|
|
|
| You are not authorized to post a reply. |
|
hjbarraza
Posts:11
 |
| 11/09/2006 5:51 PM |
 Alert
|
After reading Chapter 10: We can Manage Ourselves and Joe Flumerfelt’s “Proposal for Alternative Methodology”, I had to sit and think about the examples and lessons provided by TBE and Trading Markets. It didn’t hit me right away but after reading several other comments and talking briefly with Joe, it suddenly came to mind all different approaches to come up with the correct answer…
I’ll try to suggest an alternative (complementary) method in which organization could leverage solution development using the method of evolutionary process and collective wisdom of their own members.
Somewhere in this site I can’t remember exactly where, I read that the problem of TBE was considering only two options as the correct ones (or best ones). We can understand this for we do not expect that a business would develop all solutions but what if we could somehow have a platform to simulate o enable de the development of multiple possible solutions.
I want to combine several ideas, some of them discussed here and some others by authors many of you probably already know.
• Evolution - darwin
• Emergence – steven johnson
• Democracy – von hippel
• Genetic Algorithms
• Case Based Reasoning
============
1. Evolution
============
Now, lets think for a while on how does evolution always manages to come up with the most effective way to solve a problem. For each kind of problem it has develop a type of solution. Or I should rephrase that – for each kind of problem it didn’t develop a solution. Instead it figured out a way to solve ANY problem with only one solution: mutation.
Mutation has enabled all kind of living organism to evolve in the most varied and diverse way, thus being able to adapt themselves to different environments with very specific challenges for survival. By having small mutations a single species has been able to adapt itself to different unique environments in which it will be the most fit to survive. But how has nature come up always with the best mutation for each ecosystem?
For every natural challenge nature permits a single species to create multiple different mutations, when an evolution explosion happens, a species doesn’t mutate in a single variation but instead it creates multiple ‘enhanced’ version of itself. Yet neither the species nor nature knows which one of those many options will be the best in adapting itself to the environment. Each of this new species is slightly different from its predecessor, those who evolved into a better state of the current ecosystem will survive and will be granted with the opportunity to pass its own genetic material to further mutations, those who don’t, well…
So we have that for an organism that has to adapt itself to a new ecosystem OR sees its own ecosystem changed by any means, evolution quickly throws multiple possible “best” solutions in order to find out which of those survive. The ones that do so after a couple of generations will remain naturally as a good species.
-If I have time I’ll insert an example here, if not I hope someone does-
We’ll get back to evolution, mutation and adaptation later on, now lets talk democracy.
========================
2. Collective Knowledge
========================
Now, lets this approach at business decision-making, bear with me for a while… Imagine that as a business manager, you have to take a big decision, but you are not sure which of all options could be the best. Business compared with nature does not have eons to sit and wait for things to solve by emergence. So we have to figure out to make our own generations – iterations.
So probably we should make things run faster using the embedded power of the many. Taking in account, the we can simulate micro-evolution of possible solutions aggregating individual inputs from community members.
Taking the example of joe’s alternative method for writing the book I’ll try to adapt the main concept to create a artificial ecosystem for possible solutions to evolve (and die) using members as environment input.
For those who haven’t read joe’s proposal I’ll copy paste:
The process would work like this:
1) We create a defined period for the reception of ideas (offerings to the market). Each idea could be as simple as a short paragraph (or two) that defines the issue to be considered. It doesn't have to be a highly refined piece of literature. To the contrary, the main thing is that the idea is clear and easily understood.
2) The ideas would each be entered into the market as tradable items. Much in the same way that the stock market opens and closes each day, we would have a time based period for the members to invest in the ideas that they believe hold the best chance to make it into the book.
3) At the end of the trading period, the results are published. The ideas, which performed badly in the market, would be discarded (or simply not supported), in order to make room for the next round of new offerings.
4) A new period would be opened in which the ideas could either expand on an existing idea (think of it like stocks splitting), or to use the feedback provided by the previous round to inspire a new set of ideas.
5) Upon the entrance into the next round of trading, those ideas, which are expansions on previous ‘winners’, would begin trading at the price established by their ancestor. This way, we insure that well developed ideas (that have passed various rounds of trading) would move towards a much more stable positioning.
6) The trading would end and the process would begin again.
