<?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 Carlo Alberto on Medium]]></title>
        <description><![CDATA[Stories by Carlo Alberto on Medium]]></description>
        <link>https://medium.com/@lordkada?source=rss-9cf5c117355a------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*z6-pYSBsVimXdTtB.jpg</url>
            <title>Stories by Carlo Alberto on Medium</title>
            <link>https://medium.com/@lordkada?source=rss-9cf5c117355a------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 19:17:26 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@lordkada/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[Lurch Home: notes from an open-source smart butler — as an AI research playground]]></title>
            <link>https://lordkada.medium.com/lurch-home-notes-from-an-open-source-smart-butler-as-an-ai-research-playground-4d5cc2af432d?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/4d5cc2af432d</guid>
            <category><![CDATA[home-assistant]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[experimentation]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Wed, 03 Sep 2025 09:34:44 GMT</pubDate>
            <atom:updated>2025-09-03T09:34:44.445Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BFuCdcu4HE4tlQDm" /></figure><h3>Lurch Home: notes from an open-source smart butler — as an AI research playground</h3><p><em>Personal lab notes on AI, exploration, and radical simplicity.</em></p><p><a href="https://github.com/lordkada/lurchhome">Lurch Home</a> is an open-source “smart butler”: a minimal, privacy-first assistant that coordinates everyday tasks via natural language, schedules, and simple rules — helping run a household without heavy UIs. Born as a side-project and a nod to the Addams Family’s butler, it now serves as my research playground for agentic AI — covering tool-use protocols, memory, safety, and evaluation — tested under real-world constraints like noisy environments, multiple users, and limited compute.</p><h3>Why the home is a good lab?</h3><p>Homes are messy: overlapping routines, competing intents, interruptions, imperfect sensors, limited resources. That’s exactly why they’re interesting. If ideas hold here — <strong>under constraints</strong> — they tend to travel well to other domains. I’m not chasing demos; I want repeatable behaviors, transparent failure modes, and decisions a user can understand.</p><h3>What I’m exploring (AI-first, tech-agnostic)</h3><p><strong>Tool use &amp; protocols for orchestration</strong><br>How should an assistant decide <em>what</em> to do and <em>which</em> capability to call — reliably? I’m testing agent-to-agent / model-context protocols, planning-and-acting loops, and supervisor/worker patterns to keep behavior explainable while avoiding brittle heuristics.</p><p><strong>Memory that helps, not hallucinates</strong><br>Context spans people, places, schedules, and preferences. I’m experimenting with memory-augmented models, retrieval strategies, decay/refresh policies, and safety filters to keep recollection <strong>useful, bounded, and auditable</strong>.</p><p><strong>Multimodal perception in the wild</strong><br>Audio/video streams are noisy. I’m running ablations on lightweight pipelines, event detection, and uncertainty-aware triggers to minimize false positives and avoid invasive processing. The target is <strong>edge-friendly</strong> perception with minimal data by default.</p><p><strong>Agency with guardrails</strong><br>Autonomy is valuable only if it’s trustworthy. I’m evaluating strategies for prompt-injection resilience, capability scoping, policy checks before execution, and human-in-the-loop confirmations for sensitive actions.</p><p><strong>Evaluation that isn’t a toy</strong><br>Benchmarks are useful; <strong>household tasks</strong> are unforgiving. I’m building task-level evaluation harnesses (success criteria, latency, robustness to perturbations) plus log-driven analysis and A/B of policies, to iterate toward <em>boringly reliable</em> behavior.</p><p>and everything that comes to mind ;-)</p><h3>Method: research by doing</h3><ul><li><strong>Design → prototype → measure → iterate</strong></li><li>Blend <strong>offline tests</strong> (synthetic, controllable) with <strong>in-the-wild trials</strong> (unexpected inputs, multi-user collisions)</li><li>Prefer <strong>small, composable components</strong> so ablations are meaningful</li><li>Document assumptions, expose decisions, and keep a clear kill-switch for any autonomous path.</li></ul><h3>Simplicity as a research constraint</h3><p>A system that only its author can run isn’t useful. Lurch Home treats <strong>simplicity as a first-class constraint</strong>: clean setup, predictable configs, and graceful degradation when parts fail. That forces architectural choices that are easier to reason about — and therefore easier to study.</p><h3>Join in</h3><ul><li><strong>Explore the repo:</strong> <a href="https://github.com/lordkada/lurchhome">github.com/lordkada/lurchhome</a></li><li><strong>Jump into </strong><a href="https://github.com/lordkada/lurchhome/discussions"><strong>Discussions</strong></a><strong>:</strong> bring protocols, models, approaches, methodologies, or papers you’d like to see tested. <strong>Suggest experiments:</strong> memory policies to compare, planning loops to pit against each other, safety checks to formalize, perturbation tests to add</li></ul><p>I’m interested in collaborators who care about <strong>methods</strong> more than any single stack. The goal: move AI from <em>wow</em> to <strong>work</strong>, in a space that keeps us honest.</p><p>You rang? 😉</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4d5cc2af432d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Turning the Page: Saying Goodbye to KVStore.io]]></title>
            <link>https://lordkada.medium.com/turning-the-page-saying-goodbye-to-kvstore-io-cf0a9c562891?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/cf0a9c562891</guid>
            <category><![CDATA[journey-of-life]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[life]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Thu, 22 May 2025 09:01:24 GMT</pubDate>
