How to stop your images from being pinned to Pinterest

February 20, 2012 | 9 comments

Pinterest screenshot

Pinterest is an interesting tool, and a lot of people I know love it.

This is how it works. You maintain a set of pinboards for different kinds of images – for example, I’ve got pinboards for logos, app wireframes, and interesting visuals. If you visit a site that has an image you’d like to keep, or share, you click a “pin it” button in your browser, and that image is copied to one of your pinboards.

Obviously, not everyone loves this. It arguably pushes the envelope of fair use, and will probably torch it completely once the owners attempt to monetize. Although images are linked to their origin pages, the attribution isn’t visually striking, and it’s not like the web is actually shared under an attribution license. I’m not a lawyer, but some people might see it as copyright theft.

To answer this, Pinterest have created a meta tag that you can stick at the top of your site: meta name=”pinterest” content=”nopin”. If it’s present on the parent page, Pinterest will refuse to copy your images (although I presume you can get around that by loading an individual image in the browser). Hackers are, rightly, protesting that this sets an unmanageable precedent: imagine having to individually opt out of having your content copied by thousands of different applications by having to stick thousands of different meta tags at the top of every page. It’s dramatically unscalable.

Pinning is not a million miles away from Tumblr‘s reblogging, and it seems reasonable that there should be a meta name=”republish” content=”no” meta tag that covers all of these services.

Until then, here’s some dubious fun you could have at your own risk (if you don’t want to include the meta tag). This was pretty common about ten years ago, when hotlinking images could cause major bandwidth bills for the owner.

  • Currently, the HTTP user agent for the bot that actually copies images for Pinterest is Pinterest/0.1 +
  • The JavaScript bookmarklet works through your browser, of course, but it sets the HTTP referer for the page to a URL starting with

I’d detect Pinterest based on user agent, not referer – there are many situations where referer could be stripped out. This is true for user agent too – this isn’t guaranteed to work 100% of the time, and depends on your server setup – but there are only two parties to worry about in this scenario: your server, and Pinterest’s. (If you’re filtering the bookmarklet, you also have to worry about configuration changes in the user’s browser.)

All you need to do is filter requests by user agent in your web server’s graphics folder. If you’re running Apache with mod_rewrite, you could create an .htaccess file in your graphics folder with rules like:

RewriteCond %{HTTP_USER_AGENT} ^Pinterest.*
RewriteRule .*\.png$ copyright.png

The above rewrites any requests for PNG files to copyright.png. The contents of that graphic will be pinned to Pinterest instead of the intended image.

(PS: want to try it out? Try pinning the image illustrating this blog post.)

There shouldn’t need to be an OpenStreetMap

April 9, 2009 | 5 comments

OpenStreetMap is a project whose aim is to make a free map of the world. It’s extremely impressive: as well as searching the map in a normal way, the data is exportable via XML, PNG, JPEG, SVG and more, under a Creative Commons Attribution-Share Alike license.

But it shouldn’t need to exist.

In the US, federal government-created maps (and other data) are considered to be public information, and released freely. In the UK, such maps are subject to Crown Copyright, and the Ordnance Survey has been set up as a trading organisation that legally must make money from its efforts.

This was an archaic idea at its inception, but makes even less sense now. The economy is in dire straits, and what it should be doing is providing taxpayer-funded data for use by companies; this kind of data in particular could give British businesses a flying start. Instead, it chooses to make money from them instead, and web services are left to projects like OpenStreetMap, as well as US businesses like Google, in order to source information.

The Guardian’s Data Store is one British attempt to rectify the situation, but ideally all data in the public interest should be released in a format that is easily consumable by third-party applications. As well as helping entrepreneurs and small businesses, it’ll allow for a deeper understanding of, and participation in, how our country is run. Which can’t be a bad thing – can it?