„Kéž by internet ani neexistoval,“ myslí si skoro polovina mladých Britů. Ukázal to průzkum veřejného mínění, který provedl tamní institut British Standards Institution.

V roce 2009 vyšla na stránkách libertariánského think tanku Cato esej nazvaná The Education of a Libertarian od Petera Thiela. Pro uvedení do děje: libertariánství je&…

Linux Kernel Explorer (reverser.dev) 14:17  ↑ 119 HN Points

Check if your IP address has been observed by GreyNoise sensors. Instantly detect malicious activity, compromised devices, and security threats affecting your network.

PowerShell is a spiced up command line shell made by Microsoft. According to some people, it is a really useful and good shell alternative. Already a long time ago, we got bug reports from confused users who couldn’t use curl from their PowerShell prompts and it didn’t take long until we figured out that Microsoft had added aliases for both curl and wget. The alias had the shell instead invoke its own command called “Invoke-WebRequest” whenever curl or wget was entered. Invoke-WebRequest being PowerShell’s own version of a command line tool for fiddling with URLs. Invoke-WebRequest is of course not anywhere near similar to neither curl nor wget and it doesn’t support any of the command line options or anything. The aliases really don’t help users. No user who would want the actual curl or wget is helped by these aliases, and users who don’t know about the real curl and wget won’t use the aliases. **They were and remain pointless.** But they’ve remained a thorn in my side ever since. Me knowing that they are there and confusing users every now and then – not me personally, since I’m not really a Windows guy. Fast forward to modern days: Microsoft released PowerShell as open source on github yesterday. Without much further ado, I filed a Pull-Request, asking the aliases to be removed. It is a minuscule, 4 line patch. It took way longer to git clone the repo than to make the actual patch and submit the pull request! It took 34 minutes for them to close the pull request: > “Those aliases have existed for multiple releases, so removing them would be a breaking change.” To be honest, I didn’t expect them to merge it easily. I figure they added those aliases for a reason back in the day and it seems unlikely that I as an outsider would just make them change that decision just like this out of the blue. But the story didn’t end there. Obviously more Microsoft people gave the PR some attention and more comments were added. Like this: > “You bring up a great point. We added a number of aliases for Unix commands but if someone has installed those commands on WIndows, those aliases screw them up. > > We need to fix this.” So, maybe it will trigger a change anyway? The story is ongoing…

