<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: dfox</title><link>https://news.ycombinator.com/user?id=dfox</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 19 Nov 2026 02:56:04 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=dfox" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by dfox in "Ledger confirms spy chip implant behind wallet crypto thefts"]]></title><description><![CDATA[
<p>The root cause of this attack is that the typical cryptocurrency wallet operational model involves the key leaving the device in plain text as part of the setup/backup process. Such a thing is impossible on any normal correctly configured HSM (while creating backups is possible by some sort of multi-step process where the secret key material never exists in plaintext outside the security boundary, how exactly that is done differs significantly between vendors).</p>
]]></description><pubDate>Wed, 14 Oct 2026 09:49:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=50084937</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=50084937</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50084937</guid></item><item><title><![CDATA[New comment by dfox in "Example.com just launched the biggest redesign in decades"]]></title><description><![CDATA[
<p>It seems that they changed it again and now there is minified HTML with english message + simple script that injects the other languages and the icon, no animation. Which to me seems much more sensible.</p>
]]></description><pubDate>Tue, 06 Oct 2026 08:45:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=49975890</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49975890</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49975890</guid></item><item><title><![CDATA[New comment by dfox in "Starlink ground station in Poland hit by fire in suspected arson attack"]]></title><description><![CDATA[
<p>Note that this list is list of places where if you are enterprise starlink customer SpaceX can provide you with physical ethernet port to plug your cable into when you want to handle your traffic yourself instead of that just going directly to the internet. That is kind of obscure service that even land-based PLMN operators offer but usually do not document in this way in publicly available documentation.<p>From the perspective of network design, the fact that this certainly is not exhaustive list of places where SpaceX has physical DC presence and physical reality it comes that some of these datacenters are selected based on the fact that the ground station is somewhere nearby (I would be surprised if there is even just one place on the list where nearby means "on the roof", because there is a lot of technical, logistical and administrative reasons why you do not want to put such a thing at a roof of a datacenter).</p>
]]></description><pubDate>Thu, 24 Sep 2026 11:40:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49829254</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49829254</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49829254</guid></item><item><title><![CDATA[New comment by dfox in "What Sun got wrong"]]></title><description><![CDATA[
<p>The "two E10Ks" part is the important part there, as you need a very powerful server to support SunRay users running Solaris. At my university there was something like 200 SunRays supported by cluster of T2000s (7 nodes IIRC), it was fine for CDE, Firefox and few dtterms, completely unusable with Gnome (it took Gnome like 10 minutes to even start).<p>On the other hand I have heard of some success stories about using SunRays as RDP clients, which is actually supported configuration (you still need a Solaris box that manages the sunrays and translates RDP to ALP).</p>
]]></description><pubDate>Tue, 22 Sep 2026 15:37:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=49803075</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49803075</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49803075</guid></item><item><title><![CDATA[New comment by dfox in "Wax motor"]]></title><description><![CDATA[
<p>In washing machine application the locking action is quite fast as the heating element is PTC connected more or less directly to mains, pretty substantial amount of energy gets dumped into the wax over few AC line cycles. In that application (and also in most industrial automation situations) ~100ms is more than fast enough because when most of the effector components are AC powered you end up with ~40ms worst case actuation delay just from the physics of AC electricity. On my washing machine it is quite noticeable that when you press "start" the door gets positively locked well before the capacitive sensing "button" finishes playing its haptic feedback non-sense. (the fact that front panel of cheap-ish toploading washing machine is actually a huge touchpad with haptic feedback that is designed to look and behave like five physical buttons says many things about economics of manufacturing stuff for that somewhat hostile environment at these scales)</p>
]]></description><pubDate>Fri, 18 Sep 2026 11:55:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=49753090</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49753090</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49753090</guid></item><item><title><![CDATA[New comment by dfox in "Keys Not Included: recovering the signing keys for US driver's license barcodes"]]></title><description><![CDATA[
