreos are the "best" cookies. In terms of standards compliance DB/2 is better. In terms of performance and scalability Oracle, SQL Server, and DB/2 are all objectively "better." In terms of compared to the main open source alternative, MySQL, the two are nearly neck-and-neck with features, compliance, and performance.
"SQL is usually the #1 bottleneck in web applications, and Postgres does an excellent job of providing you with the tools necessary to manage that bottleneck." Not in my experience. Sometimes excessive database traffic causes performance issues in web apps, but many other problems show up, like Apache/nginx contending for resources with MySQL or PostgreSQL (hint: don't run those on the same server/VPS). If the code makes too many database queries, or badly-written queries (WordPress and pretty much every ORM) that's a fault in the application code, not the database, and PostgreSQL doesn't magically fix it.