<div dir="ltr"><div dir="ltr">In response to the very valuable feedback I got, including from Rus in this group but also from others outside of this group, I reworked my deep learning training material very significantly. For if anybody is interested it's available from <a href="https://www.dropbox.com/s/y5fkp10ar5ze67n/deep%20learning%20training%20rev%2013012023.zip?dl=0">https://www.dropbox.com/s/y5fkp10ar5ze67n/deep%20learning%20training%20rev%2013012023.zip?dl=0</a><br><br>Just to make sure you guys are clear about the objective of the workshop, below is a quote from the training document. <br><br>But before I give the quote, just an answer to why am I doing this workshop?<br>a) There is plenty of very good material on the internet for free to go into deep learning for people with software skills. But none (that I know of) for those without any software expertise wanting to know and use deep learning <br>b) AI is making a splash in the world right now and there are (I think) many people without software skills yearning to know more about it.<br>c) It is really very easy (I think) for a professional without software skills to acquire what's necessary to apply deep learning for applications where tabular data is available to train the deep learning. This workshop intends to do exactly that. <br><br>Now for the quote from the training material:<br><p class="MsoNormal" style="margin:0pt 0pt 0.0001pt;font-family:Calibri">"This workshop aims to provide a broad overview of deep learning principles, rather than focusing on technical details. The goal is to give you a sense of how deep learning can be used to make predictions on practical problems using labeled data in spreadsheet form. By the end of the workshop, you will have the skills to apply deep learning to address challenges in your area of expertise.<br><br>During this workshop, we will be utilizing R or Python programming languages as a means of configuring and making predictions with labeled data in spreadsheet form. However, it is crucial to note that the primary focus of the workshop is not on programming itself, but rather on the fundamental concepts and techniques of deep learning. We will provide template programs that can be used with minimal modifications, and the main emphasis will be on guidance and exercises to solidify participants' understanding of the material. For those with little to no prior programming experience, we have included an optional section to introduce basic concepts of R or Python, but it is important to note that this section is only intended to provide a basic understanding and will not make you an expert in programming."</p><p class="MsoNormal" style="margin:0pt 0pt 0.0001pt;font-family:Calibri"><br></p><p class="MsoNormal" style="margin:0pt 0pt 0.0001pt;font-family:Calibri">Pieter</p></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, 9 Jan 2023 at 20:32, Steve Smith <<a href="mailto:sasmyth@swcp.com">sasmyth@swcp.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
  
    
  
  <div>
    <p>The "celebration of the hand" project (coordinated with UNM's
      Maxwell museum of anthropology) is nearly 15 years defunct now.  
      We made a tiny bit of progress on some LANL
      small-business-assistance time from an ML researcher, a small bit
      of seed-funding and matching time on my own part.   The key to
      this was the newly available prosumer-grade laser-scanners of the
      time, as well as emerging photogrammetric techniques.<br>
    </p>
    <p>The point of the project was to augment what humans were already
      doing...  both pointing to similarities too subtle for a human to
      notice or based on dimensions a human expert wouldn't be able to
      identify directly.   I haven't touched bases on where that field
      has gone in most of that intervening time.   My
      paleontological/archaelogical partner in that endeavor went his
      own way (he was more interested in developing a lifestyle business
      for himself than actually applying advanced techniques to the
      problem at "hand" (pun acknowledged)) and then a few years later
      he died which is part of the reason for not (re)visiting it
      myself... <br>
    </p>
    <p>The relevant issue to the linked article(s) is mostly that we
      were building modelless models (model-free learning) from the data
      with the hope (assumption?) that *some* of the features that might
      emerge as being good correlates to related/identical "hands" might
      be ones that humans could detect or make sense of themselves.  
      The state of the art at the time was definitely an "art" and was
      based on the expertise of contemporary flint knappers trying to
      reproduce the patterns found in non-contemporary artifacts.   We
      did not do any actual work (just speculation guided background
      research) on pottery and textiles...  the Maxwell museum
      researcher we worked with was also "overcome by events" simply
      needing to develop the exhibits and not having time/focus to
      attend to the more researchy aspects...  her interest was more on
      textiles than ceramics or lithics... which seemed to be the area
      we likely had the least advantage over human-experts.</p>
    <p>My niece works at the U of Az Archaelogy dept cataloging the
      accellerating number of artifacts coming in.   At some point I can
      imagine an automated classification system taking over much of the
      "mundane" aspects of this work...   like a self-driving car that
      at least provides collision avoidance and lane following...<br>
    </p>
    <p><br>
    </p>
    <div>On 1/9/23 10:29 AM, Pieter Steenekamp
      wrote:<br>
    </div>
    <blockquote type="cite">
      
      <div dir="ltr">Thanks for the references, I've briefly looked at
        them and am looking forward to perusing them more closely. The
        interpretation of ML is a big thing of course. The machine gives
        you the results and it would sometimes be nice if it can be
        accompanied by some explanation of how it achieved it.<br>
        <br>
        What you describe about your work sounds very interesting
        indeed. My gut feel is that, at least with the current
        generation of AI, human ingenuity and judgement would be far
        superior to AI in recognising similarities in the "hand" of the
        artisans involved. But AI could well be a powerful tool in
        assisting the humans?  </div>
      <br>
      <div class="gmail_quote">
        <div dir="ltr" class="gmail_attr">On Mon, 9 Jan 2023 at 18:50,
          Steve Smith <<a href="mailto:sasmyth@swcp.com" target="_blank">sasmyth@swcp.com</a>>
          wrote:<br>
        </div>
        <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
          <div>
            <p>I was hoping this article would have more meat to it, but
              the main point seems highly relevant to a
              practical/introductory workshop such as the one you are
              developing:<br>
            </p>
            <p><a href="https://dl.acm.org/doi/10.1145/3544903.3544905" target="_blank">The Need for
                Interpretable Features: Taxonomy and Motivation <br>
              </a></p>
            <p><a href="https://scitechdaily.com/mit-taxonomy-helps-build-explainability-into-the-components-of-machine-learning-models/" target="_blank">https://scitechdaily.com/mit-taxonomy-helps-build-explainability-into-the-components-of-machine-learning-models/</a></p>
            <p>My limited experience with ML is that the
              convo/invo-lutions that developers go through to make
              their learning models work well tend to obscure the
              interpretability of the results.   Some of this seems
              unavoidable (inevitable?), particularly the highly
              technical reserved terminology that often exists in a
              given domain, but this team purports to help provide
              guidelines for minimizing the consequences...</p>
            <p>My main experience with this domain involved an attempt
              to find distance measures between the flake patterns in 3D
              scanned archaeological artifacts, starting with lithics
              but also aspiring to work with pottery and textiles... 
              essentially trying to recognize similarities in the "hand"
              of the artisans involved.<br>
            </p>
            <p><br>
            </p>
            <div>On 1/9/23 1:29 AM, Pieter Steenekamp wrote:<br>
            </div>
            <blockquote type="cite">
              <div dir="ltr"><a class="gmail_plusreply" id="m_4420140071018442964m_8371091482889110803plusReplyChip-0">@Russ,  </a>
                <div><a class="gmail_plusreply"><br>
                  </a>a) I appreciate the suggestion to include a simple
                  neural network that can make predictions based on
                  inputs and be trained using steepest descent to
                  optimize its weights. This would be a valuable
                  addition to my training material, as it provides a
                  foundational understanding of how neural networks
                  work. <br>
                  <br>
                  b) My focus is on providing practical training for
                  professionals with deep domain knowledge but limited
                  software experience, who are looking to utilize deep
                  learning as a tool in their respective fields. In
                  contrast, it seems that your focus is on understanding
                  the inner workings of deep learning. Both approaches
                  have their own merits, and it is important to cater to
                  the needs and goals of different learners.<br>
                  <br>
                  Pieter</div>
              </div>
              <br>
              <div class="gmail_quote">
                <div dir="ltr" class="gmail_attr">On Sun, 8 Jan 2023 at
                  21:20, Marcus Daniels <<a href="mailto:marcus@snoutfarm.com" target="_blank">marcus@snoutfarm.com</a>>
                  wrote:<br>
                </div>
                <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                  <div dir="auto"> The main defects of both R and Python
                    are a lack of a typing system and high performance
                    compilation.  I find R still follows (is used by)
                    the statistics research community more than Python.
                      Common Lisp was always better than either.<br>
                    <br>
                    <div dir="ltr">Sent from my iPhone</div>
                    <div dir="ltr"><br>
                      <blockquote type="cite">On Jan 8, 2023, at 11:03
                        AM, Russ Abbott <<a href="mailto:russ.abbott@gmail.com" target="_blank">russ.abbott@gmail.com</a>>
                        wrote:<br>
                        <br>
                      </blockquote>
                    </div>
                    <blockquote type="cite">
                      <div dir="ltr">
                        <div dir="ltr">
                          <div style="font-family:arial,helvetica,sans-serif;font-size:small;color:rgb(0,0,0)">As
                            indicated in my original reply, my interest
                            in this project grows from my relative
                            ignorance of Deep Learning. My career has
                            focussed exclusively on symbolic computing.
                            I've worked with and taught (a) functional
                            programming, logic programming, and related
                            issues in advanced Python; (b) complex
                            systems, agent-based modeling, genetic
                            algorithms, and related evolutionary
                            processes, (c) a bit of constraint
                            programming, especially in MiniZinc, and (d)
                            reinforcement learning as Q-learning, which
                            is reinforcement learning without neural
                            nets. I've always avoided neural nets--and
                            more generally numerical programming of any
                            sort. </div>
                          <div style="font-family:arial,helvetica,sans-serif;font-size:small;color:rgb(0,0,0)"><br>
                          </div>
                          <div style="font-family:arial,helvetica,sans-serif;font-size:small;color:rgb(0,0,0)">Deep
                            learning has produced so many impressive
                            results that I've decided to devote much of
                            my retirement life to learning about it. I
                            retired at the end of Spring 2022 and (after
                            a break) am now devoting much of my time to
                            learning more about Deep Neural Nets. So
                            far, I've dipped my brain into it at various
                            points. I think I've learned a fair amount.
                            For example, </div>
                          <div style="font-family:arial,helvetica,sans-serif;font-size:small;color:rgb(0,0,0)">
                            <ul>
                              <li>I now know how to build a neural net
                                (NN) that adds two numbers using a
                                single layer with a single neuron. It's
                                really quite simple and is, I think, a
                                beautiful example of how NNs work. If I
                                were to teach an intro to NNs I'd start
                                with this.</li>
                              <li>I've gone through the Kaggle Deep
                                Learning sequence mentioned earlier. </li>
                              <li>I found a paper that shows how you can
                                approximate any differentiable function
                                to any degree of accuracy with a
                                single-layer NN. (This is a very nice
                                result, although I believe it's not used
                                explicitly in building serious Deep NN
                                systems.)</li>
                              <li>From what I've seen so far, most
                                serious DNNs are built using Keras
                                rather than PyTorch.</li>
                              <li>I've looked at Jeremy Howard's <a href="http://fast.ai" target="_blank">fast.ai</a>
                                material. I was going to go through the
                                course but stopped when I found that it
                                uses PyTorch. Also, it seems to be built
                                on <a href="http://fast.ai" target="_blank">fast.ai</a>
                                libraries that do a lot of the work for
                                you without explanation.  And it seems
                                to focus almost exclusively on
                                Convolutional NNs. </li>
                              <li>My impression of DNNs is that to a
                                great extent they are <i>ad hoc</i>.
                                There is no good way to determine the
                                best architecture to use for a given
                                problem. By architecture, I mean the
                                number of layers, the number of neurons
                                in each layer, the types of layers, the
                                activation functions to use, etc. </li>
                              <li>All DNNs that I've seen use Python as
                                code glue rather than R or some other
                                language. I like Python--so I'm pleased
                                with that.</li>
                              <li>To build serious NNs one should learn
                                the Python libraries Numpy (array
                                manipulation) and Pandas (data
                                processing). Numpy especially seems to
                                be used for virtually all DNNs that I've
                                seen. </li>
                              <li>Keras and probably PyTorch include a
                                number of special-purpose neurons and
                                layers that can be included in one's
                                DNN. These include: a DropOut layer,
                                LSTM (short-long-term memory) neurons,
                                convolutional layers, recurrent neural
                                net layers (RNN), and more recently
                                transformers, which get credit for
                                ChatGPT and related programs. My
                                impression is that these special-purpose
                                layers are <i>ad hoc</i> in the same
                                sense that functions or libraries that
                                one finds useful in a programming
                                language are <i>ad hoc</i>. They have
                                been very important for the success of
                                DNNs, but they came into existence
                                because people invented them in the same
                                way that people invented useful
                                functions and libraries. </li>
                              <li>NN libraries also include a menagerie
                                of activation functions. An activation
                                function acts as the final control on
                                the output of a layer. Different
                                activation functions are used for
                                different purposes. To be successful in
                                building a DNN, one must understand
                                what those activation functions do for
                                you and which ones to use. </li>
                              <li>I'm especially interested in DNNs that
                                use reinforcement learning. That's
                                because the first DNN work that
                                impressed me was DeepMind's DNNs that
                                learned to play Atari games--and then
                                Go, etc. An important advantage of
                                Reinforcement Learning (RL) is that it
                                doesn't depend on mountains of labeled
                                data. </li>
                              <li>I find RL systems more interesting
                                than image recognition systems. One of
                                the striking features of many image
                                recognition systems is that they can be
                                thrown off by changing a small number of
                                pixels in an image. The changed image
                                would look to a human observer just like
                                the original, but it might fool a
                                trained NN into labeling the image as a
