<?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 Boris Petkov on Medium]]></title>
        <description><![CDATA[Stories by Boris Petkov on Medium]]></description>
        <link>https://medium.com/@boris-petkov?source=rss-80a1656b4edd------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*7Feso6N-qkHdVAFXj7yWsA.jpeg</url>
            <title>Stories by Boris Petkov on Medium</title>
            <link>https://medium.com/@boris-petkov?source=rss-80a1656b4edd------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 08:44:19 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@boris-petkov/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[Outcomes vs. Impact in Software Development: Understanding the Difference]]></title>
            <link>https://boris-petkov.medium.com/outcomes-vs-impact-in-software-development-understanding-the-difference-088a4bbb7898?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/088a4bbb7898</guid>
            <category><![CDATA[outcomes]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[impact]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Thu, 08 Jan 2026 08:54:35 GMT</pubDate>
            <atom:updated>2026-01-08T08:54:35.971Z</atom:updated>
            <content:encoded><![CDATA[<h4>Why focusing on both can drive meaningful progress</h4><p>In software development, the terms *outcomes* and *impact* are often used interchangeably but they represent distinct and complementary concepts that play critical roles in gauging success. While outcomes provide measurable, immediate feedback, impact reflects the broader, long-term changes brought about by achieving those outcomes. Understanding their differences, as well as how they intersect, is essential for driving meaningful progress and innovation in any field.</p><h3>What Are Outcomes?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ut6XRg0Lls23byoyY9vxsw.png" /></figure><p>Outcomes are the measurable results directly tied to an action, initiative, or project. These are tangible achievements that answer the question, “What did we accomplish?” In the software development context, outcomes are often connected to Key Performance Indicators (KPIs) and provide a window into how well specific objectives were met.</p><p>Examples of outcomes in software development might include:<br>- An increase in active daily users by 20%.<br>- A 10% improvement in code deployment speed.<br>- The successful release of a new mobile app feature.<br>- A reduction in bug resolution time by 15%.</p><p>These results often have a short-term focus and provide immediate feedback that allows teams to iterate and optimize. For instance, a development team launching a new feature might measure its adoption rate or user satisfaction within the first weeks after release. If the data suggests areas for improvement, the team can quickly implement changes, refining their work to better meet user needs.</p><p>Outcomes are invaluable because they create a sense of progress and accomplishment. They serve as benchmarks, helping teams stay aligned with project goals and giving stakeholders evidence of tangible progress. However, while outcomes are critical, they are not the whole story.</p><h4>Strengths and Limitations of Focusing on Outcomes</h4><p>The primary strength of focusing on outcomes lies in their clarity and measurability. Teams can set clear objectives, track progress, and evaluate performance against predefined metrics. This clarity fosters accountability and helps align team efforts with organizational goals.</p><p>However, an overly narrow focus on outcomes can lead to unintended consequences. For instance:<br>- Teams might prioritize short-term gains at the expense of long-term sustainability or user satisfaction.<br>- Metrics may incentivize behavior that meets the letter but not the spirit of the goal, such as achieving deployment speed targets without maintaining quality.<br>- A relentless drive to achieve outcome metrics can increase pressure on teams, potentially leading to stress and burnout.<br>- Outcomes alone may fail to capture the broader value or significance of a project, leaving the bigger picture obscured.</p><p>This is where the concept of impact comes into play, offering a complementary perspective that broadens the scope of evaluation.</p><h3>What Is Impact?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o-yNXiciPKnsA1Ip68NE5w.png" /></figure><p>While outcomes measure specific, short-term achievements, impact reflects the broader, long-term changes brought about by those outcomes. Impact seeks to answer the deeper question, “What difference did we make?” It is often intangible and extends beyond the immediate metrics, encompassing the ripple effects that arise from successful initiatives.</p><p>Examples of impact in software development include:<br>- Improved customer trust and satisfaction, leading to higher loyalty and retention rates.<br>- Enhanced accessibility of a product, resulting in greater inclusivity and societal benefit.<br>- A strengthened organizational reputation as an innovator or leader in the industry.<br>- Elevated team morale and collaboration due to a successful project launch.</p><p>Impact is typically harder to measure directly, as it may take months or even years to fully unfold. For instance, while the immediate outcome of a new feature might be increased user engagement, the broader impact could include fostering a stronger community or enabling new business opportunities for users.</p><h4>Why Impact Matters?</h4><p>Impact is crucial because it provides context and meaning to outcomes. Without impact, outcomes can feel transactional — useful but devoid of deeper significance. By focusing on impact, teams and organizations ensure that their work contributes to larger goals, such as user well-being, societal advancement, or organizational growth.</p><p>Moreover, impact often aligns more closely with an organization’s mission and values. For example, a company committed to environmental sustainability might measure the outcome of reduced server energy consumption but prioritize the impact of contributing to global carbon reduction efforts.</p><h3>The Interdependence of Outcomes and Impact</h3><p>While outcomes and impact are distinct, they are deeply interconnected. Outcomes serve as the building blocks for impact, providing the tangible achievements that enable broader change. Conversely, impact gives outcomes their significance, ensuring that short-term results contribute to long-term progress.</p><p>Considering a real-world example from software development:<br>- <strong>The outcome:</strong> A development team successfully implements a feature that reduces user onboarding time by 30%.<br>- <strong>The impact:</strong> This improvement enables more users to adopt the software, driving increased engagement, customer satisfaction, and revenue growth over time.</p><p>In this case, the immediate outcome is measurable and actionable, while the impact reflects the larger, enduring benefits that stem from that success.</p><h4>Balancing the Two</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*W27pnZ_Qe1d3sVLlDofHkQ.png" /></figure><p>To drive meaningful progress, it is essential to strike a balance between focusing on outcomes and considering their impact. This balance can be achieved through several strategies:<br>- Define clear outcomes aligned with broader goals: Ensure that the outcomes you pursue contribute directly to the desired impact. For example, if your goal is to improve user accessibility, focus on outcomes like implementing key accessibility features.<br>- Measure both short-term results and long-term changes: Use KPIs to track immediate outcomes while also developing methods to evaluate the broader impact, such as user feedback or market analysis over time.<br>- Incorporate feedback loops: Regularly assess how outcomes are contributing to impact and adjust your approach as needed to stay on track.</p><h3>Outcomes and Impact in Action</h3><p>To illustrate the interplay between outcomes and impact, let’s examine a few examples from the software development industry:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nXqETy9W1zwfmcEoAiKaiQ.png" /></figure><h4>Case Study: Enhancing User Engagement</h4><p>A social media platform launches a new recommendation algorithm designed to improve content relevance.<br>- <strong>Outcome:</strong> User engagement increases by 25% in the first three months.<br>- <strong>Impact:</strong> Over time, users report greater satisfaction with the platform, leading to higher retention rates and an increase in ad revenue.</p><h4>Case Study: Improving Development Efficiency</h4><p>A development team adopts a new CI/CD pipeline to streamline deployment.<br>- <strong>Outcome:</strong> The team reduces deployment time by 50% and resolves bugs 20% faster.<br>- <strong>Impact:</strong> These efficiencies allow the organization to bring products to market faster, gaining a competitive edge and increasing customer trust.</p><h4>Case Study: Accelerating Feature Releases</h4><p>A software company introduces an aggressive feature release schedule to respond quickly to market demands.<br>- <strong>Outcome:</strong> The team successfully delivers new features ahead of schedule, increasing the number of product updates by 40% in six months.<br>- <strong>Impact:</strong> However, the rapid release pace leads to increased technical debt and more frequent bugs, resulting in user frustration and a decline in product ratings over time.</p><h3>Conclusion</h3><p>In the fast-paced world of software development, both outcomes and impact play vital roles in achieving success. Outcomes provide the measurable, immediate results that guide day-to-day decision-making, while impact offers the long-term context that ensures those efforts contribute to meaningful progress.</p><p>By understanding and embracing the interplay between these two concepts, teams can not only achieve their immediate goals but also drive transformative change that resonates far beyond the metrics. In essence, the true measure of success lies not just in what we accomplish but in the difference we make.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=088a4bbb7898" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Programming the Mainframe in 2024]]></title>
            <link>https://medium.com/mainframe-careers/programming-the-mainframe-in-2024-6746c3641b2b?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/6746c3641b2b</guid>
            <category><![CDATA[2024]]></category>
            <category><![CDATA[vscode]]></category>
            <category><![CDATA[developer]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[mainframe]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Fri, 09 Feb 2024 20:34:19 GMT</pubDate>
            <atom:updated>2024-02-09T20:34:19.933Z</atom:updated>
            <content:encoded><![CDATA[<h4>What It Takes to Be a Mainframe Programmer in 2024</h4><h3>Introduction</h3><p>Traditionally, mainframe programming has been associated with seasoned professionals who have spent decades honing their skills. However, the demographic of mainframe programmers is undergoing a transformation. A new generation of technologists is emerging, empowered by a plethora of newly developed innovative tooling, bringing fresh perspectives, diverse skill sets and capabilities to the mainframe arena.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*OO8Mr6361DJjy99Q" /></figure><p>Mainframes, sometimes perceived as isolated and stagnant, have proven time and again their resilience by adapting to modern challenges and environments. These robust computing machines are the lifeblood of industries such as finance, healthcare, retail, and telecommunications, handling vast amounts of data with unparalleled reliability and security. Mainframes are run by programmers, and this is what it takes to be one of them in 2024.</p><h3>Mainframe Engineer</h3><p>Colloquially named just a “mainframer”, the term refers to an individual who specializes in working with mainframe computing environments. Mainframers can be programmers, system administrators or hardware installation and maintenance specialists and their expertise may range from knowing programming languages such as Assembler and COBOL to configuring and administering the operating systems and their features.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gOmg1RC4Kn_AmGRu.jpg" /></figure><p>In 2024 the mainframe is part of a bigger enterprise computing infrastructure, so beside the traditional system programming and software development, a mainframe engineer can specialize in other areas such as Hybrid Cloud or DevOps engineering and architecture. Mainframe APIs enable complex business integrations, external orchestration, and reporting tools whose designs evolve into holistic solutions. Today’s mainframers are the specialists that make those integrations possible, they are the driving force that continues the evolution of the platform.</p><h3>Mainframe Programming Languages</h3><p>By far the most popular programming language for the mainframe platform is COBOL (Common Business-Oriented Language). Although being one of the oldest, it is still widely used for coding business, finance and administrative systems and applications. COBOL <a href="https://www.thestack.technology/cobol-in-daily-use/">arguably</a> holds a record amount of lines of code in production and is still growing. Its popularity may be because its syntax is so explicit and close to spoken English.</p><p>The most performance efficient way to program the mainframe is to use Assembler. This low-level programming language is specific for the architecture of the particular microprocessor and enables direct control over the hardware, high-efficiency and close integration with the machine architecture. Many programs originally written in Assembler are still maintained and developed to enable critical system and business operations.</p><p>PL/I (Programming Language One) is a versatile programming language developed for scientific, engineering and business applications. The language was created as a response to the limitations of the existing programming languages at that time. Its original intention was to make it a more general-purpose language, capable of handling multiple use cases of different nature. Regardless of being considered one of the fundamental development tools for the platform, it remained less popular and in the shadow of its big brother — COBOL.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/936/1*WcxzqjC5ZX1rOBl4lyh3FA.jpeg" /></figure><p>In the present, mainframes come integrated with several compilers, interpreters and runtime environments such as C/C++, Java, Rexx, Node, Python among many others. New applications and automations are usually implemented using one of those, nevertheless, the mainframe programmer of today would still be expected to have a good grasp on the classics too.</p><h3>Mainframe Development Environment</h3><p>Traditionally, the mainframe platform offers a set of tools, and environments tailored to the unique requirements of these powerful computing systems, TSO (Time Sharing Option), SDSF (System Display and Search Facility) and ISPF (Interactive System Productivity Facility) being the most prominent ones. Those three, together with some other console user interfaces, are commonly called “the green screen” and are an essential part of the evolution and present of the platform.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Z3ytjXEvUAArWO20.png" /></figure><p>Being character-based programs displayed on a character-oriented terminal emulators, they often intimidate new enthusiast that are eager to become proficient with the platform. Until today, a proficient mainframe system programmer must know how to interface with the platform by using TSO, SDSF or ISPF. However, for today’s mainframe software developer, this is no longer the first they need to learn and for some it may not be necessary at all. Instead, VS Code, or less commonly known as Visual Studio Code, is becoming the IDE (Integrated Development Environment) of choice for the mainframe as well.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/936/1*_FuojMlT7u-fJYjHG8s0Vg.png" /></figure><p>This is how VS Code’s extensions marketplace looks like if one searches for the “mainframe” keyword there. An impressive number of specialized extensions from a variety of vendors and volunteers appear, providing a remarkable variety of options for one to choose when setting up their IDE. For a developer with experience in programming for distributed platforms, this would feel very familiar and a natural first step in their journey towards the secrets of the mainframe.</p><h3>Conclusion</h3><p>The landscape of mainframe programming in 2024 reflects a dynamic shift propelled by a new generation of technologists and the relentless adaptability of mainframes systems. Traditionally associated with seasoned professionals, mainframe programming now welcomes a diverse cohort armed with innovative tools and fresh perspectives. The resilience of mainframes is evident in their sustained relevance across vital industries. While traditional languages like COBOL and Assembler remain integral, today mainframes natively support powerful programming languages such as C/C++, Java, and Python. The development environment, once dominated by character-based interfaces, has evolved with the rising prominence of VS Code, symbolizing a departure from the good old “green screen”.</p><p>In 2024, a mainframe engineer, or “mainframer”, encompasses a broader skill set, extending beyond traditional programming to include expertise in Cloud, DevOps, software engineering and architecture. The mainframe’s role in the larger enterprise computing infrastructure emphasizes the importance of these specialists in driving the ongoing evolution of this powerful computing platform. As the mainframe continues to adapt, so do the professionals who contribute to its resilience and evolution, shaping the future of mainframe programming.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6746c3641b2b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/mainframe-careers/programming-the-mainframe-in-2024-6746c3641b2b">Programming the Mainframe in 2024</a> was originally published in <a href="https://medium.com/mainframe-careers">mainframe-careers</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Resilience to Antifragility]]></title>
            <link>https://medium.com/zowe/from-resilience-to-antifragility-8ef5b1d3e9de?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/8ef5b1d3e9de</guid>
            <category><![CDATA[zowe-general]]></category>
            <category><![CDATA[zowe]]></category>
            <category><![CDATA[testops]]></category>
            <category><![CDATA[zowecore]]></category>
            <category><![CDATA[resiliance]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Tue, 22 Aug 2023 20:48:50 GMT</pubDate>
            <atom:updated>2023-08-23T13:21:17.240Z</atom:updated>
            <content:encoded><![CDATA[<h4>Zowe and the Transformation of the Mainframe Platform</h4><h4>Introduction</h4><p>{<a href="https://github.com/zowe/community/blob/master/blogging/blog_guidelines.md#zowe-core">Core</a>} In the realm of technology, the ability to adapt and thrive in the face of disruption is crucial for businesses to stay competitive. Resilience has long been a sought-after quality for systems, allowing them to withstand and recover from challenges. However, as the landscape evolves, the concept of antifragility has emerged, emphasizing the capacity to not only bounce back but also grow stronger from adversity. Mainframe systems, with the help of <a href="https://openmainframeproject.org/">The Open Mainframe Project</a>’s <a href="https://zowe.org/">Zowe</a> and TestOps methodologies, can embrace this transformative journey from resilience to antifragility.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*N-rgJxx-GBIYSzCQ" /></figure><h4>Fragile vs. Robust</h4><p>Fragile systems are susceptible to deterioration, damage and destruction as a cause of minimal stress or impact. They require extra care and attention to avoid degradation and keep them within their narrow boundaries of operation. Robust systems are designed to withstand or endure various challenges, stresses or adverse conditions without breaking, deteriorating or losing functionality. Their boundaries of operation parameters are wider but not infinite and when the stress applied exceeds them, the system breaks beyond recovery.</p><h4>Resilience: The Foundation of Mainframe Systems</h4><p>Mainframe systems have served as the backbone of critical business operations for decades, renowned for their robustness and reliability. Resilience has been at the core of mainframe architecture, enabling uninterrupted operations and minimizing downtime. Organizations invested in redundant hardware, disaster recovery plans, and failover mechanisms to ensure the continuity of their mainframe systems.</p><p>However, as technology advances and businesses face increasingly dynamic environments, simply recovering from disruptions may not be sufficient to meet evolving demands.</p><h4>Antifragility: The Evolutionary Leap</h4><p>Antifragility, a concept introduced by <a href="https://en.wikipedia.org/wiki/Nassim_Nicholas_Taleb">Nassim Nicholas Taleb</a>, goes beyond resilience. It suggests that systems can benefit from volatility and uncertainty, becoming more adaptable, innovative, and robust as a result. Antifragile systems are not merely resilient; they actively learn and improve from disturbances, transforming challenges into opportunities for growth and optimization.</p><h4>Zowe: Empowering Mainframe Antifragility</h4><p>Zowe, the open-source framework developed by the Open Mainframe Project, plays a pivotal role in the journey from resilience to antifragility for mainframe systems. It facilitates the integration of modern development practices, seamless interaction with other platforms, and enhanced management capabilities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/684/0*jMFR8osEIhtcD6b0" /></figure><p>1. Integration: Zowe enables the integration of mainframe systems with modern technologies and practices. It provides a set of APIs, microservices, and open-source tools that bridge the gap between the mainframe and the broader IT ecosystem. With Zowe, mainframe applications and data can seamlessly interact with cloud platforms, microservices architectures, and emerging technologies, fostering adaptability and innovation.</p><p>2. Development: Zowe empowers developers by providing a modern development environment for mainframe applications. It enables the use of open-source tools, such as Git and Jenkins, allowing developers to embrace DevOps and agile methodologies. By adopting modern development practices, mainframe systems become more flexible, enabling rapid application delivery and iterative improvements.