<html data-lt-installed="true">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body style="padding-bottom: 1px;">
    <p><font size="5">I find myself using either MS Chat or Google Bard
        as they offer new content.</font></p>
    <p><font size="5">Still wouldn't write a paper with them. An
        outline, yes.<br>
      </font></p>
    <p><font size="5">Command line syntax questions in Windows/Linux
        have been very accurate and helpful in Chat.</font></p>
    <p><font size="5">MS is rapidly integrating AI into Windows 11, and
        Office. It will soon be a piece of cake for somebody who does
        not know how to generate charts and graphs in Microsoft Excel to
        do so easily.</font></p>
    <p><font size="5">Some companies are beginning to ban chat GPT for
        their employees. I think they believe it's no different than
        using Facebook while at work. Programmers that work for these
        companies are frustrated because they've been using ai to check
        their code and run tests, and now their bosses are telling them
        they're not allowed to.</font></p>
    <p><font size="5"><br>
      </font></p>
    <p><font size="5"><br>
      </font></p>
    <p><br>
    </p>
    <div class="moz-cite-prefix">On 6/24/2023 12:07 PM, Robert Primak
      wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:218323105.3469503.1687622847313@mail.yahoo.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div class="ydpebc6f944yahoo-style-wrap"
        style="font-family:Helvetica Neue, Helvetica, Arial,
        sans-serif;font-size:13px;">
        <div dir="ltr" data-setdir="false">The paradigm of an "<span>AI
            model they’d trained themselves that looks for telltale
            signals of ChatGPT output" has been proven time and time
            again to spew out false-positives. College professors have
            lost their jobs for making unsubstantiated accusations and
            penalizing students for allegedly using AI or ChatGPT when
            in fact the students came back with notes and drafts proving
            definitively that their work was original and not generated
            by AI. An "accuracy rate" which allows for up to 40%
            false-positives does not encourage placing any faith in AI
            models which claim to be able to detect AI produced content.
            The Swiss team was either lying, or else they were deceiving
            themselves. Put succinctly, AI can never definitively detect
            AI.</span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span>Nevertheless, there is
            a very real possibility that at least some LLMs are reusing
            data sets which were supplied with the help of AI models.
            The models might very well be feeding upon their own errors.
            And this does pose a real threat to the integrity of the
            information presented as output by all LLMs. How LLMs and
            other AI are trained does influence their behaviors when
            they are put into use.</span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span>Outsourcing any part of
            AI training does run the risk that the gig workers involved
            will take every shortcut they can to generate the illusion
            that they are producing more finished work in a shorter time
            interval. Gig workers are notorious for doing everything
            they can to avoid doing any real work.</span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span>The obvious solution is
            this: Pay qualified people to do the training work under the
            watchful eyes of in-house research staff. It's a lot more
            expensive, but this is the only way to make sure no one is
            taking thee easy way out and endangering the integrity of
            the entire project. When you take shortcuts in technology
            development, it is your end-users who invariably pay the
            price. </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span>-- Bob Primak </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div dir="ltr" data-setdir="false"><span><br>
          </span></div>
        <div><br>
        </div>
      </div>
      <div id="yahoo_quoted_7828991126" class="yahoo_quoted">
        <div style="font-family:'Helvetica Neue', Helvetica, Arial,
          sans-serif;font-size:13px;color:#26282a;">
          <div> On Saturday, June 24, 2023 at 11:43:34 AM EDT,
            <a class="moz-txt-link-abbreviated" href="mailto:jjrudy1@comcast.net">jjrudy1@comcast.net</a> <a class="moz-txt-link-rfc2396E" href="mailto:jjrudy1@comcast.net"><jjrudy1@comcast.net></a> wrote: </div>
