<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Harold Campbell on Medium]]></title>
        <description><![CDATA[Stories by Harold Campbell on Medium]]></description>
        <link>https://medium.com/@haroldcampbell?source=rss-d6f4e212dd5------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*nWUF6OJ58b_vYygo.jpg</url>
            <title>Stories by Harold Campbell on Medium</title>
            <link>https://medium.com/@haroldcampbell?source=rss-d6f4e212dd5------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 10:09:12 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@haroldcampbell/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[When a Package Never Arrives: A Case Study in Modern Delivery Failure]]></title>
            <link>https://medium.com/@haroldcampbell/when-a-package-never-arrives-a-case-study-in-modern-delivery-failure-9edb2e02785a?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/9edb2e02785a</guid>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Fri, 26 Dec 2025 21:09:09 GMT</pubDate>
            <atom:updated>2025-12-30T11:33:06.015Z</atom:updated>
            <content:encoded><![CDATA[<p>In early November, I ordered a portable generator from Amazon, intended to be shipped to a freight forwarding company in Miami and then onward to Jamaica. We were still recovering from Hurricane Melissa and didn’t have reliable power supply due the downed power lines.</p><p>Our home didn’t have Solar, so we opted to import a generator.</p><p>Typically, this should be an open and done. What happened however, was not a story about custom delays, weather impact, or logistical backlogs. It is a story about something simpler and more troubling: a “delivery” that never to have happened at all, and the humans stuck in the endless canned, process failure.</p><h3>The sequence of events</h3><p>The generator was ordered on November 1, 2025, and shipped to a freight forwarding company I decided to use for this shipment. This facility routinely handle packages daily and maintains a security cameras at their loading bays. Friends of mine have used this forwarder many times without incident.</p><p>According to Amazon’s tracking system, the package was marked “delivered” on November 13.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qdNvb5KtKASnySyZS6L3nw.png" /><figcaption>Delivery message from Amazon</figcaption></figure><p>But the generator never arrived.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1ldqn4btIMLgrgL-v9BLYw.png" /><figcaption>Generator in the back of the Amazon truck. It remained there when they departed.</figcaption></figure><p>At first, this seemed like a typical shipping error. I contacted the freight forwarder, who began an internal search of their warehouse. When days passed without results, I insisted that I be put contact with their owner. After further conversations, I was able to have them reviewed their CCTV footage and share a copy with me (via my local logistics company).</p><p>What we found raised serious questions.</p><h3>What the Video Shows</h3><p>The footage shows an Amazon delivery truck pulling into the facility. The rear door opens. A worker climbs inside the truck. Moments later, the worker exits and closes the door. The truck then leaves the premises.</p><p>At no point does the package appear on the ground, on a pallet, or in the custody of the freight forwarder.</p><p>The box with the generator never leaves the truck.</p><p>This is not a matter of theft after delivery, nor a case of a misplaced item inside a warehouse. Based on the video, the delivery simply did not occur.</p><p>Video evidence (timestamped):<br><a href="http://www.youtube.com/watch?v=jKaGTsp9B-k&amp;t=222s">www.youtube.com/watch?v=jKaGTsp9B-k&amp;t=222s</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*i6gmYTY0hQb0USafJJOEJg.png" /><figcaption>Screenshot from CC-TV footage before Amazon delivery driver departs with generator still in the truck.</figcaption></figure><h3>What does this mean?</h3><blockquote>…Amazon is not responsible for any damage, defect, material difference, or loss that occurs to items <strong>after the items are delivered</strong> to you or your freight forwarder. — About Freight Forwarders and Hand Carry Policy (<a href="https://www.amazon.com/gp/help/customer/display.html?nodeId=G201910550">amzn</a>)</blockquote><p>The CCTV footage from the freight forwarder’s premises shows the Amazon delivery vehicle. The video clearly demonstrates that:</p><ol><li>The package remained inside the Amazon truck</li><li>The package was never removed from the vehicle</li><li>The drivers departed with the package still in their possession</li></ol><p>Because the package never exited the delivery vehicle, no delivery occurred and custody was never transferred. As such, this case does not fall under freight forwarder risk or post-delivery loss. It is a failed delivery.</p><h3>Conflicting Conclusions</h3><p>When I contacted Amazon with this information, I was initially told that the matter would be reviewed. A few days later, I was informed that a refund would be issued.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uw7HLu6iCUF89AsCvbvscQ.png" /><figcaption>Email from Amazon about promised refund.</figcaption></figure><p>Shortly afterward, that decision was reversed.</p><p>In a subsequent email, Amazon stated that the item had been “delivered to the freight forwarder intact,” and therefore did not qualify for a refund. No explanation was provided for how this conclusion aligned with the video evidence showing that the package never left the delivery vehicle.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4inZCLBC79cODFXQnFXC_w.png" /><figcaption>Refund reversal by the Amazon OFM team.</figcaption></figure><p>The explanation appeared to rely entirely on internal shipping records rather than the physical reality captured on camera.</p><h3>The Larger Issue</h3><p>This case highlights a structural problem that extends beyond a single lost package.</p><p>Modern logistics systems are highly automated. When a system marks a package as “delivered,” that status can override contradictory evidence — even video evidence. Once that happens, customers can find themselves trapped in a loop where policy enforcement replaces investigation.</p><p>For customers who use freight forwarders — a common practice for international shipping — this can be particularly problematic. Responsibility becomes diffuse, and accountability difficult to establish, even when facts are clear.</p><h3>Attempts to Resolve the Issue</h3><p>I have contacted customer support multiple times, escalated the issue, and provided documentation. At one point, a customer service manager acknowledged the problem and indicated that a refund would be issued (see above). That decision was later reversed citing the freight forwarders.</p><p>I have since shared the video evidence and documentation with Amazon’s executive customer relations (ecr-replies@amazon.com) team in the hope that a human review will resolve what automated systems have not. I’ve also contact Amazon via X (<a href="http://twitter.com/AmazonHelp">@AmazonHelp</a> and <a href="http://twitter.com/AmazonHelpDesks">@AmazonHelpDesks</a>). This is after having previously tried calling the toll-free number and trying the chat bot.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QCYAUkDQAoVPxsuTbYo4CA.png" /><figcaption>Trying to get help via X</figcaption></figure><h3>Why This Matters</h3><p>My story is not just about a missing package. It is about trust in systems that increasingly govern everyday transactions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Jdc444res4BFOOX6e0VZFQ.png" /><figcaption>One of many canned responses from Amazon. This one after trying to get it escalated.</figcaption></figure><p>When evidence can be overridden by automation, and when customers have no clear path to challenge that outcome, even well-established companies risk eroding confidence.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OyH2Yc1HNrmlWKs54VcoPg.png" /></figure><p>Transparency, accountability, and the ability to correct errors are not luxuries. They are essential to maintaining trust in modern commerce.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9T0lluJsD_EoSim5Ih6-lQ.png" /><figcaption>Another response, but still no progress.</figcaption></figure><p>The question is not whether packages are sometimes lost. They are. The question is whether clear evidence of what happened still matters when it conflicts with a system’s default conclusion.</p><p>I’ve have lost several hours trying to get a resolution since discovering that the generator was taken by the Amazon driver. My family has gone weeks without having reliable power.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pNsmIozJIbel87SYhUDO4w.png" /><figcaption>Several canned messages. I’m still trying to get Amazon to review the footage.</figcaption></figure><p>For now, I am still trying to get a refund.</p><h3>Update — December 29, 2025</h3><p>On December 27, I was contacted by Amazon’s Executive Customer Relations team, who reviewed the case and approved a refund. The funds were credited to my account on December 29.</p><p>The resolution came only after escalation and the submission of CCTV footage showing that the package never left the delivery vehicle. Without that independent video evidence, it is unlikely this outcome would have been reached.</p><p>While I appreciate the eventual resolution, the experience highlights broader issues:</p><ol><li>When delivery disputes involve freight forwarders or other non-standard delivery paths, Amazon will opt to blame the third-party.</li><li>The current processes appear ill-equipped to evaluate edge cases where system records conflict with physical evidence. This is not a customer-centric approach.</li></ol><p>If Amazon’s goal is to be “Earth’s most customer-centric company,” then:</p><ul><li>It must make it easier for their teams to look beyond automated assumptions when real-world evidence tells a different story.</li><li>It must provide customers with meaningful escalation paths.</li><li><em>All customers must be treated well</em>.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9edb2e02785a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Writing better Elm code with simple List transformations]]></title>
            <link>https://medium.com/@haroldcampbell/writing-better-elm-code-with-simple-list-transformations-6993c78e4867?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/6993c78e4867</guid>
            <category><![CDATA[elm]]></category>
            <category><![CDATA[functional-programming]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Tue, 17 May 2022 15:36:48 GMT</pubDate>
            <atom:updated>2022-05-17T15:36:48.244Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>A practical walkthrough of using List.map in Elm</blockquote><p>An Elm application may have view functions that need to traverse nested records.</p><p>For instance (in Listing 1), we see a Team record that consists of a list of Board records, consisting of a list of Stage records, that also consists of a list of Card records.</p><pre>-- Listing 1</pre><pre>type alias Team =<br>    { id : String<br>    , name : String<br>    , boards : List Board<br>    }</pre><pre>type alias Board =<br>    { id : String<br>    , name : String<br>    , stages : List Stage<br>    }</pre><pre>type alias Stage =<br>    { id : String<br>    , boardID : String<br>    , name : String<br>    , cards : List Card<br>    }</pre><pre>type alias Card =<br>    { id : String<br>    , teamID : String<br>    , boardID : String<br>    , stageID : String<br>    , desc : String<br>    }</pre><p><em>View on github (</em><a href="https://github.com/haroldcampbell/elm-article-2/blob/main/src/Ex1.elm#L10-L38"><em>Ex1.elm L10 — L38</em></a><em>)</em></p><p>Our intent is to use these records to generate a view similar to figure 1, below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xOn0kR43NbY-Ibol.png" /><figcaption>Figure 1</figcaption></figure><p>The trivial approach is to simply iterate through each nested record (see Listing 2).</p><pre>-- List 2<br>-- ... style omitted for brevity</pre><pre>view : Model -&gt; Html Msg<br><strong>view</strong> (Model team) =<br>  div<br>    [...]<br>    [ h3<br>      [...]<br>      [ text (&quot;Team Name: &quot; ++ team.name) ]<br>    , viewBoardAt 1 team<br>    ]</pre><pre>viewBoardAt : Int -&gt; Team -&gt; Html Msg<br><strong>viewBoardAt</strong> boardIndex team =<br>  div<br>    [...]<br>    (team.boards<br>      |&gt; List.indexedMap<br>        (\index board -&gt;<br>          if index == boardIndex then<br>            div<br>              []<br>              [ h4 [] [ text &lt;| board.name ++ &quot; Board&quot; ]<br>              , div<br>                [...]<br>                (board.stages<br>                  |&gt; List.map<br>                    (\stage -&gt;<br>                      div<br>                        [...]<br>                        [ div<br>                          [...]<br>                          [ text stage.name ]<br>                        , div<br>                          [ onClick &lt;| NewCard stage<br>                          , ...]<br>                          [ text &quot;Add Card&quot; ]<br>                        , div<br>                          [...]<br>                          (List.map<br>                            (\card -&gt;<br>                              div<br>                                [...]<br>                                [ span<br>                                  [...]<br>                                  [ text &lt;| card.id ]<br>                                , span [] [ text card.desc ]<br>                                ]<br>                            )<br>                            stage.cards<br>                          )<br>                        ]<br>                    )<br>                )<br>              ]</pre><pre>          else<br>            text &quot;&quot;<br>        )<br>    )</pre><p><em>View the full code list on github (</em><a href="https://github.com/haroldcampbell/elm-article-2/blob/main/src/Ex1.elm"><em>Ex1.elm</em></a><em>).</em></p><p>While the approach works, it results in code that is difficult to easily understand.</p><p>This makes it harder to debug and maintain the function as the records become more complex. Additionally, testing is also difficult with this large block of code due to the hidden dependencies.</p><h3>Make it obvious</h3><p>The first step to creating more readable and more maintainable code is to make the intent obvious.</p><p>For instance we could break the viewBoardAt function into smaller separate functions (see Listing 3).</p><pre>-- Listing 3</pre><pre>viewBoardAt : Int -&gt; Team -&gt; Html Msg<br><strong>viewBoardAt</strong> boardIndex team =<br>  div<br>    [...]<br>    (updateIfIndexWithDefault<br>      (\index -&gt; index == boardIndex)<br>      (\board -&gt;<br>        viewBoard board<br>      )<br>      (\_ -&gt; text &quot;&quot;)<br>      team.boards<br>    )<br></pre><pre>viewBoard : Board -&gt; Html Msg<br><strong>viewBoard</strong> board =<br>  div<br>    []<br>    [ h4 [] [ text &lt;| board.name ++ &quot; Board&quot; ]<br>    , div<br>      [...]<br>      -- hardcoded dependency between viewBoard &amp; viewStage<br>      (List.map (\s -&gt; viewStage s) board.stages)<br>    ]<br></pre><pre>viewStage : Stage -&gt; Html Msg<br><strong>viewStage</strong> stage =<br>  div<br>    [...]<br>    [ div<br>      [ style &quot;font-weight&quot; &quot;bold&quot;<br>      ]<br>      [ text stage.name ]<br>    , div<br>      [ onClick &lt;| NewCard stage<br>      , ...]<br>      [ text &quot;Add Card&quot; ]<br>    , div<br>      [...]<br>      -- hardcoded dependency between viewStage &amp; viewCard<br>      (List.map (\card -&gt; viewCard card) stage.cards)<br>    ]<br></pre><pre>viewCard : Card -&gt; Html Msg<br><strong>viewCard</strong> card =<br>  div<br>    [...]<br>    [ span<br>      [...]<br>      [ text &lt;| card.id ]<br>    , span [] [ text card.desc ]<br>    ]</pre><p><em>View the full code list on github (</em><a href="https://github.com/haroldcampbell/elm-article-2/blob/main/src/Ex2.elm"><em>Ex2.elm</em></a><em>).</em></p><p>We can now see that we have 4 logical viewing operations:</p><ul><li><strong>viewBoardAt</strong> - Get a specific board for viewing</li><li><strong>viewBoard</strong> - Generate the view for a board and its stages</li><li><strong>viewStage</strong> - Generate the view for a Stage in the board and its cards</li><li><strong>viewCard</strong> - Generate html for Card only.</li></ul><p>This is definitely an improvement. However, we maintained a hard dependency by using the List.maps in viewBoard and viewStage.</p><p>Ideally, what I feel is better is if the viewBoard and viewStage accept html for their nested children and then they just render that Html.</p><p>This results in code that is easier to test and read.</p><h3>Make it easy to test</h3><p>In listing 4, the code has been modified to code to accept a list html for the children nodes. This breaks the dependencies since the parent functions (viewBoard and viewStage) don’t need to know how their child Html elements (stagesHtml and cardsHtml) are generated.</p><pre>-- Listing 4</pre><pre>viewBoard : Board -&gt; List (Html Msg) -&gt; Html Msg<br><strong>viewBoard</strong> board stagesHml =<br>  div<br>    []<br>    [ h4 [] [ text &lt;| board.name ++ &quot; Board&quot; ]<br>    , div<br>      [...]<br>      stagesHml<br>    ]<br></pre><pre>viewStage : Stage -&gt; List (Html Msg) -&gt; Html Msg<br><strong>viewStage</strong> stage cardsHtml =<br>  div<br>    [...]<br>    [ div<br>      [...]<br>      [ text stage.name ]<br>    , div<br>      [ onClick &lt;| NewCard stage<br>      , ...]<br>      [ text &quot;Add Card&quot; ]<br>    , div<br>      [...]<br>      cardsHtml<br>    ]</pre><p><em>View the full code list on github (</em><a href="https://github.com/haroldcampbell/elm-article-2/blob/main/src/Ex3.elm"><em>Ex3.elm</em></a><em>).</em></p><p>This would be a good place to stop… but, could we make it more expressive?</p><h3>Make it even more expressive</h3><blockquote><em>Note : </em><strong><em>The code in Listing 4 (Ex3.elm above) is Good Enough!</em></strong><em> Everything below is just for the curious.</em></blockquote><p>The viewBoardAt function could do with a sprinkling of some love.</p><p>The List.maps are not expressive enough and we can easily fix this, by trading brevity for succinctness.</p><pre>-- Listing 5</pre><pre>-- From (Ex3.elm)<br>viewBoardAt : Int -&gt; Team -&gt; List (Html Msg)<br><strong>viewBoardAt</strong> boardIndex team =<br>  updateIfIndexWithDefault<br>    (\index -&gt; index == boardIndex)<br>    (\board -&gt;<br>      viewBoard board &lt;|<br>        List.map<br>          (\s -&gt;<br>            viewStage s &lt;|<br>              List.map viewCard s.cards<br>          )<br>          board.stages<br>    )<br>    (\_ -&gt; text &quot;&quot;)<br>    team.boards<br></pre><pre>-- To (Ex4.elm)<br>viewBoardAt : Int -&gt; Team -&gt; Html Msg<br><strong>viewBoardAt</strong> boardIndex team =<br>  div<br>    [...]<br>    (updateIfIndexWithDefault<br>      (\index -&gt; index == boardIndex)<br>      (\board -&gt;<br>        viewBoard board<br>          (composeHtml board.stages<br>            (\s -&gt;<br>              viewStage s (composeHtml s.cards viewCard)<br>            )<br>          )<br>      )<br>      (\_ -&gt; text &quot;&quot;)<br>      team.boards<br>    )</pre><pre>composeHtml : List a -&gt; (a -&gt; Html Msg) -&gt; List (Html Msg)<br><strong>composeHtml</strong> list fnc =<br>  List.map fnc list</pre><p><em>View the full code list on github (</em><a href="https://github.com/haroldcampbell/elm-article-2/blob/main/src/Ex4.elm"><em>Ex4.elm</em></a><em>).</em></p><p>The addition of composeHtml makes it clearer what we are trying to do in the viewBoardAt function.</p><p><strong>Too much, maybe?</strong></p><p>Going further, we could include a full set of smaller helper functions (see Listing 6).</p><pre>-- Listing 6</pre><pre>viewBoardAt : Int -&gt; Team -&gt; Html Msg<br><strong>viewBoardAt</strong> boardIndex team =<br>  div<br>    [ style &quot;padding&quot; &quot;20px&quot;<br>    ]<br>    (team.boards<br>      |&gt; takeOnlyIndex boardIndex<br>      |&gt; bHtml<br>    )<br></pre><pre>cInnerHtml : Card -&gt; Html Msg<br><strong>cInnerHtml</strong> item =<br>  viewCard item<br></pre><pre>sInnerHtml : Stage -&gt; Html Msg<br><strong>sInnerHtml</strong> item =<br>  viewStage item (cHtml item.cards)<br></pre><pre>bInnerHtml : Board -&gt; Html Msg<br><strong>bInnerHtml</strong> item =<br>  viewBoard item (sHtml item.stages)<br></pre><pre>cHtml : List Card -&gt; List (Html Msg)<br><strong>cHtml</strong> cards =<br>  List.map viewCard cards<br></pre><pre>sHtml : List Stage -&gt; List (Html Msg)<br><strong>sHtml</strong> stages =<br>  List.map sInnerHtml stages<br></pre><pre>bHtml : List Board -&gt; List (Html Msg)<br><strong>bHtml</strong> boards =<br>  List.map bInnerHtml boards<br></pre><pre><strong>takeOnlyIndex</strong> boardIndex list =<br>  list<br>    |&gt; List.indexedMap (\index item -&gt; ( index, item ))<br>    |&gt; List.filter (\( index, _ ) -&gt; index == boardIndex)<br>    |&gt; List.map (\( _, item ) -&gt; item)</pre><p><em>View the full code list on github (</em><a href="https://github.com/haroldcampbell/elm-article-2/blob/main/src/Ex5.elm"><em>Ex5.elm</em></a><em>).</em></p><p>Now the code is easier to read and reason about. However, we have unfortunately have introduced dependencies in the helper functions that defeat our earlier gains. At this point, it’s just a perspective on taste.</p><p>Whatever your preference, writing good code requires deliberate thought.</p><p>Initially posed on <a href="https://justaddflow.substack.com/p/writing-better-elm-code-with-simple">https://justaddflow.substack.com/p/writing-better-elm-code-with-simple</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6993c78e4867" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using Elm pipelines with andThen to decode a multi-object type]]></title>
            <link>https://medium.com/@haroldcampbell/using-elm-pipelines-with-andthen-to-decode-a-multi-object-type-35896d2a9849?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/35896d2a9849</guid>
            <category><![CDATA[elm-lang]]></category>
            <category><![CDATA[elm]]></category>
            <category><![CDATA[functional-programming]]></category>
            <category><![CDATA[typescript]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Mon, 09 May 2022 02:04:25 GMT</pubDate>
            <atom:updated>2022-05-09T05:58:21.689Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>Decode the remote api responses into response types that have two parts: RequestStatus and response-specific type.</blockquote><p>This post share how I use pipelines and andThen to Decode a multi-object type.</p><p>My payload from the remote service has the following form:</p><pre>{</pre><pre>    action: &quot;UpdateCardRequest&quot;<br>    errorCode: 0<br>    message: &quot;ok&quot;<br>    successStatus: true</pre><pre>    jsonBody: {ID: &quot;M949U4J8&quot;, TeamID: &quot;J888Y3B1&quot;, BoardID: &quot;A954N8A2&quot;, StageID: &quot;U345M9U4&quot;,…}</pre><pre>}</pre><p>This response consists of two logically distinct pieces of information, a response status (<em>action</em>, <em>errorCode</em>, <em>message</em> and <em>successStatus</em>) and the context specific payload (<em>jsonBody</em>).</p><p>Because of this I want to decode the remote api response into Response types that have two parts RequestStatus and response-specific type.</p><pre>type LoadingTeamResponse<br>    = FetchTeamResponse RequestStatus Team<br></pre><pre>type LoadedTeamResponse<br>    = CreateBoardStageResponse RequestStatus Stage<br>    | CreateStageCardResponse RequestStatus Card<br>    | UpdateStageCardResponse RequestStatus Card<br></pre><pre>type alias RequestStatus =<br>    { action : String<br>    , errorCode : Int<br>    , successStatus : Bool<br>    , message : String<br>    }</pre><p>My initial approach was to use map4 from the Decoder module, which works but feels clumsy.</p><pre>postFormResponseDecoder : Decoder ApiResponse<br>postFormResponseDecoder =<br>    D.map4 (\a b c d -&gt; PostFormRedirect (RequestStatus a b c) d)<br>        (D.field &quot;action&quot; D.string)<br>        (D.field &quot;errorCode&quot; D.int)<br>        (D.field &quot;successStatus&quot; D.bool)<br>        (D.maybe (D.field &quot;jsonBody&quot; Team.teamDecoder))</pre><p>Instead, I used a helper function that allowed for a more concise api.</p><pre>onApiResponse : Decoder a -&gt; (RequestStatus -&gt; a -&gt; b) -&gt; Decoder b<br>onApiResponse fieldDecoder responseMessage =<br>    (D.succeed ApiHTTPResponse<br>        |&gt; required &quot;action&quot; D.string<br>        |&gt; required &quot;errorCode&quot; D.int<br>        |&gt; required &quot;successStatus&quot; D.bool<br>        |&gt; required &quot;message&quot; D.string<br>        |&gt; required &quot;jsonBody&quot; fieldDecoder</pre><pre>    )</pre><pre>        |&gt; D.andThen<br>            (\a -&gt;<br>                D.succeed (responseMessage (RequestStatus a.action a.errorCode a.successStatus a.message) a.jsonBody)<br>            )</pre><p>I can now write my decoders using the onApiResponse. The pattern feels more concise.</p><pre>fetchTeamResponseDecoder : Decoder LoadingTeamResponse<br>fetchTeamResponseDecoder =<br>    onApiResponse teamDecoder FetchTeamResponse<br>apiGetTeam : String -&gt; Cmd Msg<br>apiGetTeam teamID =<br>    Http.get<br>        { url = getRouteURL &quot;getTeam&quot; ++ teamID<br>        , expect = Http.expectJson LoadingApiResponse fetchTeamResponseDecoder<br>        }</pre><p>Initially posed on <a href="https://justaddflow.substack.com/p/using-elm-pipelines-with-andthen">https://justaddflow.substack.com/p/using-elm-pipelines-with-andthen</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=35896d2a9849" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Product Strategy: Sometimes it hurts]]></title>
            <link>https://medium.com/@haroldcampbell/product-strategy-sometimes-it-hurts-cc749cada0bb?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/cc749cada0bb</guid>
            <category><![CDATA[product-manager]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[product-strategy]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Sun, 24 Nov 2019 16:07:25 GMT</pubDate>
            <atom:updated>2019-11-25T13:36:56.987Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rKfX5Hi20MWy-qcHcnuLVA.png" /><figcaption>People Everywhere</figcaption></figure><p>Generally, when there are discussions about product or product strategy, it is often overlooked that regardless of the product that is being delivered, the product builders and ultimately the consumers, are all people.</p><blockquote>Putting it differently, <em>Product Strategy is all about people!</em></blockquote><p>The result of building products that challenge the status quo, will often create conflict at various levels within the organisation and in some instances, the market. This may not be a bad thing depending on how that conflict is managed.</p><p>The question remains though, <strong>why does good product strategy often result in conflict, creative or otherwise</strong>?</p><h3>Conflict: People and their beliefs</h3><p>Humans are social creatures who derive some elements of identity from their social network. Products strategies sometimes threaten this as it indirectly threatens our identity of what’s right or wrong.</p><p>Innovation often requires doing things that may be in direct conflict with people’s beliefs and expectations, moving them beyond their comfort zone.</p><p>The more disruptive the strategy, the more likely it is that we will be impacting people and the more likely it is that they will push back.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*eqjSND7c-fIg4RmaYBK-RA.png" /><figcaption>google glasses</figcaption></figure><p>An example of this was google’s “failed” wearable glasses. People felt that the new technology was invasive, pretentious and unsettling. The device quick gained wearers the term “Glasswholes” — <br>[1] <a href="https://nypost.com/2014/07/14/is-google-glass-cool-or-just-plain-creepy/">https://nypost.com/2014/07/14/is-google-glass-cool-or-just-plain-creepy/</a><br>[2] <a href="https://www.extremetech.com/extreme/183794-are-google-glass-users-glassholes-or-merely-the-misunderstood-vanguard-of-technology">https://www.extremetech.com/extreme/183794-are-google-glass-users-glassholes-or-merely-the-misunderstood-vanguard-of-technology</a><br>[3] <a href="https://techcrunch.com/2013/01/28/glassholes/">https://techcrunch.com/2013/01/28/glassholes/</a><br>[4] <a href="http://valleywag.gawker.com/glassholes-use-google-to-destroy-a-restaurants-rating-1580779355">http://valleywag.gawker.com/glassholes-use-google-to-destroy-a-restaurants-rating-1580779355</a></p><p>Remember, product strategy is all about people.</p><h3>Conflict: Organisation and its culture</h3><p>There are three levels of organisational culture. Each level may be impacted differently based on the amount of disruption/innovation that is being introduced by our product strategies.</p><blockquote>Good product strategies often change the visible parts of the organisation.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/952/1*G6BtybFCmE9mzQ9EJERqkw.png" /><figcaption>Three Levels of Culture by Edgard Schein</figcaption></figure><p>It may change the language that the organisation uses to describe itself; sometimes it changes the long held (sometimes loved) rituals, physical structures. In extreme cases, it may even change the identity of the organisation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gwzDNXAhYPlUkCYG.jpg" /><figcaption>Airbnb Rebrand. <a href="https://design.studio/work/airbnb">https://design.studio/work/airbnb</a></figcaption></figure><p><strong>Examples</strong> of this include <a href="https://design.studio/work/airbnb">Airbnb</a>, <a href="https://digiday.com/marketing/inside-pizza-huts-biggest-ever-rebrand/">Pizza Hut</a>, <a href="https://www.wired.com/2013/02/a-new-front-page-for-yahoo-suggests-a-new-strategy/">Yahoo</a>, <a href="https://www.dailymail.co.uk/sciencetech/article-7686703/Instagram-goes-global-controversial-test-hides-like-counts.html">Instagram</a></p><blockquote>Great product strategies force companies to challenge the deeper and often invisible parts of the organisation.</blockquote><p>Great product strategies take it a step further. These strategies change the shared values, and (deeper still) the shared assumptions of the company.</p><p>Collectively these organisational changes may include <em>the conscious and unconscious truths, thoughts, beliefs, mental models and related attitudes (good vs bad, right vs wrong) that are often taken for granted as truisms</em>.</p><p>Netflix and Slack are examples of companies that both created product strategies that required huge shifts to their core operating models. NetFlix pivoted from renting movies to streaming movies. Slack pivoted from being a game company to a team messaging platform. <strong><em>These weren’t comfortable changes.</em></strong></p><p>Apple and Snapchat created product strategies that challenged the core market assumptions in which they operated.</p><p>Apple’s initial foray into music required a product strategy that challenged the assumption that you could only make money selling an entire album (the prevailing truth at the time) vs. allowing a customer to buy a single song (the crazy idea).</p><p>Snapchat showed that customers would be okay with photo apps that deleted photos (the crazy idea) vs. needing always store the photos (the prevailing truth at the time).</p><h3>Change Agents: Great Product Managers</h3><p>In challenging the norms, <strong><em>product managers often venture into spaces that make people uncomfortable</em></strong>. They will knowingly effect changes that result in people’s job roles becoming redundant, removing or altering processes and workflows, etc.</p><p>Additionally, great product managers have the tenacity to fight for their idea’s survival even when everyone thinks their ideas are crazy, the time isn’t right, the technology isn’t ready, or they have had several failed experiments. <strong><em>This may sometimes feel like an uphill battle.</em></strong></p><blockquote>Tenacity is required to move an idea beyond a cool whiteboard drawing to a tangible product.</blockquote><p><strong>Examples</strong> of this include: Adobe’s shift from paid cloud to the subscription cloud, Nike moving into wearable products, SpaceX’s approach to creating rockets, Coca cola selling bottled water, Toyota investing into electric cars, Google moving into self-driving cars, etc.</p><p>Ultimately, great product managers are the ones (along with their teams) who have the boldness and courage to disrupt their organisations. They understand that the key to great products result in conflict arising from having to saying “No” to really good ideas, cool feature requests that will result in safe, but mediocre outcomes.</p><p>Their super skills includes:</p><ul><li>Having a better understanding of their customers future needs</li><li>Using frequent experiments to learn faster than their competitors about their customers’ unmet needs</li><li>Recognising that actually needing to building products to test assumptions is costly, wasteful and it’s not the answer to innovation.</li></ul><p>Cheers,</p><p>-harold</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cc749cada0bb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Agile in a Nutshell: An Executive Overview]]></title>