<p>You are assuming that the forgers care about the machine readable part at all. Vast majority of forged EU ID cards I have seen are trivially recognizable by the fact that the MRZ contains something that kinda-sorta matches the human readable part, but is syntactically invalid and has wrong checksums.</p>
]]></description><pubDate>Thu, 17 Sep 2026 12:42:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=49739904</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49739904</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49739904</guid></item><item><title><![CDATA[New comment by dfox in "Ubuntu 26.10 completes transition to Rust-based coreutils"]]></title><description><![CDATA[
<p>The problem there is that this is exactly the class of bug that does not exist in GNU coreutils because of philosophy of that project. Non-existence of such bugs proves that the impementation is not copied from AT&T code.</p>
]]></description><pubDate>Mon, 14 Sep 2026 15:21:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=49698499</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49698499</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49698499</guid></item><item><title><![CDATA[New comment by dfox in "Suica, Japan's First IC Transit Card"]]></title><description><![CDATA[
<p>At a first glance passive RFID would be a perfect match for ride photos, the whole mechanism is ridiculously fast so speed of the ride would not matter much (the technology works at highway speeds just fine) and can work over ~15 meters.<p>But there are two somewhat significant issues with the radio layer in this application: the geometry of tag antenna is somewhat critical (making it into a band would be problematic) and the back scatter is attenuated by salt water blobs (ie. people) almost perfectly.</p>
]]></description><pubDate>Thu, 27 Aug 2026 16:36:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=49467534</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49467534</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49467534</guid></item><item><title><![CDATA[New comment by dfox in "IBM Unveils Next Generation Dual-Architecture Processor for IBM Z and LinuxONE"]]></title><description><![CDATA[
<p>I suspect that is not an inherent limitation of the dual-ISA design (the HotChips slides mention some sort of bidirectional thread state mapping between arm and z), but just about there not being z/VM release that supports that (building such a thing is probably SMoP, but another question is whether that makes business sense).</p>
]]></description><pubDate>Thu, 27 Aug 2026 14:23:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49465372</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49465372</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49465372</guid></item><item><title><![CDATA[New comment by dfox in "IBM Unveils Next Generation Dual-Architecture Processor for IBM Z and LinuxONE"]]></title><description><![CDATA[
<p>IBM's mainframes do not really have a reset vector, the CPU gets initialized by some external means into whatever state the OS expects and then the clock gets enabled, so that distinction is quite moot.</p>
]]></description><pubDate>Thu, 27 Aug 2026 14:05:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=49465116</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49465116</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49465116</guid></item><item><title><![CDATA[New comment by dfox in "SIMD in the 90s: Programming Intel's Pentium MMX"]]></title><description><![CDATA[
<p>They were always required. IIRC early AMD64 CPUs did not support x87 instructions in long mode at all (causing #UD), and that support was Intel's extension in first EM64T CPUs.</p>
]]></description><pubDate>Mon, 17 Aug 2026 11:24:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=49329226</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=49329226</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49329226</guid></item><item><title><![CDATA[New comment by dfox in "My favorite keyboards"]]></title><description><![CDATA[
<p>Basically the only thing interesting about the ConcertMaster is that Carmack used it (and that it looks somewhat cool). As for the keyboard itself, it is just a basic OEM membrane keyboard and not that good one even. The speakers are lets say adequate for the time and size.<p>Interestingly, the speaker part and the keyboard part are completely separate. The "cable" consists of four separate cables (keyboard, power, line out, mic in) in a thin sleeve. Mine was supplied with AT plug on the keyboard cable and Y-adapter that converted PS2 into AT DIN and barell jack for the speaker power. The keyboard label indeed says it is powered by 9V DC, but I guess that never really happened as PS/2 is 5V no matter what various devices say.<p>Edit: And as for the supper hard to find nowadays part: I suspect that part of the reason is that the keyboard module inside the thing is ridiculously sensitive to even minor spills.</p>
]]></description><pubDate>Thu, 02 Jul 2026 11:31:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=48759770</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48759770</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48759770</guid></item><item><title><![CDATA[New comment by dfox in "Libre Barcode Project"]]></title><description><![CDATA[