7) At some point (perhaps predetermined), the ideas will be so strong that they will reach a plateau in terms of capacity for further development. At this point, the idea has won its place in the book and energies can be focused on further development of the weaker ideas.
Ok, now… lets take these ideas and place them into and democratic-evolutionary context.
1) As a business manager –or individual with specific problem- I would publish the problem and provide some background and a list of acceptance criteria for the desired solution, personally I would include a foundation upon which community members can start solution development (resembling a DNA start up.)
2) Community members will develop micro-evolution modifications that will be placed on the common scenario in which we can test each new aggregated modifications. Such scenario can be represented by a range of different tools, from process simulators to discussion forums where people validate ideas.
3) Each new development -mutation- that has been supported by the community will continue to be developed, modified or even merged with other ideas until the parameters for an acceptable solution have been achieved.
4) After feasible and acceptable solution has been reached, community should now document different approaches to solutions, even if they consider good enough, (even if the solution wasn’t used in this case).
I’ll make a pause here… to talk a little bit more about genetics and evolution. I’ve recently been learning about bio informatics and genetics thanks to several podcasts I personally recommend ‘Evolution 101’ from Zachary Moore and Leo Laporte’s “Futures in Biotech” both available in itunes store for free.
Ok now, why store so much information that seems useless? After all didn’t we all ready defined the best approach to specific problem? The not obvious answer seems to be the same of why our DNA Is full of junk-DNA. We have not, until recently learned that the so-called Junk-DNA might not be junk at all. We where simply just failing to see the reason of its existent in our Genome. I’m unsure of the numbers up to date, but almost 95% of our genome was considered “Junk”, it’s only that until recently it made no sense or seems to have no function at all.
As Surowiecki’s in WOTC states in his 9/11 example, a lot of security agencies simply discarded information that meant nothing too them, but if such information would be placed at reach of other agencies, and digested collectively, it would take shape into meaningful knowledge.
This practice –storing information- is specially required when we are talking about knowledge societies, communities in our sense.
A Potential solution might not been considered as optimal for a specific scenario, but that doesn’t mean it should be discarded. Al species carry a lot of genetic information that seems to be either useless or un-functional, but it is there for a reason. It result quite handy in future unpredicted scenarios, whether it is a dramatically change in ecosystem o a sudden change on the world economic scenario. DNA is a form of natural memory system, which stores the wisdom of eons of evolution.
Such non-implemented solutions might not seem useful for current problems, but we will sooner or later find out that such solution would fit better in a alternative scenario OR will inspire new solution developments. That is why “We” should always remember what we know.
Now, lets imagine we already have a rich common knowledge base filled up with solutions to all kinds of problems (our DNA). How do we extract such knowledge for specific problem solving (ecosystem).
My first guess (and only, there should be more) is that Case Base Reasoning is a perfect way to mine our knowledge in search for previous similar solutions which super brief explanation of the process is:
• Retrieve: extract previous possible solutions to a similar problem
• Reuse: adapt old solution to new context
• Revise: review and asses new results
• Retain: add new solution to the existent knowledge base
Ok now please try to bear even a little more with me for another idea…
What if we merge this process with genetic algorithms to conform an hybrid process of evolving and storing knowledge?
Most generic process to follow for GA would resemble something like this:
• Initialize: start multiple possible solutions with given parameters
• Discriminate: analyze option to select the fittest
• Evolve/Die: mutate or reproduce to evolve or disappear
• Resolve/Increment: select feasible solution or go through increment.
What I believe we could figure out is a process that uses both concepts that enables communities to both extract its current knowledge AND adapt their ideas into new fitter ones.
I think something like this should be a good start
1. Define Problem
2. Retrieve Knowledge
2.1. Initialize Candidate Solutions
2.2. Evolve Candidates
2.3. Discriminate Options
2.4. Retain Knowledge
2.5. Increment Cycle
3. Resolve Problem
1 - Define Problem
As stated somewhere above, and in addition of flumerfelt’s ideas, a problem should be stated and described. The idea of having this definition is for all community members to be able to have a clear idea of what the problem to be solved is. Must solutions will fail because the wrong solution is being develop or the wrong problem is being trying to solve. Such definition should include:
• Expected time periods
• Acceptance Criteria
• Number of Increments (Optional)
2 – Knowledge Retrieval
Whether it’s from an existent knowledge base – in any format- or directly from real-time collaborations of community members in this step we ‘query’ our current knowledge on similar problems of the past in search for similar solutions to adapt.