            <link>https://medium.com/hackernoon/agile-in-a-nutshell-an-executive-overview-db19f3400c1a?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/db19f3400c1a</guid>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[agile-in-a-nutshell]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Fri, 15 Jun 2018 17:36:29 GMT</pubDate>
            <atom:updated>2019-05-25T07:03:41.976Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VYD21rcS2i3oENT5HBuI6Q.png" /><figcaption>Harold Campbell, Shane Ferley and Brain Leke</figcaption></figure><h3>What’s the current problem?</h3><blockquote>Non-agile delivery methods are costly, slow and often deliver software that requires expensive rework. Additionally, these delivery methods often don’t deliver the expected customer and business value. Bummer.</blockquote><p>Some common problems that affect organisations that use non-agile delivery methods include:</p><ul><li>Expensive project deliverables</li><li>Expensive re-works</li><li>Delayed Projects</li><li>Missed market opportunities</li><li>Being unable to deliver meaningful value to clients</li><li>Lots of “Accidental” Complexity</li><li>Ballooning of costs</li><li>Inherent “Waste” embedded in rigid processes, re-enforced by organisational culture</li><li>Silo-ed teams</li><li>Poor collaboration between teams</li><li>“Code then Fix” approach to delivery</li><li>Difficult to respond “Quickly” to market changes</li><li>Long feedback loops between product delivery and expected customer value</li><li>Phased SDLC</li><li>Little to no predictability (e.g. quality, schedules, risk, costs, value, etc.)</li><li>Strong focus on process and tools, over delivering “actual” business and customer value</li></ul><h3>What is Agile?</h3><p>Hmmm. So what’s this <a href="https://hackernoon.com/tagged/agile">Agile</a> thing then?</p><blockquote>Agile is a group of delivery methods, principles and practices for effectively delivering software that leverages collaboration and customer feedback.</blockquote><p>There are various Agile Delivery Methods, like:</p><ul><li>XP</li><li>Scrum/Kanban/Scrumban</li><li>SAFe/LeSS/DAD</li><li>FDD</li><li>Crystal</li><li>Adaptive <a href="https://hackernoon.com/tagged/software-development">Software Development</a></li><li>Lean Software Development</li></ul><p>At its core however, Agile is about four things really:</p><ol><li><strong>Delivery</strong>: Focus on Income and Learning</li><li><strong>Collaboration</strong>: Focus on Trust and Team collaboration</li><li><strong>Reflection</strong>: Focus on Insights, Improvements (and celebrating success)</li><li><strong>Improvement</strong>: Focus on Experimentation and <em>actual</em> Change (which includes Data-driven decisions and Hypothesis validation)</li></ol><p>Alistair Cockburn (Co-author of the Agile Manifesto) referred to these four principles as <em>The Heart of Agile</em>.</p><p>See: <a href="http://heartofagile.com/expanding-the-diagram/">http://heartofagile.com/expanding-the-diagram/</a></p><p>Additionally, when using Agile Delivery methods, we use iterative processes to deliver small increments of value at short and regular intervals (AKA sprints, or iterations). These iterations are typically either 2 or 4 weeks long.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gWNYjT30lVkIjqJK.png" /><figcaption>Steven Thomas: Iterative Incremental</figcaption></figure><p>See: <a href="http://itsadeliverything.com/revisiting-the-iterative-incremental-mona-lisa">http://itsadeliverything.com/revisiting-the-iterative-incremental-mona-lisa</a></p><p>However, just having short iterations where no business or customer value is being delivered at the end of each iteration really doesn’t help.</p><p>We need both: An iterative process <em>and</em> incremental delivery of value.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1023/1*r8Tf9PiKiLzB3MNBr1A7Hw.png" /><figcaption>Henrik Kniberg: Iterative Incremental</figcaption></figure><p>Also see: <a href="https://www.mountaingoatsoftware.com/blog/agile-needs-to-be-both-iterative-and-incremental">https://www.mountaingoatsoftware.com/blog/agile-needs-to-be-both-iterative-and-incremental</a></p><p><strong>What do the delivery teams look like?</strong></p><p>Short answer: It depends on the “flavour” of Agile that you are using.</p><p>For instance, Kanban is structure agnostic. While Scrum, another popular methodology, has the following structure:</p><ul><li>1 x Product Owner</li><li>1 x Scrum Master</li><li>Development Team (the people that produce the actual product): QAs, Analysts, UX, Developers, etc.</li></ul><p>Going back to that iterative process and sticking with Scrum, their process may look like the figure below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s7ScHKmXvj1waKo5sjy95g.png" /><figcaption>Scrum’s delivery process</figcaption></figure><h3>Why Agile?</h3><blockquote>Staying relevant, innovating and increasing shareholder value are some of key reasons why companies undertake delivery transformation, with Agile. The benefits are similar but more wide-reaching.</blockquote><p>Additional benefits of adopting Agile Delivery methods include:</p><ul><li>Happier and more productive employees</li><li>“Healthier” delivery teams and work environments</li><li>Fast feedback from customers</li><li>Early validation of business and market assumptions</li><li>Increased innovation</li><li>More predictable product delivery costs</li><li>Shorter delivery cycles</li><li>Less risky product delivery</li><li>More predictable cycles/cadences (e.g. releases, inspect and adapt, etc.)</li><li>Continuous focus on improvement</li></ul><h3><strong>Agile Delivery Values</strong></h3><blockquote>There are only four values that guide Agile Delivery methods.</blockquote><p>The four values are:</p><ol><li>A focus on delivering <strong>working software</strong></li><li>A focus on <strong>customer collaboration</strong></li><li>A focus on <strong>individuals and their interactions</strong></li><li>A focus on <strong>responding to change</strong></li></ol><p>See: <a href="http://agilemanifesto.org/">http://agilemanifesto.org</a></p><p>These four values relate directly to the four principles that are at the Heart of Agile (described above).</p><p>Scrum, Kanban, XP, SAFe, <em>insert-shiny-new-agile-thing-here</em>, (etc.) are all just frameworks/methods that build on these four values. Different methodologies (or the various combinations) will be better suited for different types of teams , or organisations — There is NO ONE TRUE WAY!</p><p>Once your delivery methodology is based on the four values above, <em>and</em> there is a deliberate focus on delivery, collaboration, reflection and improvement, <strong><em>you will become agile</em></strong>.</p><h3>What’s required from Executives?</h3><blockquote>A shift in the organization will require a shift in thinking and most likely changes in organizational structures and processes.</blockquote><p>Executives need to lead the transformation. This is done by:</p><ul><li>Changing mindset</li><li>Shifting from Command-and-Control to Mutual Learning</li><li>Being courageous (by being vulnerable)</li><li>Adopting the Agile values and principles</li><li>Focusing on delivering value vs. meeting deadlines and plans</li><li>Being transparent</li><li>Learning from “Short” failures</li><li>Embracing the fact that the adoption of some Agile delivery practices <strong><em>will</em></strong> require structural changes within the organization (e.g DevOps).</li></ul><h3>How do you adopt Agile Methods in your organisation?</h3><p>Again, like with most things Agile, it depends.</p><blockquote>Adoption can be seen as a two-pronged transformation initiative — it will require changes in cultural principles and delivery practices.</blockquote><h4>Cultural Principles</h4><p>Culture is that intangible thing that resists change within your company. It is <em>the</em> thing that embodies the ethos of the organization. As such, it is critical that a systematic, from the inside-out approach, is taken to change cultural norms, assumptions and beliefs. This is where the “radical shift” needs to happen.