            <atom:updated>2025-05-22T09:01:24.083Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tluS-0I5XgHsZZ9ZohmFzQ.png" /><figcaption>The KVStore.io homepage</figcaption></figure><p>Today marks a bittersweet milestone: after 10 incredible years, I’ve decided to shut down KVStore.io, a project that started almost by accident and became a meaningful part of my professional and personal journey.</p><p>Back in 2015, when startups were blooming everywhere and AI was nowhere in our daily conversations, I built a simple microservice. Initially, it was nothing more than a personal tool — something to manage configurations within another project I was working on. Then came the idea: why not transform this piece of personal code into something others could use?</p><p>KVStore.io was born out of this thought, evolving from Ruby on Rails in its early days to a robust Java and React architecture in 2020. This journey wasn’t just technical — it was also deeply personal. It represented who I was: someone deeply passionate about coding, building products, and helping others simplify their tech stack.</p><p>Over time, life took me in new directions, particularly into roles that gradually moved me away from daily programming. For a couple of years, I’ve been thinking about this moment. Each time I hesitated, it was because shutting down KVStore.io felt like closing a chapter on part of my identity, a piece of my personal story.</p><p>Today, as I finally make this decision, it’s with gratitude and nostalgia. Gratitude for all I’ve learned, nostalgia for the excitement of those early startup days, and peace knowing this step represents my growth and evolving priorities.</p><p>Thank you to everyone who used KVStore.io, provided feedback, or even just supported me on this journey.</p><p>Closing this service isn’t just an end — it’s another step forward, turning the page in a story that continues to unfold.</p><p>Made with love on Earth — wishing you all the very best in your own journeys.</p><p>CA</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cf0a9c562891" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Neurosymbolic AI: Bridging Perception and Cognition]]></title>
            <link>https://lordkada.medium.com/neurosymbolic-ai-bridging-perception-and-cognition-4ed54e560812?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/4ed54e560812</guid>
            <category><![CDATA[neurosymbolic-ai]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[cognitive-science]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Tue, 02 Jul 2024 14:30:18 GMT</pubDate>
            <atom:updated>2024-07-02T16:19:11.223Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1010/0*N38LgsReJgaVN8vR.png" /><figcaption>“Le Faux Miroir” — René Magritte</figcaption></figure><p>Neurosymbolic AI, a fusion of neural networks and symbolic reasoning, may revolutionize machine intelligence. Inspired by the intricate workings of the human mind, it aligns closely with the groundbreaking ideas of cognitive scientists like Douglas Hofstadter. This cutting-edge approach brings us closer to machines that can think, reason, and understand the world.</p><p>At the core of neurosymbolic AI, is a simple and terrific idea:<strong> the essence of human intelligence unfolds through the dynamic interplay between sensory perception and cognitive processes</strong>.</p><p>My previous article, ”<a href="https://lordkada.medium.com/the-intersection-of-artificial-intelligence-and-cognitive-science-reflections-on-emergent-558dfc153a9f">The Intersection of Artificial Intelligence and Cognitive Science: Reflections on Emergent Intelligence</a>”, highlighted the critical role of abstraction and analogy in human cognition. They allow us to establish connections between seemingly unrelated ideas, facilitating the adaptation to novel situations.</p><h3>The Power of Fluid Concepts</h3><p>Hofstadter’s notion of “fluid concepts” finds a compelling analog in the way neurosymbolic AI systems can navigate between different levels of abstraction. By integrating neural networks with knowledge graphs or other symbolic representations, these systems can shift seamlessly between low-level pattern recognition and high-level reasoning.</p><p>This fluidity allows neurosymbolic AI to tackle problems that have long challenged traditional AI approaches. For instance, in areas like natural language understanding or visual reasoning, neurosymbolic systems can leverage both their ability to process raw sensory data and their capacity for abstract thought, mirroring the way humans effortlessly blend perception and cognition in their everyday interactions with the world.</p><h3>The Procedural-Neural Divide: An Ongoing Challenge</h3><p>While neurosymbolic AI aims to bridge the gap between symbolic and neural approaches, a significant divide persists. The procedural, rule-based nature of traditional reasoning systems often starkly contrasts with neural networks’ more fluid, pattern-based processing. This dichotomy presents both a challenge and an opportunity for researchers in the field.</p><p>On the one hand, symbolic systems excel at explicit, step-by-step reasoning and can provide clear explanations for their decisions — a crucial feature for many real-world applications. They embody the kind of logical thinking that Hofstadter explored in his discussions of formal systems. On the other hand, neural networks shine in their ability to handle ambiguity, recognize complex patterns, and generalize from examples, much like the intuitive leaps that characterize human thought.</p><p>The challenge lies in creating a truly seamless integration of these two approaches. How can we develop systems that can smoothly transition between explicit reasoning and intuitive pattern recognition? How can we ensure that the symbolic and neural components of a neurosymbolic system are not just coexisting, but truly collaborating and informing or evolving each other?</p><h3>Further readings</h3><p>To further explore the concept of Neurosymbolic AI, I recommend referring to this article “<a href="https://arxiv.org/abs/2305.00813">Neurosymbolic AI — Why, What, and How</a>” (arXiv:2305.00813).</p><p>The authors of this paper present a comprehensive overview of Neurosymbolic AI, covering topics such as:</p><ul><li>The integration of neural networks and symbolic reasoning</li><li>The representation and manipulation of knowledge</li><li>The development of hybrid architectures</li><li>The application of Neurosymbolic AI to various domains</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4ed54e560812" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Intersection of Artificial Intelligence and Cognitive Science: Reflections on Emergent…]]></title>
            <link>https://lordkada.medium.com/the-intersection-of-artificial-intelligence-and-cognitive-science-reflections-on-emergent-558dfc153a9f?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/558dfc153a9f</guid>
            <category><![CDATA[mind]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[cognitive-science]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Tue, 18 Jun 2024 15:41:11 GMT</pubDate>
            <atom:updated>2024-06-18T15:41:11.725Z</atom:updated>
            <content:encoded><![CDATA[<h3>The Intersection of Artificial Intelligence and Cognitive Science: Reflections on Emergent Intelligence</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/805/1*jFlxb5x7BtvrvzdySIfgNw.jpeg" /><figcaption><strong>Swans (M.C Escher, 1956)</strong></figcaption></figure><h3>Introduction</h3><p>As someone with an MSc in Computer Science and a longstanding fascination with cognitive science, particularly influenced by thinkers like Douglas Hofstadter and Daniel Dennett, I often ponder the intriguing parallels between Artificial Intelligence (AI) and cognitive science.</p><p>In this post, I’d like to explore some of these connections, particularly through the lens of large language models (LLMs) and the concept of emergent intelligence.</p><h3>Emergence of Intelligent Behavior</h3><p>One of the most captivating ideas in both cognitive science and AI is the emergence of intelligent behavior from seemingly simple interactions. In cognitive science, this is often discussed in terms of how complex mental states and processes emerge from the intricate workings of the brain’s neural networks. Similarly, in AI, we see how sophisticated behaviors and capabilities arise from the layered architectures and vast datasets that underpin models like GPT-4.</p><p>Douglas Hofstadter’s work, especially in “Gödel, Escher, Bach: An Eternal Golden Braid,” introduces how complex systems can give rise to intelligent behavior through simple, recursive processes. He uses the metaphor of strange loops to describe how self-referential systems can exhibit emergent properties, a concept that resonates deeply with the statistical emergence of intelligence in AI.</p><p>In “Fluid Concepts and Creative Analogies,” Hofstadter further delves into the mechanisms of human thought, emphasizing the fluidity and flexibility of our cognitive processes. He explores how analogies and metaphors are fundamental to human thinking, allowing us to draw connections between seemingly disparate concepts. This fluidity of thought is crucial for creativity and problem-solving, illustrating how our minds can navigate complex and abstract ideas “easily”.</p><p>In Artificial Intelligence, this concept is reflected in how models process and generate language. LLMs through their training on diverse and extensive datasets, develop a kind of “fluency” in their ability to generate contextually relevant and coherent texts. The analog thinking described by Hofstadter is similar to how Artificial Intelligence can generate creative responses based on learned patterns and structures. By processing large amounts of text data on multiple levels, LLMs can generate human-like text, understand context, and even exhibit creativity, just as our brain draws on past experiences and knowledge to form new ideas and solutions.</p><p>The statistical nature of how these models learn and operate also reflects the principles discussed in Hofstadter’s earlier work. Just as strange loops and recursive structures can lead to emergent complexity, the statistical patterns recognized and generated by LLMs lead to the emergence of coherent and contextually appropriate behaviors. This demonstrates how both human cognition and AI rely on fundamental processes that, while simple in isolation, combine to produce intelligent and adaptable behaviors.</p><h3>Cognitive Architectures and AI</h3><p>The architecture of the human brain has been a source of inspiration for AI development. Cognitive science delves into understanding the structures and functions of the brain, exploring how different regions and neural pathways contribute to various cognitive functions. This understanding has influenced the design of neural networks in AI.</p><p>For instance, recurrent neural networks (RNNs) and transformers, like those used in LLMs, are inspired by the brain’s ability to process sequences of information and maintain context over time. The self-attention mechanism in transformers allows these models to weigh the importance of different words in a sentence, akin to how our brain focuses on relevant stimuli in our environment.</p><h3>Learning and Adaptation</h3><p>Learning is another critical area where AI and cognitive science intersect. In cognitive science, learning involves the brain’s ability to adapt based on new information, forming memories, and refining skills. Similarly, AI models learn from data, adjusting their parameters to improve performance on specific tasks.</p><p>Daniel Dennett’s work on the philosophy of mind often touches on the notion of intentionality and the evolution of cognitive processes. In AI, this is reflected in the way models are fine-tuned and optimized. While current AI lacks true intentionality, the iterative process of training and adjusting AI models mimics the adaptive nature of human learning.</p><h3>Interpretable Features and Emergent Behavior</h3><p>A key aspect of LLMs is the emergence of interpretable features from large-scale training on diverse datasets. These features can be specific and understandable, similar to well-defined concepts in human thought as described by Hofstadter. For example, LLMs can develop features that trigger specific responses to certain inputs, demonstrating intelligent behavior emerging from statistical phenomena. This mirrors how our minds create and use interpretable concepts to navigate the world.</p><p>Moreover, the unsupervised learning capabilities of LLMs allow them to discover abstractions and unexpected associations, akin to how the human brain can form new connections and analogies without explicit supervision. This ability to learn and adapt in unforeseen ways is crucial for AI and human cognition, highlighting the fluidity and creativity in problem-solving.</p><h3>Ethical Considerations and Limitations</h3><p>Despite the advancements, AI and cognitive science recognize their limitations and ethical implications. In cognitive science, understanding consciousness and subjective experience remains a profound challenge. Similarly, AI grapples with issues of explainability, bias, and ethical use.</p><p>The works of Hofstadter and Dennett remind us that while we strive to emulate and understand intelligence, we must remain aware of the philosophical and ethical dimensions. As we develop more advanced AI systems, it’s crucial to consider their impact on society, ensuring they are designed and used responsibly.</p><h3>In conclusion</h3><p>The intersection of AI and cognitive science offers a rich tapestry of ideas and insights. The emergent properties of intelligent behavior, the architectural inspirations, the learning processes, and the ethical considerations all point to a deep connection between these fields. As we continue to advance AI, and our understanding of the human mind, the dialogue between these two disciplines will undoubtedly yield further fascinating discoveries and innovations.</p><p>Thank you for joining me on this reflective journey. I look forward to exploring these concepts further and engaging with your thoughts and perspectives in the comments.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=558dfc153a9f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My evolution from entrepreneurship to intrapreneurship]]></title>
