Re: [GENERAL] PostgreSQL Global Development Group - Mailing list pgsql-hackers
From | Peter Eisentraut |
---|---|
Subject | Re: [GENERAL] PostgreSQL Global Development Group |
Date | |
Msg-id | Pine.LNX.4.44.0212091847380.25355-100000@localhost.localdomain Whole thread Raw |
In response to | Re: [GENERAL] PostgreSQL Global Development Group (Robert Treat <xzilla@users.sourceforge.net>) |
Responses |
Re: [GENERAL] PostgreSQL Global Development Group
Re: [GENERAL] PostgreSQL Global Development Group |
List | pgsql-hackers |
Robert Treat writes: > I think we've already shown why it doesn't hurt to market to the > converted. I'll add that if you compare the 7.2 press release with the > 7.3 press release, you'll see none of the technical content was removed. Compare the 7.3 release notes, written for the most part by Bruce Momjian and revised by a couple of other developers, to the "press release", written by people who were obviously ill-informed. Release notes: Schemas Schemas allow users to create objects in their own namespace so two people or applications canhave tables with the same name. There is also a public schema for shared tables. Table/index creationcan be restricted by removing permissions on the public schema. Press release: Schemas PostgreSQL now joins the handful of ORDBMS's to support the SQL 92 Schema specification, improvingboth enterprise database management and security through the use of namespaces. This not only removes all information about the actual use of schemas, it contains completely bogus information, because SQL 92 is obsolete, there is no "SQL Schema specification", and none of this has to do with being an ORDBMS. And besides, whose hands were used to do the counting? Release notes: Drop Column PostgreSQL now supports the ALTER TABLE ... DROP COLUMN functionality. Press release: <void> Release notes: Table Functions Functions returning multiple rows and/or multiple columns are now much easier to use thanbefore. You can call such a "table function" in the SELECT FROM clause, treating its output like a table.Also, PL/pgSQL functions can now return sets. Press release: Table Functions PostgreSQL version 7.3 has greatly simplified returning result sets of rows and columns in databasefunctions. This significantly enhances the useability of stored procedures in PostgreSQL, and will makeit even easier to port Oracle applications to PostgreSQL. Again, this removes all details about how the feature can be used, and again it inserts completely bogus information. There are no "sets of columns", and PostgreSQL does not have stored procedures. Also, it makes it look as though PostgreSQL exists merely to reimplement Oracle. Release notes: Prepared Queries PostgreSQL now supports prepared queries, for improved performance. Press release: - Prepared queries for maximized performance on common requests. I'm curious to know how the marketing department determined that this is, in fact, the maximal performance. Release notes: Dependency Tracking PostgreSQL now records object dependencies, which allows improvements in many areas."DROP" statements now take either CASCADE or RESTRICT to control whether dependent objects are alsodropped. Press release: - Enhanced dependency tracking for complex databases. Again, all relevant information dropped, replaced by marketing fluff. Release notes: Privileges Functions and procedural languages now have privileges, and functions can be defined to runwith the privileges of their creator. Press release: Security Advances In response to community demands, PostgreSQL has added schema, function, and other permissionsand settings to increase the database administrator's granular control over security. Information dropped, replaced by broad and repetitive verbiage. But at least they didn't write, "in response to market pressures". And my personal favorite is this: Release notes: Internationalization Both multibyte and locale support are now always enabled. Press release: - Supports data in many international characters sets (UNICODE, EUC_JP, EUC_CN, EUC_KR, JOHAB, EUC_TW, ISO8859-1 ECMA-94, KOI8, WIN1256, etc...) That is just plain wrong. Support for various character sets is years old. -- Peter Eisentraut peter_e@gmx.net
pgsql-hackers by date: