Real-Time Comment Analysis

2001 Roadless Rule Revision

The docket in numbers: how the pile of submissions was collected, deduplicated, and cleaned — volume, length, duplication, and the organized campaigns running. Sentiment lives on the analysis home; topics on Topics in Depth.

2026-08-20 – 2026-10-01 · 307,291 comments to date
307,291
Comments collectediEvery public submission pulled from regulations.gov for this docket, before any processing — the raw count.
51,552
Unique commentsiThe distinct things the docket was told. Three kinds of repetition are folded into one comment each: identical copies, template letters with an identical body under personalized openings and signatures, and campaign letters built from the same template but reworded. Every 'unique' figure on every page of this analysis is this number. Repetition is detected on comments of at least 100 characters — shorter ones carry too little text to compare.
255,739
Repeated submissionsiSubmissions that repeat another comment — an identical copy, a template letter, or a reworded campaign letter: collected minus unique.
18
Topics trackediThe number of topic categories the classifier can assign to a comment (a comment can raise several).
304,982
Classified for sentimentiComments read by the model and labeled for their position — supports, opposes, or neutral. Identical duplicates inherit their group's label.

Comment length distribution

How long the comments run — a proxy for substance versus one-line form submissions.

17,084
34,168
51,251
68,335
2505007501k1k1k1k2k2k2k2k3k3k3k3k4k4k4k4k5k5k+
Comment length (characters, upper bound of bin)Bar height = number of comments

About the comments

Repetition is folded away one kind at a time, so each row is what still stands after that kind. The bottom row is the “unique comments” figure used on every page of this analysis. Repetition is detected on comments of at least 100 characters — shorter ones carry too little text to compare meaningfully.

83% repeated
Collectedall submissions
307,291
After identical copiesthe same text submitted more than once
306,545−746 folded into 746
After template lettersan identical body under personalized openings and signatures
306,541−4 folded into 4
After small familiesthree to nine copies of one letter, caught by shared text
302,741−3,800 folded into 1,081
Unique commentswhat remains
51,552

Organized campaigns

Beyond exact duplicates, comments are grouped into campaigns — sets built from the same template, whether copied word-for-word or lightly reworded. These are the last two rows of the funnel above: 251,818 submissions folded into 625, each counted once in the “unique comments” total. Below is who is running them and how large they are.

625 distinct campaigns

Exact form letters — 8 submissions share a verbatim template block, folded into 4

21,570“keep the roadless rule in place”
21,308“keep the roadless rule in place”
157“keep the roadless rule in place”
81“keep the roadless rule in place”
78“you to keep the roadless rule”
77“keep the roadless rule in place”

Reworded templates — 251,810 submissions match a paraphrased version (caught by meaning, not exact words), folded into 621

46,106Please save this magical land. grew up always seeing nature forests in all beautiful things. It would be a tragedy that because we think that there is…
10,033Special Areas; Roadless Area Conservation Public Comment,Last year, more than 625,000 people from across the United States wrote to the USDA, urging t…
454am a resident of Doral in the state of Florida and respectfully urge you to withdraw this proposal and retain the Roadless Rule's national protect…
242Dear Joshua White, USDA Forest Service Acting Director of Ecosystem Management Coordination:<br/><br/>I'm respectfully writing to oppose the US Fo…
140strongly oppose the proposed repeal of the 2001 Roadless Rule.<br/><br/>Our national forests are public lands, our shared birthright, and they should…
128Dear Joshua White, <br/><br/>I am born and raised here in the Tongass. spent my childhood playing, subsisting and enjoying the Tongass National Forest…

Sharing & methodology

These results are meant to be shared.

Please cite us. This analysis is the intellectual property of Roadless.org and Nicholas Holshouser. You are welcome to reproduce it — every reproduction must include this citation:
Roadless.org and Nicholas Holshouser, “2001 Roadless Rule Revision — Public Comment Analysis,” https://roadless.org.

How it works. Sentiment and topic classification is performed by a large language model (an open-weight Qwen 2.5 model) that reads each comment and labels its position — supports, opposes, or neutral — and the topics it raises. Lexical analysis, which finds duplicate and template submissions and scores each comment for originality, uses TF-IDF and n-gram phrase matching, sentence-transformer embeddings, and syntactic and stylometric features (built with spaCy, scikit-learn, and sentence-transformers).