            <link>https://lordkada.medium.com/my-evolution-from-entrepreneurship-to-intrapreneurship-54f96ab75e7b?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/54f96ab75e7b</guid>
            <category><![CDATA[longlife-learning]]></category>
            <category><![CDATA[careers]]></category>
            <category><![CDATA[innovation]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Wed, 15 Dec 2021 09:51:50 GMT</pubDate>
            <atom:updated>2021-12-15T09:51:50.532Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PHPJlMSTushbEDxJ4nAZ0Q.jpeg" /></figure><p>I’m thrilled to announce that today I joined the brand new <strong>Strategic Innovation department at </strong><a href="https://www.pagopa.gov.it/"><strong>PagoPA SpA</strong></a>, a state-owned company devoted to innovating digital services in Italy and supervised by the Prime Minister.</p><p>For all of you who know me and my adventures, this is a very big leap in my professional and personal life; it looks much more than just “changing my skin”.</p><p>I have thought a lot about this so highly impacting metamorphosis, but I suddenly realized that is not the first time in my life.</p><p>In fact, it took different shapes in the past. For example, when I’ve:</p><ul><li>started my first business and I still was at the University (from student to entrepreneur),</li><li>turned me (and my ventures) into a “software product man” (from consulting to software production),</li><li>flown to the United States and then started my business there (from the “comfort zone” of the place where you were born and raised to a completely “new world”).</li></ul><p>All of these moments marked a dramatic change in my life. More than I initially understood and much more long-lasting than expected.</p><p>Every single time, I needed to put my skills at the challenge and learned new stuff, faced my fears, embraced the change, and re-framed my mindset. I had to adapt day by day, in every single situation, continuously happening and re-happening. I have unexpectedly become a lifelong learner.</p><p>And today, this is coming again.</p><p>Welcome change, welcome life.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=54f96ab75e7b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AngelBrain, a cyber salad for the Raspberry PI]]></title>
            <link>https://lordkada.medium.com/angelbrain-a-cyber-salad-for-the-raspberry-pi-67566a4708ef?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/67566a4708ef</guid>
            <category><![CDATA[security]]></category>
            <category><![CDATA[neural-networks]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[raspberry-pi]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Wed, 10 Feb 2021 17:51:16 GMT</pubDate>
            <atom:updated>2021-02-11T05:24:10.773Z</atom:updated>
            <content:encoded><![CDATA[<p>Today, I want to write about my latest side project, AngelBrain, a sort of guardian of my house built on a very cheap and constrained hardware, a Raspberry PI 3 B+.</p><h3>Preface</h3><p>I’ve always been looking for something new to create, or to build since I was very young. One of the earliest creations I remember was a red lamp, mounted on the rear of my bicycle, turning on when I braked. I was 10 years old and it was so raw and fragile. Nevertheless, it caught the interest of the kids in the neighborhood, so I spent the next two weeks “enhancing” their bikes too. Not so bad for my pockets.</p><p>In the following years, I have found myself working in “many variants&quot; of that red braking light. Some of them were just “side projects&quot;, others became real businesses. In all these cases, my desire to build something was (and still is) the actual propeller; and I’ve been always learning a lot.</p><p>And even this time, I had a lot of fun and learned so much :-)</p><h3>The idea</h3><p>In recent years, there has been a huge increase in the supply of surveillance cameras you can buy and install on your own. For a few bucks, you may get great HD devices transmitting videos straight to your mobile phone, wherever you are. Indeed, their presence has enormously grown in our cities and homes: I installed 4 of them to cover the perimeter when I moved to my new house!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/273/0*di0XGwFwG1YwNHIj" /><figcaption>Security cameras you can buy with “pocket money”</figcaption></figure><p>Usually, these cameras can be configured with a dedicated app. Most of them are nowadays integrated with Google or Alexa, and they have their own premium services, like the ability to save the video stream in the cloud other than leveraging a local SD memory.</p><p>One of the most interesting features, at least for my own needs, are “triggers” based on motion detection. Basically, if “something moves” then your camera can start recording or may send an alert. When you leave your house, eg. for your holidays, you will be alerted in the unfortunate event someone would break inside.</p><p>But this great functionality miserably fails if you live with a pet or you put these devices outside, as I did. In fact, a falling leaf, a quick change in the lights, or a flying bee in front of their lenses would suddenly trigger an alert.</p><p>Believe me, I was flooded by a lot of false alarms. So I deactivated these triggers and started periodically checking my cams. In the very beginning even more times per day, but quickly a lot less frequently…!</p><p><strong>So, the idea popped out of my mind</strong>: let’s build something watching my cameras and alerting me only in very specific situations, like a man showing up… if I had this, I could considerably reduce or even eliminate any false alarm!</p><p>I was so thrilled, I quickly started to think about it and I realized I’d need to put together something able to read the camera video streaming, analyze frames, and communicate, maybe sending messages?</p><p>The idea for “AngelBrain”, my new project, was set.</p><h3>The software</h3><p>In the following spare time, I started implementing the “core” of AngelBrain, a <em>C++</em> program able to read a video streaming and classify the frames using a neural network.</p><p>Since I didn’t want to make this code too complex, I also decided to send these raw events, the classifications, to a <em>Java</em> server; the latter would be then responsible to filter, aggregate the events, and ultimately communicate with a simple Telegram chat.</p><p>The “core” design is quite simple and its key components are:</p><ul><li>An RTSP video stream <strong>grabber</strong> based on the <a href="https://ffmpeg.org/">FFmpeg</a> library (I started with GStreamer but I’ve found it too complex for my purposes)</li><li>A <strong>classifier</strong> built with <a href="https://opencv.org/">OpenCV</a> and <a href="https://pjreddie.com/darknet/yolo/">YOLO V3</a> neural network</li><li>A <strong>messenger</strong> that communicates with the Java server by leveraging the MQTT protocol provided by the <a href="https://www.eclipse.org/paho/">PAHO MQTT</a> library</li></ul><p>I was able to have a prototype of the software running on my PC in a matter of few week-ends.