GPT Tools and Thoughts on Education

 In my prior post it was evident that I was quite impressed by the LaMDA chatbot. Since that experience I encountered a number of other surprises: 

First was that some of the Google guys of LaMDA started Character.ai and made it possible for anyone to experiment with building their own bot to see what happens. I did it, and you can check the last iteration of the bot "Front" here (it is worth your while to create a login).  [Sidebar: my first bot implementation was as Yoda of my novel "The Yoda Machine", but the network knew nothing of my novel and lots unrelated about the Star Wars character, so it was untrainable for my purposes.]

Shortly after that I encountered chatGPT the bot made by Openai.com and opened to the public for experimentation which you can try here (it is worth your while to create a login). I have tried that a fair bit. Smart move on their part as they get the whole world to train their network.

Last but not least I stumbled upon a post in NY Post by professor Darren Hick who encountered his first provable case of plagiarism-by-bot and documented its uncovering. He detailed it further in his FB page.

My conclusion from all this is that GPT for now is nothing more than a language manipulator limited by the factual information it has. It writes incredibly nicely structured gibberish from what it knows to satisfy the request it receives. That is how  Prof. Hick uncovered his guilty student. ChatGPT was writing BS that could be uncovered by somebody knowledgeable of the subject matter and not distracted by the form. Any ignorant person would be in the opposite position. Any lazy student would not check what the bot wrote. 

The problem professor Hick correctly identified is that as time goes on ChatGPT will acquire the necessary information (about Hume in his case), if anybody chats with the bot about it (Hume), so that, eventually, it will be able to write something credible even to him.

This brings me to a subject that I have touched repeatedly in this blog over many years which is the use of mind maps as a means of communication, summarization and organization. In the example above professor Hick asked students to "write a 500-word essay on the 18th-century philosopher David Hume and the paradox of horror" and chatGPT spit out 500 words of beautifully worded gibberish.

What if the request had been to produce a mindmap that summarizes concepts and supporting and dissenting arguments for Hume's Paradox of Horror with each node containing no more than 20 words?  I tried my hand at a very abbreviated and unresearched version of it below inadequate for a student. I am not a student to trying to pass his course (72yo, semi-retired SW developer, former prof, and still a researcher).

In my opinion, these outcomes would follow such a request:

as of today, no tool like chatGPT could produce something like this 

Students would have the incentive to learn to summarize their thoughts instead of embellishing words for the sake of increasing the work-count

structuring thoughts as in an outline is the key to effective verbal and written communication

Anyone could use the best of those maps as a cheat sheet for quick learning about Hume

Hume's Paradox of Horror (bogus) Map 

Another thought regarding tools. 

In the 70s, before PCs and Visicalc and lastly, Excel were either invented or became popular, the use of any such tools for an assignment would have been considered cheating. Eventually, people caught up with technology (I taught using Excel and fundamentals of database systems at Seattle University in 1983) and today we take it for granted that much math work not only can but should be done with spreadsheets. So it will be with text generators and it will become necessary to figure out how to grade the intelligent use of tools such as chatGPT. Clearly in this case the student not only did not know the subject matter but did not bother to find out what garbage chatGPT had created, so she failed on multiple fronts due to laziness, but kudos for trying to use a new tool. 

As time goes on perhaps being able to give appropriate instructions to a bot so that it can generate the MINIMUM amount of text to COMPREHENSIVELY explain the matter will become the rule. As with word processing versus handwriting, why write if you don't need to but you can think effectively? We are not there but it will come.

Full disclosure: Much of this post is written with dictation and tools for automatic error corrections since I have familial tremor of the hands that makes writing a painful chore

Working Backwards

New entrepreneurs seeking funding from angel investors often appear surprised by the multitude of considerations and requirements they must satisfy to get funded. If they can keep both the big picture and the details in perspective, the puzzle is not so difficult to solve.

Working backwards from the investors' requirements and preferences one can create a proposal that will "sell" provided that all the underlying reasoning, projections and plan are supported and convincing. Conversely, if you cannot make a credible case that your venture meets the investors' criteria time may be better spent seeking other forms of financing.

You can navigate the roadmap below opening  and closing various branches to look into the details and reasoning behind them, or hide them to stay focused on specific HOW and WHY of various aspects of the problem.

