Add more info sources

[MDeLuise]

Discussed in #90

^{Originally posted by MDeLuise December 19, 2023}

Avoid duplicated feature requests

• There are no open or closed feature requests that are related to this request

Description

It would be useful to enhance the app by incorporating additional information sources. This not only provides users with more comprehensive data but also serves as a backup in the event that the Trefle API service becomes unavailable or the user does not have access to network.

Solution

During initial exploration, I came across this service as a potential information source for plants. Notably, the database can be downloaded in CSV format from here. A feasible approach would be to parse this file during the app's startup, allowing for the storage of all species contained within it in the app's database.

What are alternatives?

Other services could be used, if you find any please add them in this conversation.

Additional context

Currently, the app uses only Trefle species and "local" species. Consequently, when searching for a species, the user is presented with a maximum of two choices for the same species. To accommodate the expansion of services for retrieving species information, a decision must be made regarding the handling of duplicated species. One approach is to offer only the most detailed entry to the user. Alternatively, all duplicated species from the employed services could be provided, each species tagged by service — using labels such as "Trefle," "Custom," "USDA," and so on.

[MDeLuise]

I've done some initial investigation into additional datasets that could complement the Trefle service.
I came across the WFO Plant List, which provides both a REST API and downloadable datasets. I'm currently exploring whether the datasets are available at this link or at this link.

While I still need to verify the quality and quantity of the data provided, this seems like a promising starting point.

Here are a few thoughts on how we could leverage this resource:

• We could query the REST API directly, since they do not currently require an API key, but they can be a bit slow. I've also discovered that the dataset and server can be self-hosted, which is something I'll be looking into further.
• It would be beneficial to parse the downloaded datasets so that users can search for plants offline or without relying on the Trefle API. However, since the downloaded files are quite large, I'll need to determine the best approach for handling them. One possibility could be providing two different types of backend images — one with the offline information and one without.

I'll continue to explore these options and I'll update this issue on new findings.

[krillin666]

There's also this: https://open.plantbook.io/

[nickgnat]

I don't think there's an API for it, but https://www.gardenia.net/ has a lot of info for each plant and is very searchable

[WallK]

I use https://powo.science.kew.org/ a lot to understand more about my plants
Not sure about APIs, but checklists are downloadable

[justcomplaining]

There's also (with a community aspect):

OpenFarm, https://openfarm.cc/en

GrowStuff, https://www.growstuff.org/

For a larger overview of software and data projects connected to agriculture there's a list here: https://github.com/brycejohnston/awesome-agriculture

[pbiswal]

Hey,

Love this project and looking for ways to contribute back.

I came across Pl@ntNet, which has an API to find information by image.

I wish we could improve the add new plant flow, take an image, and find the scientific information automatically. I will create a feature request later on this.

Thanks