[PRE REVIEW]: rollama: An R package for using generative large language models through Ollama

[editorialbot]

Submitting author: !--author-handle-->@JBGruber<!--end-author-handle-- (Johannes B. Gruber) Repository: https://github.com/JBGruber/rollama Branch with paper.md (empty if default branch): Version: 0.0.3 Editor: Pending Reviewers: Pending Managing EiC: Arfon Smith

Status

Status badge code:

HTML: <a href="https://joss.theoj.org/papers/8edaa677d5ce04288dbe65c196a881af"><img src="https://joss.theoj.org/papers/8edaa677d5ce04288dbe65c196a881af/status.svg"></a>
Markdown: [![status](https://joss.theoj.org/papers/8edaa677d5ce04288dbe65c196a881af/status.svg)](https://joss.theoj.org/papers/8edaa677d5ce04288dbe65c196a881af)

Author instructions

Thanks for submitting your paper to JOSS @JBGruber. Currently, there isn't a JOSS editor assigned to your paper.

@JBGruber if you have any suggestions for potential reviewers then please mention them here in this thread (without tagging them with an @). You can search the list of people that have already agreed to review and may be suitable for this submission.

Editor instructions

The JOSS submission bot @editorialbot is here to help you find and assign reviewers and start the main review. To find out what @editorialbot can do for you type:

@editorialbot commands

[editorialbot]

Hello human, I'm @editorialbot, a robot that can help you with some common editorial tasks.

For a list of things I can do to help you, just type:

@editorialbot commands

For example, to regenerate the paper pdf after making changes in the paper's md or bib files, type:

@editorialbot generate pdf

[editorialbot]

Software report:

github.com/AlDanial/cloc v 1.90  T=0.03 s (930.0 files/s, 111495.2 lines/s)
-------------------------------------------------------------------------------
Language                     files          blank        comment           code
-------------------------------------------------------------------------------
Markdown                         7            203              0            845
R                               12             96            218            450
SVG                              1              4              1            362
TeX                              1             34              0            299
YAML                             3             18              5             78
Rmd                              4            178            549             17
-------------------------------------------------------------------------------
SUM:                            28            533            773           2051
-------------------------------------------------------------------------------

Commit count by author:

    46  JBGruber
    17  Maximilian Weber

[editorialbot]

Paper file info:

📄 Wordcount for paper.md is 1243

✅ The paper includes a Statement of need section

[editorialbot]

License info:

🟡 License found: GNU General Public License v3.0 (Check here for OSI approval)

[editorialbot]

:point_right::page_facing_up: Download article proof :page_facing_up: View article proof on GitHub :page_facing_up: :point_left:

[arfon]

@editorialbot query scope

@JBGruber – due to the relatively small size of this code, the editors will now discuss if it meets the substantial scholarly effort criterion for review by JOSS. We should get back to you sometime next week.

[editorialbot]

Submission flagged for editorial review.

[editorialbot]

Reference check summary (note 'MISSING' DOIs are suggestions that need verification):

OK DOIs

- 10.48550/arXiv.2401.00284 is OK
- 10.31235/osf.io/sthwk is OK
- 10.31235/osf.io/u9nft is OK
- 10.1073/pnas.2305016120 is OK
- 10.48550/arXiv.2303.16854 is OK
- 10.31219/osf.io/cx752 is OK
- 10.31235/osf.io/x58kn is OK
- 10.18653/v1/S17-2088 is OK
- 10.1038/d41586-023-01295-4 is OK
- 10.1145/3543873.3587368 is OK
- 10.1177/00491241221134527 is OK
- 10.5255/UKDA-SN-5790-2 is OK
- 10.48550/arXiv.2307.02179 is OK
- 10.48550/arXiv.2312.14972 is OK
- 10.1080/19312458.2023.2261372 is OK
- 10.1017/pan.2023.20 is OK
- 10.5117/CCR2023.1.003.CHAN is OK
- 10.1037/met0000542 is OK

MISSING DOIs

- No DOI given, and none found for title: SentimentGPT: Exploiting GPT for Advanced Sentimen...
- No DOI given, and none found for title: Large Language Models are Zero-Shot Reasoners
- No DOI given, and none found for title: Power Hungry Processing: Watts Driving the Cost of...
- No DOI given, and none found for title: Is a prompt and a few samples all you need? Using ...
- No DOI given, and none found for title: Automated Annotation with Generative AI Requires V...
- No DOI given, and none found for title: Leveraging Label Variation in Large Language Model...
- No DOI given, and none found for title: ChatGPT-4 Outperforms Experts and Crowd Workers in...
- No DOI given, and none found for title: Self-Consistency Improves Chain of Thought Reasoni...
- No DOI given, and none found for title: Chain-of-Thought Prompting Elicits Reasoning in La...
- No DOI given, and none found for title: A Prompt Pattern Catalog to Enhance Prompt Enginee...
- No DOI given, and none found for title: Can ChatGPT Reproduce Human-Generated Labels? A St...
- 10.1162/coli_a_00502 may be a valid DOI for title: Can Large Language Models Transform Computational ...
- No DOI given, and none found for title: Can ChatGPT Understand Too? A Comparative Study on...
- 10.31219/osf.io/rvy5p may be a valid DOI for title: Testing the Reliability of ChatGPT for Text Annota...
- No DOI given, and none found for title: Large language models and political science
- 10.18653/v1/2023.emnlp-main.575 may be a valid DOI for title: Modeling Legal Reasoning: LM Annotation at the Edg...
- No DOI given, and none found for title: reticulate: Interface to ’Python’

INVALID DOIs

- None

[JBGruber]

Thank you for taking the paper into consideration.

We see the scholarly effort not just in the size of the codebase, but also in making this fast moving field available to R users through a familiar interface and our learning materials consisting of web tutorials and the video lecture. As we describe in the paper, the package fills a gap in the toolkit of researchers wanting to use open source large generative language models. Currently, our fields (communication, political science and sociology) start to embrace generative AI for text annotation, albeit through the use of Openai's proprietary API. As researchers who care about open science and reproducibility, we see this as a big problem and the package is a way to nudge less technical users to embrace better research practices by making them easier to use and understand.

[arfon]

@JBGruber - thanks for your submission to JOSS. Unfortunately, after review by the JOSS editorial team we've determined that this submission doesn't meet our substantial scholarly effort criterion.

One possible alternative to JOSS is to follow GitHub's guide on how to create a permanent archive and DOI for your software. This DOI can then be used by others to cite your work.

[arfon]

@editorialbot reject

[editorialbot]

Paper rejected.