<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: tristanj</title><link>https://news.ycombinator.com/user?id=tristanj</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 25 Sep 2026 20:17:59 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=tristanj" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by tristanj in "Google’s Project Suncatcher to put ML infrastructure in space"]]></title><description><![CDATA[
<p>1) The chips don't reach 220C. The 220C is the temperature at the hot end of the heat pump. The chips are on the cold end of the heat pump.<p>2) The International Space Station has used a dual-loop ammonia/water-based heat pump to cool the station temperatures. It's been in place for several decades. Heat pumps are a proven technology.<p>Other satellites have also used heat pumps, such as SES-17 in geostationary orbit  <a href="https://www.esa.int/Applications/Connectivity_and_Secure_Communications/Very_High_Throughput_Satellite_ready_to_pump_heat" rel="nofollow">https://www.esa.int/Applications/Connectivity_and_Secure_Com...</a></p>
]]></description><pubDate>Thu, 24 Sep 2026 20:47:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=49836539</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49836539</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49836539</guid></item><item><title><![CDATA[New comment by tristanj in "Making Tailscale Faster"]]></title><description><![CDATA[
<p>Tailscale takes two minutes to setup and you can add more devices with zero configuration.<p>WireGuard takes 30 mins to an hour to set up, you'll need to configure port forwarding, DDNS, create keys for each device, and add them to each device manually. But you have 100% control.<p>Performance-wise, I haven't noticed a difference. My internet connection maxes out way before Tailscale hits any performance limits.<p>Tailscale wins for convenience.</p>
]]></description><pubDate>Thu, 24 Sep 2026 01:57:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=49825256</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49825256</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49825256</guid></item><item><title><![CDATA[New comment by tristanj in "ArXiv receives multiyear commitments to support it as an independent nonprofit"]]></title><description><![CDATA[
<p>The preprint papers on arXiv are like 99% the same as the published versions, except they're free instead of locked behind a multi-thousand dollar/year paywall.</p>
]]></description><pubDate>Thu, 24 Sep 2026 00:57:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=49824825</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49824825</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49824825</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>I'm not advocating for helicopters at all - it's an analogy that clearly went over your head.<p>It's about cost comparison: SF Muni's operating costs (in terms of costs per passenger mile) are so bloated they're comparable to chartering a helicopter.<p>This should not be possible, and it's a strong signal of poor fiscal management. SF Muni operates at $3.62 per passenger mile. For comparison, Transport for London (TfL) operates at around $0.50/passenger mile and Shanghai Metro operates at $0.11/passenger mile.</p>
]]></description><pubDate>Wed, 23 Sep 2026 22:33:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=49823518</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49823518</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49823518</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>The numbers aren’t made up. The Federal government publishes these statistics annually, they're widely used by policy analysts, and they're simple to calculate yourself.<p>The Federal government spends ~$17.5B/year on maintaining and constructing the interstate highway system. Americans drive 852 billion miles/year on it. The cost per Vehicle Mile Traveled (VMT) is 2.1 cents / mile.<p>If we look at all government spending (Federal, State, Local) on highways, roads, and supporting infrastructure (which includes traffic management systems), that's around $285B/year and 3.28 trillion miles, or 8.7 cents per Vehicle Mile Traveled.<p>The cost to the government to provide and maintain roads is on the order of several cents per mile traveled per year. Roads are simply much more cost-effective to build and maintain than dedicated public transit infrastructure.</p>
]]></description><pubDate>Wed, 23 Sep 2026 22:12:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=49823305</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49823305</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49823305</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>But you can't ignore the financials. SF Muni's $3.62 cost per passenger mile is atrocious. At that price point, it's cheaper to charter a helicopter. I'm not kidding. You could literally charter a 20-seat Airbus H175 helicopter to fly people around for $2.50 per passenger mile. And you'd get there faster.<p>If public transit is less cost-effective than taking a helicopter, then something is clearly broken.</p>
