<?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 Darell Kastl on Medium]]></title>
        <description><![CDATA[Stories by Darell Kastl on Medium]]></description>
        <link>https://medium.com/@darell-kastl?source=rss-4d9e28dbbd8a------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*yUeB2di2dmeX3eRctA3JvA.png</url>
            <title>Stories by Darell Kastl on Medium</title>
            <link>https://medium.com/@darell-kastl?source=rss-4d9e28dbbd8a------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 21:31:33 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@darell-kastl/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[MetalCore Has the Data. Lee Caplin Could Help Build the Market]]></title>
            <link>https://medium.com/the-investors-handbook/metalcore-has-the-data-lee-caplin-could-help-build-the-market-2b1011d782dc?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/2b1011d782dc</guid>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[robotics]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[stock-market]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Tue, 21 Jul 2026 17:22:28 GMT</pubDate>
            <atom:updated>2026-07-22T01:40:36.929Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>4.1 million records</strong> sounds like a major technology milestone. It is not yet a business model.</p><p>Data becomes commercially valuable when someone can use it to make a faster or better decision. For MetalCore, that could mean helping an exploration company review a property, find overlooked historical information or decide where limited capital should be spent next.</p><p><a href="https://www.newsfilecorp.com/release/304193">NRED has already assembled significant scale. MetalCore’s dataset grew from more than <strong>2.7 million records</strong> to over <strong>4.1 million</strong>, including approximately <strong>1.5 million geochemistry records</strong>, <strong>1.4 million oil and gas well records</strong>, more than <strong>800,000 deposit records</strong> and nearly <strong>200,000 geothermal records</strong>. The initial early-access group attracted <strong>920 registrations out of 1,000 available places</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EXiPqC0kVABv3ebKe0BcMA.png" /></figure><p>Those numbers show data and early interest. The next challenge is turning both into a product customers understand, trust and pay for.</p><p>That is where Lee Evan Caplin may matter. <a href="https://www.newsfilecorp.com/release/305915">His appointment to NRED Advisory Board was directly linked to MetalCore commercialization, intellectual-property strategy and strategic partnerships</a>. Caplin brings more than <strong>40 years of experience</strong> across technology ventures, IP and emerging businesses.</p><p>MetalCore has built the technical foundation. Caplin may help NRED build the market around it.</p><h3>Mining companies buy outcomes, not records</h3><p>A junior explorer may want a rapid property review before signing an option agreement. A consultant may want to combine public geological records with private client data. A larger mining company may want to rank targets across several projects.</p><p>These customers have different budgets and workflows, which means MetalCore first commercial offer should probably be simple.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8Jk0ppzpnJKT4HfC3AT5SA.png" /></figure><p>One possible starting point would be a structured property assessment showing the most relevant geological records, comparable occurrences, missing data and areas requiring further review. That gives the customer a clear input, a visible output and a result that can be measured.</p><p>Did MetalCore reduce weeks of manual research? Did it uncover a forgotten assessment report? Did it change the order in which targets were reviewed? Did it help reject a weak property before more capital was committed?</p><p>Those are outcomes a customer may pay for. The <strong>4.1 million records</strong> become supporting evidence rather than the product itself.</p><h3>The market for geoscience software already exists</h3><p>NRED does not need to prove that mining companies will purchase geological software.</p><p><a href="https://investors.bentley.com/news-releases/news-release-details/bentley-systems-enters-105-billion-agreement-acquire-seequent/">In 2021, Bentley Systems agreed to acquire geoscience-software company Seequent in a transaction valued at approximately <strong>US$1.05 billion</strong>. At the time, Seequent had more than <strong>430 employees</strong>, operated from <strong>16 offices</strong> and served users in over <strong>100 countries</strong></a>.</p><p>By 2025, Seequent said its products were used by <strong>eight of the world’s ten largest mining companies</strong> and customers across <strong>150 countries</strong>. Its strategy focuses on a familiar problem: geological information is often scattered across teams, formats and applications.</p><p>The comparison does not mean MetalCore is another Seequent. It shows that customers will pay for platforms that turn complex subsurface information into usable decisions.</p><p>MetalCore is much earlier. Its task is to identify one part of the exploration workflow where its data integration and probabilistic scoring can create a clear advantage.</p><p>The potential market may be broad, but the first product should be focused.</p><h3>Caplin may help define that first product</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/732/1*xuo8CijQzWphYZqKya2ZOg.png" /></figure><p>The important commercial questions now go beyond marketing.</p><p>Who should use MetalCore first? Should NRED charge per property, per user or through an enterprise licence? Should early customers receive direct software access or a completed analytical report? Which functions should remain proprietary, and which should connect with software mining teams already use?</p><p><a href="https://www.newsfilecorp.com/release/298327/NovaRed-Mining-Files-NonProvisional-U.S.-Patent-Application-for-AIDriven-Mineral-Evaluation-and-Transaction-Management-Platform">NRED U.S. patent application suggests a wider ambition. The filing describes systems for parcel-level mineral evaluation using integrated geological data, probabilistic scoring, property information and document-verification features.</a></p><p>That is broader than a geological search tool. But a broad technology vision does not require a broad first product. NRED could protect the larger platform while commercializing one practical use case at a time.