          <div><br>
          </div>
          <div><br>
          </div>
          <div>
            <div id="yiv3946723560">
              <style>#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 p.yiv3946723560MsoNormal, #yiv3946723560 li.yiv3946723560MsoNormal, #yiv3946723560 div.yiv3946723560MsoNormal
        {margin:0in;font-size:11.0pt;font-family:"Calibri", sans-serif;}#yiv3946723560 h1
        {margin-right:0in;margin-left:0in;font-size:24.0pt;font-family:"Calibri", sans-serif;font-weight:bold;}#yiv3946723560 h2
        {margin-right:0in;margin-left:0in;font-size:18.0pt;font-family:"Calibri", sans-serif;font-weight:bold;}#yiv3946723560 h3
        {margin-right:0in;margin-left:0in;font-size:13.5pt;font-family:"Calibri", sans-serif;font-weight:bold;}#yiv3946723560 a:link, #yiv3946723560 span.yiv3946723560MsoHyperlink
        {color:#0563C1;text-decoration:underline;}#yiv3946723560 span.yiv3946723560EmailStyle17
        {font-family:"Arial", sans-serif;font-variant:normal !important;color:windowtext;text-transform:none;text-decoration:none none;vertical-align:baseline;}#yiv3946723560 span.yiv3946723560Heading1Char
        {font-family:"Calibri", sans-serif;font-weight:bold;}#yiv3946723560 span.yiv3946723560Heading2Char
        {font-family:"Calibri", sans-serif;font-weight:bold;}#yiv3946723560 span.yiv3946723560Heading3Char
        {font-family:"Calibri", sans-serif;font-weight:bold;}#yiv3946723560 span.yiv3946723560bylineby--3i70z
        {}#yiv3946723560 p.yiv3946723560bylineauthor--g26rn, #yiv3946723560 li.yiv3946723560bylineauthor--g26rn, #yiv3946723560 div.yiv3946723560bylineauthor--g26rn
        {margin-right:0in;margin-left:0in;font-size:11.0pt;font-family:"Calibri", sans-serif;}#yiv3946723560 span.yiv3946723560screen-reader-text
        {}#yiv3946723560 span.yiv3946723560imagecredit--1fj0h
        {}#yiv3946723560 span.yiv3946723560slideradclosetext--1dj37
        {}#yiv3946723560 .yiv3946723560MsoChpDefault
        {font-family:"Calibri", sans-serif;}#yiv3946723560 filtered {}#yiv3946723560 div.yiv3946723560WordSection1
        {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 filtered {}#yiv3946723560 ol
        {margin-bottom:0in;}#yiv3946723560 ul
        {margin-bottom:0in;}</style>
              <div>
                <div class="yiv3946723560WordSection1">
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      style="font-size:12.0pt;font-family:inherit,
                      serif;color:#111111;text-transform:uppercase;"><a
                        rel="nofollow noopener noreferrer"
                        target="_blank"
                        href="https://www.technologyreview.com/topic/artificial-intelligence/"
                        moz-do-not-send="true"><b><span
                            style="font-family:Courier
                            New;color:white;border:none windowtext
                            1.0pt;padding:0in;text-decoration:none;">ARTIFICIALThis
                            is scary INTELLIGENCE</span></b></a></span></p>
                  <h1
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;margin:0px;outline:0px;"><span
                      style="font-family:Arial, sans-serif;color:white;">paid
                      to train AI are outsourcing their work… to AI</span></h1>
                  <p
style="margin:0in;vertical-align:baseline;font-style:inherit;font-variant:inherit;font-weight:inherit;font-stretch:inherit;line-height:inherit;outline:0px;"><b><span
                        style="font-size:18.0pt;font-family:inherit,
                        serif;color:white;border:none windowtext
                        1.0pt;padding:0in;">It’s a practice that could
                        introduce further errors into already
                        error-prone models.</span></b><span
                      style="font-size:18.0pt;font-family:inherit,
                      serif;color:white;"></span></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      class="yiv3946723560bylineby--3i70z"><b><span
                          style="font-size:15.0pt;font-family:inherit,
                          serif;color:white;border:none windowtext
                          1.0pt;padding:0in;">By </span></b></span><b><span
                        style="font-size:12.0pt;font-family:Arial,
                        sans-serif;color:white;"></span></b></p>
                  <ul style="margin-top:0in;" type="disc">
                    <li
style="color:white;margin-top:0in;margin-bottom:0in;vertical-align:baseline;font-style:inherit;font-variant:inherit;font-weight:inherit;font-stretch:inherit;line-height:inherit;outline:0px;display:inline-block;"