The AI gold rush is starting to drain one of the tech industry’s oldest and most reliable resources: the humble hard drive. New reports reveal that **high-capacity HDDs are now on backorder for up to two years** , as the world’s largest cloud providers rush to feed the insatiable storage needs of artificial intelligence systems. According to **TrendForce** and **DigiTimes** , demand for nearline enterprise HDDs, the massive drives that power hyperscale data centers, has exploded far beyond manufacturers’ production capacity. Companies such as **Google, Microsoft, and Oracle** are buying every terabyte they can find to support AI inference workloads, pushing global supply chains to the brink. ### **AI Inference Creates a Data Avalanche** The AI boom has shifted from training to **inference** , the phase where large language models and other neural networks actually perform tasks, and that’s where data volumes multiply exponentially. Unlike static storage for trained models, inference workloads constantly generate, process, and move data across servers. Every AI response, search query, image, or chatbot query generates gigabytes of data that must be stored somewhere. TrendForce notes that **nearline HDDs** , long considered the backbone of data center storage, are now facing **“severe shortages”** due to this unexpected spike in demand. As a result, **lead times for 30TB and 32TB hard drives have stretched from a few weeks to more than 52 weeks** , and in some cases, orders placed today might not arrive until **late 2026**. That’s not just a minor logistics hiccup; it’s a tectonic shockwave through the data storage market. ### **Western Digital: Prices Going Up, Shipments Slowing Down** The situation is so extreme that **Western Digital** has officially raised prices across its entire HDD lineup, citing “unprecedented demand” and a strain on manufacturing capacity. The company is also increasing its use of **ocean freight** to move products between regions, which could add **6 to 10****weeks** to shipping times for external and enterprise drives. For consumers and small businesses, that means **longer wait times and higher prices,** not because of gaming or video editing demand, but because **AI models are hogging your hard drive space**. ### **SSD Adoption Accelerates, Even for Cold Data** In a surprising twist, this shortage is forcing a **rapid industry pivot to QLC NAND-based SSDs** (quad-level cell flash). While SSDs are faster and more energy efficient than HDDs, they’re also more expensive and typically used for **“hot” or “warm” data** files that need to be accessed quickly. “Cold” data, on the other hand, usually lives on cheaper hard drives since it’s rarely touched. But with no HDDs to buy, even **cold data storage** is being offloaded to SSDs. That’s a bold and costly move: TrendForce warns that it’s **“not financially prudent”** to use SSDs for long-term archiving. Still, **cloud providers have no choice**. They’re now reconfiguring entire systems to handle SSD storage for workloads that traditionally relied on mechanical disks. And there’s another problem: **QLC NAND production is fully booked through 2026** , meaning SSD prices could soon follow HDDs upward. **QLC NAND to Overtake TLC by 2027** Industry analysts now predict that QLC NAND will **overtake TLC NAND** (triple-level cell flash) as the dominant storage technology by **early 2027**. The switch isn’t just about performance; it’s about sheer survival. QLC NAND packs more bits per cell, offering larger capacities at lower cost, albeit with slightly reduced endurance. For AI-scale workloads, that’s a tradeoff worth making. In fact, some NAND producers, like **SanDisk** , have already raised prices by **up to 50%** , after initially warning of smaller increases earlier this year. The ripple effect will likely hit consumer-grade SSDs next, especially since mainstream drives often use the same QLC supply chain as enterprise models. ### **The AI Effect: When Machines Outpace Manufacturing** This is the latest example of **AI outpacing the physical world’s ability to keep up**. Factories that make HDDs, NAND chips, and DRAM modules were never designed for the blistering scale of modern AI data centers, facilities that can consume **hundreds of petabytes** of storage each month. Just as the world witnessed a memory crunch earlier this year, with **DRAM prices doubling** due to AI-driven demand, the same story is now unfolding in storage. In the words of one industry insider: “AI money doesn’t wait for anyone.” Manufacturers are now prioritizing hyperscalers, clients willing to pay premium prices, over smaller enterprise buyers and consumers. This has left **ordinary users, small firms, and even national labs waiting months** for standard storage gear. ### **A Two-Year Wait for 32TB Drives** At the moment, **30TB Seagate Exos Mozaic+ drives** are still available for enterprise clients, but anything larger is practically impossible to find. Drives in the **32TB to 36TB** range, critical for cloud-scale storage, are now on **two-year backorder** lists. Even companies that rely on long-term archiving, such as video streaming services or research institutions, are being told to expect **delivery dates stretching into 2027**. For data centers planning extensive upgrades, that’s a serious operational nightmare. ### **SSD Prices Expected to Rise 5–10% in Q4 2025** TrendForce anticipates that **enterprise SSD contract prices** will rise between **5% and 10% in Q4 2025** , as hyperscalers rush to lock in capacity before NAND inventories tighten further. Given that all DRAM and NAND makers are now selling nearly everything they can produce to AI clients, the typical **2–3 month buffer** in supply chains has collapsed to just a few weeks. And that scarcity doesn’t just hit the data center giants, it **trickles down to regular consumers** , driving up the cost of gaming SSDs, NAS drives, and even external storage. ### **AI Eats the Storage World** What’s happening now may mark the start of a new era in storage economics, one where **AI infrastructure dictates the pace of manufacturing** , not consumer demand. Hard drives, once taken for granted, are now precious commodities in the race to power smarter machines. By 2026, expect AI training and inference farms to account for the majority of global HDD and SSD shipments. Analysts suggest that the shortage could continue until **2027** , when new fabrication plants and expanded HDD assembly lines finally come online. Until then, both **consumers and corporations may have to adapt to longer waits, higher costs, and shifting technologies**.

Vědci do dnešních dnů předpokládali, že za Faradayovým efektem stojí výhradně elektrická složka světla. Ta magnetická byla považována za příliš slabou a

V 81 letech dnes zemřel Jimmy Cliff, jamajský zpěvák a velký popularizátor reggae hudby, který ji pomohl dostat se do celého světa. Svou kariéru zahájil v šedesátých letech, posléze ho proslavily hity jako Wonderful World, Beautiful People a You Can Get It If You Really Want. Působil také jako herec.