                                banana rather than, say, an automobile,
                                which is what it really is. To address
                                this problem people have developed
                                Generative Adversarial Networks (GANs)
                                which attempt to find such weaknesses in
                                a neural net during training and then to
                                train the NN not to have those
                                weaknesses. This is a fascinating
                                result, but as far as I can tell, it
                                mainly shows how fragile some NNs are
                                and doesn't add much conceptual depth to
                                one's understanding of how NNs work. </li>
                            </ul>
                            <div>I'm impressed with this list of things
                              I sort of know. If you had asked me before
                              I started writing this email I wouldn't
                              have thought I had learned as much as I
                              have. Even so, I feel like I don't
                              understand much of it beyond a superficial
                              level. </div>
                            <div><br>
                            </div>
                            <div>
                              <div>So far I've done all my exploration
                                using Google's Colab (Google's Python
                                notebook implementation) and Kaggle's
                                similar Python notebook implementation.
                                (I prefer Colab to Kaggle.) Using either
                                one, it's super nice not to have to
                                download and install anything!</div>
                              <div><br>
                              </div>
                            </div>
                            <div style="color:rgb(34,34,34);font-family:Arial,Helvetica,sans-serif">
                              <div dir="ltr">
                                <div dir="ltr">
                                  <div>
                                    <div dir="ltr">
                                      <div>
                                        <div dir="ltr">
                                          <div>
                                            <div dir="ltr">
                                              <div>
                                                <div dir="ltr">
                                                  <div>
                                                    <div dir="ltr">
                                                      <div>
                                                        <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                        </div>
                                                      </div>
                                                    </div>
                                                  </div>
                                                </div>
                                              </div>
                                            </div>
                                          </div>
                                        </div>
                                      </div>
                                    </div>
                                  </div>
                                </div>
                              </div>
                            </div>
                            <div>I'm continuing my journey to learn more
                              about DNNs. I'd be happy to have company
                              and to help develop materials to teach
                              about DNNs. (Developing teaching materials
                              always helps me learn the subject being
                              covered.)</div>
                            <div><br>
                            </div>
                          </div>
                          <div>
                            <div dir="ltr">
                              <div dir="ltr">
                                <div>
                                  <div dir="ltr">
                                    <div>
                                      <div dir="ltr">
                                        <div>
                                          <div dir="ltr">
                                            <div>
                                              <div dir="ltr">
                                                <div>
                                                  <div dir="ltr">
                                                    <div>
                                                      <div dir="ltr">
                                                        <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr">
                                                          <div dir="ltr"><span></span>--
                                                          Russ Abbott  
                                                                       
                                                                       
                                                                  <br>
                                                          Professor
                                                          Emeritus,
                                                          Computer
                                                          Science<br>
                                                          California
                                                          State
                                                          University,
                                                          Los Angeles<br>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                        </div>
                                                      </div>
                                                    </div>
                                                  </div>
                                                </div>
                                              </div>
                                            </div>
                                          </div>
                                        </div>
                                      </div>
                                    </div>
                                  </div>
                                </div>
                              </div>
                            </div>
                          </div>
                          <br>
                        </div>
                        <br>
                        <div class="gmail_quote">
                          <div dir="ltr" class="gmail_attr">On Sun, Jan
                            8, 2023 at 1:48 AM glen <<a href="mailto:gepropella@gmail.com" target="_blank">gepropella@gmail.com</a>>
                            wrote:<br>
                          </div>
                          <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> Yes, the
                            money/expertise bar is still pretty high.
                            But TANSTAAFL still applies. And the
                            overwhelming evidence is coming in that
                            specific models do better than those trained
                            up on diverse data sets, "better" meaning
                            less prone to subtle bullsh¡t. What I find
                            fascinating is tools like OpenAI
                            *facilitate* trespassing. We have a
                            wonderful bloom of non-experts claiming they
                            understand things like "deep learning". But
                            do they? An old internet meme is brought to
                            mind: "Do you even Linear Algebra, bro?"