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/639/1*-gGIkCOvqgeimhhE9ehemQ.png" /><figcaption>AngelBrain core at work</figcaption></figure><p>The <em>Java</em> server, hosted on <a href="https://www.heroku.com/">Heroku</a>, receives these events, applies some basic logic, and sends the alert in case of a “significative event” as a Telegram chat message.</p><p>The main blocks of the Java server are:</p><ul><li>the free MQTT broker add-on from Heroku</li><li>a message interpreter from the “core”, with some basic and configurable filtering logic</li><li>the <a href="https://github.com/pengrad/java-telegram-bot-api">Java client</a> to Telegram Bot API</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_GhQ729E_Hnjm0lNLE3y9w.png" /><figcaption>Some of the events received by the MQTT broker on Heroku</figcaption></figure><p>The prototype version of my AngelBrain was ready, it was smoothly running on my computer… but I wanted something more… I just wanted to have this code running on somewhere else than my computer (I am talking about the “core”, so the grabber and the neural network classifier)…</p><p>… and this is where I have met the true challenge!</p><h3>Porting AngelBrain to the Raspberry PI 3 B+</h3><p>Just after I run the prototype of AngelBrain, I was playing with a Raspberry PI, nicely boxed and I started thinking it was the “perfect platform” to run my software: a low consumption energy device, super compact, and very cheap! Something that can be turned on and simply forgotten!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0r2nM-aaQgD-CdIC6geIMw.jpeg" /><figcaption>My Raspberry PI 3 B+, nicely boxed, equipped with a Broadcom BCM2837B0, Cortex-A53 (ARMv8) 64-bit, VideoCore IV, SoC running at 1.4GHz, and 1GB LPDDR2 SDRAM.</figcaption></figure><p>But how to make it really work? I needed to take my C++ code, the libraries, compile everything for that platform, add the neural network weights. I felt I had to squeeze and save even the single clock pulses and bits of my AngelBrain… and so I did, but it was not an easy trip!</p><p>My first move was to kick out the Raspbian OS and replace it with a custom kernel, highly optimized. A friend of mine suggested <a href="https://www.yoctoproject.org/">Yocto</a>:</p><blockquote>The project provides a flexible set of tools and a space where embedded developers worldwide can share technologies, software stacks, configurations, and best practices that can be used to create tailored Linux images for embedded and IOT devices, or anywhere a customized Linux OS is needed.</blockquote><p>Yocto is a beast. It’s very powerful and potentially limitless, but it comes at a price: the learning curve is very steep, it requires a lot of time, and building your configuration looks like “trial and error.” At least in my own experience.</p><p>Then you have to deal with the bundled version of the libraries. It happened that Yocto was shipping older versions. Sometimes it’s not a big problem. Sometimes it is huge.</p><p>After some time spent tuning the versions, rebuilding on my computer, testing, I was finally able to build my first OS image, flash the SD card, and turn on my RPI.</p><p>Ta-daaaa…. my trip to Hell was just beginning…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/340/0*Uv7RB2aywSoqvJps.jpg" /></figure><p>Yes, because I experienced so many problems, core-dumps, and unexpected behaviors that I was on the verge of giving up several times.</p><p>The first biggest problem was with GStreamer. The pipeline I was smoothly running on my PC just silently died on the RPI. I tried so many different things and approaches, but they never worked and I was not really able to replicate the problem out of my RPI. I spent weeks reading documentation, browsing forums, formulating hypotheses, trying to create some code to test. Nothing.</p><p>In the end, I was so desperate that I decided to re-implement the video grabber from scratch moving to FFmpeg. But this chapter was so exhausting, that I put the project on hold for several months. I had to recharge my batteries.</p><p>Then I got new energy and started again.</p><p>After re-implementing part of my AngelBrain to introduce FFmpeg, I had to review the set of Yocto configurations and recipes to purge GStreamer (the easy part) and to include FFmpeg. In the meantime, a new version of Yocto was released: Gatesgarth. I decided to update it too. Luckily it wasn’t so hard.</p><p>Something started working, the software didn’t die, but it was not running as expected. At least for the effort that is costed: it was not detecting anything.</p><p>I started debugging again… and found the source of the new problem.</p><p>The RPI is very slow. When I say slow, I really mean it. And decoding an RTSP video stream requires some “juice”: you can’t think to be late in calling the FFmpeg methods to read the stream. Otherwise, everything literally turns into a marmalade. And your neural network won’t work. Simple as this. Period.</p><p>So I had to re-think my “core” again, splitting the video capturing code from the neural network classification, introducing a multi-threaded producer/consumer pattern.</p><p>New debug, tests, compilations, and image build. Luckily this time it went better… much better: my RPI started recognizing and sending events to the <em>Java</em> server :-)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mAcOiD48YImRgnx7UnyvVA.png" /><figcaption>Inside my RPI 3B+ showing the dump of classified events</figcaption></figure><p>Ok, it’s very slow, a few frame recognitions per minute, but it’s fairly enough if you think that the RPI 3 B+ costs about 40$ :-)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=67566a4708ef" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Ready for a quick tour of the Java SDK for KVStore.io?]]></title>
            <link>https://lordkada.medium.com/ready-for-a-quick-tour-of-the-java-sdk-for-kvstore-io-4adbbaedef4e?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/4adbbaedef4e</guid>
            <category><![CDATA[api]]></category>
            <category><![CDATA[java]]></category>
            <category><![CDATA[sdk]]></category>
            <category><![CDATA[storage]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Mon, 06 Jul 2020 18:06:05 GMT</pubDate>
