Where's the Beef

BeefCAMPus 2007 has been really great so far

We launched the first BeefCAMPus in Mallorca in April 2007. 15 participants solved 9 real cases. Great stuff. You can read some of the testimonials at our new BeefCAMPus website . From Jacob Bar's and my perspective the most interesting challenge is the balance between SOLVING THE CASES and LEARNING methodology. Using real cases in real-time exercises is extremely complex both for us as trainers and certainly for the trainees. But using pre-studdied cases takes some of the authenticity out of the workshop.

In Cyprus (at the TII Summer School) we got the opportunity to work two days with 23 participants from 15 countries and started out by giving priority to methodology: guess what the trainees wanted?? Real case! So we worked with a case of mushroom culture and disinfectant technology. Really interesting. Jacob has tuned his fabulous search engine and it is now a power tool in the hands of the trained driver !!

The second Mallorca event takes place in late October 2007 and here we have reduced the number of cases and participants to be able to use real cases but also have time to work with methodology. We have universities from 3 countries and two startup companies, one of which is a gazelle with an exceptional growth over the latest 5 years.

We have agreed to deliver 3 more workshops in 2007: one in Canarias where we will be focusing on 7 startup companies; one in Portugal (with a mixed group of university tech transfer officers and startup companies) and finally one in Mallorca in December. 2008 promises to be even more interesting.

Stay tuned/come back and find out what we learnt.

New medical search engine provides a wealth of nanotechnology information

This is an interesting twist to a classical search engine; but JBEngine already has this facility (as does Highwire, which is included in JBEngine)

New medical search engine provides a wealth of nanotechnology information: "A new search engine provides medical professionals, researchers and the general public with a more efficient and targeted way to search PubMed for the latest, most relevant medical literature to answer medical queries."

(Via Nanowerk News.)

Science, Biology for the Rest of Us

Sometimes, if not always, I/we need to check what scientists tell us. They talk science language; we talk business. So we need "translators". Access Excellence is such a site, highly recommended!

Access Excellence, launched in 1993, is a national educational program that provides health, biology and life science teachers access to their colleagues, scientists, and critical sources of new scientific information via the World Wide Web. The program was originally developed and launched by Genentech Inc., and in 1999 joined the National Health Museum, a non-profit organization founded by former U.S. Surgeon General C. Everett Koop as a national center for health education. Access Excellence will form the core of the educational component of the National Health Museum Website that is currently under development. .

If, say, you wish to know about Transfer and Cloning of the Insulin Geneyou just go to http://www.accessexcellence.org/RC/VL/GG/transfer_and.html

I like the graphics or Visual Library. I pull down an illustration and ask the scientist to explain how and where her/his invention fits in. Works well.

Biotechnology explained

More often than not, the Tech Transfer officer meets a biotechnology invention. It can be difficult for us non-scientists to really understand all the science involved.

The National Health Center (US) has created an interesting Graphics Gallery and a wider overview of issues in biotechnology and human health related issues.

Very useful to quickly get up-to-date on terms such as Retrovirus or details of Microscopy.

Where's the Beef CAMPus?

Well, in Mallorca, of course!
This weblog covers most of the topics of my "Where's the Beef" course - mainly a one-day course. More than 200 professionals have participated in one of these courses over the last 18 months. Here's the announcement of a new version: a week-long training camp:Click here to read more

Commercialisation CAMPus Workshops: 5 days of practical work in Mallorca in March and April 2006

“Where's the Beef” – the course and methodology on Evaluation of Commercial Potential in Inventions and Innovation projects - has been offered to many groups over the latest five-six years. The format is a one-day course covering cases, background theory and methodology, but little time for real exercises and discussion.

Based on feedback, I have developed the concept of Beef CAMPus, a whole week concerned with building your own clear-cut action plans about how to commercialise a portfolio of concrete cases. How to build the business case of your own Technology Opportunities, and, as usual, a lot of learning.

Beef CAMPus will be offered on the island of Mallorca, which, apart from its natural beauty and friendliness, is easy and cheap to reach from almost anywhere in Europe. Together with local partners we offer the best training facilities.

The course is developed and managed together with Jacob Bar, the creator of the JBEngine.

NEW: Since Spring 2007 we have re-designed the format - have a look at our new BeefCAMPus website

XVII: Search with independent entry points

I have been giving courses on "Where's the Beef" methods in 4 different countries over the last month. Faced with the tough questions and feedback from participants, I found or realized a few things about searching, which I hadn't thought of before.

If you click on the category Search Engines on this blog, you will see my earlier articles, in which I try to explain the needs of the profession,and , for example, why JBEngine is a good engine.

I have developed a new method in which I combine JBEngine with other good search engines to get a fast impression of a given invention, patent or new technology. The key in the method is to use several INDEPENDENT entry points to Internet, Extranets and Intranets.

