<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: sarpdag</title><link>https://news.ycombinator.com/user?id=sarpdag</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 12 Sep 2026 08:10:20 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=sarpdag" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by sarpdag in "Measuring Claude 4.7's tokenizer costs"]]></title><description><![CDATA[
<p>Since the Opus 4.7 release. I hit my 5 hour window limit second time on claude code max plan, which never happened before. I am not happy for sure.</p>
]]></description><pubDate>Sat, 18 Apr 2026 16:45:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47817297</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=47817297</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47817297</guid></item><item><title><![CDATA[New comment by sarpdag in "Schedule tasks on the web"]]></title><description><![CDATA[
<p>I can't pick the effort for the tasks run on Claude Web. I have a feeling Claude is using low or medium effort on those tasks, and I observe clear quality differences with the task ran on my local claude code, which uses high effort.</p>
]]></description><pubDate>Fri, 27 Mar 2026 15:26:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=47543857</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=47543857</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47543857</guid></item><item><title><![CDATA[New comment by sarpdag in "Dark patterns: Buying a Bahncard at Deutsche Bahn"]]></title><description><![CDATA[
<p>Sadly, Deutsche Bahn has become, for me, a symbol of how Germany is going downhill.<p>I got a BahnCard 25 for 3 months, and one month before it expired, it automatically renewed for a full year. Even though I canceled it, I still had to pay for the entire year.<p>I bought a train ticket to Austria, and they announced strikes just five days before my trip. They didn’t refund my money directly. I had to fill out a form in person at DB, and I only got 80% of the money back.<p>The trains are unreliable; they are often delayed.</p>
]]></description><pubDate>Thu, 09 Oct 2025 18:28:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=45531320</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=45531320</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45531320</guid></item><item><title><![CDATA[Ask HN: Why don't we have a mobile app to screen for skin cancer?]]></title><description><![CDATA[
<p>We already have the data, knowledge, and technology to build this. Why can’t I just use my phone camera to check if a spot on my skin looks concerning? It wouldn’t need to give a full diagnosis—just suggest whether I should see a doctor about a specific area.<p>If such apps exist, why aren’t they widely available?</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=45383834">https://news.ycombinator.com/item?id=45383834</a></p>
<p>Points: 2</p>
<p># Comments: 5</p>
]]></description><pubDate>Fri, 26 Sep 2025 07:36:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=45383834</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=45383834</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45383834</guid></item><item><title><![CDATA[New comment by sarpdag in "Ask HN: Cursor is using 269,738 tokens to edit 1200 token file"]]></title><description><![CDATA[
<p>Actually they are charging for 269000 tokens. Looks like they changed how they charge accounts.<p>The screenshot for my dashboard: <a href="https://ibb.co/2Vy5CRJ" rel="nofollow">https://ibb.co/2Vy5CRJ</a><p>Cursor Version: 1.2.4 (Universal)</p>
]]></description><pubDate>Fri, 18 Jul 2025 12:48:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=44604124</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=44604124</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44604124</guid></item><item><title><![CDATA[Ask HN: Cursor is using 269,738 tokens to edit 1200 token file]]></title><description><![CDATA[
<p>I started to track my Cursor usage via https://cursor.com/dashboard?tab=usage.<p>The pricing and usage of tokens hugely changed recently. To test I gave a simple task to Cursor which would use a 192 line swift code, including comments. This file has 1213 tokens according to https://platform.openai.com/tokenizer .<p>Here is my prompt: "There is a bug when i click on a raw it plays the audio and it deletes it afterwards, even though without clicking the delete button."<p>The Cursor dashboard says it used 269,738 tokens, and records to usage accordingly. I am a pro plan user.<p>Is there anyone else having similar issues?</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44603443">https://news.ycombinator.com/item?id=44603443</a></p>
<p>Points: 5</p>
<p># Comments: 4</p>
]]></description><pubDate>Fri, 18 Jul 2025 11:22:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=44603443</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=44603443</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44603443</guid></item><item><title><![CDATA[New comment by sarpdag in "OpenAI wins $200M U.S. defense contract"]]></title><description><![CDATA[
<p>A journey from nonprofit to military contract.</p>
]]></description><pubDate>Tue, 17 Jun 2025 15:41:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=44300472</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=44300472</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44300472</guid></item><item><title><![CDATA[New comment by sarpdag in "Show HN: MidWord – A Word-Guessing Game"]]></title><description><![CDATA[
<p>It was fun. I don't know if I would play again or not, but played 2 times without losing focus.</p>
]]></description><pubDate>Tue, 10 Jun 2025 21:09:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=44241460</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=44241460</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44241460</guid></item><item><title><![CDATA[New comment by sarpdag in "Lines of code that beat A/B testing (2012)"]]></title><description><![CDATA[
<p>I really like multi armed bandit approach, but struggles with common scenarios involving delayed rewards or multiple success criteria, such as testing ecommerce search with number of orders and GMV guardrails.<p>For simple, immediate-feedback cases like button clicks, the specific implementation becomes less critical.</p>
]]></description><pubDate>Tue, 14 Jan 2025 13:06:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=42696696</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=42696696</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42696696</guid></item><item><title><![CDATA[New comment by sarpdag in "IMG_0001"]]></title><description><![CDATA[
<p>Authentic and nostalgic, I am enjoying watching.</p>
]]></description><pubDate>Wed, 04 Dec 2024 05:51:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=42314790</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=42314790</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42314790</guid></item><item><title><![CDATA[New comment by sarpdag in "Notes on OpenAI's new o1 chain-of-thought models"]]></title><description><![CDATA[
<p>I have tried the "mad cow" joke on o1-mini and it is still failing to explain correctly, but o1-preview correctly states "The joke is funny because the second cow unwittingly demonstrates that she is already affected by mad cow disease."</p>
