Timeline for Fabricated data in posts.xml for multiple/all data dumps
Current License: CC BY-SA 4.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 30 at 17:09 | history | bounty awarded | Bryan Krause | ||
| Aug 27 at 21:22 | history | edited | M-- | CC BY-SA 4.0 | added 1 character in body |
| Aug 21 at 4:36 | history | undeleted | user314146 | ||
| Aug 21 at 4:35 | history | deleted | user314146 | via Vote | |
| Aug 20 at 0:05 | history | edited | DLosc | CC BY-SA 4.0 | Fixed typos |
| Aug 19 at 19:17 | comment | added | NoDataDumpNoContribution | @dan1st Thanks for reminding me that nobody is stopping me. :) | |
| Aug 19 at 16:17 | comment | added | dan1st | @NoDataDumpNoContribution If you really want that, feel free to do so - nobody stops you. | |
| Aug 19 at 15:44 | comment | added | NoDataDumpNoContribution | Maybe also compute with previous data dumps (the parts where no edits occurred) and random sampling of the changed/new parts could do the trick. You only need to find a discrepancy once (or a few times) to lose faith in the whole process. | |
| Aug 18 at 21:44 | comment | added | anon | I have wondered the same thing. I did a full diff of several of the small-medium sites where there is less data volume, and there doesn't appear to be tampering beyond the additions. BUT, your comments prove as useful data points for the Company to see how un-announced tinkering with the data dump erodes trust. People no longer trust the data in the data dump. | |
| Aug 18 at 21:16 | comment | added | dan1st | @NoDataDumpNoContribution If only a few (e.g. <50) posts were tempered with by changing the content, that would be hard to check properly as it would probably be fairly unlikely to stumble across one of these posts by randomly sampling only a few posts. That being said, the current one was sufficiently obvious/blatant that I don't think they made any other modifications to that data dump. | |
| Aug 18 at 21:09 | comment | added | NoDataDumpNoContribution | Btw. could also other content be altered or removed or added and how could we find out. Maybe comparing random samples with the website? | |
| Aug 18 at 14:52 | history | edited | anon | CC BY-SA 4.0 | added 7519 characters in body |
| Aug 14 at 18:22 | history | edited | bobble | CC BY-SA 4.0 | it's = it is |
| Aug 14 at 18:21 | history | edited | dan1st | CC BY-SA 4.0 | fix typos |
| S Aug 14 at 18:11 | history | answered | anon | CC BY-SA 4.0 | |
| S Aug 14 at 18:11 | history | made wiki | Post Made Community Wiki by anon |