                      class="yiv3946723560bylineauthor--g26rn"><b><span
                          style="font-size:15.0pt;font-family:inherit,
                          serif;"><a rel="nofollow noopener noreferrer"
                            target="_blank"
                            href="https://www.technologyreview.com/author/rhiannon-williams/"
                            moz-do-not-send="true"><span
                              style="border:none windowtext
                              1.0pt;padding:0in;text-decoration:none;">Rhiannon
                              Williams</span><span
                              class="yiv3946723560screen-reader-text"><span
                                style="color:blue;border:none windowtext
                                1.0pt;padding:0in;text-decoration:none;">archive
                                page</span></span></a></span></b></li>
                  </ul>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      style="font-size:12.0pt;font-family:Arial,
                      sans-serif;color:white;">June 22, 2023</span></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      style="font-size:12.0pt;font-family:inherit,
                      serif;color:#111111;border:none windowtext
                      1.0pt;padding:0in;"><img
                        style="width:22.0in;min-height:12.3802in;"
                        id="yiv3946723560Picture_x0020_3"
                        src="cid:Vqf8OfFIWa4nXH9z7UMS"
                        yahoo_partid="1.2" alt="anonymous workers at a
                        conveyor belt and strings of words flow out one
                        end" moz-do-not-send="true" width="2112"
                        height="1189" border="0"></span><span
                      style="font-size:12.0pt;font-family:inherit,
                      serif;color:#111111;border:none windowtext
                      1.0pt;padding:0in;"></span></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      class="yiv3946723560imagecredit--1fj0h"><span
                        style="font-size:12.0pt;font-family:inherit,
serif;color:#616568;text-transform:uppercase;letter-spacing:.75pt;border:none
                        windowtext 1.0pt;padding:0in;">STEPHANIE
                        ARNETT/MITTR | GETTY</span></span><span
                      style="font-size:12.0pt;font-family:inherit,
                      serif;color:#111111;"></span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">A significant
                      proportion of people paid to train AI models may
                      be themselves outsourcing that work to AI, a new
                      study has found. </span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">It takes an incredible
                      amount of data to train AI systems to perform
                      specific tasks accurately and reliably. Many
                      companies pay gig workers on platforms like <a
                        rel="nofollow noopener noreferrer"
                        target="_blank"
href="https://www.technologyreview.com/2020/12/11/1014081/ai-machine-learning-crowd-gig-worker-problem-amazon-mechanical-turk/"
                        moz-do-not-send="true"><span style="border:none
                          windowtext 1.0pt;padding:0in;">Mechanical Turk</span></a> to
                      complete tasks that are typically hard to
                      automate, such as solving CAPTCHAs, labeling data
                      and annotating text. This data is then fed into AI
                      models to train them. The workers are poorly paid
                      and are often expected to complete lots of tasks
                      very quickly. </span></p>
                  <div style="border:none;border-top:solid black
                    1.0pt;padding:0in 0in 0in 0in;">
                    <h2
style="vertical-align:baseline;border:none;padding:0in;border-right-style:initial;border-bottom-style:initial;border-left-style:initial;border-right-color:initial;border-bottom-color:initial;border-left-color:initial;font-style:inherit;font-variant:inherit;font-stretch:inherit;line-height:inherit;padding:0px
                      0px;outline:0px;"><span
                        style="font-family:inherit,
                        serif;color:#111111;">Related Story</span></h2>
                  </div>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><a rel="nofollow
                      noopener noreferrer" target="_blank"