I am experimenting with this method of communication. It allows you to switch between details and summary views.  Your comments would be greatly appreciated. Is this method effective for you or not? In either case why? Thanks Email me

Working Backwards

Entrepreneurship, Not for Everyone, Salesmanship Is for Everyone

It may be because of where and with whom I hang out, but these days Entrepreneurship seems to be the magic word of the day. Policy makers and economists tell us that with more of it the country will be more competitive, jobs will be created, we will stand up to China, we'll fix the national debt, we'll have a credible national medical insurance, etc. If only we had a Steve Jobs at every corner, the American Dream will be assured. It is all true, but it does not follow that it is true for all.

The message to high school and college grads these days seems to be: it's easy, be an entrepreneur, be a Jobs, or a Zuckerberg, or a Gates,  have an idea and make it a Google. All true and as possible like winning Olympic medals, entering the NBA, winning the lottery, but let's be honest with the youngsters who struggle with figuring out how make a living for their future.  The odds of all of that are pretty low and it takes much more than an idea.

To keep the discussion short, I developed the roadmap below as a way to look at the principles and the details of what I think is in play. As in some of my other posts, I try to synthesize the issues and the logical relationships to be considered to reach a clearer understanding.  The objective is to not waste energy in impossible quests (see "Working Backwards" and "Angelcalc" for when not to bother seeking funding from angel investors, or "Should I Get a Patent?", etc.)

This roadmap, as in other cases has no absolute answers or recommended solutions, only considerations to be explored and used in seeking a best answer for oneself (notice ONEself not ALLself). 

One, and only one of many conclusions, is that a temperament for risk-taking is mandatory for entrepreneurship. Salesmanship on the other hand is a skill fundamental to entrepreneurship, BUT also applicable to all other activities in human activity and commerce.

So, the question for which unfortunately I have no answer is: Why salesmanship is not a primary skill taught at all high schools, colleges and graduate schools?  Why as a society we leave it to Xerox, IBM, Proctor & Gamble and similar companies or to multi-level marketing companies to impart that training with a planned and focused process?  The Introduction to Dale Carnegie 's How to Make Friends and Influence People pointed out that tragic circumstance decades ago.  Why are we still at the same point?
Why is it so hard to find a high school or college grad with even only a cursory idea of how to sell an honest product to a respected customer that has a need?
Perhaps it is time for some bright entrepreneur to design and roll out an effective online sales training program for all those recent grads that are not finding the job they trained for or the countless unemployed displaced by technological change.


Marco Messina

Patents and Due Diligence

I frequently run into investors that seem to find a great deal of confidence in the fact that the company the are doing due diligence on has an "issued" patent.  They seem to believe that once the PTO issues it we are in Safe Land.  I wish I could be that optimistic, instead I often find myself "raining on the parade" suggesting that there are still big questions to be addressed:

Markets covered

If the projected market is global, but the patent is only issued in the US, what will the cost be to cover other countries?  

Is there still time to file abroad in desired markets? 

The rest of the world works on the basis of "first to file", so if someone invented well after the US inventor, but filed first in the country in question, it would be quite hard (not impossible) and expensive to contest the foreign filing.

Cost and means cost of enforcement

The PTO issues a patent but does no enforcement. Protection and enforcement of the rights implicit in the patent are up to the inventor/holder: Does the holder have the means to enforce its rights?  No cash to pay for litigation is about good as no patent.

If a company is granted a permanent irrevocable exclusive license to the patent by the inventor, the holder is the one that has to protect it through litigation, unless the right to prosecute infringers is granted along with the license, which normally isn't since the licensor is expected to protect the patent rights as consideration for the royalties received.  Does the holder have the ability, financial means and will to protect the patent rights? If not and the company does not either, it may have no means to prosecute infringers and in practice have no patent at all.

How "real" is the patent?

This is the question that seldom seems to be considered. In "Patents: what do they mean to you" I referenced the debacle of Research In Motion (RIM the maker of Blackberry) whose issued patent had one claim  invalidated years after being issued.

Another interesting case is that of so called "bogus patents" as this "Must Read" case reported by ReadWriteWeb.com: 

The notorious U.S. patent 6,411,947, a broad "method" for automatically classifying and responding to email inquiries known as the Firepond/Polaris patent, has finally been invalidated after 12 years on the books. (continue)

The warning here is: if it looks to you to be too easy, too obvious to be patentable, have an expert check the details, not just validate that the patent is issued. If it does not quack like a duck, it may not be one regardless of the stamp put on by the PTO or it may be so only for a short while.

Are patents useful?

Of course they are.  They certify to a good degree the novelty of an idea if not to its economic value. By virtue of the prior art research done, they attest to the difficulty of finding competitors.  Competitors could well exist that have prior art but never bothered to file a patent and they could come out later as they did for RIM.