]]></description><pubDate>Wed, 23 Sep 2026 08:11:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=49813037</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49813037</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49813037</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>In my opinion, the primary driver is the astronomical cost of housing in the SF bay area (which is 3x the national average), which drives up the cost of all labor. Public transit is infrastructure heavy, very labor intensive, and the high cost of labor makes it very expensive.<p>In the bay, low-skilled / manual laborers spend >50% of their income on housing. Many can't afford to live here, so they demand higher wages or leave, which drives up the price of labor even further.<p>If sufficient housing were built and the Bay Area had the same housing prices as the rest of the US, then 1) the supply of labor would go up 2) wages could fall by 30-40% without a quality of life decrease. Businesses could reduce their labor costs by 20%, which would improve the QoL of everyone in the area, including manual laborers. When local businesses cut costs, then everything costs less (food, housing, transport), and QoL goes up.<p>Then run this process iteratively and you will see massive reductions in prices everywhere, the cost of everything will go down by 30%+ percent.<p>For the case of BART, they spend ~$800 million / year on labor, so reducing cost here has the highest RoI. They would save hundreds of millions per year.</p>
]]></description><pubDate>Wed, 23 Sep 2026 07:06:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49812632</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49812632</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49812632</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>The best formula for public transit is the Rail plus Property business model. In Hong Kong, the MTR Corporation buys up land adjacent to a planned future station, then builds the station afterwards. Then makes money from renting out the properties adjacent to the station.<p>JR East in Japan operates in the same way: it makes the majority of its profit from the real estate, malls, and hotels it operates adjacent to its train stations.</p>
]]></description><pubDate>Wed, 23 Sep 2026 06:25:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=49812397</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49812397</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49812397</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>Yeah I was shocked as well when I learned about the financials. I'll the subsidy per passenger mile to the table as well.<p>I remember talking to a VC a few years back, and he vehemently explained why he will never ever fund companies related to public transit, for this reason. Public transit was one of the two sectors he would not invest in (the other being materials science).</p>
]]></description><pubDate>Wed, 23 Sep 2026 05:15:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=49811930</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49811930</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49811930</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>Sure, but cars aren't subsidized at $18 dollars per trip. The cost to the government of providing and maintaining roads is on the order of several cents per vehicle mile traveled.<p>For BART, that number is:<p>$1.07B operating expense / 892.5M million passenger miles  traveled = $1.20 per mile. Then subtract out the fare per mile and the government is paying $0.84/mile for people to ride BART.</p>
]]></description><pubDate>Wed, 23 Sep 2026 04:54:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=49811803</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49811803</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49811803</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>Obviously yes, you have to since the vast majority of expenses (>70%) are fixed costs. If you only exclude them and only look at marginal costs, they're profitable.<p>But you cannot run a transit system on marginal costs, so using that comparison is also misleading.</p>
]]></description><pubDate>Wed, 23 Sep 2026 04:43:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=49811740</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49811740</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49811740</guid></item><item><title><![CDATA[New comment by tristanj in "Transit rewards"]]></title><description><![CDATA[
<p>The reason there's not more public transit is because the economics are terrible. Looking at the financials:<p><pre><code>  Cost per passenger trip:
    BART: $1.07B operating expenses / 59.5M trips = $16.92
    SF Muni: $1.13B / 160M trips = $7.06
    Caltrain: $225M / 9.1M trips = $24.73

  Passenger Fare collected per trip:
    BART: $259M / 59.5M trips = $4.35
    SF Muni: $111M / 160M trips = $0.69
    Caltrain: $58.7M / 9.1M trips = $6.45

  Subsidy per trip:
    BART: $12.57
    SF Muni: $6.37
    Caltrain: $18.27

  Cost per passenger mile:
    BART: $1.07B / 892.5M passenger miles = $1.20/mile
    SF Muni: $1.13B / 312M = $3.62/mile
    Caltrain: $225M / 177.5M = $1.27/mile

  Subsidy per passenger mile:
    BART: $1.07B / 892.5M passenger miles = $0.84/mile
    SF Muni: $1.13B / 312M = $3.26/mile
    Caltrain: $225M / 177.5M = $0.94/mile
</code></pre>
Every single ride requires a $6 to $18 subsidy, paid for by tax dollars.<p>For comparison, federal spending for the interstate highway system is about 2-3 cents per passenger mile.</p>
]]></description><pubDate>Wed, 23 Sep 2026 04:23:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49811620</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49811620</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49811620</guid></item><item><title><![CDATA[Can the US Build a DJI? A Teardown of Two DJI Drones]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.arenaphysica.com/publications/dji-teardown">https://www.arenaphysica.com/publications/dji-teardown</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49810113">https://news.ycombinator.com/item?id=49810113</a></p>
<p>Points: 25</p>
<p># Comments: 14</p>