href="https://www.technologyreview.com/2023/04/04/1070938/we-are-hurtling-toward-a-glitchy-spammy-scammy-ai-powered-internet/"
                      moz-do-not-send="true"><span
                        style="font-size:10.5pt;font-family:inherit,
                        serif;color:blue;border:none windowtext
                        1.0pt;padding:0in;text-decoration:none;"><img
                          style="width:22.0in;min-height:12.3802in;"
                          id="yiv3946723560Picture_x0020_2"
                          src="cid:5aW0omu0H7MBAAXnOjAQ"
                          yahoo_partid="1.3" alt="&quot;&quot;"
                          moz-do-not-send="true" width="2112"
                          height="1189" border="0"></span></a><span
                      style="font-size:10.5pt;font-family:inherit,
                      serif;color:#111111;"></span></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      style="font-size:10.5pt;font-family:inherit,
                      serif;color:#111111;"><a rel="nofollow noopener
                        noreferrer" target="_blank"
href="https://www.technologyreview.com/2023/04/04/1070938/we-are-hurtling-toward-a-glitchy-spammy-scammy-ai-powered-internet/"
                        moz-do-not-send="true"><b><span
                            style="font-size:11.0pt;font-family:Calibri,
                            sans-serif;border:none windowtext
                            1.0pt;padding:0in;"></span></b></a></span></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><b><u><span
                          style="font-size:10.5pt;font-family:inherit,
                          serif;color:blue;border:none windowtext
                          1.0pt;padding:0in;"><a rel="nofollow noopener
                            noreferrer" target="_blank"
href="https://www.technologyreview.com/2023/04/04/1070938/we-are-hurtling-toward-a-glitchy-spammy-scammy-ai-powered-internet/"
                            moz-do-not-send="true"><span
                              style="color:blue;">We are hurtling toward
                              a glitchy, spammy, scammy, AI-powered
                              internet</span><span
                              style="font-size:11.0pt;font-family:Calibri,
sans-serif;color:windowtext;border:none;font-weight:normal;text-decoration:none;"></span></a></span></u></b></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      style="font-size:10.5pt;font-family:inherit,
                      serif;color:#111111;">  </span></p>
                  <p
style="margin:0in;vertical-align:baseline;font-style:inherit;font-variant:inherit;font-weight:inherit;font-stretch:inherit;line-height:inherit;outline:0px;"><span
                      style="font-size:10.5pt;font-family:inherit,
                      serif;color:#111111;">Large language models are
                      full of security vulnerabilities, yet they’re
                      being embedded into tech products on a vast scale.</span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">No wonder some of them
                      may be turning to tools like ChatGPT to maximize
                      their earning potential. But how many? To find
                      out, a team of researchers from the Swiss Federal
                      Institute of Technology (EPFL) hired 44 people on
                      the gig work platform Amazon Mechanical Turk to
                      summarize 16 extracts from medical research
                      papers. Then they analyzed their responses using
                      an AI model they’d trained themselves that looks
                      for telltale signals of ChatGPT output, such as
                      lack of variety in choice of words. They also
                      extracted the workers’ keystrokes in a bid to work
                      out whether they’d copied and pasted their
                      answers, an indicator that they’d generated their
                      responses elsewhere.</span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">They estimated that
                      somewhere between 33% and 46% of the workers had
                      used AI models like OpenAI’s ChatGPT. It’s a
                      percentage that’s likely to grow even higher as
                      ChatGPT and other AI systems become more powerful
                      and easily accessible, according to the authors of
                      the study, which has been shared on <a
                        rel="nofollow noopener noreferrer"
                        target="_blank"
                        href="https://arxiv.org/pdf/2306.07899.pdf"