</p><p>Caplin’s background may be useful because his role is not to improve the geological model. It is to help translate complex technology into something customers and partners can understand, adopt and eventually purchase.</p><h3>The first 1,000 users can help build the business model</h3><p>The <strong>920 early-access registrations</strong> are useful, but their long-term value depends on user behaviour.</p><p>How many people complete onboarding? How many analyze a real property? Which functions bring them back? Which users request private-data uploads, downloadable reports or collaboration tools? How many would move from free access to a paid pilot?</p><p>Those answers could reveal whether MetalCore should begin as a subscription platform, a per-property service, an enterprise product or a hybrid model.</p><p>Wilmac also gives NRED a real-world case study. The company holds an option to earn a <strong>70% interest</strong> in the <strong>16,078-hectare</strong> project, approximately <strong>10 kilometres west</strong> of the Copper Mountain Mine. MetalCore has already been used to organize historical information related to Wilmac’s copper-gold-platinum opportunity and documented associated-mineral indicators.</p><p>Wilmac does not prove that MetalCore is commercially validated. It can show what the platform changed: which data was added, which targets moved in priority and how fieldwork tested those conclusions.</p><p>That type of evidence may be more persuasive to a customer than another database milestone.</p><h3>The next important number is not 5 million</h3><p>MetalCore may soon pass <strong>5 million records</strong> and eventually <strong>10 million</strong>. Those figures will show scale, but they will not answer the commercial question.</p><p>The numbers that matter next are active users, completed property analyses, repeat usage, paid pilots and customer conversions.</p><p>NRED has not announced those milestones yet. The positive signal is that it has appointed someone whose stated role is to help build toward them.</p><p>MetalCore already has <strong>4.1 million records</strong>, <strong>920 early registrations</strong>, a patent strategy and a real exploration property where its workflow can be tested.</p><p>The data layer exists.</p><p>Lee Caplin could help NRED turn it into a market.</p><p>📩 <strong>Join </strong><a href="https://investorshandbook-newsletter-c72f94.beehiiv.com/"><strong>Investor’s Handbook Digest</strong></a><strong> — get the best investing, markets, and wealth-building insights each week.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2b1011d782dc" width="1" height="1" alt=""><hr><p><a href="https://medium.com/the-investors-handbook/metalcore-has-the-data-lee-caplin-could-help-build-the-market-2b1011d782dc">MetalCore Has the Data. Lee Caplin Could Help Build the Market</a> was originally published in <a href="https://medium.com/the-investors-handbook">Investor’s Handbook</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[$700,000 Per Interceptor: A Ukrainian Drone Maker Moves Into Air Defense And Changes The Math Of…]]></title>
            <link>https://medium.com/the-geopolitical-economist/700-000-per-interceptor-a-ukrainian-drone-maker-moves-into-air-defense-and-changes-the-math-of-866ea62f3370?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/866ea62f3370</guid>
            <category><![CDATA[robotics]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[materials]]></category>
            <category><![CDATA[infrastructure]]></category>
            <category><![CDATA[business]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Wed, 08 Jul 2026 19:40:22 GMT</pubDate>
            <atom:updated>2026-07-08T19:40:22.654Z</atom:updated>
            <content:encoded><![CDATA[<h3>$700,000 Per Interceptor: A Ukrainian Drone Maker Moves Into Air Defense And Changes The Math Of War</h3><p><strong>Fire Point shows a new stage of defense-tech: not only long-range strikes, but radar, interceptors, command systems, electronics, production, and the cost of one interception.</strong></p><p>In defense, sometimes the most important number is not range or speed. It is the cost of one shot.</p><p>If an attacking missile costs less than the weapon used to shoot it down, air defense quickly becomes not only a military problem, but an economic one. And that is where Fire Point’s story becomes more interesting than a normal headline about another drone maker.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/711/1*TzhKG2eyMKFJuqC-jHyCVA.png" /></figure><p><a href="https://www.globaldefensecorp.com/2026/07/08/ukraines-top-strike-drone-maker-fire-point-moves-into-ballistic-missile-defence/">Global Defense Corp writes that Ukrainian company Fire Point, known for long-range strike drones, is moving into ballistic missile defence through joint German-Ukrainian production of the FP-7.X Freyja system.</a> The company already makes the FP-1 deep-strike drone, the FP-2 and the FP-5 Flamingo cruise missile, and is now trying to enter the other end of the same war: not only striking far away, but intercepting what comes back.</p><p>The article includes several numbers that show the scale. Fire Point-produced drones, according to co-founder and chief designer Denys Shtilerman, carry out about 60% of Ukrainian strikes inside Russia. The upgraded FP-1 now reportedly flies up to 1,677 miles instead of 1,025 miles, the FP-2 can carry a 440-pound warhead over 230 miles, and Zelensky has spoken about FP drones with a range of up to 3,000 km.</p><p>This no longer looks like a small workshop. It looks like a producer sitting near the center of the long-range strike economy.</p><p>But Freyja is a different logic. The point is not to reach a target deep behind the front line. The point is to reduce the cost curve of interception.</p><p>Fire Point wants to shoot down a ballistic missile for less than $1 million per shot. The article points to a target of around $700,000 for the FP-7.X, which is meant to operate at an altitude of about 15 miles, versus roughly $3.8 million for a Patriot PAC-3. If that math works even partly, this is not just a new missile. It is an attempt to make air defense more serial, more repeatable, and less financially exhausting.