Should inventors file them? Of course, but with the awareness that they grant no explicit protection. They only give one the right to spend money in litigation  to protect the rights implicit in the patent. 

Should investor value them?  Certainly, but, in my view, subject to the above considerations and making sure that due diligence includes looking carefully under the hood.

Marco Messina

Patents - What do they mean to you?

If you are interested in this subject you probably fit one of the following classes:

1. You are an expert angel investor with IP due diligence experience - your comments would be most welcome.
2. You are a potential or new angel investor (bless you for helping to grow our country). In this case you might be interested in one (mine) practical view of what protection a patent gives you.
3. You are an entrepreneur that could use some financial help to get your invention (IP) to market (bless you for helping to grow our country too). In this case, you should know what questions to expect from your angel investor and have good answers.

This post attempts to share, for the benefit of the last two groups, my experiences over twenty-five years as an inventor, patent holder, investor in technology ventures, an acquirer of intellectual property (IP), licensor of IP, entrepreneur commercializing my own or someone else's IP. If you want legal opinions or legal advice on this, call your attorneys. If they are IP specialists they'll have legal details and perspective far more reliable than mine. However, beware: because they are so familiar with the domain, they will often presume that you understand the nuances of IP legal protection principles and legal practice, not necessarily the practical aspects of it, which is what I am focusing on here. Mine is a "business" view of patents, not a "legal" view and therefore focuses on what is practical not only what is legally rightful.

What does a US patent give the inventor?
In simple terms, the RECOGNITION by the US government that he/she is PRESUMABLY the inventor of a certain concept, product or process and therefore has the EXCLUSIVE RIGHT to practice the invention for a certain period of time (e.g. 17 years) without competition. Note that contrary to public perception it does not give the inventor any real protection (with one exception) unless (s)he takes steps to enforce that right. The exception is that the Customs Office will do their best to stop an infringing product from entering the US if they are provided with a suitable request, documentation and proof of patent infringement. Otherwise, the inventor is solely responsible to enforce his patent rights, which implies taking legal action against an infringer generally at some considerable protracted cost.

Against the above background then a number of questions arise:

Can one afford to sue the infringer?
Practically speaking, only if the practice of the invention has a cash flow worth protecting in an amount that covers the expenses, distraction, aggravation, etc. associated with a litigation. Winning the case may not produce cash flow or a significant pay off; it may produce only an injunction against the infringer, which by the way does not preclude another one from coming along.

What if in the course of litigation one discovers that other aspects of the product infringe on patents held by the infringer or someone else?
This is very common in the electronics industry where frequently patents have great value to counter-sue and eventually settle by reciprocal licensing. Much of this jousting is now going on and reported in the press involving Apple and their iPad against various competitors among which only one (the smallest and weakest) was actually sued for infringement. In these cases there may only be legal expenses and the benefit of upsetting a competitor's product development road map.

If one has a validly issued patent, how much protection does it provide?
Better than none to be sure, but hardly a guarantee. The case of RIM (Research In Motion), makers of the Blackberry, is very instructive (details): It started in 2000 when RIM was a startup with strong beliefs that their issued patents covered their products well. After a surprising chain of events, by 2006 they settled for $650 million with NTP (plaintiff) after a court injunction forced the Blackberry network dark for one day (RIM by then had a $2 billion business at risk of going dark worldwide, hardly a desirable bargaining position). Since 2006 and continuing to this date "patent reexamination" action by RIM has sought to void NTP's prior art claims; NTP is responding in kind. One can hardly imagine the costs involved (at $500 or more per hour). Needless to say RIM investors had been confident of their IP position and of the issued patents. Conclusion: even ISSUED patents are no guarantee since prior art can be submitted at any time and re-examination requested.

Is a patent much ado about nothing?
Definitely not. A Provisional Patent filing, if properly written and searched, tells potential investors that there may be more than just an idea. It says that the company and the inventor understand the IP implications of their business and have invested time and money to protect their innovation. If a patent is issued, it tells investors that a pretty solid stake is in the ground that proves innovation, at least from a theoretical and PTO standpoint. If potential competitors exist, they are obviously not easy to find and on the time-line may be behind the current inventor and therefore not innovators in the PTO sense. Of course they could still come out later with proof of "prior art" and open the RIM type can of worms.