            <atom:updated>2020-07-07T05:11:54.985Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*g6KuPlSn8qaAqoMY" /><figcaption>The Java SDK for KVStore.io is on <a href="https://github.com/kvstore-io/sdk-java">GitHub.com</a></figcaption></figure><p><strong>TL; DR</strong> <em>You will learn to interact with KVStore.io, storing and retrieving data, by writing a very few lines of Java code thanks to the provided SDK.</em></p><p><strong>Disclaimer</strong></p><p>The current version of the KVStore.io Java SDK library is <strong>Release Candidate (RC)</strong>. Please, refer to the GitHub repository to report issues (see the links at the bottom of this page)</p><p><strong>Introduction</strong></p><p><a href="https://www.kvstore.io/">KVStore.io</a> is a simple storage service based on a “key/value” data modeling: you can easily organize data in <em>dictionaries</em>, where your <em>keys</em> map to <em>values, </em>both of which are simple strings. Furthermore, you may create one or more dictionaries that are called <em>collections.</em></p><p>Although very simple, this approach is very flexible and powerful, because it does not imply any specific data format or use-case scenario. You could use it to store simple sequences of characters read from a hardware sensor or complex JSON structures generated from a micro-service.</p><p>KVStore is data agnostic and so its interfaces are (RESTful API/SDK). For further details, see the <a href="https://www.kvstore.io/#/documentation">documentation</a>.</p><p><strong>Obtain your secret token</strong></p><p>To start using KVStore.io, you need to <a href="https://www.kvstore.io/#/signup">signup</a> for the service and to obtain your secret token (API key). Don’t worry, the “forever free” plan is more than enough to start learning or for simple use cases.</p><p>Once logged, you locate your secret token in your private home page, like in the image below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*UCrVLaBmSe9n8GEq.png" /></figure><p><strong>Create your Java program</strong></p><p>Let’s start by getting the KVStore Java SDK. Of course, the easiest way to install it is by using a <em>build automation tool</em>, like Maven or Gradle. Otherwise, you will need to download the <a href="https://repo1.maven.org/maven2/io/kvstore/sdk-java/1.0.0-RC/sdk-java-1.0.0-RC.jar"><em>jar</em></a><em> </em>and later include it in your classpath while compiling and executing your code (eg. <em>javac -cp &lt;path_to_sdk/sdk-java-1.0.0-VERSION.jar&gt; Main.java</em>).</p><p>For Maven, add this dependency to your pom.xml:</p><pre>&lt;dependency&gt;<br>    &lt;groupId&gt;io.kvstore&lt;/groupId&gt;<br>    &lt;artifactId&gt;sdk-java&lt;/artifactId&gt;<br>    &lt;version&gt;1.0.0-RC&lt;/version&gt;<br>&lt;/dependency&gt;</pre><p>For Gradle, add the following to your build.gradle:</p><pre>compile group: &#39;io.kvstore&#39;, name: &#39;sdk-java&#39;, version: &#39;1.0.0-RC&#39;</pre><p>For the sake of simplicity, let’s create a simple Java application from scratch using Maven. Open your terminal and paste this command:</p><pre>mvn archetype:generate -DgroupId=io.kvstore.demo -DartifactId=sdk-demo -Dversion=1.0-SNAPSHOT -DarchetypeArtifactId=maven-archetype-quickstart</pre><p>It will create a folder “sdk-demo” containing an empty app just after you will confirm the operation by typing “Y”.</p><pre>$ mvn archetype:generate -DgroupId=io.kvstore.demo -DartifactId=sdk-demo -Dversion=1.0-SNAPSHOT -DarchetypeArtifactId=maven-archetype-quickstart -q<br>Confirm properties configuration:<br>groupId: io.kvstore.demo<br>artifactId: sdk-demo<br>version: 1.0-SNAPSHOT<br>package: io.kvstore.demo<br> Y: : Y</pre><p>Now jump into the new folder “sdk-demo”, edit the <em>pom.xml </em>file using your preferred editor, and paste the KVStore SDK dependency just between the <em>&lt;dependencies&gt;</em> tag.</p><p>Your final <em>pom.xml</em> should look like:</p><pre>&lt;project xmlns=&quot;<a href="http://maven.apache.org/POM/4.0.0">http://maven.apache.org/POM/4.0.0</a>&quot; xmlns:xsi=&quot;<a href="http://www.w3.org/2001/XMLSchema-instance">http://www.w3.org/2001/XMLSchema-instance</a>&quot;<br>  xsi:schemaLocation=&quot;<a href="http://maven.apache.org/POM/4.0.0">http://maven.apache.org/POM/4.0.0</a> <a href="http://maven.apache.org/maven-v4_0_0.xsd">http://maven.apache.org/maven-v4_0_0.xsd</a>&quot;&gt;<br>  &lt;modelVersion&gt;4.0.0&lt;/modelVersion&gt;<br>  &lt;groupId&gt;io.kvstore.demo&lt;/groupId&gt;<br>  &lt;artifactId&gt;sdk-demo&lt;/artifactId&gt;<br>  &lt;packaging&gt;jar&lt;/packaging&gt;<br>  &lt;version&gt;1.0-SNAPSHOT&lt;/version&gt;<br>  &lt;name&gt;sdk-demo&lt;/name&gt;<br>  &lt;url&gt;<a href="http://maven.apache.org/">http://maven.apache.org</a>&lt;/url&gt;<br>  &lt;dependencies&gt;<br><strong>    &lt;dependency&gt;<br>      &lt;groupId&gt;io.kvstore&lt;/groupId&gt;<br>      &lt;artifactId&gt;sdk-java&lt;/artifactId&gt;<br>      &lt;version&gt;1.0.0-RC&lt;/version&gt;<br>    &lt;/dependency&gt;</strong><br>    &lt;dependency&gt;<br>      &lt;groupId&gt;junit&lt;/groupId&gt;<br>      &lt;artifactId&gt;junit&lt;/artifactId&gt;<br>      &lt;version&gt;3.8.1&lt;/version&gt;<br>      &lt;scope&gt;test&lt;/scope&gt;<br>    &lt;/dependency&gt;<br>  &lt;/dependencies&gt;<br>&lt;/project&gt;</pre><p>Edit now the file <em>src/main/java/io/kvstore/demo/App.java</em> and paste (please, remember to replace the PUT_YOUR_API_KEY with your own secret token):</p><pre>package io.kvstore.demo;</pre><pre>import io.kvstore.sdk.KVStore;<br>import io.kvstore.sdk.clients.CollectionsClient;<br>import io.kvstore.sdk.clients.ItemsClient;<br>import io.kvstore.sdk.clients.KVStoreClient;<br>import io.kvstore.sdk.exceptions.KVStoreException;</pre><pre>public class App {</pre><pre>public static void main(String[] args) {<br>        KVStoreClient kvStore = KVStore.instance(&quot;PUT_YOUR_API_KEY&quot;);</pre><pre>CollectionsClient collectionsClient = kvStore.collectionsClient();</pre><pre>collectionsClient.create(&quot;test&quot;);<br>        ItemsClient itemsClient = kvStore.itemsClient();</pre><pre>itemsClient.put(&quot;test&quot;, &quot;mykey&quot;, &quot;my value!&quot;);<br>        ItemsClient.ItemValue item = itemsClient.get(&quot;test&quot;, &quot;mykey&quot;);</pre><pre>System.out.println(item);</pre><pre>}</pre><pre>}</pre><p>The code shows the basic operations:</p><ol><li>instance the client by calling the static instance method and assign to the <em>kvStore</em> variable (KVStoreClient kvStore = KVStore.instance(&quot;PUT_YOUR_API_KEY&quot;))</li><li>obtain the collections handler (CollectionsClient collectionsClient = kvStore.collectionsClient())</li><li>create the “test” collection (collectionsClient.create(&quot;test&quot;))</li><li>obtain the items handler (ItemsClient itemsClient = kvStore.itemsClient())</li><li>store “my value!” with the key “mykey” (itemsClient.put(&quot;test&quot;, &quot;mykey&quot;, &quot;my value!