</p><p>That matters because modern war increasingly looks like a factory contest. Drones, missiles, interceptors, guidance electronics, motors, composites, launch canisters, batteries, wiring harnesses, and sensors cannot merely exist. They have to be produced in volume, quickly, and more cheaply than the enemy expects.</p><p>One expensive interceptor can be a technological marvel. But if the threat is mass-produced, the answer needs industrial math.</p><p>Freyja, as described by Fire Point, is meant to combine the interceptor with a European radar, tracking, and command layer. The article mentions Hensoldt, talks with Thales on radar, Leonardo on tracking, and Kongsberg on command and control. So this is no longer “one missile.” It is a system: detection, tracking, fire control, interceptor, communications, launch hardware, and a decision loop that has to work in seconds.</p><p>This is where defense-tech stops looking like a software startup story. Even if the core innovation sits in guidance or cost engineering, it still needs production lines, test stands, precision machining, high-temperature materials, cables, connectors, radar components, power supplies, and thermal management.</p><p>Missile defense is not only an algorithm. It is a factory floor, a supply chain, and the ability to repeat a complex product dozens or hundreds of times.</p><p>What interests me here is the changing role of Ukraine. Fire Point does not look like a classic old-school defense giant. But the war has created conditions where new defense manufacturers can grow much faster than a peacetime bureaucratic system would normally allow. That does not make the path simple or perfect. But it shows how quickly battlefield demand can turn an engineering team into an industrial player.</p><p>At the physical layer, this story again runs into metals. An air defense system is radar arrays, power electronics, cabling, grounding, motors, actuators, telemetry, control cabinets, cooling, launch equipment, and mobile platforms. Copper is not the headline hero here, but it sits inside wiring, power distribution, electronics, motors, and radar infrastructure.</p><p>The same goes for aluminum, titanium, specialty alloys, rare earth magnets, graphite, ceramics, and energetic materials. Defense capacity is not only engineers. It is a materials pipeline.</p><p>In this kind of defense economy, early copper exploration companies such as NovaRed Mining no longer look like a side topic, but part of the material chain. The point here is not a finished ton of copper tomorrow, but the early work before demand for electrified defense systems becomes even tighter. The Trojan-Condor Corridor adds five mineral tenures covering roughly 4,573.82 ha to the Wilmac land position, while the broader work revolves around claims, soil data, geophysics, and target selection.</p><p><a href="https://novaredmining.com/">For defense-tech, where radar, command systems, launchers, and mobile platforms require more electronics and power infrastructure, the copper story becomes less peripheral.</a></p><p>Fire Point shows that the defense industry is changing in two directions at once. On one side, cheaper long-range drones are pressuring old air-defense systems. On the other, the same kind of engineering teams are starting to look for cheaper interceptors so they do not lose the economics of protection.</p><p>This is not a clean symmetry. It is the reality of a war where both sides are learning to count cost per strike and cost per intercept.</p><p>$700,000 per interceptor is not a victory or a guarantee yet. It is a target that still has to be proven through production, testing, and real interceptions. But the appearance of that target matters.</p><p>It shows that the future of air defense will not be judged only by whether it can shoot down a missile. It will also be judged by whether it can do it cheaply enough, often enough, and fast enough to survive war at factory speed.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=866ea62f3370" width="1" height="1" alt=""><hr><p><a href="https://medium.com/the-geopolitical-economist/700-000-per-interceptor-a-ukrainian-drone-maker-moves-into-air-defense-and-changes-the-math-of-866ea62f3370">$700,000 Per Interceptor: A Ukrainian Drone Maker Moves Into Air Defense And Changes The Math Of…</a> was originally published in <a href="https://medium.com/the-geopolitical-economist">The Geopolitical Economist</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[La vitesse du rail se cache dans une armoire au bord de la voie]]></title>
            <link>https://medium.com/@darell-kastl/la-vitesse-du-rail-se-cache-dans-une-armoire-au-bord-de-la-voie-789dca4bb462?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/789dca4bb462</guid>
            <category><![CDATA[money]]></category>
            <category><![CDATA[infrastructure]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[energy]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Thu, 02 Jul 2026 17:34:47 GMT</pubDate>
            <atom:updated>2026-07-02T17:34:47.768Z</atom:updated>
            <content:encoded><![CDATA[<p>Dans le rail, la vitesse est souvent vendue avec de beaux mots : modernisation, rail numérique, meilleure capacité. Mais dans la pratique, elle se trouve dans des choses très concrètes — armoires de signalisation, câbles, liaisons radio, alimentation électrique, systèmes de contrôle et équipements installés au bord de la voie. Un train peut circuler plus souvent non pas parce que le rail est soudain devenu plus intelligent. Mais parce que l’ensemble du système sait mieux où se trouve ce train et ce qu’il est autorisé à faire ensuite.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*Q8VuI_iJScVySxamOiZo0Q.png" /></figure><p><a href="https://www.railwaygazette.com/infrastructure/2026/07/02/queensland-awards-first-sunshine-coast-etcs-contract/">Railway Gazette écrit qu’Alstom a obtenu le premier contrat pour équiper le réseau principal du Queensland avec le système European Train Control System. Il s’agit d’un package de 114 millions de dollars australiens dans le cadre d’un accord-cadre de 354 millions de dollars australiens pour la phase 1 du projet The Wave. Alstom doit concevoir, fournir, tester et mettre en service les équipements trackside Onvia Control pour une exploitation ETCS Level 2 sur le Sector 1 North.