2.1 - Initialize Candidate Solutions
Gather a list of several candidate solutions that seem the fittest or most similar to current context.
2.2 - Evolve Candidates:
Let the interested community members develop candidates in different approaches based on individual knowledge and opinion
2.3 - Discriminate Options
Candidates are placed into market and performance is tested. By the end of the iteration we should have ‘enhanced’ or developed candidates that will make it to the next iteration and those which will die.
2.4 - Retain Knowledge
Disregarding if candidates ‘survived’ or not, community should decide whether to keep portions of a develop idea, complete o not at all. Such information should be then fed to the common knowledge base.
2.5 - Increment Cycle
Now that some information has been processed, the business manager, or knowledgeable member should Asses whether a optimal solution has been achieved or if further increments should take place. Remember this decision is also affected by the parameters established in the Problem Definition in step one.
3 - Resolve Problem
That’s right… you implement the solution and document it in order to enrich the knowledge base.
You might notice (or not) that I changed the order of several items must notable the “Evolve’ and “Discriminate” steps, the reason to do so is because in comparison of an ordinary genetic algorithm we have already had a first discrimination round when we retrieved knowledge in step 2.0 “Knowledge Retrieval”
Well, I guess this kinda gives a very rough idea of how we could combine these concepts and ideas to actually apply a little bit of democratic business decision making. As you might surely noticed I’m not a native English speaker, hence my English is far away from perfect. Never the less I hope I made some sense out of all this mumbo jumbo.
Oh I would like to add one final note, of course in order this kind of method to work a healthy population of community members is imperative, but such topic does not belong hear. I’ll just point out a few questions.
I believe that in order community to work a sense of belonging is imperative. A Sense of community must be developed; these are a few questions that I haven’t still answered to my self but would help in order to find out how to create such sense
• Why members must realise they are part of community
• Why members must have a clear definition of the communities goals
• Why members must know why they are of the community
o What does this –being part of specific group- mean
• Why Members must be aware that their actions (or lack of) affect the communities environment, hence themselves
|
|
|
|
|
joe_flum
Posts:151
|
| 11/12/2006 1:41 AM |
Alert
|
H,
Wow, you've definately touched on a lot of material here. You've got some great ideas.
Perhaps it would help spur on more discussion if you could separate up the main points and move them into different posts. It may be just a little too much information for us to make sense of (even though the ideas are logical and straight forward).
either way, good work.
cheers,
joe |
|
|
|
|
hjbarraza
Posts:11
|
| 11/16/2006 7:47 PM |
Alert
|
Whoa,
sorry for not following my post earlier guys, i went out of town for some days then i slept some time and worked some time. Anyway I've re-read the post and I agree that definitely is a great deal of material for a single, post. I'll try to distribute it and develop the ideas. |
|
|
|
|
1ofus
Posts:103
|
| 11/24/2006 7:04 PM |
Alert
|
Wow, what a great post at the start here! I have learnt so much and have some great podcasts to look forward to as well.
I am quite overwhelmed as to how best to explore, exploit and use the ideas set out. Hence, I will have to sleep on it before returning here. However, to provoke other thoughts, should this not go into the Chapter text somehow?
A more generic point, and likely not best placed here (Joe, can you please indicate where it should go), I just marvel at the way this whole project is developing. Isaac, is there any way that this whole set of contributions in Forums and in the Chapter discussions can become a book about a creative process of writing in groups?
Just a thought, the product here can be actually two distinct books -
1. the content produced by this community
2. the exchanges (annotated and perhaps enlivened by some participant stories) in the Forums
Just late night doodling!
Lilly |
|
|
|
|
joe_flum
Posts:151
|
| 11/25/2006 1:38 PM |
Alert
|
Lilly, I actually think that we shouldn't separate out the two things into different books. I think it would provide fascinating insight to have the "process" interleaved between the theory. Our content is one thing, but then to be able to see how we came to that content and how the group debated it would be extremely fascinating.
If you get the chance, take a look at Douglas Hofstaeder's book, "Godel, Escher & Bach". It has a wonderful alternating chapter structure that takes you in and out of dense theory and demonstrative allegories. The structure helps increase greatly the joy of reading the book and really shows you what he's talking about. |
|
|
|
|
|
|
|
|
|
|
|