                            >8^D<br>
                            <br>
                            On 1/8/23 01:06, Jochen Fromm wrote:<br>
                            > I have finished a number of Coursera
                            courses recently, including "Deep Learning
                            & Neural Networks with Keras" which was
                            ok but not great. The problems with deep
                            learning are<br>
                            > <br>
                            > * to achieve impressive results like
                            chatGPT from OpenAi or LaMDA from Goggle you
                            need to spend millions on hardware<br>
                            > * only big organisations can afford to
                            create such expensive models<br>
                            > * the resulting network is s black box
                            and it is unclear why it works the way it
                            does<br>
                            > <br>
                            > In the end it is just the same old back
                            propagation that has been known for decades,
                            just on more computers and trained on more
                            data. Peter Norvig calls it "The
                            unreasonable effectiveness of data"<br>
                            > <a href="https://research.google.com/pubs/archive/35179.pdf" rel="noreferrer" target="_blank">
                              https://research.google.com/pubs/archive/35179.pdf</a><br>
                            > <br>
                            > -J.<br>
                            > <br>
                            > <br>
                            > -------- Original message --------<br>
                            > From: Russ Abbott <<a href="mailto:russ.abbott@gmail.com" target="_blank">russ.abbott@gmail.com</a>><br>
                            > Date: 1/8/23 12:20 AM (GMT+01:00)<br>
                            > To: The Friday Morning Applied
                            Complexity Coffee Group <<a href="mailto:friam@redfish.com" target="_blank">friam@redfish.com</a>><br>
                            > Subject: Re: [FRIAM] Deep learning
                            training material<br>
                            > <br>
                            > Hi Pieter,<br>
                            > <br>
                            > A few comments.<br>
                            > <br>
                            >   * Much of the actual deep learning
                            material looks like it came from the Kaggle
                            "Deep Learning <<a href="https://www.kaggle.com/learn/intro-to-deep-learning" rel="noreferrer" target="_blank">https://www.kaggle.com/learn/intro-to-deep-learning</a>>"
                            sequence.<br>
                            >   * In my opinion, R is an ugly and /ad
                            hoc/ language. I'd stick to Python.<br>
                            >   * More importantly, I would put the
                            How-to-use-Python stuff into a preliminary
                            class. Assume your audience knows how to use
                            Python and focus on Deep Learning. Given
                            that, there is only a minimal amount of
                            information about Deep Learning in the
                            write-up. If I were to attend the workshop
                            and thought I would be learning about Deep
                            Learning, I would be disappointed--at least
                            with what's covered in the write-up.<br>
                            > <br>
                            >     I say this because I've been
                            looking for a good intro to Deep Learning.
                            Even though I taught Computer Science for
                            many years, and am now retired, I avoided
                            Deep Learning because it was so
                            non-symbolic. My focus has always been on
                            symbolic computing. But Deep Learning has
                            produced so many extraordinarily impressive
                            results, I decided I should learn more about
                            it. I haven't found any really good
                            material. If you are interested, I'd be more
                            than happy to work with you on developing
                            some introductory Deep Learning material. <br>
                            > <br>
                            > -- Russ Abbott<br>
                            > Professor Emeritus, Computer Science<br>
                            > California State University, Los
                            Angeles<br>
                            > <br>
                            > <br>
                            > On Thu, Jan 5, 2023 at 11:31 AM Pieter
                            Steenekamp <<a href="mailto:pieters@randcontrols.co.za" target="_blank">pieters@randcontrols.co.za</a>
                            <mailto:<a href="mailto:pieters@randcontrols.co.za" target="_blank">pieters@randcontrols.co.za</a>>>
                            wrote:<br>
                            > <br>
                            >     Thanks to the kind support of
                            OpenAI's chatGPT, I am in the process of
                            gathering materials for a comprehensive and
                            hands-on deep learning workshop. Although it
                            is still a work in progress, I welcome any
                            interested parties to take a look and
                            provide their valuable input. Thank you!<br>
                            > <br>
                            >     You can get it from:<br>
                            >     <a href="https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0" rel="noreferrer" target="_blank">https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0</a>
                            <<a href="https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0" rel="noreferrer" target="_blank">https://www.dropbox.com/s/eyx4iumb0439wlx/deep%20learning%20training%20rev%2005012023.zip?dl=0</a>><br>
                            > <br>
                            <br>
                            -- <br>
                            ꙮ Mɥǝu ǝlǝdɥɐuʇs ɟᴉƃɥʇ' ʇɥǝ ƃɹɐss snɟɟǝɹs˙ ꙮ<br>
                            <br>
                            -. --- - / ...- .- .-.. .. -.. / -- --- .-.
