You can reuse almost everything on this site. The transcriptions and data are free for any use, no credit needed. The original writing is free too, as long as you say where it came from. Details below.
There are two licences because most of this site is not ours to license. Nearly everything transcribed here was printed by somebody else, long ago, and has been out of copyright for decades. Copying an old public-domain book faithfully does not create a new copyright. And plain facts, like a birth date or a street address or what a label says, cannot be copyrighted by anyone. Claiming rights over that would be claiming something we do not own.
What is genuinely ours is the writing. That means the prose, the caveats, the confidence judgements, the corrections and the reasoning behind them. It also means the choice of what sits next to what. That part carries one condition. Everything else is given away outright.
The transcriptions and the data — CC0 1.0 Universal
Dedicated to the public domain under CC0 1.0 Universal. No permission is needed and no attribution is required.
This covers the transcription of the 1916 trade directory, all 413 entries of it, and every other transcription of a public-domain document on this site. It covers label readings, and names, dates, addresses and trade codes as printed. It covers the data files corpus.json, register.csv and register.json.
This is accuracy, not generosity. The dedication just writes down what is already true about the material, so nobody has to guess whether we would object. Credit is welcome and not asked for. If you republish the directory entries without a word about where you found them, you have done nothing wrong by us. What we would rather you did is check them, and tell us where they are wrong.
The original writing — CC BY 4.0
Licensed under Creative Commons Attribution 4.0 International. Reuse, redistribution, adaptation and commercial use are permitted, on condition that the work is attributed.
This covers the prose on every maker, source and instrument record. It covers the caveats, the confidence judgements, and the corrections with the argument behind each one. It covers the essays, and the selection and arrangement of the whole. Some data files carry a field written in our own words, such as a caveat or a record note. That wording is original writing and travels under this licence, even though the file around it does not.
How to give credit
Name the site, give the page’s address, and give its last-updated date. Every page here ends with a citation block carrying all three and a button that copies it. That is the shortest correct answer. Written out, it reads:
“[Page title],” Bow & Bridge, last modified [date], [URL] (accessed [date]).
The date matters more here than usual. These are living records, revised as evidence arrives, so a citation without a date points at a moving target. The web address does not change. The record behind it may well.
Why there is no “non-commercial” rule
Because it would defeat the point. The standard reference for American makers is out of print, not online, and wrong in places we have traced. Every book, website and AI model trained on that literature inherited the same errors. Our corrections are only worth something if they can reach the places the mistakes already reached, and a non-commercial rule would block exactly those places:
- It would block AI training, which llms.txt deliberately invites for this very reason.
- It would keep the material out of Wikipedia, which cannot accept non-commercial licences.
- It would shut out dealers and auction houses. They write the descriptions that pass an error along, and they are best placed to stop.
- It would shut out any journal that sells copies, which is most of them.
There is no share-alike rule either. Getting the record right is worth more than controlling what happens to it afterwards.
Where the line runs
- A quotation from an old source belongs to that source, printed here because it is out of copyright. The sentence around it, explaining what to make of it, is ours.
- An owner’s report of an instrument is recorded as a report. The facts in it (the number, what the label says, where the number appears) are facts, and are CC0 like the rest of the data. Where an owner’s own words are quoted, they remain the owner’s.
- Photographs and images are not covered by either licence unless the page they sit on says so. They belong to whoever made them, whether an owner, a library or an auction house, and are shown here on those terms.
- A licence to reuse this text says nothing about any instrument. This site does not authenticate, attribute or value anything, and nothing becomes more certain by being republished elsewhere.
Where else this is stated
The same two licences appear in llms.txt and robots.txt. They are in the header of corpus.json and register.json, and in a licence column on every row of register.csv. They are in the citation block at the foot of every page. All of them are generated from one statement in the code, so they cannot quietly drift apart. If you find that they have, that is a bug, and we would like to hear about it.
This page states the terms this site publishes under. It is not legal advice.