</p><p>3. Management: Zowe simplifies mainframe management tasks through a unified interface and standardized APIs. It provides a centralized platform for monitoring, administration, and troubleshooting, enhancing operational efficiency. With streamlined management processes, organizations can respond quickly to evolving needs and leverage real-time insights to optimize their mainframe systems.</p><h4>TestOps: Enabling Antifragility on the Mainframe</h4><p>TestOps, an extension of DevOps practices, emphasizes continuous testing throughout the software development lifecycle. By integrating testing into the development process, organizations can identify and address issues early, ensuring higher-quality software and faster delivery cycles. TestOps enables organizations to embrace antifragility by actively learning from testing feedback and improving system resilience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*H4Hkq8bkv4XRffDy" /></figure><p>Zowe, combined with TestOps methodologies, offers organizations the opportunity to transform their mainframe systems from resilient to antifragile:</p><p>1. Comprehensive Testing: TestOps enables organizations to implement comprehensive testing strategies for mainframe systems. By incorporating automated testing tools and practices, organizations can identify vulnerabilities, optimize performance, and ensure the stability of their mainframe applications.</p><p>2. Continuous Improvement: TestOps fosters a culture of continuous improvement and learning from testing feedback. Organizations can analyze testing results, gather insights, and implement optimizations to enhance the robustness and performance of their mainframe systems over time.</p><h4>Antifragility in Mainframe Systems</h4><p>Mainframe systems embody many antifragile principles and often incorporate various features and practices to increase their resilience and adaptability. IBM’s z/OS operating system includes several characteristics that contribute to their resilience:</p><ol><li>Hardware and Software Redundancy</li><li>High Availability and Failover</li><li>Dynamic Capacity Expansion</li><li>Performance Monitoring and Analytics</li><li>Rolling Updates and Maintenance</li><li>Security and Encryption</li><li>Transaction Logging and Recovery</li><li>Disaster Recovery and Data Replication</li></ol><p>In <a href="https://diginomica.com/eating-elephant-nedbanks-digital-transformation-built-around-devops-and-mainframe-resilience">Nedbank’s digital transformation</a> story they talk about their DevOps journey from robustness to resilience hinting to the key objective “<em>to automate as much as possible</em>”. With Zowe and TestOps they would be leaping towards their next big evolution of mainframe antifragility.</p><h4>Conclusion</h4><p>In the face of a rapidly evolving technological landscape, organizations must transition from traditional resilience to antifragility to thrive. Zowe and TestOps methodologies provide a powerful enabler for mainframe systems to embark on this transformation. By embracing integration, modern development practices, and comprehensive testing, organizations can empower their mainframe systems to adapt, innovate, and continuously improve. Through this journey, mainframe systems become antifragile, leveraging the strength of resilience while actively learning and growing from disruptions.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ef5b1d3e9de" width="1" height="1" alt=""><hr><p><a href="https://medium.com/zowe/from-resilience-to-antifragility-8ef5b1d3e9de">From Resilience to Antifragility</a> was originally published in <a href="https://medium.com/zowe">Zowe</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Personal Access Tokens for the Zowe API Mediation Layer]]></title>
            <link>https://medium.com/zowe/personal-access-tokens-for-the-zowe-api-mediation-layer-53e383ff1e61?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/53e383ff1e61</guid>
            <category><![CDATA[zowecore]]></category>
            <category><![CDATA[personal-access-tokens]]></category>
            <category><![CDATA[zowe]]></category>
            <category><![CDATA[mainframe-modernization]]></category>
            <category><![CDATA[zowe-apiml]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Wed, 09 Nov 2022 23:03:25 GMT</pubDate>
            <atom:updated>2022-11-09T23:03:25.638Z</atom:updated>
            <content:encoded><![CDATA[<p>{<a href="https://github.com/zowe/community/blob/master/blogging/blog_guidelines.md#zowe-core">Core</a>} Personal Access Tokens (PAT) are ubiquitous today where most of the applications that we use or develop consume and rely on remote services that require secure and independent authentication. The main reason one would want to use a PAT is to avoid having to send the username and password as part of the authentication request. Oftentimes one would also want to limit the access of an authenticated application to just a subset of the full scope that the associated user account has. However, for The <a href="https://openmainframeproject.org/">Open Mainframe Project</a>’s <a href="https://zowe.org/">Zowe</a> and the mainframe platform, the most desirable feature of PATs is that they live independently of the associated mainframe account credentials and this is why they were implemented in the Zowe API Mediation Layer.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*kZcQ6_yKwlD-eb9eE_OknA.png" /></figure><p>The personas that would most benefit from this new feature of the Zowe API Mediation Layer are the system integrators, the people that design and construct the mainframe DevOps infrastructure for their organization. You may need to enable secure authentication for your hybrid cloud environment to access the mainframe services. You may also want to provide instructions for your fellow mainframe developers on how to set up their workstations to be able to easily access and work with their source code there. <a href="https://github.com/marketplace/ca-endevor-bridge-for-git">Endevor® Bridge for Git by Broadcom</a> is a good example of how this enhancement can be used to authenticate users with PATs instead of their account credentials.</p><p>Constructing DevOps infrastructure, however, is just one of the use cases that could leverage the use of PATs, in fact there are many other integrations that could benefit as well. Any client side applications that consumes REST APIs through the Zowe API Mediation layer would have yet another secure and convenient way of authentication.</p><p>With the release of Zowe 2.3 and the API Mediation Layer version 2.3.1, the Zowe users will have the following APIs available to manage PATs:</p><ul><li>User APIs: generate, validate, invalidate a specific token, invalidate all tokens</li><li>Security Administrator APIs: invalidate all tokens for a user, invalidate all tokens for a service, evict non-relevant tokens and rules</li></ul><p>The authentication to access the PAT APIs can be achieved via the other existing API ML methods, i.e. via username and password or a client certificate.</p><p>The base paths to all API ML Gateway PAT endpoints would have the following structure: https://{gatewayUrl}:{gatewayPort}/gateway/api/v1/</p><p>The calls to all those APIs would return code 200 in case of a successful operation or 401 in case of failure. All cached values involved in their operations are protected by the API ML client certificate.