For instance, I will use JBEngine together with Kartoo, Edgar and my own Rollyo to understand a new topic. JBEngine would be used to look for market reports and patents, to mention an example. Kartoo would give me an impression of interlinkages between websites (I want pages which DO NOT link to each other), Edgar to see what the US Stock Exchange "thinks" about the topic and Rollyo to see if a technology has been "buried" in the databases of the Rollyo sites.

If the hits from these INDEPENDENT sites give me sufficient information to think well about an opportunity I trust them. This method has the advantage compared to Googling, where one starts with a given search string and then follow the hyperlinks "deeper and deeper" into the web. One problem with this approach is that one has to rely on the quality of the first search string. Another is that the hits are biased by Google or Yahoo's highest ranked pages - and those hits are biased by money and popularity, which are not necessarily what one wishes to use as selection criteria for good hits. That is why INDEPENDENT hits which point in the same direction (good!) are indicators of something worthful, whereas conflicting indications might help one to form new hypotheses about the given topic (ie. reformulate search to test the hypothesis).

Using independent hits gives one feedback which makes it easier to "brainstorm" to refine search strings: for instance, a "controlled release of drugs in a cardiovascular stent" turns out to point to "drug-eluting stent" as the best keyword.

Thinking graphically, the INDEPENDENT search method (we could call it that..) gives one hits at level 1 of many sites, whereas, I see the classical Google search as a method whereby one is guided many levels deep before one finds anything. In the Independent search, one finds possible new search strings at the same level 1, which then guide a dive into level 2 in these independent sites. Chances are that different writers will use different words and form opinions independently of each other. That's good to look through all the useless pages on the web.

US Patents "grabbing" (data mining)

In an earlier article (see the same category as this), I have referred to how WIPO and espac@net allow some rather nifty ways to rapidly download and browse patent information. For some time, I have been looking for ways to the same with USPTO but have not been successful: you get only patent (application) number, and title.

Well, I found a solution for my Macintosh (and I'm sure you Windows and Linux guys can find a similar solution at Bruce Pokras' web site). It works!

Patent Grabber creates a set of files on my hard disk. With Mac I can then use the internal index search engine, Spotlight, to find the files with hits that match my interest (Windows users will use Copernic, I guess).

Patent Grabber also works with WIPO and EPO files. To create the files from downloads from those respective web sites, you need to isolate the information containing country code and patent number (it takes a bit of copy-paste & find-replace in Excel, for instance, to create such a list. Fast!

Update of JBEngine

Jacob Bar has refined some of his templates. Here are four new:

Strategic Partner Finder under  Company Information.

For finding Possible strategic partners. (lets say that you have a new invention like: biodegradable plastic and you want to find a strategic partner. In this case use the following keywords: biodegradable plastic etc.).

Business Model Finder under Product Information, to find a business model for your invention.

Strategy Finder under Market Information, to build a strategy for you product or service

Distribution Channels Finder under Market Information, to find a distribution channels for your product/technology or service.

(XVIII) Patent Class Analysis

Patent Class Analysis
A special set of sources is patent databases. Sources such as esp@cenet, WIPO, USPTO are really useful. I use a method of quickly browsing these databases in regard to a given invention or company and then collect information about the competition, “prior art” etc. This involves using your internet browser, a good spreadsheet, a good database and how to jump around in these applications to quickly get you the overview which is so critical for “Where’s the Beef”.

It's amazing how much information there is in such files. The key to using them in practice is to employ methods which will allow you to quickly form an opinion. Once you have spotted a Patent Class or a Search Pattern which suits your needs and clients, then you should try to subscribe to these saved searches. There are many ways. WIPO allows you to subscribe to RSS which is a really useful method.

I generally believe Microsoft's products are not very useful. The Office series is of great help for patent mining, if you do it right. I show some tricks.

Once you have spotted patents, classes or other patterns in your findings, you'll find references to important persons, companies, prior art links etc. This information is really important for quick-and-dirty judgements.

