Huh, that's an interesting take - honestly I don't really know what to do about that :/
It's definitely not our intention to send that message - Apollo Client should be a good fit for everybody.
The biggest adoption hurdle I could think of is that we only offer a normalized cache.
From what we're seeing everybody sooner or later has a good use for it, and switching from a non-cache solution to a normalized cache can cause a lot of pain when it's done too late, so we don't give people the "without" choice from the start.
That said, I've been looking at your docs, and you seem to make any kind of cache completely optional. That's a bold move the other direction ^^
I'll be keeping an eye out on how people start using that :)
Make sure to stop by the GraphQL Discord - most of the maintainers of the different cache libraries are over there and we're always up for a chat!
lead engineer of unchained engine here. still using apollo client for client projects but i stopped using apollo server because you guys started to put important security features behind your paywall for example whitelisting persistent queries (pql). that makes it difficult trust-wise.
I'm an engineer on the client team, so I'm a bit fuzzy on the Apollo Server details and history, so please bear with me :)
I guess you're talking about [Safelisting Persisted Queries](https://www.apollographql.com/docs/graphos/platform/security...) here.
Afaik, if you want to use the full integration with the GraphOS platform, you'll need an account with us. The feature is supported on the free account with a rate-limit, and if you would need more, you could go with the "developer" account, which we launched this summer, which is quite reasonably priced and comes with starting credits: https://www.apollographql.com/pricing
That said, Apollo Server is open source and at no point you are locked into our platform - it's just that our platform takes a lot of work away from you, allowing you to manage those whitelists easier.
If you want to manage your whitelist yourself, it's totally doable - I just did a quick google search and have come up with more than a dozen articles describing various ways of doing it.
Just be aware that there is a lot of coordination between builds and deployments going on, quickly leading to very complex setups - which is why we offer the platform integration in the first place.
So I think the idea was not to lock anyone out of these features, just create a quick pit of success by making the GraphOS integration really easy - but I can see where your feelings are coming from. I hope we can earn back your trust over time!
It's definitely not our intention to send that message - Apollo Client should be a good fit for everybody.
The biggest adoption hurdle I could think of is that we only offer a normalized cache. From what we're seeing everybody sooner or later has a good use for it, and switching from a non-cache solution to a normalized cache can cause a lot of pain when it's done too late, so we don't give people the "without" choice from the start.
That said, I've been looking at your docs, and you seem to make any kind of cache completely optional. That's a bold move the other direction ^^ I'll be keeping an eye out on how people start using that :)
Make sure to stop by the GraphQL Discord - most of the maintainers of the different cache libraries are over there and we're always up for a chat!