“He said that it was simpler and more effective just to act, without seeking explanations, and that by talking about my experience and by thinking about it I was dissipating it.”
Carlos Castaneda

Tales of Power

3 Sources of Bias in AI

n

” … Math really is pure, has certain truth, it’s eternal, it would be the same without any particular sentient species looking at it. That’s because math is an abstraction that doesn’t exist in the real world. Computation is a physical process.

… the first source of AI bias:  unintentionally uploading the implicit human biases that pervade our culture … There’s no real way to fix this without fixing our culture first, so we need to compensate for it when we design our systems.

… the second source of AI bias is poorly-selected training data for machine learning, or poorly reasoned rules … when we detect these errors we can fix them, so we can expect that these types of AI biases should iteratively improve and hopefully eventually disappear

… the third source of AI bias is evil programmers.  Or corporations, or governments … The way to deal with this is to insist on the right to explanation, on due process.  All algorithms that affect people’s lives should be subject to audit.”

source

Posted in Intake, Intellect Run Amok, outside, Tech Stuff | Tagged | You are welcome to add your comment

On Ceptr and Wagn and WordPress

n

This post, in addition to its subject matter, is offered as an experiment / demonstration for:

  1. This format of communication / expression (even though it isn’t published where it should be) as having intrinsic value as a collaborative tool.
  2. Coordinated, distributed and asyncronous work while potentially reducing the need for yet another meeting (face-to-face or online).
  3. If this was discussed in (yet another) meeting, the contents may have not made it into this written form and been available only to the participants of the meeting,
  4. That the written and published form creates a referencable / sharable / discussion point resource for others who would not have attended the (yet another) meeting.

This is a followup to a conversation with Arthur and Jarod about creating a collaborative work (codenamed: make.ceptr.org)

Personally

I’d like to clarify that in an ideal situation I would prefer to not get involved “under the hood”. I enjoy applying these tools more than creating them. I’ve been using WordPress for 10 years and have formed a satisfactory relationship with it, in that empowers me to express myself and create things I’d like to see in the world (that no one else has created or will create for me). Given that there aren’t technical resources available in the Ceptr team, I am offering to create what I can.

Wagn

I am curious about Wagn and what it can do. I apperciate the approach it takes and maybe in another context (not after a year of challenging coding work) I would be interested in exploring what can be done with it. However it feels to me like a technology and not yet a usable tool. I can imagine that for someone who is closer to a developer mindset it is more of a tool. But I am not of that mindset.

Lets assume that an in depth suitability analysis of Wagn as a technology to create a clearly outlined solution were conducted and gave a clear resounding yes. If I am the one who has to now create with it I would have to enter a learning curve which 1) I am not interested in and 2) would divert energy from creating a solution for Ceptr, to investing in Wagn.

I do, however, want to dedicate a few more sentences to Wagn as a technological manifestation.

Solutions and Technologies

Wagn highlights a fault-line I have been observing for many years in IT and more endemically in open-source. A technology is not a product. An engine is not a car. Technology is only one dimension of creating good applications that provide valuable solutions. RSS was an oh-so-simple protocol/technology for connecting people that wasn’t properly productized. Facebook did fill that void with an exploitative product.

I would like to avoid this shortcoming in Ceptr.

I realize that Ceptr & Holochain ARE technologies. Ceptr & Holochain are on par with developing an operating system like Linux. For Linux to “succeed” it took many others who built upon it distributions that were targeted for more specific contexts / applications.

My understanding is that a similar view is currently shaping Ceptr & Holochain. That others will adopt Holochain and build solutions upon it. But I am not convinced that can happen, or that if even if it did, that it would necessarily be be a good thing.

Yes Ceptr can be framed, for example, as a Blockchain alternative. But, in my mind, that isn’t true. Ceptr is designed for a different mindset … heck … lets say it out loud … to a different heart-set. Those who see and embrace it as a Blockchain alternative will not be manifesting that different heart-set. They will be applying pressure to get Holochain to do their bidding. The question is what will give? Will Holochain bend their mind-set and bring them closer to its mindset? Or will Holochain itself be bent to be framed into their expectations?s

This is why I believe that it is up to us, the makers of Ceptr & Holochain to also create applied solutions to really give them life. When we do, please, please, please … lets avoid making Wagns.

