Bug #103062 | MySQL Replication extract from 5.7 is non-consistent with Reference Manual. | ||
---|---|---|---|
Submitted: | 22 Mar 2021 16:07 | Modified: | 6 Apr 2021 12:33 |
Reporter: | Jean-François Gagné | Email Updates: | |
Status: | Verified | Impact on me: | |
Category: | MySQL Server: Documentation | Severity: | S3 (Non-critical) |
Version: | OS: | Any | |
Assigned to: | CPU Architecture: | Any |
[22 Mar 2021 16:07]
Jean-François Gagné
[22 Mar 2021 16:40]
Jean-François Gagné
Looking more closely into this, the content might be the same, but the fact that everything is in a single page vs in sub-pages confused me. And if I was confused, then I guess some other people might get confused. If would be good to simplify this.
[22 Mar 2021 17:31]
MySQL Verification Team
Thanks for the report Bogdan
[6 Apr 2021 10:47]
Margaret Fisher
Posted by developer: Thanks for the comments! As you said in the update the content is exactly the same, the extracts are created from the original source for the full manual. The extracts have been around for a while and my understanding is that they are created as a more lightweight and modular way for users to engage with themes in the reference manual, which is pretty large. The different page divisions are because the top level division of the material is higher for the extracts. So the build process creates subtopics where the same content in the reference manual would all be on one page. Personally, I think this helps make the extracts easier to parse and search, and helps reduce the cognitive load for users in the case of the extracts, which supports what I believe is their intended purpose. I hope that explanation is okay! I don't think it's likely that many people will compare the extract directly to the reference manual in this way. So I think the benefits of the more modular structure outweigh the chances of confusion here, and we wouldn't change it for that reason.
[6 Apr 2021 12:33]
Jean-François Gagné
Thanks for the reply Margaret. > I don't think it's likely that many people will compare the extract directly to the reference manual in this way. Allow me to try changing your mind on this. I ended-up comparing the doc because I searched Google for "slave_type_conversions" (or one of ALL_LOSSY, ALL_SIGNED, ...) and the two first results are the reference manual and the replication extract. Realizing that I was reading twice almost the same thing, this lead me to compare the docs and open this bug. IMHO, the combination of having a replication extract and having different doc structure makes it hard, when arriving from Google, to see that these two are the same doc. The different section numbering is also increasing this confusion. I will let you decide is this is worth fixing (I could argue both ways). Thanks.