Will PocketBase Survive?(github.com) |
Will PocketBase Survive?(github.com) |
There's no ability to set null as a value of an item despite null being a perfectly valid data type in SQLite (which is what Pocketbase uses).
Having to arbitrarily assign a MAGIC CONSTANT value (like -1 for example) to represent a "lack of information" in a column is kludgy at best, and impossible at worst - e.g. if all floating point values have meaning in a column, you're out of luck. The only workaround is to create an unnecessary additional boolean column isNumFilled to indicate if the data in the numerical column is valid.
This is the situation for a lot of open source projects.
In fact, ~~development velocity has slowed recently~~ PocketBase has been in a feature freeze recently.
But like with any project like this, you need to weigh up the pros and cons of whatever tech stack you choose.
PocketBase at least is MIT licensed, so you can fork and fix/add new features if you want. If you’re running a successful business with PocketBase as a backend, it might be sensible to do this anyway, and potentially hire a Go dev or consultant to do whatever work you need doing, rather than hoping Gani has it on his priority list.
Obviously everyone has a price, so maybe I’m wrong, but even then you’ve not solved the “bus factor” problem.
Asking “will it survive” is the wrong question. It’s already there, works, and is used by thousands of people. Whether it gets updates or not depends entirely on whether Gani wants to and is able to keep updating it. No one can answer that except Gani, and he’s saying “no promises”.