&quot;))</li><li>read back its value and assign to the <em>item</em> variable (ItemsClient.ItemValue item = itemsClient.get(&quot;test&quot;, &quot;mykey&quot;))</li><li>print the value of the <em>item</em> variable to the console</li></ol><p>Now we’re very close to the finish line! We just need to compile the code:</p><pre>$ mvn compile</pre><p>and then execute it:</p><pre>$ mvn exec:java -Dexec.mainClass=&quot;io.kvstore.demo.App&quot; -q</pre><pre>Item(value=my value!, created_at=Mon Jul 06 19:29:15 CEST 2020, updated_at=Mon Jul 06 19:29:15 CEST 2020)</pre><p>Yeah, as expected we can see the output of our System.out :-)</p><p><strong>Conclusions</strong></p><p>In this short tutorial, we have installed the Java SDK thanks to Maven and created a sample app connecting and leveraging some of the KVStore.io features, such as creating a collection, storing and then retrieving some data.</p><p>Useful links:</p><ul><li>The <a href="https://www.kvstore.io/#/documentation/sdk-java">official documentation</a> on the main website,</li><li>GitHub repository of the <a href="https://github.com/kvstore-io/sdk-java">Java SDK</a> (report issues here)</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4adbbaedef4e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[KVStore.io, a letter for my beloved users :-)]]></title>
            <link>https://lordkada.medium.com/kvstore-io-a-letter-for-my-beloved-users-9eccd13d7d29?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/9eccd13d7d29</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[storage]]></category>
            <category><![CDATA[serverless]]></category>
            <category><![CDATA[developer]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Thu, 25 Jun 2020 14:22:03 GMT</pubDate>
            <atom:updated>2020-06-26T07:19:59.807Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/852/1*R9aaMYc_Oo24N24nmKCc6Q.jpeg" /></figure><h3>KVStore.io, a letter to my beloved users :-)</h3><p>After 5 years of honored but concealed service, I have some news about <a href="https://www.kvstore.io/">KVStore.io </a>that I’m thrilled to share with you.</p><p>I hope you will enjoy it.</p><p>Good reading,<br>Carlo Alberto (aka <a href="https://twitter.com/lordkada">lordkada</a>)</p><h3>New “everything”</h3><p>If you recall the old “design”, it was the web version of a computer terminal showing a shell. The logo was drawn with simple characters, the website, and even its functionalities were available as simple commands typed from the keyboard. I loved that approach a lot :-)</p><p>And I was not alone, because, during this long period, I have collected many positive feedbacks. That “design” was recognized as really distinctive and with a “character”.</p><p>But among these appreciations, I have also noticed people leaving the website with a sense of frustration; they simply didn’t have enough time to spend, or the will, with a service that was simply looking like a “hack”.</p><p>Furthermore, I was relying a lot on the RapidAPI platform (formerly known as Mashape) to deliver API, docs, and offer commercial plans, but in the end, it didn’t fit my users’ needs, ultimately you!</p><p>For all the reasons above (and an extra time due to the Covid-19 lockdown we have suffered here in Italy), I have decided to re-design the whole service and publish the brand new website: <a href="https://www.kvstore.io/">https://www.kvstore.io</a>.</p><p>There, you will find all the <a href="https://www.kvstore.io/#/documentation">documentation</a>, the <a href="https://www.kvstore.io/#/login">login</a> to your private dashboard and the set of subscription <a href="https://www.kvstore.io/#/pricing">plans</a> currently available.</p><h3>What about RapidAPI?</h3><p>If you have signed KVStore.io through RapidAPI, don’t worry you may continue to use it as you have done in the past. I will continue to support the traffic coming from that source.</p><p>Unfortunately, there’re some bugs (?!) on their platform that are preventing me to update my endpoints documentation.</p><p>Moreover, they have considerably limited the quantity and quality of documentation I can post; I can’t create “wiki” pages or have a real announcements section. I just can’t, but my users need it!</p><p>So I want to strongly suggest all of you referring to the KVStore.io website for the <a href="https://www.kvstore.io/#/documentation">documentation</a> (guides, tutorials, quickstart) and API. They will be updated to the best of my efforts.</p><h3>New plans</h3><p>Along the “forever free”, I have added the “basic” and the “pro” <a href="https://www.kvstore.io/#/pricing">plans</a>. These paid plans provide some new features, like the “public read” or the “customer support” (more coming in the future), and are the tangible way to actively support this project and to make it live “forever”.</p><p>Please take note that there’s also a plan reserved for no-profit organizations (not listed in the pricing page, there’s only a footnote); it’s equivalent to the “basic” plan, but it’s free. To subscribe to this plan, you need to contact me in private.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9eccd13d7d29" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Letterina a Babbo Natale Sindaco (di Milano)]]></title>
            <link>https://lordkada.medium.com/letterina-a-babbo-natale-sindaco-di-milano-87d02c7e0654?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/87d02c7e0654</guid>
            <category><![CDATA[responsabilita]]></category>
            <category><![CDATA[natale]]></category>
            <category><![CDATA[milano]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Wed, 09 Dec 2015 10:58:47 GMT</pubDate>
            <atom:updated>2015-12-09T16:23:33.526Z</atom:updated>
            <content:encoded><![CDATA[<p>Caro Sindaco,</p><p>con l’avvicinarsi del Natale quest’anno ho deciso anche io di spedire la mia letterina e così, come fanno molti bimbi, anche io ho impugnato “carta e penna” per scriverla.</p><p>Solo che invece di mandarla a Babbo Natale, ho deciso di inviarla a te che ti considero un po’ il papà di tutti noi cittadini. E ti chiedo subito scusa per il tono colloquiale, ma penso che me lo perdonerai. Infatti dietro questo “tu” sono sicuro che riconoscerai tutto il profondo rispetto che sinceramente mi anima.</p><p>Devo anche confessarti che sono stato davvero combattuto se scrivertela, un pò perché sono tanti anni che non ne mando una ed un po’ anche perché, come dico sempre ai miei due bimbi, guardando al proprio “cuoricino” dobbiamo sentire di meritarci almeno un po’ quello che chiediamo. Perlomeno negli intenti, quelli più sinceri.</p><p>Ma poi, ancora dubbioso, ho anche pensato che quello che ti voglio chiedere, ciò che desidero, non è un dono per me, almeno non solo. Ho quindi preso il coraggio di scriverti ed eccomi qui.</p><p>Prima però di chiederti cosa desidero più di ogni altra cosa per quest’anno, fammi raccontare un piccolo episodio e ti chiedo scusa se ti rubo troppo tempo. Ma per me è davvero importante per farti comprendere meglio quello che ti voglio chiedere.</p><p>Ti esorto quindi di leggere con la pazienza del “padre”.