</a></p><p>Il est facile d’appeler cela simplement un contrat de signalisation. Mais ETCS Level 2 n’est pas un vieux feu au bord de la voie. C’est un système dans lequel le train reçoit son autorisation de mouvement grâce à un échange continu de données. Autrement dit, la sécurité et la capacité ne dépendent plus seulement de l’acier, des traverses et de l’horaire, mais aussi du software, des liaisons radio, de l’équipement embarqué, des armoires trackside et des centres de contrôle.</p><p>Ce qui m’intéresse ici, ce n’est pas tellement le nom technologique, mais le fait qu’Alstom ne travaille pas dans un terrain vierge. Le Sector 1 North comprend une partie de l’existing North Coast line et une nouvelle ligne pour la Sunshine Coast. Cela signifie qu’il faut mélanger un ancien réseau avec une nouvelle infrastructure. Il faut installer un nouveau système qui fonctionne avec l’infrastructure existante, les travaux civils, Queensland Rail, le futur horaire passagers et les vraies contraintes du terrain.</p><p><a href="https://www.indexbox.io/blog/queensland-awards-first-etcs-contract-to-alstom-for-main-line-network/">IndexBox, qui reprend les détails du contrat, indique que la phase 1 du projet Sunshine Coast prévoit une ligne double voie de 19 km entre Beerwah et Caloundra. Une future phase 2 pourrait ajouter 18,8 km jusqu’à Birtinya et Maroochydore.</a> Ce n’est pas simplement “ajouter du software”. C’est de la voie, des équipements, de l’alimentation, des tests, de l’intégration, du travail réglementaire et beaucoup d’installation sur le terrain.</p><p>Les Jeux olympiques et paralympiques de Brisbane 2032 ajoutent une pression séparée. Avant ce type d’événement, les villes aiment parler de capacité et d’héritage transport. Mais à l’intérieur, c’est toujours une liste de tâches beaucoup plus terre à terre : intervalles plus courts, plus de trains dans l’horaire, moins de bottlenecks manuels, réaction plus rapide des dispatchers et un système qui ne doit pas se désorganiser au moment du pic.</p><p>Le rail numérique semble presque léger. En réalité, il ajoute une couche supplémentaire à une infrastructure déjà lourde. À côté des voies apparaissent des cabinets, des sensors, des balises, des telecom shelters, des power feeds, des backup systems, des onboard units et du software certifié pour la sécurité. Tout cela ne doit pas seulement être installé. Il faut aussi le vérifier, le certifier, le connecter et l’entretenir pendant des années.</p><p>Sur ce point, le rail ressemble un peu au réseau électrique. Davantage de capacité ne signifie pas toujours plus de béton ou plus d’acier. Parfois, la capacité s’achète par une gestion plus précise du mouvement physique. Dans le grid, ce sont les transformers, le switchgear, les sensors et le dispatch software. Dans le rail, ce sont la signalisation, les canaux radio, les centres de contrôle et des équipements qui doivent fonctionner sans erreur quand le train roule déjà sur la ligne.</p><p>Le cuivre n’est pas le personnage principal ici, mais sans lui, ce rail numérique ne se ferme pas : signalling cabinets, power feeds, control wiring, telecom shelters, onboard electronics et sous-stations reposent toujours sur le métal. C’est aussi pour cela que les premières cibles copper comptent bien avant que quelqu’un commande du câble pour un nouveau projet ferroviaire. Sur le Wilmac Copper-Gold Project en Colombie-Britannique, développé par <a href="https://novaredmining.com/">NovaRed Mining</a>, North Lamont avait auparavant montré jusqu’à 1 125 ppm de cuivre dans les sols. Ce n’est pas une ressource et ce n’est pas une preuve de future mine, mais c’est le type de signal précoce à partir duquel commence une vérification de terrain plus coûteuse.</p><p>Cette nouvelle ne parle pas seulement d’Alstom ni seulement du Queensland. Elle parle du fait que l’ancienne infrastructure devient plus intelligente d’une manière très physique. Pas par magie logicielle, mais grâce à des équipements qu’il faut fixer, connecter, tester et laisser fonctionner au bord de la voie.</p><p>Le rail de demain sera plus rapide non pas parce qu’il sera devenu moins physique. Mais parce que chaque armoire au bord de la voie, chaque câble et chaque signal devront fonctionner avec plus de précision que l’ancien horaire ne le permettait.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=789dca4bb462" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Battery No Longer Wants to Be Just a Battery]]></title>
            <link>https://medium.com/@darell-kastl/the-battery-no-longer-wants-to-be-just-a-battery-b11caeb53b3b?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/b11caeb53b3b</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[mining]]></category>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[energy]]></category>
            <category><![CDATA[industrialautomationparts]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Thu, 25 Jun 2026 18:28:46 GMT</pubDate>
            <atom:updated>2026-06-25T18:28:46.497Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>At ees Europe 2026, the most interesting part was not how many MWh could be packed into a container. The more interesting shift was that the battery market started speaking the language of grids, data centres and electrical stability.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/898/1*1e96UJezUWJonfPDKc4eDw.png" /></figure><p>A few years ago, battery storage was often sold in a very simple way: here is the container, here are the cells, here are the MWh, here is the charge-and-discharge curve. In 2026, that story already feels too narrow. At ees Europe in Munich, manufacturers were not just showing new boxes filled with batteries. They were showing full energy nodes — with PCS, grid-forming, DC-coupling, black-start, VPP, AI optimisation and architecture built for data centres.</p><p>What interests me here is not only that Fluence introduced Smartstack 10MWh. Although the number is strong: around 680MWh per acre and up to 40% lower balance-of-plant costs compared with standard DC blocks. The bigger point is that this kind of density is becoming a response to land, grid and permitting pressure. When the footprint does not grow but capacity does, this is no longer just engineering. It is a way to fit more electrical flexibility into the same physical limits. (<a href="https://www.energy-storage.news/ees-europe-2026-grid-forming-pcs-dc-coupling-ci-and-data-centre-solutions-among-product-launches-and-market-debuts/?utm_source=chatgpt.com">energy-storage.news</a>)</p><p><strong>A battery no longer just stores electricity. It sells confidence in time.