]]></description><pubDate>Fri, 13 Sep 2024 07:49:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=41528965</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=41528965</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41528965</guid></item><item><title><![CDATA[New comment by sarpdag in "Show HN: Sorbaba – A Search Engine Where Users Can Change the Results"]]></title><description><![CDATA[
<p>I got the domain in 2010 to build a different search engine, but sadly I always postponed.</p>
]]></description><pubDate>Mon, 09 Sep 2024 18:44:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=41491951</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=41491951</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41491951</guid></item><item><title><![CDATA[New comment by sarpdag in "Show HN: Sorbaba – A Search Engine Where Users Can Change the Results"]]></title><description><![CDATA[
<p>I would let the user choose to see the new content vs old but upvoted. Also It is important to publish the ranking behind it, if it is not straight forward.</p>
]]></description><pubDate>Mon, 09 Sep 2024 18:21:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=41491682</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=41491682</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41491682</guid></item><item><title><![CDATA[New comment by sarpdag in "Show HN: Sorbaba – A Search Engine Where Users Can Change the Results"]]></title><description><![CDATA[
<p>Thank you. Please let me know what is missing or should be better from your perspective.</p>
]]></description><pubDate>Mon, 09 Sep 2024 16:05:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=41489924</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=41489924</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41489924</guid></item><item><title><![CDATA[New comment by sarpdag in "Show HN: Sorbaba – A Search Engine Where Users Can Change the Results"]]></title><description><![CDATA[
<p>Search engines not only control what information you see but also how it’s ranked. Today, information is tightly controlled by a few major platforms. AI tools like ChatGPT or Perplexity typically provide a single answer to a question, even if they include sources.<p>Platforms like Reddit are excellent for complex discussions, but they’re overkill for simple queries like “What’s Michael Jackson’s age?” and not efficient for straightforward information needs. Your profile is tied to the questions you ask.<p>I am building a proof of concept that puts both the creation and ranking of information in the hands of users, supported by AI for speed.<p>AI will gather answers from various sources, offering a range of perspectives, articles. One answer per source, rather than unifying them into a single response. The World Wide Web is wonderful. Knowledge sources should remain distributed. People should visit the original source to get the full context.<p>Users can contribute their own answers and points of view, allowing others to see multiple perspectives and real world expertise.<p>Instead of algorithms deciding what ranks highest, the community will upvote or downvote answers. The best content will rise based on real user feedback, not hidden ranking systems.<p>Please let me know what do you think about it?</p>
]]></description><pubDate>Mon, 09 Sep 2024 15:44:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=41489674</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=41489674</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41489674</guid></item><item><title><![CDATA[Show HN: Sorbaba – A Search Engine Where Users Can Change the Results]]></title><description><![CDATA[
<p>Article URL: <a href="https://sorbaba.com/">https://sorbaba.com/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=41489659">https://news.ycombinator.com/item?id=41489659</a></p>
<p>Points: 4</p>
<p># Comments: 7</p>
]]></description><pubDate>Mon, 09 Sep 2024 15:43:09 +0000</pubDate><link>https://sorbaba.com/</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=41489659</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41489659</guid></item><item><title><![CDATA[New comment by sarpdag in "Ask HN: iOS App Stuck in Review for 4 Weeks – Any Advice?"]]></title><description><![CDATA[
<p>I will try immediately.</p>
]]></description><pubDate>Mon, 12 Feb 2024 16:40:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=39346968</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=39346968</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39346968</guid></item><item><title><![CDATA[Ask HN: iOS App Stuck in Review for 4 Weeks – Any Advice?]]></title><description><![CDATA[
<p>Hi HN,<p>I submitted my iOS app to the App Store 4 weeks ago(14th Jan), and it's still "In Review." Attempts to get clarity from Apple haven't yielded much. I'm looking for advice:<p>Has anyone faced similar long review times? How did you handle it?
Any tips for expediting the review or getting more info from Apple?
Suggestions to avoid such delays in future submissions?
Would really appreciate any insights or experiences you can share. Thanks!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39346883">https://news.ycombinator.com/item?id=39346883</a></p>
<p>Points: 2</p>
<p># Comments: 2</p>
]]></description><pubDate>Mon, 12 Feb 2024 16:35:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=39346883</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=39346883</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39346883</guid></item><item><title><![CDATA[New comment by sarpdag in "The Boyer-Moore Fast String Searching Algorithm"]]></title><description><![CDATA[
<p>I implemented it in Go for fun. I expected Go's standard lib will beat it but for long substring searches it perform up to 10 times faster than `strings.Index`.<p><a href="https://github.com/sarpdag/boyermoore" rel="nofollow">https://github.com/sarpdag/boyermoore</a><p>Go's strings package depends on the substring size tries different approaches like Rabin Karp algorithm, (<a href="https://en.wikipedia.org/wiki/Rabin%E2%80%93Karp_algorithm" rel="nofollow">https://en.wikipedia.org/wiki/Rabin%E2%80%93Karp_algorithm</a>)</p>
]]></description><pubDate>Sat, 24 Apr 2021 19:40:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=26927703</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=26927703</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=26927703</guid></item><item><title><![CDATA[New comment by sarpdag in "Show HN: Type-aware code search for Go"]]></title><description><![CDATA[
<p>Currently searches on most starred 100 Go repositories on Github including big code bases like Kubernetes, Ethereum... and returns results mostly under a second.<p>I couldn't find a similar solution so building and experimenting on it.<p>Let me know what you think!</p>
]]></description><pubDate>Tue, 20 Apr 2021 14:06:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=26874863</link><dc:creator>sarpdag</dc:creator><comments>https://news.ycombinator.com/item?id=26874863</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=26874863</guid></item></channel></rss>