Briefly yesterday, linkedin.com was down with Cloudflare's internal server error page.
I don't think that's necessarily a bad review for Azure services, there are other explanations, like organizational inertia, contractual agreements, team preference.
You can feel the pain of GitHub migrating to Azure from afar. With a beast that LinkedIn is, there's got to be a good prize for someone at the end of the road before they pull the trigger.
They are not retiring the API. Nginx Ingress is one of the many projects that implements this API, and you are free to migrate to another implementation.
I dabbled in this area, there are poster layout generation attempts that use gen ai to come up with an initial layout plan, and even feed the visualization back into the llm for iterative fine-tuning.
I was intrigued, but couldn't make it work reliably. Perhaps I forgot to add "make it look nice" to my prompts.
This tool mainly differs around how identities are structured and how the evaluation works.
Identities represent specific entities in your application. For example pricing plan can have it's own identity, organization users belong to can be separate identity, users themselves are represented as identity objects.
You define as many identity types as you need through the interface or the API, then create actual identities through the API from you server-side application.
Now the flags evaluation part is where it differs from most other tools, as you define the hierarchy at evaluation time.
Let's say you have Premium product identity, Mailstorm Inc. organization identity, and Alicia end user identity. Alicia belongs to Mailstorm Inc. organization which is subscribed to Premium plan.
When Alicia is using the app, you could evaluate feature flags for her by sending following identity IDs: premium_id, mailstorm_id, alicia_id (order is important)
For each feature flag evaluated it will then take the first override it encounters among the provided identities, or fall back to the default environment value.
This gives a lot of flexibility and control on flag evaluations.
Briefly yesterday, linkedin.com was down with Cloudflare's internal server error page.
I don't think that's necessarily a bad review for Azure services, there are other explanations, like organizational inertia, contractual agreements, team preference.
You can feel the pain of GitHub migrating to Azure from afar. With a beast that LinkedIn is, there's got to be a good prize for someone at the end of the road before they pull the trigger.
reply