Change name

[Robinlovelace]

SI is taken and si is therefore not allowed under CRAN policies: https://cran.r-project.org/package=SI

Options:

• sim
• spint
• any others?

sims is taken: https://cran.r-project.org/package=sims

[Nowosad]

sim is fine; simodels could be an alternative...

[Robinlovelace]

Thanks Jakub, I was thinking you'd have good ideas on this : ) FYI I discussed this package today with @adamdennett and @fcorowe in the context of a short chapter on SIMs. My plan: get the package tidied up and submitted to CRAN by end of the month.

[Nowosad]

https://cran.r-project.org/web/packages/changer/index.html -- I used this package to change a name of some of my packages in the past -- very straightforward

[Robinlovelace]

Nice! I will give that a try!

[fcorowe]

sim is fine; simodels could be an alternative...

Yes, sim would make sense - in reference to the models. Like it!

[Robinlovelace]

Tried again with the winbuilder and... Should have checked the package name:

Conflicting package names (submitted: sim, existing: SIM [https://bioconductor.org/packages/3.15/bioc])

[Robinlovelace]

available::available("sim")
Urban Dictionary can contain potentially offensive results,
  should they be included? [Y]es / [N]o:
1: 
── sim ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Name valid: ✔
Available on CRAN: ✔ 
Available on Bioconductor: ✖
Available on GitHub:  ✖ 
Abbreviations: http://www.abbreviations.com/sim
Wikipedia: https://en.wikipedia.org/wiki/sim
Wiktionary: https://en.wiktionary.org/wiki/sim
Sentiment:???

[fcorowe]

What about sim_models?

[Robinlovelace]

simodels sounds the best of the suggestions above to me. _ is not allowed in pkg names. si.models could work also..

[Robinlovelace]

And for completeness:

available::available("sim_models")
Urban Dictionary can contain potentially offensive results,
  should they be included? [Y]es / [N]o:
1: N
── sim_models ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Name valid: ✖
Available on CRAN: ✔ 
Available on Bioconductor: ✔

[Robinlovelace]

Reopening only as I've thought of another possibility: tidysim or tidysi

Pros and cons:

• tidysim is short and captures the fact this is about data processing and 'tidying' well but implies it is a bit tied to the tidyverse (which it is with dplyr dependency and tidyeval)
• tidysi is even shorter but potentially confusing
• simodels implies that the package is about models, which it is not currently the case, still my favourite as it has the most 'does what it says on the tin' name minimising ambiguity.

Thoughts? Least of our worries atm and default is stick with simodels but any feedback on this welcome!

[fcorowe]

I see where you are coming from. I would have thought that a natural evolution for the package in the near future would be to incorporate more modelling framework and estimation procedures. If that is the expectation for the package, I would go for simodels.

[adamdennett]

Only one comment here - not sure if it's either for or against - but simodel was the name of Stewart Fotheringham's SIM package which I think was eventually converted into visual basic by someone in Japan. At least that's where I found it. I know there are some in the community who have views about that academic. Simodels is reminiscent. How about something like SIMple or variations thereof - SIMplr etc.?

[Robinlovelace]

That is interesting, didn't know about simodel! I think if its in the distant past and only hardcore SIM enthusiasts know about it it's not too much of a risk. I'm not a fan of mixed uppercase lowercase words. I like simplr though!

[Robinlovelace]

But the 'does what it says on the tin' nature of simodels makes it still my fave...

[adamdennett]

yeah, I quite like simodels