</strong></p><p>Sungrow went in another direction — C&amp;I storage, on-grid/off-grid, AC-coupled and DC-coupled configurations. The PowerStack 255CS has a 125kW PCS, 257kWh or 514kWh per unit, and can scale up to 6MWh through 24 units. This is no longer only a utility story. It is about factories, sawmills, weaker grids, backup power, demand control and situations where businesses do not want to wait until the grid becomes perfect.</p><p><a href="https://www.energy-storage.news/ees-europe-2026-grid-forming-pcs-dc-coupling-ci-and-data-centre-solutions-among-product-launches-and-market-debuts/?utm_source=chatgpt.com">Trina Storage added another layer: a 13.8MVA/25MWh integrated DC+AC solution, 10ms full-power response, grid-forming and black-start capability. It also pointed to a 40MW/160.48MWh deployment in Romania, with target commissioning in Q3 2026.</a> These details show where the market is going: a battery project can no longer be separated from the inverter, transformer, cooling, controls and grid code.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/937/1*Guhj7ZJG4UBiIcLhyFXRFA.png" /></figure><p>And the most interesting signal, in my view, came from the data centre theme. Envision presented its AI Power System for AI data centres: coordination between generation, storage, grid interaction and computing loads, 800VDC power architecture, 2.5MW LFP storage, a 2.5MW solid-state transformer and sodium-ion batteries for fast fluctuations in AI loads. This shows very clearly that AI infrastructure is starting to require not just more electricity, but electricity with the right speed of response.</p><p><strong>AI does not live in the cloud. It lives where PCS, transformers, cables and substations can handle the load.</strong></p><p>And this is where the battery storage market quietly crosses into the metals market. Grid-forming PCS, DC-coupling, 800VDC, solid-state transformers, switchgear, busbars, substations, cables — none of this appears out of thin air. Behind every “smart” energy system is a very physical base: copper, aluminium, power electronics, land, interconnection and a long supply chain.</p><p>That is why, in the copper-gold space, it is worth looking not only at operating mines but also at earlier-stage projects that are still moving from maps toward drill targets. <a href="https://novaredmining.com/">NovaRed is working on the Wilmac Copper-Gold Project in British Columbia</a> — 16,078 hectares in the Quesnel porphyry belt, approximately 10 km west of the Copper Mountain Mine. In 2026, the company plans expanded soil sampling, IP/AMT surveys across the North Lamont, West Lamont, Wilmac and Plume grids, as well as an initial drill program subject to an approved permit.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u0KeTfKso39PgxsiTTMRoQ.png" /></figure><p>You can look at ees Europe 2026 as a battery products exhibition. But I do not think that is quite right anymore. It was really an exhibition about how electricity is becoming more complex: it has to be not only clean, but controllable, fast, dense, local, backed up and ready for AI loads.</p><p>Battery storage used to sell the pause between sunlight and evening demand. Now it sells stability for an entire economy.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b11caeb53b3b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Amazon ne compte plus seulement les cartons]]></title>
            <link>https://medium.com/@darell-kastl/amazon-ne-compte-plus-seulement-les-cartons-5a0a5dcc292e?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/5a0a5dcc292e</guid>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[money]]></category>
            <category><![CDATA[energy]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Thu, 18 Jun 2026 14:03:13 GMT</pubDate>
            <atom:updated>2026-06-18T14:03:13.964Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>La nouvelle étape d’Amazon dans l’automation ressemble à une autre histoire de warehouse efficiency. Mais en réalité, l’entrepôt commence à devenir un système d’exploitation qui gère non seulement les produits, mais aussi le temps humain.</strong></p><p>Amazon sait depuis longtemps transformer un entrepôt en machine. Les robots déplacent les étagères, les conveyor belts transportent les cartons, les scanners enregistrent chaque mouvement, et le software voit le flux de commandes plus vite qu’un manager ne peut traverser les zones. Mais la nouvelle de Business Insider est intéressante non pas parce qu’Amazon ajoute encore plus de robotics. Cette fois, l’entreprise essaie d’automatiser non seulement le mouvement du package, mais aussi la décision de savoir où une personne doit travailler.</p><p>Il s’agit de Full Facility Load Balancing, ou FFLB. Selon BI, le système analyse les package volumes, les forecasts et d’autres operational signals, puis recalcule environ toutes les 3 minutes le nombre de personnes nécessaires dans différentes parties de l’entrepôt. Si une zone paraît overstaffed et qu’une autre commence à prendre du retard, le système recommande de déplacer des workers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/976/1*DJ4upocHiQ_tkDeiULwf_w.png" /><figcaption><a href="https://www.businessinsider.com/amazon-warehouse-automation-moving-workers-labor-hours-robots-2026-6?utm_source=chatgpt.com">https://www.businessinsider.com/amazon-warehouse-automation-moving-workers-labor-hours-robots-2026-6</a></figcaption></figure><p>L’entrepôt du futur ne déplace pas seulement les cartons. Il commence à déplacer les décisions.</p><p>Ce qui m’intéresse ici, c’est que l’automation entre dans une phase plus silencieuse. Avant, tout était visuellement clair : un robot à la place d’un heavy lift, un robotic arm à la place du manual sorting, un autonomous mobile robot à la place d’un cart. Maintenant, l’automatisation devient moins spectaculaire, mais plus profonde. Elle s’installe dans la staffing logic. Elle calcule les idle stations, l’overstaffing, le workload imbalance et les minutes vides entre les tâches.</p><p>Selon les estimations internes vues par Business Insider, Amazon voit un potentiel de recover roughly $193 million annually in labor costs et de cut almost 7 million labor hours each year. En même temps, l’entreprise précise que ces chiffres ne doivent pas être lus comme un résultat garanti — il s’agit d’un hypothetical modeling, pas d’une économie mesurée en temps réel. Et cette précision est importante. Dans le thème de l’automation, il est très facile de passer de “system can improve flow” à “robots replace everyone”, alors que la réalité est plus complexe.