It should be noted that claiming infringement while holding an issued patent has risk. The "infringer" may turn up to have prior art and that may invalidate the patent. To wit, I had occasion of working with a small manufacturer who was an outstanding and recognized innovator but never filed a patent. He explained that, not being interested in having investors or selling the company, patents to him were a cost and of no value. His strategy was to practice whatever innovative process he devised without fear. If anyone came to try and stop him he depended on his meticulously documented prior art files from many years back to trump the action. Furthermore he had no interest in licensing his own inventions and felt that since issued patents files are open to the public, they are more risky than helpful unless one deals with fundamental inventions. By those strategies, his company never grew to dominate his markets, but he was happy with his lifestyle company; my lesson was to beware that prior art occasionally may come out only when one "kicks the beehive".

Why uncertainty cannot be eliminated
In the US the PTO operates on the doctrine of "first to invent". This means that anyone can come and claim to be THE inventor of anything if he can show with a preponderance of evidence prior art precedent to that of any patent filed. If a patent had been issued, reexamination is the cure. Documentary proof of prior art can be from most kind of documents, preferably lab books (numbered non-removable pages), dated and witnessed affidavits, etc. Elsewhere in the world the "first to file" rule is followed therefore it is imperative to not delay a viable filing and one can depend on the value of an issued patent to a greater degree.  Even there a patent can be contested with suitable evidence that the prior art was "published" and therefore in the public domain prior to the filing of the issued patent. Demonstrations of preponderance of evidence of prior art are not easy to do, so an issued patent is still a strong stake in the ground, but...


Conclusion
Beware: from a business viewpoint, the value of a patent is not an absolute one. It depends on the circumstances of the business, the market, the product, the objectives of the inventor, the objectives of investors and many other factors. Deciding to file a patent (the inventor) or to assign a value to a patent (the investor) is a problem without an optimal solution. Clear understanding of options, implications and trade offs is the best one can achieve in reaching an entirely subjective decision.
My "Should I get a patent" Roadmap may help with that analysis. Read more about my Roadmaps

By the way for my own IP I still file patents, but I am cognizant of how thin the ice is that I stand on.  If I cannot justify a costly fight to defend it, my patent is not more than a feather in my cap.

Marco Messina

Roadmaps and Knowledgebase Navigation (KnowNav™)

For nearly ten years I have researched effective uses of "mind mapping" for a variety of communication and collaboration purposes. Part of my research and usability testing was done as Director of Technology Programs at Maricopa Colleges Small Business Development Center in Phoenix, AZ. The opportunity provided by the Maricopa College SBDC to publicly test the results of my research is gratefully acknowledged.

I dislike the term "mind map" originally employed for this technique. I believe it is a confusing term with new-age overtones that has hindered its acceptance and that of derivatives and tools to implement it. For that reason, and for the particular use I make of it, I prefer the terms Roadmap and Roadmapping, which are more descriptive of the knowledge domain navigation process enabled by the map.

Roadmapping is particularly powerful to navigate complex knowledge domains where a problem solution is sought which cannot be readily optimized. In that case, optimizing the exploration and reasoning process is the next best thing. I call this process Knowledgebase Navigation or KnowNav™.

KnowNav™ allows users to explore alternatives and their associated facts through the prism of their personal preferences. To this goal, the domain expert lays out in a roadmap, as objectively as possible, the alternatives and all the judgmental considerations that could be made about each. Users then can navigate the knowledge-base, in an almost random-walk fashion, intuitively attaching relative personal preferences to each of the considerations identified by the expert.

The roadmap's power comes from:

1. ability to taxonomically organize the knowledge-base hiding the overwhelming size of the data set
2. ability to open and close branches at will thereby switching from macro to micro views
3. keep open only the branches relevant to the user at any given moment
4. express options and features in one line statements thereby breaking down the complexity of any concept

The expert's challenge involves:

1. definition of the alternative navigation paths
2. taxonomic organization
3. break down complex ideas until they can be defined in one-liners

Following are some simple and abbreviated examples open to all. Some more detailed roadmaps are available to my consulting clients:

1. Angels' Due Diligence
2. Should I get a Patent?
3. Should I file for a Trademark?
4. Should I apply for a SBIR Grant?

Very large data sets and complex issues can be summarized and detailed so that the end user can navigate them with ease. More details on these methods are contained in my forthcoming book "KnowNav™and Roadmaps: Simply Reasoning About Complex Issues"

I use this technique because, folding and unfolding branches of the roadmap (read carefully the instructions at the top of each map), you can switch from a "30,000 foot" summary view to "detail" view of different parts of the issue and navigate the knowledgebase with continuous reference to the general context.

All content and roadmaps copyright of Marco Messina 2006-2010 KnowNav™ is a trademark of Marco Messina