Every analytics tool that 3 years ago had only clickops query builders, now has a "Make with AI" feature for building visualizations. These work for obvious use cases, but their limitation is the fact that they are built on proprietary technology that frontier models are not trained on. You need to add skills and integrations in order to get adequate results - the existing analytics services are working on these, but it's an uphill battle against inertia.
SQL and Grafana are open standards. Frontier models are already proficient with them, and they're the tools I want my mobile analytics platform to be based on, because doing so gives me amazing flexibility and portability.
This kind of tool didn't exist yet, so I built it over the last few weeks. It's designed to work completely from your AI coding harness, including set up and dashboard creation.
If you’re building a mobile app and want an analytics tool that works natively with your development work flow, this is for you.
I agree with you, although I think your example does perhaps display some missing social understanding, even if the intention was good-natured.
> the white balance is so poorly executed on this everyone looks like they're suffering from a severe case of jaundice. It's all I can focus on and it's not helping your film
It's a bit much right? There is a way to give strong negative feedback without sounding affected. You could have communicated the same thing with the slightly toned down: "The white balance is really poorly executed unfortunately. It's all I can focus on and it's not helping your film."
I liked the original article by the author as an example of giving negative feedback.
I really like this. There are clear shibboleths which identify the author as a person who deeply respects and cares for the readers of error messages, and their experiences. It makes me hopeful for the future of software when I see that there are others. Thanks for sharing.
The initial call to deprecate this page was quite obviously driven by usage telemetry (i.e. "engagement" metrics), rather than actually speaking to users about what they like and don't like. I suspect that companies sometimes believe they can automate their way into quality designs, by relying on analytics, rather than empathizing deeply with their customers.
In some Asian vernacular Englishes (e.g. Indian, Singaporean), this is means "yes, right" / "I agree". I used to experience your reaction to people saying "correct" to me while I talk (i.e. "fuck you") but since then, my understanding has increased, and I interpret it to mean "yes, right" / "I agree".
Additionally, it may be that people who say this aren't native speakers of an Asian vernacular English, but are accommodated (i.e. Linguistic Accommodation) to it and so reproduce it in other contexts (e.g. in conversations with you).
Every analytics tool that 3 years ago had only clickops query builders, now has a "Make with AI" feature for building visualizations. These work for obvious use cases, but their limitation is the fact that they are built on proprietary technology that frontier models are not trained on. You need to add skills and integrations in order to get adequate results - the existing analytics services are working on these, but it's an uphill battle against inertia.
SQL and Grafana are open standards. Frontier models are already proficient with them, and they're the tools I want my mobile analytics platform to be based on, because doing so gives me amazing flexibility and portability.
This kind of tool didn't exist yet, so I built it over the last few weeks. It's designed to work completely from your AI coding harness, including set up and dashboard creation.
If you’re building a mobile app and want an analytics tool that works natively with your development work flow, this is for you.
If any questions or interest, reach out to me at [email protected].