</p><p>Questa mattina, proprio pochi minuti prima dell’apertura dei cancelli della nostra scuola posti in una strettissima stradina, tanto piccola da essere adatta ad elfi, gnomi oltre che ai nostri piccoli bambini, si sono affollati operai e mezzi per il rifacimento del manto stradale (quella stessa stradina che vede andare avanti ed indietro tante auto di Vigili Urbani e di grandi camion, quelli con le ruote grandi grandi che piacciono tanto ai miei due piccoli mostri; già, perché in quel viottolo c’è probabilmente un distaccamento dell’acquedotto e chissà cosa altro, visti i mezzi che transitano).</p><p>Preoccupato di quanto stesse accadendo, riesco ad aprire il cancello di ingresso in modo che i bambini potessero entrare nel cortile interno invece che accumularsi tra camion e ruspe. Cerco quindi i bidelli o qualcuno per segnalare la situazione, magari per aprire un altro ingresso sul fronte-strada, sfortunatamente non trovo nessuno. Mi dirigo allora di gran carriera al parcheggio interno dei Vigili Urbani certo di trovare qualcuno a cui riferire la situazione. Proprio là dove me li ero immaginati, trovo proprio un paio di Agenti che stavano conversando a meno di 5 metri di distanza.</p><p>Vedi caro Sindaco, sinceramente credo di essere una persona ben educata e credo di essermi rivolto alle Agenti in modo adeguato. Purtroppo la reazione è stata particolare, quantomeno imbarazzante, riferito proprio da alcuni genitori che avevano assistito alla scena. Ad ogni modo ed in fin dei conti spero che sia stata colpa mia e delle mie scarse doti di comunicazione. Mi tranquillizza davvero pensarla così.</p><p>Ma in tutta questa storia ahimè, c’é un nodo fondamentale, un <em>leitmotiv</em> che costantemente riconosco al di là del singolo episodio e che mi ha spinto a scriverti ed a vincere tutte le mie diffidenze.</p><p>Si tratta di un aspetto essenziale, una cosa che cerchiamo di instillare nei nostri figli con un grande “sforzo educativo” nonostante i numerosissimi e quotidiani contro-esempi che purtroppo incontrano in ogni angolo della nostra città, sin dentro le mura scolastiche.</p><p>Sono convinto che sarai anche tu d’accordo con me.</p><p>Per questo Natale, mio caro Sindaco, ti chiedo in dono per tutti noi un regalo vero e sincero:</p><p>il senso di responsabilità.</p><p>Grazie tante,<br>Carlo Alberto Degli Atti</p><p>PS: davvero mi auguro che tu riceva in tempo questa mia letterina</p><p>PS2: spero che i folletti tuoi aiutanti abbiano fatto scorta di questa dote così rara</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=87d02c7e0654" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Store user-generated content from Jekyll/GitHub Pages or similar…]]></title>
            <link>https://lordkada.medium.com/store-user-generated-content-from-jekyll-github-pages-or-similar-ded76aabf17?source=rss-9cf5c117355a------2</link>
            <guid isPermaLink="false">https://medium.com/p/ded76aabf17</guid>
            <category><![CDATA[api]]></category>
            <category><![CDATA[jekyll]]></category>
            <category><![CDATA[github]]></category>
            <dc:creator><![CDATA[Carlo Alberto]]></dc:creator>
            <pubDate>Fri, 20 Nov 2015 10:55:42 GMT</pubDate>
            <atom:updated>2018-11-06T08:22:07.318Z</atom:updated>
            <content:encoded><![CDATA[<p>Hey, are you enthusiastic about developing your brand new website using a static content generator like <a href="https://jekyllrb.com/">Jekyll</a> or <a href="https://gohugo.io/">Hugo</a>?</p><p>And what about replacing your old-fashioned blogging engine to leverage the lean and clean, off-the-shelf <a href="https://pages.github.com/">GitHub Pages</a>?</p><p>Great… I can feel your excitation… I really do! The world is fast changing and the pace is accelerating… like your heart (and mine too :-)) !</p><p>But.. wait…</p><p>Did you ever consider that maybe in the near future you could need something like a simple contact form or a feedback page? And with a static site, not backed-up by a piece of code and a sort of database.. hmm… maybe you’re on a slippery slope… ;-)</p><p>“<em>PLEASE STOP DUDE</em>, <em>I DIDN’T COME DOWN WITH THE LAST DROP OF RAIN!</em>&quot; (I swear I can feel your thoughts, solid as rocks…)… is this something you’re thinking right now?</p><p>Yes I know, Google is your friend, you can just type something like “Jekyll form” or whatever you picked up… and you’re done.. just flooded with a tons of solutions… awesome!</p><p>But please (there’s always a moment like this… I love it!), stop again!</p><p>Just spend a few of seconds and let me ask you this question: are you really gonna replacing your carbon-fuelled steam train with something from the stone age? Is that true?</p><p>Ok, if you’re good with that, I’m fine too… really… just go on… it was really a pleasure to meet you and sorry for taking your time!</p><p>But (second time guys… I’m a very adversative person) if not, maybe you are curious enough to deserve more… really!</p><p>This is why I created <a href="http://www.kvstore.io/">kvstore.io</a>, a super simple key/value storage service with a cool feature called “<em>client-side-only environment</em>”: by activating this mode and injecting a single line of code in your HTML page, you can then easily save any kind of user-data coming from your static website to a real storage engine!</p><p>And do you know what does it mean having “true storage” in your hands instead of a just <em>easy-but-no-more-than-a-web-form</em>?</p><p>Let me give you some examples:</p><ul><li>organise your data in collections, e.g. to differentiate the user-generated contents from your private configuration settings</li><li>manage (query/update or do whatever you want) your data using our RESTful API, leveraging the language/environment you love</li><li>store data using different formats (e.g. a JSON for a feedback page and a plain-text for a contact form)</li><li>pre-process data before storing (e.g. client-side encryption)</li></ul><p>and much, much more…</p><p>So, what’s next?</p><p><strong>Easy… go to </strong><a href="http://www.kvstore.io/"><strong>www.kvstore.io</strong></a><strong> and just “signup”…<br>Try yourself, it’s a child’s play!</strong></p><p>Enjoy!</p><p>@lordkada</p><p>PS: at the time of writing you may want to know that</p><ol><li>currently you need an invitation code, but if you follow the instructions on the website (typing “signup”) and you just drop me a note, I will be happy to provide one for you!</li><li>the service works (I use it for my own stuffs) and there’s a steady growing user base, but it is still under “heavy” development</li><li>I’m currently offering just the “basic” plan, which is and will stay free, but it is limited to up to 100 key/value pairs</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ded76aabf17" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>