</p><h3>User Personal Access Token APIs</h3><h3>Generate a Token</h3><p>An authenticated user can generate a token by calling the following endpoint:</p><pre>POST /auth/access-token/generate</pre><pre>Body: {<br>    “validity”: &lt;days&gt;,<br>    “scopes”: [“&lt;serviceId_1&gt;, &lt;serviceId_2&gt;”]<br>}</pre><p>Where the validity period is specified in number of days and the scope is a list of services the PAT would be valid for. The token value is returned in the response body as plain text.</p><h3>Validate a Token</h3><p>An authenticated user can validate a token by calling the following endpoint:</p><pre>POST /auth/access-token/validate</pre><pre>Body: {<br>    “token”: “&lt;token&gt;”,<br>    “serviceId”: “&lt;serviceId&gt;”<br>}</pre><p>Where a combination of a token value and a service ID is used to validate the access token. The same validation is also performed every time a token is used to access a service.</p><h3>Revoke a Token</h3><p>An authenticated user can revoke a token by calling the following endpoint:</p><pre>DELETE /auth/access-token/revoke</pre><pre>Body: {<br>    “token”: “&lt;token&gt;”<br>}</pre><p>Where a token value is specified to be revoked. The PAT is then stored by the Catching Service under the invalidTokens key and thus effectively is being revoked.</p><h3>Revoke All Tokens</h3><p>An authenticated user can revoke all their tokens by calling the following endpoint:</p><pre>DELETE /auth/access-token/revoke/tokens</pre><pre>Body: {<br>    “timestamp”: “&lt;timestamp&gt;”<br>}</pre><p>Where a timestamp value specifies the moment, before which all PATs for the authenticated user would be revoked. The provided user rule is stored by the Catching Services under the invalidUsers key and thus effectively revoking all tokens for the user.</p><h3>Security Administrator Personal Access Token APIs</h3><p>An authenticated user with privileged access to SAF resources is able to invalidate all tokens based on criteria established by rules about user IDs or services scopes. These criteria would define the level of access control and would restrict access in advance allowing for prompt reactions against security breaches.</p><h3>Revoke All Tokens for a User</h3><p>A call to the following endpoint would allow the revocation of all tokens issued for a user prior to the specified timestamp.</p><pre>DELETE /auth/access-token/revoke/tokens/users</pre><pre>Body: {<br>    “userId”: “&lt;userId&gt;”,<br>    “timestamp”: “&lt;timestamp&gt;”<br>}</pre><p>Where the user ID refers to the user in question and the timestamp value specifies the moment, after which all PATs for the relevant user would be revoked. The provided user rule is stored by the Catching Services under the invalidUsers key and thus effectively revoking all tokens for the user.</p><h3>Revoke All Tokens for a Service</h3><p>A call to the following endpoint would allow the revocation of all tokens issue for a service prior to the specified timestamp.</p><pre>DELETE /auth/access-token/revoke/tokens/scope</pre><pre>Body: {<br>    “serviceId”: “&lt;serviceId&gt;”,<br>    “timestamp”: “&lt;timestamp&gt;”<br>}</pre><p>Where the service ID refers to the service in question and the timestamp value specifies the moment, after which all PATs for the relevant service would be revoked. The provided service rule is stored by the Catching Services under the invalidScopes key and thus effectively revoking all tokens for the service.</p><h3>Evict Non-Relevant Tokens and Rules</h3><p>A call to the following endpoint would allow the eviction of all invalidate tokens that have expired and the related to them rules.</p><pre>DELETE /auth/access-token/evict</pre><p>The main purpose of the eviction is to maintain the cache size within reasonable parameters. The token verification processes require the validation of all rules, including those no longer applicable. As such, maintaining the cache size within reasonable limits would reduce the processing cost for stored rules that are no longer relevant. In case of a successful eviction call, the response is an empty body with a status code of 204. In case the eviction fails due to insufficient permissions, the status code is 403.</p><h3>Use of PAT for Authentication</h3><p>Ultimately, the purpose of the Personal Access Tokens is to be used for authentication as part of a Single Sign On operation, in which an authorized service is accessed by a user that has a valid token at the time of the call. There are two ways how to provide the token value in the request:</p><ul><li>Using a secure HttpOnly cookie with the name personalAccessToken</li></ul><pre>GET /&lt;allowed-service&gt;/api/v1/request HTTP/1.1</pre><pre>Cookie: personalAccessToken=eyJhbGciOiJSUzI1NiJ9…<br>HTTP/1.1 200<br>…</pre><ul><li>Using a request header with the name PRIVATE-TOKEN</li></ul><pre>GET /&lt;allowed-service&gt;/api/v1/request HTTP/1.1</pre><pre>PRIVATE-TOKEN: eyJhbGciOiJSUzI1NiJ9…<br>HTTP/1.1 200<br>…</pre><p>In the examples above, the return codes indicate successful authentication. In case the authentication fails for any of the possible reasons, i.e., expired token, access to unauthorized service, incorrect token value, the X-Zowe-Auth-Failure error header is set and passed in the response. The error message contains a reason for why the provided authentication is not valid.</p><p>Reference: <a href="https://github.com/zowe/docs-site/blob/docs-staging/docs/user-guide/api-mediation/api-mediation-personal-access-token.md">Zowe API Mediation Layer Personal Access Token</a></p><p>If you enjoyed this blog, and would like to learn more about Zowe, check out more <a href="https://medium.com/zowe">Zowe blogs here</a>. You are also welcome to ask questions and join the conversation on the <a href="https://openmainframeproeject.slack.org/">Open Mainframe Project Slack Channel</a> #Zowe-dev, #Zowe-user or #Zowe-onboarding. If this is your first time using the Open Mainframe Project slack channel register <a href="https://slack.openmainframeproject.org/">here</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=53e383ff1e61" width="1" height="1" alt=""><hr><p><a href="https://medium.com/zowe/personal-access-tokens-for-the-zowe-api-mediation-layer-53e383ff1e61">Personal Access Tokens for the Zowe API Mediation Layer</a> was originally published in <a href="https://medium.com/zowe">Zowe</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Measuring Zowe’s Success]]></title>
            <link>https://medium.com/zowe/measuring-zowes-success-200b42b93609?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/200b42b93609</guid>
            <category><![CDATA[zowe]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[broadcom]]></category>
            <category><![CDATA[open-mainframe]]></category>
            <category><![CDATA[zowe-general]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Tue, 06 Sep 2022 18:38:26 GMT</pubDate>
            <atom:updated>2022-09-06T18:38:26.771Z</atom:updated>
