With the proliferation of gene patents and the increasing profusion of biotech patents and licenses with overlapping and competing rights, the ability to interpret and filter intellectual property (IP) has never been more important. Last month's announcement by Australian startup CAMBIA, and its initiative BIOS (Biological Innovation for Open Society), of the creation of an open-access patent database collating IP data from several national patent offices promises to radically improve that process.

In the good old days, looking for a patent involved rifling through card indexes, perusing dog-eared index books and squinting at microfiche. Today, all this has been superseded by the introduction of shiny new web-based interfaces and Boolean search algorithms. You might think that this would have made identifying relevant IP in patent databases a relatively simple task. You'd be wrong.

In reality, searching for a biotech patent has become an inexplicably frustrating and convoluted process. There is no streamlined and universal approach for searching patents filed at the various national and international patent offices. And the three main repositories of English-language filings—the European Patent Office (EPO), the US Patent and Trademark Office (USPTO) and the World Trade Intellectual Property Organisation's Patent Cooperation Treaty—offer databases with online search tools that all work differently and display different results. All require the user to enter numbers, dates and keywords in different formats. Each interface has its own idiosyncrasies, such as failing to recognize certain characters (for instance, apostrophes, slashes or hyphens) or requiring the user to pad a patent number with leading zeros. When you go wrong, online help is unhelpful; trial and error is often the most fruitful approach.

Worst of all, each patent of interest must be downloaded and printed one page at a time—even though it might be 100 pages long. The explanation is not some technical difficulty, or even a lack of funding or resources. According to the EPO, “this was done as a voluntary restriction at the request of the commercial [patent search firm] operators.” In other words, searching and accessing patents has been made difficult purposely so that patent search firms can more readily charge clients for searches.

Fortunately, help is now at hand. CAMBIA's 'Patent Lens' is a freely accessible IP database that contains 2.5 million patents from the USPTO, EPO and PCT, together with a powerful search engine (http://www.bios.net/daisy/bios/patentlens.html). The interface makes possible searches of the full text of patents from all these patent databases. The database can only become more useful as its coverage is extended elsewhere (for instance, to Japan), but its intrinsic value is already clear. It is estimated that underexploitation of technical information (an estimated 80% of which is published in patent documentation and nowhere else) costs European industry alone $20 billion each year—simply because the inability to access relevant patent information results in duplication of effort or the creation of products that overlap with prior art.

Patents were conceived as bargains struck between the inventor and the state, not between patent offices and patent search firms. For biotech in particular, researchers, tech transfer offices and company executives need a facile means of establishing the novelty of their offerings and the nature of their competitors' inventions. In this respect, CAMBIA's Patent Lens is a giant leap in the right direction