                        moz-do-not-send="true"><span style="border:none
                          windowtext 1.0pt;padding:0in;">arXiv</span></a> and
                      is yet to be peer-reviewed. </span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">“I don’t think it’s the
                      end of crowdsourcing platforms. It just changes
                      the dynamics,” says Robert West, an assistant
                      professor at EPFL, who coauthored the study. </span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">Using AI-generated data
                      to train AI could introduce further errors into
                      already error-prone models. Large language models
                      regularly present false information as fact. If
                      they generate incorrect output that is itself used
                      to train other AI models, the errors can be
                      absorbed by those models and amplified over time,
                      making it more and more difficult to work out
                      their origins, says Ilia Shumailov, a junior
                      research fellow in computer science at Oxford
                      University, who was not involved in the project.</span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">Even worse, there’s no
                      simple fix. “The problem is, when you’re using
                      artificial data, you acquire the errors from the
                      misunderstandings of the models and statistical
                      errors,” he says. “You need to make sure that your
                      errors are not biasing the output of other models,
                      and there’s no simple way to do that.”</span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">The study highlights
                      the need for new ways to check whether data has
                      been produced by humans or AI. It also highlights
                      one of the problems with tech companies’ tendency
                      to rely on gig workers to do the vital work of
                      tidying up the data fed to AI systems.  </span></p>
                  <p
style="vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;outline:0px;"><span
                      style="font-size:12.0pt;font-family:Times New
                      Roman, serif;color:black;">“I don’t think
                      everything will collapse,” says West. “But I think
                      the AI community will have to investigate closely
                      which tasks are most prone to being automated and
                      to work on ways to prevent this.”</span></p>
                  <p style="vertical-align:baseline;"
                    class="yiv3946723560MsoNormal"><span
                      class="yiv3946723560slideradclosetext--1dj37"><span
                        style="font-size:12.0pt;font-family:inherit,
                        serif;color:#111111;border:none windowtext
                        1.0pt;padding:0in;">hide</span></span><span
                      style="font-size:12.0pt;font-family:inherit,
                      serif;color:#111111;"></span></p>
                  <h3
style="margin:0in;vertical-align:baseline;font-style:inherit;font-variant:inherit;font-stretch:inherit;line-height:inherit;outline:0px;"><span
                      style="font-size:18.0pt;font-family:inherit,
                      serif;color:#111111;">by <a rel="nofollow noopener
                        noreferrer" target="_blank"
                        href="https://www.technologyreview.com/author/rhiannon-williams/"
                        moz-do-not-send="true"><span style="border:none
                          windowtext
                          1.0pt;padding:0in;text-decoration:none;">Rhiannon
                          Williams</span></a></span></h3>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:12.0pt;font-family:Arial,
                      sans-serif;">  </span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:12.0pt;font-family:Arial,
                      sans-serif;">  </span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;">John Rudy</span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;">  </span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;">781-861-0402</span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;">781-718-8334  cell</span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;">13 Hawthorne Lane</span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;">Bedford MA</span></p>