            <content:encoded><![CDATA[<p>{<a href="https://github.com/zowe/community/blob/master/blogging/blog_guidelines.md#zowe-ecosystem">Ecosystem</a>} Zowe is a vendor neutral project, owned and managed by The Open Mainframe Project, an umbrella .org bringing together a number of open source initiatives on the mainframe all under the parent of The Linux Foundation. Broadcom’s Open Mainframe value stream is dedicated to the success of <a href="https://zowe.org/">Zowe</a> and the <a href="https://openmainframeproject.org/">Open Mainframe Project</a> as a whole. At Broadcom our vision for this involves the dedication of our high-performing teams as a main ingredient to the recipe of success (<a href="https://medium.com/modern-mainframe/software-development-for-the-mainframe-high-performing-teams-a53a074886d8">my article here</a> describes how we achieve this). The Linux Foundation itself proposes several ways to measure success for open source programs (<a href="https://www.linuxfoundation.org/blog/measure-open-source-programs-success-2/">see this great article</a>) but a challenge we set teams at Broadcom, is how to measure objectively our success amongst the wider breath of contributors. This is something we are very proud of, together with Zowe being independently recognized as the <a href="https://medium.com/modern-mainframe/mainframe-devops-award-goes-to-zowe-505b201fa842">Best DevOps for Mainframe</a> solution in 2021.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5G-_keLd5WkXQvk5YpIEhg.jpeg" /></figure><p>Success is subjective, it depends on the perspective and the person evaluating it, unless it is properly defined, unless the goal is clear. Having objective indicators for concrete achievements could remove ambiguities, unconscious biases and serve as a real-time measure of one’s progress. There are several ways to measure success that the LF propose in <a href="https://www.linuxfoundation.org/blog/measure-open-source-programs-success-2/">their article</a> for open source programs, among which are the quantitative measurements of:</p><ul><li>number of contributors</li><li>number of pull requests</li><li>issues submitted</li><li>commits per contributor</li><li>number of external adopters</li><li>number of projects in the program</li></ul><p>The Linux Foundation goes beyond just numbers and argues in <a href="https://github.com/todogroup/guides/blob/master/measuring-your-open-source-program.md">their blog post</a> why<strong> identifying particular goals</strong>, <strong>setting them as objectives</strong> and <strong>following up on their achievement is the key.</strong></p><p>With Zowe we do not go much further from the prescribed recipe and we collect <a href="https://www.metrics.zowe.org/">similar metrics</a> to those recommended so far. We are proud to see how fast the number of Zowe’s client side downloads are growing and the continued trend of steady interest in the server-side components. Furthermore, we pay special attention to how our community is developing by watching the number of members in our <a href="https://slack.openmainframeproject.org/">Slack workspace</a>, how many GitHub unique contributors we have and how many compliant products companies register with us.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2K3RrcwO4IbD2KHmeuRSpQ.png" /></figure><p>The question now that we all ask ourselves is… is this enough, does it show how successful is Zowe? Well, it certainly does. The <strong>number of downloads a product has is a direct indicator of how many potentially active users it has</strong> (we cannot be completely certain as none of Zowe’s components use telemetry). The thing is, Zowe is not the typical open source project: Zowe is not <a href="https://data.firefox.com/dashboard/user-activity">Firefox</a>, which is mostly a client side end-user application, whose aim is to attract as many active users as possible so it could commercialize on them. Just like the <a href="https://www.kernel.org/">Linux kernel</a>, <strong>Zowe is an enabler, an essential mainframe infrastructure software bundle, whose goal is to open the platform to the modern developer, system administrator and end-user.</strong></p><p>The Zowe project is in the phase where it is already sufficiently developed to be interesting for users to deploy in production and this puts the focus to success on its adoption. We follow the progress of this trend by tracking the number of <a href="https://www.openmainframeproject.org/all-projects/zowe/conformance">Zowe Conformant</a> products that are advertised on the Open Mainframe portal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wX22C6YEBHgUlT8ywqfG6A.png" /></figure><p>Zowe’s recent release of version 2.x included changes to the conformance criteria by tightening the minimum security features an onboarded service should support in order to issue a certification badge for it. Introducing the new requirements automatically removed some badges from the chart above, nevertheless, the efforts of both the Zowe community and the software development companies helped put the amount back to the levels from before the Zowe v2 release.</p><p>Having a conformant API is not only a technical achievement, <strong>it is also a sign of commitment on behalf of the mainframe vendor, it is a sign of trust in Zowe</strong>. That commitment means the developer invested in onboarding and verifying their work against Zowe’s criteria, it means their users can have a consistent experience and added value. The total number of conformant APIs is a strong indicator for engagement — a direct consequence of the real use cases already present for those services.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/792/1*LFXFuXekB0RpXIV5iX8-pw.png" /></figure><p>In the end, if Zowe’s goal was to attract conformant extenders then the total number of conformant APIs would be the ultimate measure for its success. Its fundamental value proposition, however, is to transform the way technology professionals interact with the mainframe. Extensions are seen as important requirement for its success, conformance is seen as necessary to streamline the experience for the consumers but they are not its goal. Would Zowe have been successful if it had an infinite number of conformant APIs and no customers able or willing to use it? Would this accelerate the modernization, automation and self-service use cases related to the mainframe?</p><p>The answer to those questions is most probably yes but not necessarily. Watching the number of conformant APIs and actively working towards increasing them contributes to the success of Zowe but it is not the main goal of all our efforts. Our aim is to transform the mainframe platform and we can only claim success when we see the real world stories of the Zowe’s adopters.</p><h3>Learn more:</h3><p>If you enjoyed this blog check out more <a href="https://medium.com/zowe">Zowe blogs here</a>. Or, ask a question and join the conversation on the <a href="https://openmainframeproject.slack.com/">Open Mainframe Project Slack Channel</a> #zowe-client-sdk, #zowe-cli, #zowe-dev, #zowe-user, or #zowe-onboarding. If this is your first time using the Open Mainframe Slack Channel register <a href="https://slack.openmainframeproject.org/">here</a>.</p><p>Zowe is owned and managed by the <a href="https://www.openmainframeproject.org/">Open Mainframe Project</a>, which is a <a href="https://linuxfoundation.org/">Linux Foundation</a> project.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=200b42b93609" width="1" height="1" alt=""><hr><p><a href="https://medium.com/zowe/measuring-zowes-success-200b42b93609">Measuring Zowe’s Success</a> was originally published in <a href="https://medium.com/zowe">Zowe</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Governing of Zowe Engineering]]></title>
            <link>https://medium.com/zowe/governing-of-zowe-engineering-1234fad5b120?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/1234fad5b120</guid>
            <category><![CDATA[zowe-community]]></category>
            <category><![CDATA[zowe-general]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software-engineering]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Wed, 26 Jan 2022 19:39:17 GMT</pubDate>
            <atom:updated>2022-02-02T11:22:47.043Z</atom:updated>