</p><p>Amazon décrit FFLB comme un outil destiné à aider les managers à réagir plus vite aux conditions changeantes, et non à les remplacer. Mais la logique elle-même montre quand même un changement : le manager ne regarde plus simplement le floor. Il regarde le floor à travers le software.</p><p>La robotisation a commencé avec un bras métallique. Maintenant, elle atteint l’emploi du temps humain.</p><p>Le fragment le plus intéressant dans BI concerne Container Build. C’est une fonction d’entrepôt où les employés placent les packages dans des outbound carts et containers avant l’expédition. Amazon, selon les documents, appelle cette zone la “single largest labor automation opportunity”. Une analyse de 97 fulfillment centers a trouvé 48 sites travaillant sous les productivity targets et représentant environ 309 000 excess labor hours per month. Séparément, encore 266 000 labor hours per month étaient liées à des stations où il y avait peu ou pas de work in progress. Ensemble, cela représente environ 6.9 million hours par an.</p><p>Ces chiffres sont importants non pas parce qu’ils sont définitifs. Ils sont importants comme manière de penser. Amazon ne cherche pas seulement les “grandes” victoires de l’automation. Elle cherche les microfissures dans ses opérations : où une personne attend, où le flux est irrégulier, où un container n’est pas prêt, où le package volume change plus vite que le shift plan.</p><p>C’est ainsi que l’entrepôt commence à ressembler à un algorithme vivant. Chaque zone devient un process segment. Chaque minute devient un data point. Chaque worker devient une partie d’un dynamic balancing system. Cela ne signifie pas forcément un entrepôt entièrement sans humains. Au contraire, Amazon parle toujours de collaboration between workers and robots. Mais le rôle de la personne à l’intérieur de cette collaboration change.</p><p>Amazon a déjà montré jusqu’où cette logique peut aller dans les advanced fulfillment centers. Dans son site de Shreveport, d’une superficie de plus de 3 million square feet et réparti sur 5 floors, l’entreprise a réuni Sequoia, Proteus, Sparrow, Cardinal, Robin et d’autres robotics systems. Amazon affirme que Sequoia peut stocker plus de 30 million items, que Sparrow travaille avec plus de 200 million unique products, et que les next-generation facilities peuvent réduire les fulfillment processing times up to 25%.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*olfVkQZ9uxkrR5-TcdToGQ.png" /><figcaption><a href="https://www.aboutamazon.com/news/operations/amazon-fulfillment-center-robotics-ai?utm_source=chatgpt.com">https://www.aboutamazon.com/news/operations/amazon-fulfillment-center-robotics-ai</a></figcaption></figure><p>Quand le software voit chaque minute, une simple pause commence à ressembler à une dépense.</p><p>Et il y a encore une autre couche, qui apparaît rarement dans les titres. Un tel entrepôt, ce n’est pas seulement de l’AI, des dashboards et des robots. C’est de l’électricité. Ce sont des chargers, motors, sensors, cameras, control cabinets, cables, busbars, power distribution, networking equipment, backup systems et cooling. Plus un warehouse devient “intelligent”, plus il devient une infrastructure électrique.</p><p>C’est pourquoi cette automation story a une vraie couche materials. Un robot ne fonctionne pas dans un cloud abstrait. Il roule sur un sol en béton, se recharge depuis le réseau, transmet des signaux à travers du hardware et dépend de métaux que les titres financiers mentionnent rarement. Dans cette partie physique, le copper reste l’un des matériaux de base de la logistique électrifiée.</p><p>Dans le thème plus large du copper-gold, cela mène aux early exploration stories. NovaRed Mining travaille sur le Wilmac Copper-Gold Project en Colombie-Britannique — 16 078 hectares dans la ceinture porphyrique de Quesnel, à environ 10 km à l’ouest de Copper Mountain Mine. Son planned 2026 field program comprend expanded soil sampling, 4 IP/AMT geophysical surveys sur les grids North Lamont, West Lamont, Wilmac et Plume, ainsi qu’un initial drill program envisagé pour l’automne 2026 sous réserve d’un approved drill permit. C’est encore une étape précoce, mais c’est ainsi que la future base matérielle de l’automation economy apparaît d’abord sur le terrain — comme soil data, geophysics et drill targets.</p><p>La conclusion de cette nouvelle est simple : Amazon n’automatise plus seulement l’entrepôt. Elle automatise la perception du temps à l’intérieur de l’entrepôt. Avant, la question était : comment livrer le package plus vite ? Maintenant, la question devient plus dure : comment faire en sorte qu’aucune personne, aucune station et aucune minute ne restent sans rôle.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5a0a5dcc292e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Gas Station Is No Longer About Gasoline]]></title>
            <link>https://medium.com/@darell-kastl/the-gas-station-is-no-longer-about-gasoline-9b848ad078d3?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/9b848ad078d3</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[energy]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Wed, 17 Jun 2026 14:18:11 GMT</pubDate>
            <atom:updated>2026-06-17T14:18:41.546Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>YPF and Tesla are talking about charging networks, but behind this news there is something else: the road is slowly becoming part of the energy system.</em></strong></p><p><a href="https://www.reuters.com/business/energy/argentinas-ypf-tesla-explore-ev-charging-energy-storage-collaboration-2026-06-16/">The news about YPF and Tesla is easy to read as another electric vehicle story. Argentina’s YPF signed a letter of intent with Tesla to explore possible cooperation in fast charging and energy storage. This came after YPF CEO Horacio Marin visited Tesla’s Gigafactory in Texas, where they discussed energy infrastructure, electric mobility and technology.</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/743/1*l3C-LLLdKAtlGgBjV1WeQw.png" /></figure><p>But what caught my attention here was not the word Tesla itself. That word always takes over the headline anyway. The more interesting part is that a major fuel company is looking at charging networks as the next layer of its business. If a car no longer stops only for gasoline, then what should stand at the old gas station?