(XVII) Where's the Beef: Search tricks II

Search the Right Places II: Knowledge Management Tricks
Building “Where’s the Beef” models takes practice and there is a lot of tricks which may make life much easier for you. In the clinic I normally give  advice and tips & tricks about how to take -cheap- shortcuts (you can buy many/most information and get poor in the process. Question is: will you also get richer?):

• Which patent databases to use and how to mine them. JBEngine does the start; the rest is up to you.
• How to find specific case material about the “value chain”. JBEngine is powerful.
• How to use indirect sources of information such as Edgar SEC filings, to get information of a whole industry. Because of the paranoic legal situation for “publicly traded” corporations, i.e. on the stcok exchange, there is no limit to what upper management discloses about their own business, and thereby the competition.
• Brainstorm keywords from different perspectives. Mindmapping, functional/substantial substitution; laguage versions (English is not the –only- world language).
• How to use mindmap search engines such as Kartoo
• How to use the new RSS/xml standards (Web 2!) to monitor certain sources of information in specific terms.
• How to “Roll” your own search engine (pack) with software such as Rollyo (mainly Yahoo). Use TII’s search pages as well.
• How to use “indexing” to store and find “value chain relevant” past findings: Copernic for PC; Spotlight for Mac
• How to build Knowledge Management "libraries" which cover prior work but prepares for future work with the same group of/type of researchers.
• How to screen sources
• How to mine sources
• How not to trust the Internet and how to distinguish when (Advanced Cell...)

In later articles I shall cover each of these topics.

(XVI) Jacob Bar: Help me

Jacob Bar: Help me
With the Internet and databases it has become so much easier to find many of the answers to the above questions, to “populate” our models. ”Google It”, many TO practitioners now would say. That’s fine, but the problem is that the Internet (data and its complexity) is growing at an enormous speed and indexing is not keeping pace. It’s hard to choose what to believe when you ”Google It”. The advantage of Internet searching is that it puts an emphasis on the need for links, and that the practitioner must jump from one to another piece of information to patch together a final verdict. Don’t believe the first hit(s).

Jacob Bar, an Israeli gentleman with many years of experience in the theory of natural language and as special advisor to the Israeli Minister of Science & Technology, has developed a search engine, which is particularly well adapted to the needs of the TO practitioner. Jacob, who is a also a lover of good jokes, has called his engine: JbhelpMe.com or the Jbengine.

Jacob says that he has looked at the way we pose questions, when we do commercial appraisal and has reduced the multitude of questions to a few hundred ”natural language” queries. In a sense, he has qualified or multiplied the 10 dimensions of COAP (and other good business planning tools) by asking many more detailed questions.

Secondly, Jbengine searches not only the Internet, the few billion html-pages (and other protocols) which are publicly available via browsers. His engine also looks into certain Extranet sources, the large reservoir of ”internal” databases of different organizations. This gives the user exceptional search power of 700+ billion pages and also power to get to the core faster. This area of search changes its nature rapidly, so this verdict may be outdated in a few months. You will find Extranet sources, but then you will to have to pay to get the real stuff: what you find is only the teaser. A good example is Mindbranch

Finally, JbhelpMe collects information relevant to my query and presents it in one package. So I get Google feedback alongside information links from other search engines. For example, a search concerns only whether a given, keyword matches one or more patents. Jbengine returns information from several patent databases in one window, so it gives me faster and better overview.

The only real problem with Jacob’s useful work is that it is not always updated. I will show you ways to work around that problem.

(XV) Where's the Beef: Search Engines

Search the right places
As per estimate in 2005, 5,000,000 terrabytes of information exist in the world, only 2-3% of which are currently indexed and searchable. Google CEO Schmidt's timing for when all of the world's information will be searchable?  300 years, according to a presentation he did in the summer of 2005.
Google indexes 3% per year, which doesn’t cover Extranet sources!! Clearly we need an engine which starts by looking the right places first.

So what to do about it? At the moment there is no plug-and-play solution. No one engine delivers what we want. There is a multitude of search engines and methods of using computers, which are necessary and helpful to the tech transfer practitioner.

The Good Search Engine searches not only the Internet, but also Extranet and in some cases, Intranets (requires “co-authentication”). The Good Engine uses experts to pre-select sites, known to be most relevant. Innovation & tech transfer professionals have to focus to be efficient. You will have to build your own subset of good sources for your line of business.

Each engine has its special language. And most users specify <5 keywords, when they search. What to do when we get for example 153,000 hits or more? The practitioner knows how to customize the Good Engine in terms of special templates, known to produce best results, and set up with 20-30 keywords and using Boolean language and other methods to focus the searches.

The Good Engine, customized as above, must deliver results in easily reviewable formats, not only with headlines and links, as is mostly the case at present.

WYSINWYG: What You See Is NOT What You Want to Get.  The right results may be down the pile behind the paid-for advertisers. Google and Yahoo are biased in their sorting, must be, it’s part of the algorithm (any system has a bias).  The problem(s) are many: normally much too many hits (imagine what it may be when it all is indexed!), and it’s  WYSINWYWG.

The Good Engine makes its own biases, confirmed by peers of the trade. The practical - and affordable- solution is to combine different approaches. One of the best available tools is JBEngine, available through TII Click here. I developed that website and have included some other good hints and sources for searching, including Rollyo, whic I shall review later . There are other good sources and tools, some of which are not fully integrated in JBEngine.