            <content:encoded><![CDATA[<h4>An Open Mainframe Story of a High Performing Project</h4><p>{<a href="https://github.com/zowe/community/blob/master/blogging/blog_guidelines.md#zowe-core">Core</a>} The <a href="https://www.openmainframeproject.org/">Open Mainframe Project</a>’s <a href="https://www.zowe.org/">Zowe</a> initiative that aims to deliver an integrated and extensible open-source framework for z/OS. Like macOS or Windows, z/OS comes with a set of APIs and OS capabilities that applications can build upon and also includes some utilities out of the box. While the technical aspects of Zowe are quite exciting and are the topic of a great number of articles, in this one the goal is to summarise and briefly analyze how the project is governed and with what purpose.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*99nAK-Gjc859YBgK" /></figure><p>Zowe is an agile and open-source software development project. Many mainframe users would approach such projects with a certain amount of doubts about the quality and security an open-community undertaking would be able to sustain. Mainframe users have one of the toughest requirements for performance, reliability, security and adaptability in the industry and they would not accept anything lesser. This is why Zowe’s community invests considerable efforts in governing its software development processes.</p><p>Zowe’s active contributors and maintainers are organised within squads that act and exercise in alignment with the key principles of the <a href="https://www.scrum.org/resources/what-is-scrum">Scrum</a> and <a href="https://www.scaledagileframework.com/">SAFe</a> frameworks. It is part of the Open Mainframe Project, which is governed by the Linux Foundation. It also has its own governing body named the <strong>Zowe Technical Steering Committee</strong> (TSC) supported by the <strong>Zowe Advisory Council</strong> (ZAC) among others.</p><p>To say it simply, the main purpose of those entities is to facilitate and streamline the engagement with the community for both users and developers. In a little more detail, TSC is responsible for the overall planning, development and technical feedback assessment while ZAC is oriented more towards onboarding new customers and promoting the project to future users and vendors while evaluating proposals for new projects and the value of the currently developed ones. Together, they create the foundation for a strong community of high-performing software developers and mainframe users.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*zdC0Ka-FatY6HQ3MPEidcg.png" /></figure><p>While ZAC’s work is essential for the success of Zowe, this article is focused more on how TSC and in general, how the proper governance of an open-source project may enable and influence the performance of its software developers. Zowe is a high-performing agile project and its squads are already producing a tremendous amount of value for its community. Those squads, however, do not exist in a void, and when investing their time to solve common challenges, they need the presence of mind to consider the best possible solution for the project. Since their results impact directly the whole project’s integrity, the TSC, its members and guests come together, discuss and decide how best to steer the Zowe project for the wellbeing of everyone.</p><p>The Zowe community is diverse in terms of experts and specialists, which makes it a great source of competencies for making informed decisions. The different workgroups are incepted by the TSC and are tasked to serve the currently prioritised needs of the project and its users. To name a few, Zowe currently has permanent Software Architecture and Software Security workgroups with members of the community getting organised in them to provide guidance and make decisions to support the important achievements of their peer software developers and users. TSC is the body that oversees their work and may yield the decision to create or disband them as the need for them appears.</p><p>In the past few months, the TSC has been busy working hard on the very challenging topic of reviewing and updating Zowe’s principles, guidelines and best practices recommendations. The principles and the general <a href="https://github.com/zowe/community/tree/master/Technical-Steering-Committee/guidelines">guidelines</a> are essential for the smooth running of the project, what is more interesting here is to analyse the need for recommended <a href="https://github.com/zowe/community/tree/master/Technical-Steering-Committee/best-practices">best practices</a> and see how they might be shaping the work of the existing squads and any newly created ones.</p><p>Some of the interesting categories listed at the time of editing this article are the <a href="https://github.com/zowe/community/blob/master/Technical-Steering-Committee/best-practices/quality.md">quality</a> and <a href="https://github.com/zowe/community/blob/master/Technical-Steering-Committee/best-practices/security.md">security</a> recommendations. The analysis and potential suggestions that one would make to the so defined documents could be a topic of separate work, the mere existence of those documents sends a strong message of what the project’s leaders are passionate about. Those policies are a work-in-progress and Zowe, and specifically, the TSC would happily welcome more contributors. Peeking at the open <a href="https://github.com/zowe/community/pulls">pull requests</a> in the <a href="https://github.com/zowe/community">community repository</a> one could find another interesting piece dedicated to the <a href="https://github.com/zowe/community/pull/1313">build and test practices</a> that talks about the ever so important practice of continuous integration. This change is proposed by a non-member of the TSC, which is a great example of the openness of Zowe, especially within its governing bodies.</p><p>One area where the TSC recognises opportunities for improvement is the engagement with independent contributors. The Zowe project benefits from multiple internships and onboarding programs but those implicitly involve participation within an existing squad. It’s well known that the mainframe learning curve for beginners is rather steep, which may be a good justification for requiring contributors to work within squads where support and expertise would be available for them from day one. This, however, should not become a barrier and the TSC faces the challenge to enable and promote an open environment without compromises.</p><p>Comparing Zowe’s leadership practices to the <a href="https://opensource.guide/leadership-and-governance/">Open Source Guide for Leadership and Governance</a> one can see how the TSC goes beyond and spends a significant effort in recommending contemporary agile practices for high performing software development teams. The message sent out is to not compromise with quality and security, and while it includes concrete suggestions and recommendations for actions and best practices, it also suggests openness and invites others to contribute. This is how Zowe’s leadership strives to maintain its strong commitment to the core principles of the mainframe platform that has retained its relevance for so many years serving the world.</p><p>If you enjoyed this blog checkout more <a href="https://medium.com/zowe">Zowe blogs here</a>. Or, ask a question and join the conversation on the <a href="https://openmainframeproeject.slack.org/">Open Mainframe Project Slack Channel</a> #Zowe-dev, #Zowe-user or #Zowe-onboarding. If this is your first time using the OMP slack channel register <a href="https://slack.openmainframeproject.org/">here</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1234fad5b120" width="1" height="1" alt=""><hr><p><a href="https://medium.com/zowe/governing-of-zowe-engineering-1234fad5b120">Governing of Zowe Engineering</a> was originally published in <a href="https://medium.com/zowe">Zowe</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Software Development for the Mainframe & High-Performing Teams]]></title>