</p><p>To achieve this, there are a few core principles that should be adopted and promoted within the company. Specifically, these include:</p><ul><li>Servant Leadership</li><li>Coaching — Process, technical, leadership and behavioral</li><li>Continuous learning — Promoted at the organization, team and individual-level; and is budgeted for and factored in during planning</li><li>Frequent reflection — Retrospectives, Sprint/Iteration reviews, etc.</li><li>Individual feedback — Effective/impact feedback, speed-dating feedback, etc.</li><li>Estimations as guides not commitments — shift from date-driven planning to value-driven planning. Engineering effort estimates <em>should be</em> seen only as guides and not “holy” commitments that can’t be changed. This promotes transparency and ensures that any change in facts, assumptions and dependencies, or knowledge and experience (e.g. changes to the team), that will impact delivery are shared as soon as they are known</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jIbSbZlngtPfYNS5MYX2tA@2x.jpeg" /><figcaption>Harold Campbell and Brain Leke: All Estimates are lies!</figcaption></figure><ul><li>Data-driven decisions — PDCA/Build, measure, learn/OODA models</li><li>Incremental delivery of value — see above</li></ul><h4>Delivery Practices</h4><p>At the other end of the spectrum are delivery practices. These changes refer directly to how we deliver software.</p><p>These practices typically include:</p><ul><li>Engineering practices — adopting Test-driven development, Pairing, having a code repository were developers commit code frequently (e.g. daily vs weekly), etc.</li><li>DevOps — Practices that focus on getting fast build feedback, reporting, automated testing, improving code quality, continuous integration and continuous delivery of software</li><li>Co-location — As much as possible, having the development team together or being able to collaborate easily</li><li>Small batches in queues</li><li>Work-In-Progress limits</li><li>Self-organising teams</li><li>Continuous team/self assessments</li><li>Transparency over Cover-Your-Ass</li><li>Iterative software development — see above</li><li>Delivery team provides estimates for engineering effort — The people doing the work make the estimates</li><li>Empiricism — Data-driven decisions using hypothesis validation (e.g. with Plan-Do-Check-Act/Build-Measure-Learn)</li></ul><h3>Next Steps</h3><blockquote>Pick a method and start. It’s doesn’t really matter if you start with Scrum or Kanban or whichever. Simply start.</blockquote><p>What is important however, regardless of your organisation’s maturity, is that you have a process that is focused on consistently improving your delivery, collaboration and reflection processes.</p><p>Special thanks to Shane Ferley and Brain Leke for their contributions and feedback in creating this piece!</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fupscri.be%2Fdde502%3Fas_embed%3Dtrue&amp;dntp=1&amp;url=https%3A%2F%2Fupscri.be%2Fhackernoon%2F&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=upscri" width="800" height="400" frameborder="0" scrolling="no"><a href="https://medium.com/media/3c851dac986ab6dbb2d1aaa91205a8eb/href">https://medium.com/media/3c851dac986ab6dbb2d1aaa91205a8eb/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=db19f3400c1a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/hackernoon/agile-in-a-nutshell-an-executive-overview-db19f3400c1a">Agile in a Nutshell: An Executive Overview</a> was originally published in <a href="https://medium.com/hackernoon">HackerNoon.com</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why experimentation matters in organisational transformation]]></title>
            <link>https://medium.com/@haroldcampbell/experiment-driven-transformation-aed2cfaea046?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/aed2cfaea046</guid>
            <category><![CDATA[digital-transformation]]></category>
            <category><![CDATA[management-and-leadership]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[transformation]]></category>
            <category><![CDATA[management]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Sun, 25 Mar 2018 10:25:32 GMT</pubDate>
            <atom:updated>2018-03-30T10:45:31.796Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*i7znJu3LtGAcJ64VUzIssw.png" /><figcaption>source: phys.org</figcaption></figure><p>I think that true agility comes from building a culture that focuses on using experimentation rather than relying on a big-bang Agile transformation.</p><p>The rhetoric that we often hear is that, “We are going Agile…”, “We have adopted Agile…”, etc. and then fast-forward a few years, “…Agile is a waste of time and money”, “…this Agile thing isn’t suited for us”, etc. It is more often than not, that Agile transformations fail. But what exactly does that mean? In this post, I’ll talk about why Agile transformations fail and then present an alternative perspective for a successful organisational transformation.</p><h4>What is an Agile transformation?</h4><p>I’d like to position an Agile transformation as: adopting and promoting activities that focus on inspecting and adapting; identifying and reducing waste; promoting staff safety; and increasing collaboration across the organisation so as to increase the organisation’s ability to deliver value to its customers faster.</p><p>To be clear, I’m not referring to, or promoting the use of any particular Agile methodology. Instead, I am making the distinction that there are a sequence of activities that an organisation can undertake to increase its ability to become more “agile”. If the organisation is able to replace its old, non-agile operating model and is also able to achieve its desired outcome through the sustained usage of those sequence of activities, then we have a successful Agile Transformation.</p><h4>Why do Agile Transformations fail?</h4><p>So what causes the transformations to fail? Where do the wheels come unhinged?</p><p>Dan North wrote an interesting article titled, “<a href="https://dannorth.net/2018/01/26/in-praise-of-swarming/">In Praise of SWARMING</a>”. In it, he highlighted some of the consequences of treating Agile methodologies (e.g. SAFe, LeSS, Dad, Scrum, Kanban, etc.) as holy, sacred artefacts — never to be challenged or modified.</p><p>Elsewhere I have seen other debates that echo similar sentiments; which method is better, <a href="https://www.google.co.za/search?q=scrum+vs+kanban">Scrum vs. Kanban</a>, <a href="https://www.google.co.za/search?q=safe+vs+less+vs+dad">SAFe vs. LeSS vs. blah blah blah</a>. If we are focusing on transformation, then I think that the above debate regarding which method is better, is the wrong conversation.</p><p>Dan touched on this when he wrote that:</p><ol><li>Having the ability and credibility to influence across the organisation; and,</li><li>Having leadership that is consistent, invested and resilient throughout the transformation process.</li></ol><blockquote>I believe that Agile transformations fail when executive leadership isn’t able to break the entrenched culture.</blockquote><p>The effort required to change an organisation’s culture is often underestimated. It is not as simple as hiring some external consultants, embedding some <a href="https://www.agileconnection.com/article/role-agile-coach">Agile coaches</a>, sending the teams on certification courses, and “presto!”: Agile.</p><p>Culture is the “invisible hand” that actively re-enforces the old ways of thinking (and operating model) within the organisation and it is enforced at the middle-management layer. It is the intangible thing that often prevents the successful, wholesale adoption of continuous learning, new engineering practices and various Agile principles.</p><blockquote>I have found that the breakdown of Agile transformations happen at this middle-management layer.</blockquote><p>At the middle-management level, we find people who typically apply pressure to the teams to meet the objectives (implied or otherwise) that are set by the executive leadership above. Due to their own biases, fears and their unspoken vulnerabilities, middle-managers can derail Agile transformation initiatives without actively realising that they are doing so.</p><p>This leads us to the question, <em>what’s the role of the executive leadership?</em></p><p>The answer? Instead of focusing on the efficacy of Agile transformation, focus on providing structures and initiatives that promote experimentation and organisational learning underpinned by a rabid emphasis on customer-centricity.