BOLD system made their own package. Is this one still needed?

[salix-d]

Hello everyone,

I just learned that BOLD system made their own r package.

It seems to have more a lot more functionalities, including the ability to fetch of private records and analyse the data.
From my quick look on it, the way to fetch the data is different from this package as it reflects the structure of their new API (v5).
I haven't try getting taxonomy data with their package yet, and the website indicates it's still using the v4 API though, which this package uses.

So, if you could try out their package and give your feedback from it (comparing it with this one), I would greatly appreciate it.
I would mainly like to know if their package covers all your needs or if there are still functionalities this package is better suited for.
With that feedback, I'll be able to plan what to do next.

Thank you!

[mpadge]

@salix-d This is not an answer to your question, but might nevertheless give you some useful insight. The following graphs come from https://github.com/ropensci-review-tools/repometrics, and running:

data <- repometrics_data("/<local>/<path>/<to>/bold")
repometrics_dashboard(data)

Number of new issues opened

Number of issue comments

They both indicate a gradual decline, and so might plausibly be interpreted to support a decision to archive this pkg?

---

Footnote: That repometrics pkg is a very rapidly changing WIP, and currently just demonstrates the kinds of things possible. No guarantee of ability to reproduce those results in the future ...)

[salix-d]

I thought it meant the package was doing well without issues 😅

Thanks for the tool!

[noamross]

This kind of thing is always challenging. Despite what I think in an increase focus in OSS and R on getting people to look at prior art, it's very common for competing packages to spring up and very rare for projects to coordinate/merge. And that's not neccessarily bad! Sometimes there are different needs and both should be maintained and/or it makes sense for there to be succession. But I'd encourage you (if you haven't already), to reach out the BOLDconnectR team to discuss possible collaboration. As {bold} has been around 14 years and has a number of reverse dependencies in the taxonomic ecosystem, there are probably lessons to be learned. Maybe it makes sense to integrate some components, maybe they can give insight as to expected API migration, or maybe they just want to do their own thing and that's fine. Given the timing, it's also possible that the package coming off CRAN was part of their motivation, and it would be good to discuss, as well.

[salix-d]

I think the reason for it stems more from their new API; the structure is pretty different.
They were aware of this package and that I was its maintainer; will still try to discuss.

Re: different needs:
That's exacly why I've opened this discussion, to see if that was the case.

[RobLBaker]

For packages that do have dependencies on bold - or depend on packages that themselves depend on bold - what are your suggestions? Is there a timeline for either getting bold back on CRAN or a different approach? If you'd like help, I'm happy to lend a hand towards whatever the most workable solution ends up being.