<p>Do not do this unless you do not have any other choice. Preferrably use whatever native barcode support of the printer involved, if it does not have that, just generate the barcode as vector image or bitmap with a resolution that is a integer fraction of the printers resolution. Generating correct Code128 as a SVG is about the same amount of work as generating the correct input for some sort of barcode font (the hard part is determining the switches between character sets, not generating bars from bytes).</p>
]]></description><pubDate>Fri, 26 Jun 2026 10:27:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48684870</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48684870</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48684870</guid></item><item><title><![CDATA[New comment by dfox in "Libre Barcode Project"]]></title><description><![CDATA[
<p>The fact that this is standard practice does not mean that it is not perverse. It kind of works sanely for plain Code39 (and even then you will see effects of doing that in weird places, like VAG stamping human readable VIN on a chassis, including the Code39 start/stop symbols), once you start using barcode fonts for Code128-derived symbologies (ie. UPC/EAN) the whole thing becomes a pointless exercise.</p>
]]></description><pubDate>Fri, 26 Jun 2026 10:20:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=48684809</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48684809</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48684809</guid></item><item><title><![CDATA[New comment by dfox in "Bohemia Interactive: Cold War Assault Remastered Source Code on GitHub"]]></title><description><![CDATA[
<p>It had, another thing is that it used somewhat questionable system for CD key validation, which is stubbed out in this release (<a href="https://github.com/BohemiaInteractive/CWR/blob/main/engine/Poseidon/Security/Serial.cpp" rel="nofollow">https://github.com/BohemiaInteractive/CWR/blob/main/engine/P...</a>).<p>One thing is 120bit RSA (readily broken with a graphing calculator at the time of release), another thing is the provenance of the RSA implementation code in the original binary.</p>
]]></description><pubDate>Thu, 25 Jun 2026 12:22:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=48672316</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48672316</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48672316</guid></item><item><title><![CDATA[New comment by dfox in "The time the x86 emulator team found code so bad they fixed it during emulation"]]></title><description><![CDATA[
<p>The most important fix in SP1 for Office 2007 was fixing exactly that in Excel. Doing ridiculous amount of 4 byte reads made it basically unusable on network filesystems.</p>
]]></description><pubDate>Tue, 16 Jun 2026 13:00:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=48554591</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48554591</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48554591</guid></item><item><title><![CDATA[New comment by dfox in "How Shamir's Secret Sharing Works"]]></title><description><![CDATA[
<p>One point is that there is no reason for the entire secret to be one element of the underlying field, it can very well be a n-tuple of elements of a smaller field, with GF(2^8) being the somewhat obvious choice if you do not expect ridiculous numbers of shares, no need to deal with bignum math.</p>
]]></description><pubDate>Tue, 26 May 2026 09:22:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=48277214</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48277214</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48277214</guid></item><item><title><![CDATA[New comment by dfox in "Hyperpolyglot Lisp: Common Lisp, Racket, Clojure, Emacs Lisp"]]></title><description><![CDATA[
<p>> cannot start with digit<p>Not only it can, but both CL and Emacs Lisp actually defines primitives with names that start with digit.</p>
]]></description><pubDate>Tue, 19 May 2026 09:02:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=48190976</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48190976</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48190976</guid></item><item><title><![CDATA[New comment by dfox in "Google Cloud fraud defense, the next evolution of reCAPTCHA"]]></title><description><![CDATA[
<p>It is far from being universal. And the more annoying part of that is that there are at least 4 incompatible "standards" as to the format of the QRcode.</p>
]]></description><pubDate>Thu, 07 May 2026 12:46:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=48048756</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=48048756</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48048756</guid></item><item><title><![CDATA[New comment by dfox in "5x5 Pixel font for tiny screens"]]></title><description><![CDATA[
<p>IIRC the really cheap Casio Organizers/DataBanks of 90's used 5x5 font. And then my ex used something like that on linux in order to fit a ridiculous amount of xterms onto 14" CRT (somewhat absurd feat with her congenital vision defect).</p>
]]></description><pubDate>Wed, 22 Apr 2026 17:31:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=47866650</link><dc:creator>dfox</dc:creator><comments>https://news.ycombinator.com/item?id=47866650</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47866650</guid></item></channel></rss>