</p><blockquote>Agile transformation initiatives should be guided by experimentation that increase organisational learnings as the company strives to find better ways to serve its customers.</blockquote><h3>Cultural change through experimentation</h3><p>Before diving-in, it’s worthwhile setting context regarding experimentation. Experimentation requires performing the following steps:</p><ol><li>Formulating a hypothesis,</li><li>Testing that hypothesis by performing some activity,</li><li>Observing the activity and its outcome,</li><li>Evaluating if the hypothesis was validated or invalidated,</li><li>Returning to step 1. where we formulate a new hypothesis or update the hypothesis,</li><li>Repeat.</li></ol><p>Key to a valid experiment, is the limiting of the number of variables that are under observation at any given time. Persons familiar with empiricism will recognise the above steps and it’s practical application as parts of the Observe/Orientate/Decide/Act (OODA), the Plan/Do/Check/Act (PDCA), and finally the Build/Measure/Learn loops</p><h4>Why does it matter?</h4><p>Empiricism is a useful method for individuals, teams and organisation to adopt and apply lean practices in their <strong>inspect and adapt</strong> cycle. It provides us with a proven method to continuously improve both the outcome of the processes, as well as the processes themselves.</p><h4>How does this relate to culture?</h4><p>If executive leadership adopts experimentation upfront as a transformation principle <em>in conjunction with Agile transformation initiatives, </em>it provides both the middle-mangers (and their teams) with a safer environment, where quantified “failure” is accepted. It further provides managers with an avenue to surface their biases, fears and vulnerabilities, without them feeling threatened which makes it less likely that they will derail the transformation initiatives.</p><blockquote>In accepting that people fail, we must also be willing to support them so that we can help them to learn from the failure.</blockquote><p>This subtle shift in operating model sends a powerful message that <strong>there aren’t a few people at the top that know everything — no one does</strong>. It also sets the expectations that no single group is expected to have all the answers. Similarly, a culture that focuses on experimentation can:</p><ul><li><strong>Reduce turf wars and political manoeuvring between middle-managers. </strong>Middle-managers use strategic planning sessions as a tool to predict the future, after which, they move into operational mode where pressure/threats/bonuses are leveraged to motivate the staff to meet the future prediction. As is often the case, on realising that the prediction was wrong, they create a new predictions; repeat. At its core, this is a causal loop, which results in conflict that materializes as turf wars, etc. Experimentation breaks this loop (since we don’t try to predict the future). Instead we focus on learning and using those learnings to build products that delight our customers.</li><li><strong>Stimulate collaboration and reduce organisational silos.</strong> By vigorously promoting experimentation as a tool to realise and improve customer-centricity, middle-managers are more open to asking for, giving, and receiving help from people that are external to their units (to run experiments). They are also less likely to kill ideas/experiments that didn’t originate within their units, as the focus is on serving the customer.</li><li><strong>Leverage the collective intelligence of the entire organisation.</strong> Traditional organisations try to rely on a few “Stars” within the organisation to make all the right calls. In organisations that have a culture of experimentation, multiple, cheap experiments originating all across the organisation result in better, naturally occurring decisions that are rooted in empiricism.</li></ul><h4>How do you contain business risk when doing experiments?</h4><p>Firstly, let’s acknowledge that there is no 100% guaranteed way to prevent failure. By simply operating and being in business we will be exposed to failure — regardless of whether we are performing experiments or not. What’s important is defining structures that can provide us with fast feedback when things go off-track. Empiricism by its very nature helps with this.</p><p>But still, empiricism or not, we have to contain loss, right? Here are five ways for managing risk:</p><ol><li><strong>Risks and value trade-offs</strong>. Create a <a href="http://johnnyholland.org/2012/06/creating-a-shared-understanding/"><em>shared understanding</em></a><em> </em>of the risks of not doing the experiment vs. value that will be realised if experiment is successful. The easiest way to create a <em>shared understanding</em> is to visualise the assumption (flip-charts, whiteboards, etc.) then discuss and challenge the assumptions. More information about creating “shared understanding” can be found in this 2009 <a href="http://www.cognexus.org/Rotman-interview_SharedUnderstanding.pdf">interview [pdf]</a> of Jeff Conklin, the author of <em>Dialogue Mapping: Building Shared Understanding of Wicked Problems</em>.</li><li><strong>Minimum viable experiment (MVE)</strong>. Instead of a minimum viable product, instead identify the <a href="https://medium.com/@megak/the-mve-minimum-viable-experiment-b2498fda9f93">smallest possible experiment</a> that can be done to validate your assumptions and hypothesis. Regarding your MVE, the conversations should include discussions pertaining to: business/customer value, cost, risks, assumptions, issues, dependencies, knowledge and skills (required to execute the experiment).</li><li><strong>Exit criteria</strong>. Beyond having clear goals for performing the experiment, we need to manage scope creep. Having clear exit criteria in the form of <strong>Failure/Success states</strong> ensures that we remain within scope and budget.</li><li><strong>Time-box</strong>. <a href="https://www.agilealliance.org/glossary/timebox/">Time-boxing</a> an experiment sets the upper-bounds for how long the experiment will run. It is used as another method to curtail budget overruns, and potential scope creep.</li><li><strong>Disagree and commit</strong>. An easy way to prevent analysis paralysis and sabotage is simply to adopt the <a href="https://kellblog.com/2017/11/21/handling-nagging-endless-disagreements-with-the-disagree-and-commit-and-new-information-principles/"><em>Disagree and Commit</em></a><em> principle</em>. The principle allows us to cut through conflict by ensuring that people engage in meaningful dialog and then, commit to the success of the experiment.</li></ol><h3>Bringing Experiment-Driven Transformation to your organisation</h3><p>By this point, a natural question would be, <em>how do I start Experiment-Driven Transformation (EDT) at my organisation?</em></p><p>This short answer is, <em>it depends</em>.</p><p>Most organisational transformations begin at the top (i.e. executive leadership) and then trickle down through the ranks. I think ideally, EDT should be approached in a similar manner. This goes back to having leadership that has the influence and credibility to affect change (especially at the board-level) and can also drive the adoption EDT.</p><p><strong>But what if you are not an executive?</strong></p><p>While it may be a bit more difficult, you can still introduced EDT to your organisation, but it will depend on your <em>sphere of influence</em>. As such, focus on using empiricism to change things that are in your sphere of influence and then cascade <em>that</em> behaviour of empiricism downwards. Further, whenever possible, have a bias towards small experiments as successful experiments tend to be small and have few moving parts.</p><h4>Closing thoughts</h4><p>Transformational change isn’t hard. It only becomes difficult when we force the organisation to change unnaturally. By its very nature, experimentation allows us to learn and grow, and in so doing, welcome change through validated learning.</p><p><strong>Thanks:<br></strong>Special thanks to <a href="https://twitter.com/BrianLekebrian">@BrianLekebrian</a> and <a href="https://twitter.com/KevinTrethewey">@KevinTrethewey</a> for all the help and feedback.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aed2cfaea046" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[@MerlinVsCancer]]></title>