</p><p>YPF has something that is hard to copy quickly: locations, routes, drivers’ habit of stopping, and an existing presence along the roads. For gasoline, that was convenient. For electricity, it may become even more important, because fast charging needs more than just a point on a map. It needs power brought to the site.</p><p>This is where the old gas station starts to change meaning. It used to be the end of the fuel chain: a car arrived, filled up, and drove away. Now it can become a small energy node. It needs cables, transformers, distribution equipment, batteries and peak-management systems. Otherwise, several fast chargers running at the same time can become not a service, but a problem for the local grid.</p><p>That is why the energy storage part of this news does not look secondary. In fact, without batteries, fast charging often runs into peak-load problems. With batteries, the site can work more smoothly: store energy earlier, release it when demand rises, and avoid pulling the grid to its limit every time. This is no longer just charging for EVs. It is a small model of how transport starts living inside the electrical system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/953/1*GJ29PTL9x5RZd9F1RVWZUw.png" /></figure><p>For Argentina, this is especially interesting because there is no magical jump from intention to a finished network. A letter of intent is only the beginning of the conversation. Next come sites, permits, suppliers, substations, equipment, money and time. Are old fuel locations ready to take on a new electrical load? And who will rebuild the infrastructure faster — the auto market, or the energy system around it?</p><p>Copper is not in the headline of this story, but it sits inside the body of the system. Fast charging does not work without copper. Energy storage does not work without power cables and electronics either. If a gas station becomes an energy node, it asks for more than software and a logo on the charger. It asks for metal that can carry the current.</p><p>One early junior example from the copper side of this long story is <a href="https://novaredmining.com/"><strong>NRED</strong></a>. At Wilmac, the company is working with copper-gold geology in British Columbia, while materials on the Lamont zones have mentioned geophysical signals, intrusive bodies and copper-in-soil up to 1,125 ppm at North Lamont. This is still a very early layer, far from a roadside charging station. But this is what the beginning of the chain looks like before it passes through ore, processing, cables and finished infrastructure.</p><p>I think this news matters not because it is loud. No one has shown a map of the future network yet, the number of stations, or the size of the batteries. There is only an intention. But sometimes an intention shows the direction clearly: the gas station is no longer just a place to sell fuel.</p><p>It is becoming a place where the road asks for electricity. And electricity, if we are being honest, never arrives by itself — it arrives through metal.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9b848ad078d3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Scalpel Became Infrastructure]]></title>
            <link>https://medium.com/@darell-kastl/the-scalpel-became-infrastructure-9ebc8f1b4a9a?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/9ebc8f1b4a9a</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[copper]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[robotics]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Thu, 11 Jun 2026 15:16:38 GMT</pubDate>
            <atom:updated>2026-06-11T15:16:38.487Z</atom:updated>
            <content:encoded><![CDATA[<p>The Scalpel Became Infrastructure</p><p>Da Vinci 5 in Europe looks like a medical story. But behind the robotic arm sits a much heavier layer — hospitals, electricity, sensors, cables, and materials.</p><p>In news about robotic surgery, it is easy to look only at the machine itself. A sleek robotic arm, a surgeon’s console, 3D imaging, precise movements, technology that seems to bring the operating room into the future. Da Vinci 5 sounds exactly like that: a new platform, more computing power, better ergonomics, and more data for the physician.</p><p>But something else caught my attention in this story. Robotic surgery does not live in the air. It enters real hospitals, real operating rooms, procurement systems, training, maintenance, power supply, and technical support. And the smarter the machine becomes, the more physical infrastructure appears around it.</p><p><a href="https://investor.intuitivesurgical.com/news-releases/news-release-details/intuitives-da-vinci-5-surgical-system-receives-ce-mark">Intuitive received CE mark approval for da Vinci 5 in Europe in 2025 for adult and pediatric use in minimally invasive procedures, including urologic, gynecologic, and general laparoscopic procedures</a>. On its product page, the company describes the platform as a system with more than 150 design innovations and 10,000x computing power compared with previous capabilities. That sounds like medtech. But behind those words sits a simple question: what has to be ready inside a hospital so that such a system does not become just an expensive object in the corner?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/918/1*zFDJcnX4vgd-HAXHDAj0Aw.png" /></figure><p>Intuitive’s financial reporting shows the scale well. In the first quarter of 2026, the company placed 431 da Vinci systems globally, including 117 in Europe. The total da Vinci installed base as of March 31, 2026 was approximately 11,395 systems, including 2,257 in Europe. This is no longer a story about one hospital or one demonstration. It is a network of operating rooms gradually becoming part of robotic medicine.</p><p>And here it is worth thinking not only about the surgeon, but also about the room around them. The robot needs stable power. The console needs data transmission. Imaging needs computing. The arms need sensors, drives, circuit boards, cables, and service. Who usually sees this part? The patient sees the result. The surgeon sees the tool. But underneath it all works a material chain that is far less visible in presentations.</p><p>I think there is a parallel here with the wider technology world. We often talk about AI, robots, automation, and precision as if the future is becoming lighter. In reality, it often becomes more complex. More data means more electronics. More automation means more components. More hospitals with robotic systems means more service networks, spare parts, engineers, and materials.