Over the previous weeks, now we have seen renewed dialogue/concern within the MySQL neighborhood round claims that “Oracle has stopped creating MySQL” or that “MySQL is being deserted.” These issues have been amplified by graphs exhibiting an obvious halt in GitHub commits after October 2025, in addition to by weblog posts and discussion board discussions that interpreted these indicators at face worth.
As somebody who has publicly analyzed the MySQL repository exercise and who works every day with MySQL at Percona, I need to clearly separate what the information really exhibits from what it doesn’t.
This submit shouldn’t be learn as an uncritical protection of Oracle. We regularly disagree with Oracle’s choices, and we are saying so brazenly. However equity issues — particularly when concern, uncertainty, and doubt (FUD) begin affecting clients and the broader ecosystem.
The Declare: “Oracle Stopped Committing to MySQL After October 2025”
We lately had a shocking query from our neighborhood. Has MySQL actually been deserted? Additionally they included the graph that was shared in Otto Kekäläinen’s submit.
This conclusion is normally drawn from GitHub exercise graph on the general public https://github.com/mysql/mysql-server/ repository, which certainly exhibits lengthy durations with no seen commits.
The graph itself just isn’t fallacious. However the interpretation is incomplete.
The Lacking Context: How MySQL Is Truly Developed
The error lies in assuming that MySQL is developed on GitHub, which isn’t the case. For a few years, Oracle has adopted a selected workflow wherein real-time engineering takes place in personal, closed repositories. GitHub serves solely as a public mirroring and publication platform slightly than an lively improvement workspace. Consequently, code is launched to the general public in massive, consolidated “code drops” that align with official releases, as a substitute of showing as incremental every day commits.
In different phrases:
GitHub is an asynchronous publication mirror, not the event system of document.
This implies:
- A scarcity of incremental commits on GitHub doesn’t indicate an absence of improvement
- Lengthy quiet durations are anticipated between launch pushes
- Sudden massive commit bursts are the traditional launch mechanism
This improvement mannequin just isn’t new, it has been this fashion for a few years. Can one argue that this isn’t a “actually open supply improvement mannequin”? Perhaps, however in the long run, the identical graph that was pulled on 21/01/2026 (after current releases 9.6.0, 8.4.8, 8.0.45) doesn’t look deserted anymore.
The MySQL “abandonment” narrative is an ideal reminder that metrics are solely pretty much as good as our understanding of the techniques they measure. A flatlining GitHub graph isn’t at all times a pulse verify for a dying mission; usually, it’s simply the silence of an engine operating behind a closed door. Whereas we will debate the transparency of Oracle’s improvement mannequin, we shouldn’t mistake a unique workflow for an absence of labor. It’s not at all times because it appears, and it’s a mistake to guage a guide by its cowl, or a database by its mirror.