            <link>https://medium.com/modern-mainframe/software-development-for-the-mainframe-high-performing-teams-a53a074886d8?source=rss-80a1656b4edd------2</link>
            <guid isPermaLink="false">https://medium.com/p/a53a074886d8</guid>
            <category><![CDATA[mainframe]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[zowe]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[broadcom]]></category>
            <dc:creator><![CDATA[Boris Petkov]]></dc:creator>
            <pubDate>Tue, 01 Jun 2021 21:15:32 GMT</pubDate>
            <atom:updated>2021-06-01T21:15:32.958Z</atom:updated>
            <content:encoded><![CDATA[<h4>A case study in modernisation</h4><p>One of our goals at Broadcom’s Mainframe Software Division or MSD is to enable the present-day developer to work on the mainframe platform using modern tools and technologies that support high-performing teams. We bring a great deal of innovation to our partners and customers, but by today’s standards, are we a high-performing team ourselves? Even when judged against the toughest industry standards, I believe we are. Here’s why:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*gCdDK1YmTrVSXRLlRud4Jw.jpeg" /></figure><p>A generic definition of a high-performing team is that of a group that brings together members committed to mutual growth and personal success. Common attributes of such teams may include participation, accountability, clarity, interaction, flexibility, localization, creativity and quickness. Contemporary agile software development practices provide enablers for some of these attributes. For example, the process of breaking down user stories together as a team allows for knowledge sharing and fast cycle times. Stand-up meetings, pair-programming sessions, retrospectives directly encourage interaction and participation, provide clarity and fuel a sense of accountability.</p><p>High-performing teams need technological enablers; fast cycle times imply quick feedback from automated tests and deployment, which require a robust CI/CD pipeline. DevOps and lately AIOps are the disciplines that support those efforts and one cannot imagine a modern development process without them. Many of the historical reasons why DevOps has taken longer to emerge on the mainframe have been overcome in recent years, and today’s mainframe developers have no excuse but to lead with best practices.</p><p>My recent professional experience took me on a journey from a professional services-oriented development team at NCR Corp., through the fast-paced highly performing agile teams of Microsoft, to the in-house development department of the Barclays bank ending up in Broadcom’s MSD. The broad perspectives I gained helped me understand the reasons behind the classical waterfall methodology and sophisticated CI/CD pipelines enabled by the latest automation technologies offering agility, short feedback cycles, and scalability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zPij85VzTmpZ4iO0uMMoCQ.jpeg" /></figure><p>As part of the Open Mainframe Project — a multi-company initiative promoting the use of open source software on the mainframe, I have had the privilege of witnessing a DevOps transformation on the platform. I was very pleasantly surprised when I realised that a mainframe software development house used tools and implemented practices comparable to those seen in cutting-edge technology giants such as Microsoft. The combination of agile methodologies and DevOps techniques enables us to foster and enjoy an environment for truly high-performing teams.</p><p>In Broadcom’s MSD, we started our transformation with the people. Our development teams benefit of a dedicated structure of agile coaches and scrum masters that work hand-in-hand to promote the behaviors of high-performing teams. Participation, accountability, clarity, interaction, flexibility, localization, creativity, quickness are all covered in our practices. They are supervised and managed by trained professionals and frequently discussed and challenged during retrospectives.</p><p>Quickness in particular, however, also needs an accessible and reliable DevOps infrastructure. Mainframe environments are unique in many ways and for that reason are often kept siloed which can result in reduced flexibility and slower acceptance of changes. Well, not for us! Since we are designing the tools for DevOps on the Open mainframe, we revolutionise the platform to make it appealing to the present-day developer and system programmer.</p><p>We make and use modern, feature-rich development tools for both traditional mainframe programming environments such as Assembler and COBOL as well as for the much more modern and widely popular Java and JavaScript. On top, we are building and maintaining fully automated and frictionless CI/CD pipelines using open tools such as Jenkins and, lately, GitHub Actions that execute automated testing and validating on the mainframe using the open source Zowe CLI and API Mediation Layer. And we are not stopping there; we are already looking forward and experimenting with extensive usage of containerisation to smoothen our setup and deployment processes even further.</p><p>One differentiator I see for the first time in my career is how Broadcom has structured our mainframe software engineering management organisation. My previous experience has been that most line managers in software development are promoted high-performing senior engineers. They usually work with small teams and act as both people managers and technical leads. I observed this pattern in all my previous jobs but not in Broadcom’s MSD, where we pay extra attention to people management.</p><p>Our line managers all have formal training, they lead slightly bigger teams but are leaving the technical side to the experts while they focus on the bigger picture. Don’t get me wrong, we are still very engaged with technical stuff: constantly observing, participating, sharing opinions and giving feedback. However, we aim to truly enable and support an environment that promotes self-governing and autonomous teams who are engaged with their decisions and their success.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vlVYhrMcpTuDLDBczKCy3w.jpeg" /></figure><p>So, are mainframe software development processes compatible with high-performing agile teams? Well, they are if they embrace the open mainframe as a platform that is not much different and special than any other enterprise computing environment. If one is familiar with the latest developments in IT and has the agile mindset, there is no barrier against achieving the same results as those seen in any other big software development house. For this last year that I have been with Broadcom’s Open Mainframe team, I must say it has been an eye-opening experience.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a53a074886d8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/modern-mainframe/software-development-for-the-mainframe-high-performing-teams-a53a074886d8">Software Development for the Mainframe &amp; High-Performing Teams</a> was originally published in <a href="https://medium.com/modern-mainframe">Modern Mainframe</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>