RailcarIDBot
RailcarIDBot builds a link-only index of public freight-car photos so a railfan who looks up a car number can find pictures of that car on other sites. Every result sends the visitor to the original page.
User-Agent
RailcarIDBot/1.0 (https://railcarid.com/bot; hello@railcarid.com)
What we store
- The photo page URL, its title, and the reporting mark and car number we read from it.
- The photographer's name or handle, the license, and the date taken when the source lists one.
- A thumbnail URL only when the photo's license allows reuse (for example Creative Commons on Wikimedia Commons). The thumbnail loads from the source's own servers. We never download, copy, or re-host image files.
Sources and rules
- Wikimedia Commons (active): read through the official MediaWiki API, never by crawling pages. One request at a time with a pause between requests,
maxlag=5, and back-off on errors, following Wikimedia's API etiquette and User-Agent policy. - Flickr (not active): only through Flickr's API under its terms, never by scraping.
- Other photo sites (not active): only after we've contacted the site owner, and only where robots.txt allows. We obey
DisallowandCrawl-delay, wait at least 10 seconds between requests to one site, and never log in or get around a login, paywall, or CAPTCHA. - Sites whose rules forbid automated access are not crawled at all.
- Indexed links are re-checked about every 90 days. Dead links are dropped.
Opt out
Site owners and photographers can ask us to stop indexing them or to remove specific photos. Use the opt-out form. We suppress the listing the same day we read the request and purge it within 72 hours. A site that opts out is never crawled again . Copyright notices can also go through the takedown page. Questions about the indexer: hello@railcarid.com.