]]></description><pubDate>Wed, 23 Sep 2026 00:32:20 +0000</pubDate><link>https://www.arenaphysica.com/publications/dji-teardown</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49810113</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49810113</guid></item><item><title><![CDATA[New comment by tristanj in "PDF Forgeries Are Surprisingly Rare (2022)"]]></title><description><![CDATA[
<p>I think a big reason for this is that PDF editing tools are just pretty terrible.<p>PDFs are meant to be a presentation document format, not an editable document format. I have some experience editing PDFs, and it's difficult to make significant modifications. Editing small sections of text or replacing a date is fine, but once you try to start editing entire paragraphs or changing the layout, it quickly becomes unmanageable. The kerning looks weird, the spacing between words looks off, and when you look into it you realize the the sentence you're trying to edit is actually split across a dozen text objects. For these it becomes almost impossible to edit without destroying the spacing, so you just give up.<p>And the approved way to edit these PDFs is via Adobe Acrobat Pro, which costs $20/month (ugh), which I'm not willing to pay for.<p>So I think this is why we just don't see that many PDF forgeries. The editing tools just aren't very good. Maybe it's different now we can just ask Claude to edit our PDFs for us.</p>
]]></description><pubDate>Tue, 22 Sep 2026 04:48:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49796892</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49796892</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49796892</guid></item><item><title><![CDATA[New comment by tristanj in "Smart Ring Maker Oura, Backers Seek $2.2B in US IPO"]]></title><description><![CDATA[
<p>I believe OURA is living on borrowed time and only exists because Apple hasn't yet entered the smart ring market. It's a repeat of the Fitbit situation. Before the Apple Watch launched in 2015, Fitbit had 30% of the wearables market. Then the Apple came in, and Fitbit started bleeding users: high end ones to the Apple Watch, and low end ones to cheap Chinese competition. Fitbit eventually collapsed and got sold in a fire sale to Google.<p>Apple is already rumored to be working on their own smart ring. Samsung is already in the market. This is one reason why Oura is in a rush to IPO, they need liquidity for early investors + cash to compete before Apple comes in and takes a huge slice of their market.</p>
]]></description><pubDate>Tue, 22 Sep 2026 04:12:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=49796715</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49796715</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49796715</guid></item><item><title><![CDATA[New comment by tristanj in "RSA-896"]]></title><description><![CDATA[
<p>There is no bounty, RSA labs ended the $75,000 reward in 2007.</p>
]]></description><pubDate>Sun, 20 Sep 2026 04:09:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49772437</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49772437</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49772437</guid></item><item><title><![CDATA[New comment by tristanj in "RSA-896"]]></title><description><![CDATA[
<p>But Anthropic isn't paying for the electricity and cooling. They don't run their own data centers, they rent compute from providers who cover those costs.<p>That's entirely why they can blow compute on the fun projects like this. If they had to pay extra for the electricity, they wouldn't do it.</p>
]]></description><pubDate>Sun, 20 Sep 2026 03:25:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=49772276</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49772276</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49772276</guid></item><item><title><![CDATA[New comment by tristanj in "RSA-896"]]></title><description><![CDATA[
<p>If you've already paid for and reserved a whole cluster of GPUs, any idle capacity is capacity you've already paid for. Using it is effectively free. So might as well use it to solve fun math puzzles.<p>Though, it would make more financial sense to mine crypto.</p>
]]></description><pubDate>Sun, 20 Sep 2026 02:52:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49772137</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49772137</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49772137</guid></item><item><title><![CDATA[New comment by tristanj in "Apple M6 Pro Achieves the Highest Single-Core CPU Score in Geekbench 7"]]></title><description><![CDATA[
<p>Internal leaks, these happen frequently. Employees have access to these devices for months prior to launch. Someone ran a benchmark and uploaded it on a machine they weren't supposed to.</p>
]]></description><pubDate>Sat, 19 Sep 2026 09:03:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=49764709</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49764709</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49764709</guid></item><item><title><![CDATA[New comment by tristanj in "Apple M6 Pro Achieves the Highest Single-Core CPU Score in Geekbench 7"]]></title><description><![CDATA[
<p>That was a mistake and I updated it - only the M2 Max benchmarks leaked prior to announcement, not the M2.<p>Those M2 Max Geekbench benchmarks appeared on November 30, 2022. Apple announced the Apple M2 Max chip on January 17, 2023.<p>Similarly, those M4 Geekbench benchmarks for Mac were leaked on October 6, 2024, prior to Apple announcing the product. Apple announced the M4 Macs on October 28, 2024.</p>
]]></description><pubDate>Sat, 19 Sep 2026 09:02:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=49764708</link><dc:creator>tristanj</dc:creator><comments>https://news.ycombinator.com/item?id=49764708</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49764708</guid></item></channel></rss>