</p><p>One junior example in the copper theme is <a href="https://novaredmining.com/">NRED</a>. Its Wilmac Copper-Gold Project in British Columbia does not need to be directly tied to da Vinci 5 or to hospitals. It is early-stage exploration, where geological signals still need to be tested. But the logic of the material world is similar: before copper becomes part of cables, circuit boards, power systems, hospital equipment, or the broader electrical system, it first has to be found and proven in the ground.</p><p>Robotic surgery looks clean and almost silent. The surgeon sits at the console, the system mirrors the movements, and the instruments work inside a small space in the body. But how much metal, energy, and technical discipline are needed to make that moment look so calm? How much invisible work has to happen before a robot in a hospital begins to function like a normal part of medicine?</p><p>Da Vinci 5 in Europe is not only about the next stage of surgery. It is also another sign that precision is becoming infrastructure. And the smaller the movement of the instrument inside the patient becomes, the larger the world of materials, electricity, and maintenance that has to stand behind it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9ebc8f1b4a9a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Le plan français pour 45 000 km de réseaux cache une question dérangeante]]></title>
            <link>https://medium.com/@darell-kastl/le-plan-fran%C3%A7ais-pour-45-000-km-de-r%C3%A9seaux-cache-une-question-d%C3%A9rangeante-3dad0cad540b?source=rss-4d9e28dbbd8a------2</link>
            <guid isPermaLink="false">https://medium.com/p/3dad0cad540b</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[supply-chain]]></category>
            <category><![CDATA[infrastructure]]></category>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Darell Kastl]]></dc:creator>
            <pubDate>Fri, 29 May 2026 18:39:12 GMT</pubDate>
            <atom:updated>2026-05-29T18:39:12.837Z</atom:updated>
            <content:encoded><![CDATA[<p><em>L’Europe veut plus d’électricité, plus de véhicules électriques, plus de pompes à chaleur et plus d’usines sans énergies fossiles. Mais derrière cette belle formule d’indépendance énergétique se cache une réalité plus simple et plus dure : y aura-t-il assez de métaux pour construire tout cela ?</em></p><p>L’électrification ressemble souvent à une promesse politique venue d’un avenir plus propre. Moins de carburants importés. Plus d’électricité domestique. Plus d’emplois. Mais si l’on retire les mots de présentation, il reste une image très physique : câbles, transformateurs, bornes de recharge, sous-stations, pompes à chaleur, raccordements industriels et des milliers de kilomètres de nouvelles lignes.</p><p>La France mise précisément sur cette réalité physique. <a href="https://www.reuters.com/business/energy/edf-invest-240-million-french-electrification-plan-2026-05-26"><strong>Reuters</strong></a> rapporte qu’Emmanuel Macron a signé un pacte national d’électrification avec de grandes entreprises afin de doubler la part de l’électricité domestique dans le mix énergétique à 60% d’ici 2030. Le plan implique 6 000 entreprises, plus de 600 000 emplois, jusqu’à 10 milliards d’euros de soutien public par an jusqu’en 2030, 240 millions d’euros d’investissement de la part d’EDF et 45 000 km de lignes de transport et de distribution par RTE et d’autres opérateurs de réseau.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/911/1*z_3N5f_3rqIMgNKdlwRBwg.png" /></figure><p>Ce n’est pas seulement une politique énergétique. C’est la liste de ce qu’il faudra réellement installer sur le terrain. Stellantis prévoit plus d’un milliard d’euros pour une nouvelle génération de véhicules électriques dans son usine de Mulhouse à partir de 2029. La France veut doubler le nombre de sites de recharge pour véhicules électriques, produire un million de pompes à chaleur et doubler la production de radiateurs électriques d’ici 2030. Tous ces objectifs reposent sur une même base : une infrastructure électrique qui exige du cuivre, de l’acier, de l’aluminium, des équipements et de longues chaînes d’approvisionnement.</p><p>Le cuivre reste presque invisible dans cette histoire, mais sans lui le plan perd rapidement sa forme. Il se trouve dans les câbles, les moteurs, les transformateurs, les systèmes de recharge, les sous-stations et les équipements électriques industriels. Quand un pays parle de 45 000 km de nouveaux réseaux, il ne parle pas seulement d’ingénierie. Il parle aussi de matériaux qui doivent être découverts, financés, extraits, transformés et livrés à temps.</p><p>C’est pourquoi l’électrification ramène l’attention vers le secteur minier, même si ce mot n’apparaît pas toujours dans les discours politiques. Les nouveaux réseaux ne se construisent pas sans métaux. Les nouveaux raccordements industriels ne fonctionnent pas sans conducteurs. Les nouveaux sites de recharge pour véhicules électriques ont besoin d’une infrastructure riche en cuivre. Et si l’Europe veut construire plus vite, elle aura besoin d’un pipeline plus long de futures sources d’approvisionnement dans des juridictions stables.</p><p>NovaRed Mining entre dans cette discussion non pas comme une histoire française, mais comme un petit exemple de ce qui existe à l’autre bout de l’électrification. En mai 2026, NRED a ajouté Ed Kostenski à son conseil consultatif — un entrepreneur avec plus de 40 ans d’expérience dans les équipements miniers, le financement de projets, le développement d’infrastructures et la croissance industrielle internationale. Pour le secteur copper-gold à un stade précoce, ce type d’expérience a du sens : avec le temps, les projets de ressources ne dépendent pas seulement de la géologie, mais aussi des équipements, de l’accès, de la logistique, du financement et de la capacité à exécuter des plans industriels complexes.</p><p>Le plan français pour 2030 ressemble à une histoire d’indépendance énergétique. Mais sous cette ambition se cache une autre question, moins confortable pour les grands titres : le monde peut-il extraire et livrer assez de matériaux pour l’économie électrique qu’il a déjà promis de construire ? Car l’avenir peut paraître propre dans les discours, mais il se construit avec des métaux très réels.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3dad0cad540b" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>