Timeline for Is it a bad practice to have a "record status" column in a database table?
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 11, 2018 at 17:30 | vote | accept | ADTC | ||
| May 9, 2017 at 9:29 | history | tweeted | twitter.com/StackSoftEng/status/861875661639745536 | ||
| May 6, 2017 at 3:19 | answer | added | Archimedes Trajano | timeline score: 1 | |
| Apr 27, 2017 at 8:31 | history | edited | ADTC | CC BY-SA 3.0 | added 172 characters in body |
| Mar 9, 2016 at 21:44 | answer | added | Mike | timeline score: 4 | |
| Mar 9, 2016 at 20:25 | answer | added | Jules | timeline score: 4 | |
| Mar 9, 2016 at 18:59 | answer | added | kevinskio | timeline score: 1 | |
| Mar 9, 2016 at 12:35 | comment | added | Doc Brown | The problem with the unique constraints (which you already named) is exactly why history tables are often preferable - you can keep the unique key constraints on the original tables, and do not add them on the history table.Moreover, separate history tables allow easier to use specific (DB dependent) storage options for them, so they are often better in terms of storage, not worse. When you have lots of those tables, the triggers and history tables should not be hand-written, but generated, that will solve the problem how to keep them "up-to-date". | |
| Mar 9, 2016 at 12:23 | answer | added | Phill W. | timeline score: 7 | |
| Mar 9, 2016 at 7:38 | answer | added | Justin Cave | timeline score: 9 | |
| Mar 9, 2016 at 7:14 | history | asked | ADTC | CC BY-SA 3.0 |