                  <p class="yiv3946723560MsoNormal"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;"><a rel="nofollow noopener noreferrer"
                        ymailto="mailto:jjrudy1@comcast.net"
                        target="_blank"
                        href="mailto:jjrudy1@comcast.net"
                        moz-do-not-send="true"
                        class="moz-txt-link-freetext">jjrudy1@comcast.net</a></span></p>
                  <p class="yiv3946723560MsoNormal"><img
                      style="width:1.2916in;min-height:1.1927in;"
                      id="yiv3946723560Picture_x0020_1"
                      src="cid:w1MTgGmzoW0FsWL1wJ4P" yahoo_partid="1.4"
                      moz-do-not-send="true" width="124" height="115"
                      border="0"><span
                      style="font-size:10.0pt;font-family:Arial,
                      sans-serif;"></span></p>
                  <p class="yiv3946723560MsoNormal">  </p>
                </div>
              </div>
            </div>
            ===============================================<br>
            ::The Lexington Computer and Technology Group Mailing List::<br>
            Reply goes to sender only; Reply All to send to list.<br>
            Send to the list: <a ymailto="mailto:LCTG@lists.toku.us"
              href="mailto:LCTG@lists.toku.us" moz-do-not-send="true"
              class="moz-txt-link-freetext">LCTG@lists.toku.us</a>     
            Message archives: <a
              href="http://lists.toku.us/pipermail/lctg-toku.us/"
              target="_blank" moz-do-not-send="true"
              class="moz-txt-link-freetext">http://lists.toku.us/pipermail/lctg-toku.us/</a><br>
            To subscribe: email <a
              ymailto="mailto:lctg-subscribe@toku.us"
              href="mailto:lctg-subscribe@toku.us"
              moz-do-not-send="true" class="moz-txt-link-freetext">lctg-subscribe@toku.us</a> 
            To unsubscribe: email <a
              ymailto="mailto:lctg-unsubscribe@toku.us"
              href="mailto:lctg-unsubscribe@toku.us"
              moz-do-not-send="true" class="moz-txt-link-freetext">lctg-unsubscribe@toku.us</a><br>
            Future and Past meeting information: <a
              href="http://LCTG.toku.us" target="_blank"
              moz-do-not-send="true">http://LCTG.toku.us</a><br>
            List information: <a
              href="http://lists.toku.us/listinfo.cgi/lctg-toku.us"
              target="_blank" moz-do-not-send="true"
              class="moz-txt-link-freetext">http://lists.toku.us/listinfo.cgi/lctg-toku.us</a><br>
            This message was sent to <a
              ymailto="mailto:bobprimak@yahoo.com."
              href="mailto:bobprimak@yahoo.com." moz-do-not-send="true"
              class="moz-txt-link-freetext">bobprimak@yahoo.com.</a><br>
            Set your list options: <a
              href="http://lists.toku.us/options.cgi/lctg-toku.us/bobprimak@yahoo.com"
              target="_blank" moz-do-not-send="true"
              class="moz-txt-link-freetext">http://lists.toku.us/options.cgi/lctg-toku.us/bobprimak@yahoo.com</a><br>
          </div>
        </div>
      </div>
      <br>
      <fieldset class="moz-mime-attachment-header"></fieldset>
      <pre class="moz-quote-pre" wrap="">===============================================
::The Lexington Computer and Technology Group Mailing List::
Reply goes to sender only; Reply All to send to list.
Send to the list: <a class="moz-txt-link-abbreviated" href="mailto:LCTG@lists.toku.us">LCTG@lists.toku.us</a>      Message archives: <a class="moz-txt-link-freetext" href="http://lists.toku.us/pipermail/lctg-toku.us/">http://lists.toku.us/pipermail/lctg-toku.us/</a>
To subscribe: email <a class="moz-txt-link-abbreviated" href="mailto:lctg-subscribe@toku.us">lctg-subscribe@toku.us</a>  To unsubscribe: email <a class="moz-txt-link-abbreviated" href="mailto:lctg-unsubscribe@toku.us">lctg-unsubscribe@toku.us</a>
Future and Past meeting information: <a class="moz-txt-link-freetext" href="http://LCTG.toku.us">http://LCTG.toku.us</a>
List information: <a class="moz-txt-link-freetext" href="http://lists.toku.us/listinfo.cgi/lctg-toku.us">http://lists.toku.us/listinfo.cgi/lctg-toku.us</a>
This message was sent to <a class="moz-txt-link-abbreviated" href="mailto:dking65@kingconsulting.us">dking65@kingconsulting.us</a>.
Set your list options: <a class="moz-txt-link-freetext" href="http://lists.toku.us/options.cgi/lctg-toku.us/dking65@kingconsulting.us">http://lists.toku.us/options.cgi/lctg-toku.us/dking65@kingconsulting.us</a>
</pre>
    </blockquote>
    <div class="moz-signature">-- <br>
      Drew King</div>
  </body>
  <lt-container></lt-container>
</html>