For anyone interested in reading more on this I recommend The Inmates are Running the Asylum by Alan Cooper.

WordPress

I would like to keep this short:

  1. WordPress, in addition to being an application / product has been, for me, a vivid example of distributed collaboration. I have enjoyed watching (though not as an active participant) how it grew as a collaborating community. WordPress, to a large extent, was used to create itself. I would like to bring that morphic field into Ceptr.
  2. Part of what I bring with WordPress is … what I bring with WordPress … the way I personally create with and use it. I would like to try to demonstrate the difference between engineering a theoretical process vs. a gradual unfolding and emergence of process. This can happen when we enter a cycle in which 1) a tool is being used; 2) the usage provides feedback; 3) the feedback can be converted in short iterations into changing the tool itself.
  3. If the unfolding experience pleases us we can apply it with other online contexts (websites and what not) because WordPress can be used for those too.
  4. WordPress, in my experience, brings with it many subtle assets. If we opt to use it I will try to make a conscious effort to highlight them as we encounter them. This touches on what it truly takes to elevate a technology into a solution.
  5. WordPress is fully open and documented, lock-in to it comes not from it but from the gains and satisfaction of using it. It is not the Hotel California or Google … you can always leave.
  6. WordPress can potentially be an agile, living mockup-tool for us to experiment in what we want collaboration to look like … so that when the time comes we can be better informed about how to build alternatives in Holochain.

Actionables

  1. Review the beginning of this post and reflect on its format as a collaborative tool.
  2. Communicate contents to others who you feel are vested in the next question.
  3. Do you feel you have sufficient information to make a decision about using WordPress to create a collaborative work space? If not, what is missing?
  4. Do you want to move forward with this experiment?
Posted in Ceptr, outside | You are welcome to add your comment

Ceptr CEO

n

A few days ago I was invited to listen in on a conversation where some Ceptr friends tried to deal with questions on organizational structure and whether whatever organizational structure there was needed a CEO and/or COO. I am going to try to zoom in on the CEO question as a reflection on the wider context.

I will start with what wasn’t brought up in the meeting at all: what is a CEO? Because this wasn’t brought up the conversation seemed to gravitate towards an unspoken but conventional assumption of what a traditional CEO (at least in my mind). I would vaguely describe it as someone who is brought in to introduce and sustain structure and order through authority (in this case authority that would be willfully given). Given some of the core narratives that have shaped Ceptr this felt like an inconsistency at least, possibly even a betrayal of values. Ceptr is about creating a technology which evokes dyamic and adaptive self-organization – an idea inspired by nature. I would want to see the organism that is Ceptr live in such a way. A forest doesn’t have a CEO, why does Ceptr need one? Can centralized authority be a foundation for creating a system of distributed authority?

Can there be another kind of CEO? These thoughts are inspired by the work of Frederic Leloux and his work Reinventing Organizations and by some of the inspiration I experienced in the existing examples introduced in Better Companies: Equal Footing.

The primary responsibility of a CEO is to listen to what wants to be. It is not to figure out or decide where an organization needs to go. It is to listen to listen carefully to what an organizations is asking to become. The primary action of a CEO is to direct attention and energy throughout the organization so that it can gracefull move as a whole towards that which wants to be. It isn’t to make sure that everyone is executing a predetermined plan. It is to notice and share with others that by bringing together our wills, resources and gifts we have created emergent conditions and that something is emerging.

Such a CEO cannot exist in a hierarchical and centralized organization. Such a CEO resonates with a different form of organization. If there needs to be a process of recruitment (whether its direct hiring or onboarding) then that process needs to be informed by the emerging story not by some centralized decision about recruitment which falls on the shoulders of one person. If there needs to be a process of funding then that too should not fall upon the shoulders of one person to direct and control, but is shaped to be in service of what wants to emerge. Maybe this CEO is a modern-day shaman-like figure … an intimate keeper of story.