                            ... . / -.-. --- -.. .<br>
                            FRIAM Applied Complexity Group listserv<br>
                            Fridays 9a-12p Friday St. Johns Cafe   / 
                             Thursdays 9a-12p Zoom <a href="https://bit.ly/virtualfriam" rel="noreferrer" target="_blank">
                              https://bit.ly/virtualfriam</a><br>
                            to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">
http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
                            FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">
                              http://friam-comic.blogspot.com/</a><br>
                            archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" rel="noreferrer" target="_blank">
                              https://redfish.com/pipermail/friam_redfish.com/</a><br>
                              1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" rel="noreferrer" target="_blank">
                              http://friam.383.s1.nabble.com/</a><br>
                          </blockquote>
                        </div>
                        <span>-. --- - / ...- .- .-.. .. -.. / -- ---
                          .-. ... . / -.-. --- -.. .</span><br>
                        <span>FRIAM Applied Complexity Group listserv</span><br>
                        <span>Fridays 9a-12p Friday St. Johns Cafe   /
                            Thursdays 9a-12p Zoom <a href="https://bit.ly/virtualfriam" target="_blank">https://bit.ly/virtualfriam</a></span><br>
                        <span>to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a></span><br>
                        <span>FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" target="_blank">http://friam-comic.blogspot.com/</a></span><br>
                        <span>archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a></span><br>
                        <span> 1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" target="_blank">http://friam.383.s1.nabble.com/</a></span><br>
                      </div>
                    </blockquote>
                  </div>
                  -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . /
                  -.-. --- -.. .<br>
                  FRIAM Applied Complexity Group listserv<br>
                  Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays
                  9a-12p Zoom <a href="https://bit.ly/virtualfriam" rel="noreferrer" target="_blank">https://bit.ly/virtualfriam</a><br>
                  to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
                  FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">http://friam-comic.blogspot.com/</a><br>
                  archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" rel="noreferrer" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a><br>
                    1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" rel="noreferrer" target="_blank">http://friam.383.s1.nabble.com/</a><br>
                </blockquote>
              </div>
              <br>
              <fieldset></fieldset>
              <pre>-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom <a href="https://bit.ly/virtualfriam" target="_blank">https://bit.ly/virtualfriam</a>
to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" target="_blank">http://friam-comic.blogspot.com/</a>
archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a>
  1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" target="_blank">http://friam.383.s1.nabble.com/</a>
</pre>
            </blockquote>
          </div>
          -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. ---
          -.. .<br>
          FRIAM Applied Complexity Group listserv<br>
          Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p
          Zoom <a href="https://bit.ly/virtualfriam" rel="noreferrer" target="_blank">https://bit.ly/virtualfriam</a><br>
          to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
          FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">http://friam-comic.blogspot.com/</a><br>
          archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" rel="noreferrer" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a><br>
            1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" rel="noreferrer" target="_blank">http://friam.383.s1.nabble.com/</a><br>
        </blockquote>
      </div>
      <br>
      <fieldset></fieldset>
      <pre>-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .
FRIAM Applied Complexity Group listserv
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom <a href="https://bit.ly/virtualfriam" target="_blank">https://bit.ly/virtualfriam</a>
to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" target="_blank">http://friam-comic.blogspot.com/</a>
archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a>
  1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" target="_blank">http://friam.383.s1.nabble.com/</a>
</pre>
    </blockquote>
  </div>

-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. .<br>
FRIAM Applied Complexity Group listserv<br>
Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom <a href="https://bit.ly/virtualfriam" rel="noreferrer" target="_blank">https://bit.ly/virtualfriam</a><br>
to (un)subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">http://friam-comic.blogspot.com/</a><br>
archives:  5/2017 thru present <a href="https://redfish.com/pipermail/friam_redfish.com/" rel="noreferrer" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a><br>
  1/2003 thru 6/2021  <a href="http://friam.383.s1.nabble.com/" rel="noreferrer" target="_blank">http://friam.383.s1.nabble.com/</a><br>
</blockquote></div></div>