            <link>https://medium.com/@haroldcampbell/merlinvscancer-da01bbaeadb7?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/da01bbaeadb7</guid>
            <category><![CDATA[multiple-myeloma]]></category>
            <category><![CDATA[health]]></category>
            <category><![CDATA[cancer]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Sun, 05 Nov 2017 19:11:03 GMT</pubDate>
            <atom:updated>2017-11-05T19:20:46.088Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ahRMou-_k6TrVFuxFeMPjA.png" /></figure><p>So we’ve started. Its’ a trek to raise awareness about <strong>Multiple Myeloma</strong>, and also a journey to raise funds to save our mom. Her current treatment is not working as it should, and she needs to undergo a more rigid treatment.</p><p>I honestly don’t know where this journey will take us, but, so far we’ve created:<br>* a gofundme page — <a href="http://bit.ly/2zlRipi">http://bit.ly/2zlRipi</a><br>* a twitter page — <a href="http://twitter.com/merlinvscancer">@merlinvscancer</a></p><p>Our mom (Merlin) was diagnosed with Multiple Myeloma on May 2017. We found out about it after she did spinal surgery, a battle to save her mobility. Right now, we are just focusing on getting her the help that she needs. This is in the form of stem cell treatment in Cuba.</p><p>What do we need from you? Please help to fund her treatment and simply share this post to raise awareness.</p><p>Thanks</p><p>-h</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=da01bbaeadb7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Borrow vs. Lend]]></title>
            <link>https://medium.com/@haroldcampbell/borrow-vs-lend-89072f8f161?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/89072f8f161</guid>
            <category><![CDATA[english-language]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[english]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Sat, 28 Oct 2017 07:42:29 GMT</pubDate>
            <atom:updated>2017-10-28T07:42:29.333Z</atom:updated>
            <content:encoded><![CDATA[<blockquote><em>Hey friend. Let’s explore why saying, “Borrow me your pen” is incorrect.</em></blockquote><p>If a word is in the subject, we say it is in the <em>Nominative Case</em>. Similarly, if the word is in the object, we say it is in the <em>Objective Case</em>.</p><p>So this follows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/510/1*B9-kxUJjxyj5O0JjdhWkgg.png" /></figure><p>Nouns keep their form regardless of case: subject or object.</p><p>Pronouns are different. Most pronouns have one form for the subject (the Nominative Case) and another form when used as an object (i.e. the Objective Case). <strong><em>I</em></strong> in the subject becomes <strong><em>me</em></strong> when it goes in the object — we’ll see more shortly.</p><p>Likewise, “he” becomes “him”, “she” becomes “her”, etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/550/1*08kxADqFIE_9Ga8Fuif8QA.png" /></figure><p>Interestingly, “you<em>”</em> remains the same regardless of case or tense.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/849/1*e-RM75uzgiSPLlgI6TN3zw.png" /></figure><h4>Of Giving and Receiving</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/401/1*Z6tNQMfbWeNLCnpkO9x_rw.png" /></figure><p>In the example below, “I” is the recipient of the pen.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/510/1*LhMBEf88uZJo3gYUbSWrTQ.png" /></figure><p>Note that “I” is the subject for an implied “your”, which is the indirect object of the sentences.</p><p>For instance, the first sentence could be rewritten as:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/419/1*gkxh_2XLqVdSgrPiHkXyqg.png" /><figcaption>Direct vs indirect objects</figcaption></figure><p><strong>More fun</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/554/1*pCbNUtwwX_cXV_Wd1MCN1A.png" /></figure><p>In the case above, “me” is now the object (ie. “I” being used in the objective case). “Me” is referring to the person that wants the pen, and as such, they are the receiver of the pen.</p><p>Once again there is an implied party that has the pen: <em>Can you</em> give…Please give me <em>your</em>…</p><p>Based on above, we can finally look at the correct use of “me” and “I” with <strong><em>Lend or Borrow</em></strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*tGw8Fnc0oZgA1JoqZ4lxyA.png" /><figcaption>Lending — I as the giver</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*cvPbBko0R_JX-sVqFqeoLg.png" /><figcaption>Lending — I as the receiver</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*NhwELl9XwC3U8tKziJMjLQ.png" /><figcaption>Borrow — I as the receiver</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/560/1*8vVVu8Ma1bvlxYMWPQE-rA.png" /><figcaption>Borrow — I as the giver, which becomes “me”</figcaption></figure><p>Hopefully this makes distinguishing between <em>lend</em> and <em>borrow</em> a bit easier, and also shows why saying, “<em>Borrow me</em> your pen” is incorrect.</p><p>References: Brighter Grammar (Book 1), P. A. Ogundipe et. al.; English Grammar, Rose and Purkis</p><p>Walk good.</p><p>-h</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=89072f8f161" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Yesterday, the words bubbled upwards from the the dark depths of my being]]></title>
            <link>https://medium.com/to-the-point/yesterday-the-words-bubbled-upwards-from-the-the-dark-depths-of-my-being-35602ae0e4f?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/35602ae0e4f</guid>
            <category><![CDATA[life]]></category>
            <category><![CDATA[poetry]]></category>
            <category><![CDATA[creativity]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Fri, 28 Apr 2017 17:24:31 GMT</pubDate>
            <atom:updated>2017-04-28T17:41:32.752Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y-9ArPjswfmoZIq6t6eIBg.png" /><figcaption>Crisco/Corbis @ <a href="http://www.sciencemag.org/news/2015/07/why-screams-are-so-scary">http://www.sciencemag.org/news/2015/07/why-screams-are-so-scary</a></figcaption></figure><p>They sprang forward with a life of their own. Demanding to be heard.</p><p>Rushing out of my mouth, catapulting forward, upheaving everything in their path. My body convulsed.</p><p>In their wake? A trail of hurt and anguish, broken promises and pain.</p><p>I was shocked and surprised, as was she. Not by the devastation I had wrought.</p><p>No.</p><p>Instead, I was shocked by her kindness. “Perhaps I am for loving a fool like you...”</p><p>How often do we say things that hurt other people that we love? Things that we wish we could instantly take back. How often are we the target of such vitriol?</p><p>Yesterday, she reminded me that we don’t have to respond with hurt for hurt.</p><p>What will you do tomorrow?</p><blockquote>Enjoyed reading this? Hit the ❤️ to show your support.</blockquote><p>Cheers</p><p>-harold</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=35602ae0e4f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/to-the-point/yesterday-the-words-bubbled-upwards-from-the-the-dark-depths-of-my-being-35602ae0e4f">Yesterday, the words bubbled upwards from the the dark depths of my being</a> was originally published in <a href="https://medium.com/to-the-point">To the Point</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Dear Medium,]]></title>
            <link>https://medium.com/@haroldcampbell/dear-medium-f8d4eaaa4e86?source=rss-d6f4e212dd5------2</link>
            <guid isPermaLink="false">https://medium.com/p/f8d4eaaa4e86</guid>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[support]]></category>
            <dc:creator><![CDATA[Harold Campbell]]></dc:creator>
            <pubDate>Thu, 15 Sep 2016 17:47:12 GMT</pubDate>
            <atom:updated>2016-09-15T17:47:12.833Z</atom:updated>
            <content:encoded><![CDATA[<p>Dear <a href="https://medium.com/u/504c7870fdb6">Medium</a>,</p><p>How do I prevent your iOS app from auto-refreshing?</p><p>Is there a button that I can press, or a box I can tick to disable this feature?</p><p>I know you had my best interest at heart when you designed this feature, but honestly, it’s kind of frustrating.</p><p>Thanks for listening, and hopefully, thanks for assisting.</p><p>Kind regards,</p><p>Harold</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f8d4eaaa4e86" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>