If I look at Ceptr in this context, it already has a defacto CEO. If I try to imagine this defacto CEO being placed in a traditional CEO role, I can see him rejecting that notion … which is what seems to be happening. It goes against his nature. A traditional notion of CEO would undermine him and undermine Ceptr. His rejection of it is an act of leadership. I believe, the challenge we are left with is not to find a CEO, but to find a way to create conditions for an organization to emerge around the CEO we already have.

I would like to also touch lightly on the subject of man/woman. I believe we need to better discern between man/woman and masculine/feminine. I believe we need to give rise to feminine qualities and that confusing that with woman can be distracting. I would very much like to create and partake in an organization that is shaped by a better feminine-masculine mix (to avoid male dominance). I  believe our defacto CEO is a good example of this.

 

 

 

 

Posted in AltEco, Business, Ceptr, Community, outside | You are welcome to read 1 comment and to add yours

Frederic Leloux – Reinventing Organizations

n

I should have posted this a long time ago. I didn’t because I felt that I was too distanced from any organizational work to truly relate and engage with this work. But I have mentioned it numerous times and decided it would be easier for me to have it here as a reference.

This is Frederic Leloux talking about an evolution in thinking about organizations:

 

This is his website where his book is available. The books is also offered in a pay-what-feels-right-after-reading model.

I also noticed that the home page now includes an invitation to make the book freely accessible  “For networks engaged in societal and environmental transformation”.

 

Posted in AltEco, Business, Intake, outside | You are welcome to add your comment

Justin Searls: How to Program

n

I enjoyed this talk by Justin Searls who highlights what should be an obvious subject (for anyone who programs). There is a gap between knowing the semantics of a programming environment and being able to program with it to create something useful. Justin sheds some interesting light on this fundamental subject.

Though I still live in my intimate bubble with software at my fingertips, I am no longer actively involved in any “communal” making of software. When I do program it is because I want to do something and I have no one else who can program for me. Programming is an edgy experience for me. On one hand I appreciate the potential power of programming. On the other hand I don’t like doing it … and this talk touches on a lot of what I don’t like about it. (It is also another indicator to me that Ruby on Rails is … interesting).

How to Program from Test Double on Vimeo.

I am fresh off a year-long, focused programming effort. During this time I explored,  within my own private bubble, applying some of Christopher Alexander’s ideas about unfolding wholeness in the context of creating software. At the heart of Alexander’s view is that to achieve wholeness we need to create wholeness at every step of the way. Nature does not create separate parts which are then assembled into wholes. Am embryo in a womb is always whole, it is never in a temporary state where parts need to come together (fingers are not created and then attached to form a hand).

I found places to apply and express wholeness on many levels …  from code constructs (a single line of code, a function, a class) through to underlying processes and overall design. At every point I aspired to have something whole, sensible and working (even if not necessarily “producing tangible results”). It made me wish I’d known of these ideas when I was involved in software professionaly. I would want to explore these ideas in more depth and in the context of collaborative work.

For me, there is a subtle fault line in Justin’s talk. I sensed it when he qualified some of his choices as “personal preference”, evoking a sense of openness and pluralism instead of asserting “rightness” or “wholeness”. Alexander offers a parallel from his world of architecture using an example of a door. If we say a door is  3 feet wide, 8 feet tall, made of wood, painted green with brass hinges … these “facts” will not be disputed. But if I say that moving the door 3 inches to the left will give the room more life, that will be written off as opinion and just a matter of taste. Alexander’s work is an attempt to show that this is an error. That there is an empirical (thought not necessarily quantifiable) truth there, as true as the “more factual” attributes. If we are to get better at making rooms (or writing code) we need to learn to see and recognize this “wholeness” so that we can get better at creating it.

I feel that unfolding wholeness can be a meta-process that can embrace Justin’s observations and give them a deeper and more profound home.

 

Posted in Intake, Open Source, outside, Tech Stuff | You are welcome to add your comment