The best Hacker News stories from Show from the past day
Latest posts:
Show HN: Automatic prompt optimizer for LLMs
Show HN: WasmGPT – “ChatGPT” in the browser, no WebGPU and no server needed
Using threaded emscripten to speed up the generation and offload the main loop. No SIMD or other optimizations. Might work faster with #enable-experimental-webassembly-features enabled.<p>Tested in x86 Chrome and Firefox, Apple Silicon Safari<p>Run it yourself: <a href="https://github.com/lxe/ggml/tree/wasm-demo">https://github.com/lxe/ggml/tree/wasm-demo</a><p>Thanks, <a href="https://github.com/ggerganov/ggml">https://github.com/ggerganov/ggml</a>,
Show HN: WasmGPT – “ChatGPT” in the browser, no WebGPU and no server needed
Using threaded emscripten to speed up the generation and offload the main loop. No SIMD or other optimizations. Might work faster with #enable-experimental-webassembly-features enabled.<p>Tested in x86 Chrome and Firefox, Apple Silicon Safari<p>Run it yourself: <a href="https://github.com/lxe/ggml/tree/wasm-demo">https://github.com/lxe/ggml/tree/wasm-demo</a><p>Thanks, <a href="https://github.com/ggerganov/ggml">https://github.com/ggerganov/ggml</a>,
Show HN: WasmGPT – “ChatGPT” in the browser, no WebGPU and no server needed
Using threaded emscripten to speed up the generation and offload the main loop. No SIMD or other optimizations. Might work faster with #enable-experimental-webassembly-features enabled.<p>Tested in x86 Chrome and Firefox, Apple Silicon Safari<p>Run it yourself: <a href="https://github.com/lxe/ggml/tree/wasm-demo">https://github.com/lxe/ggml/tree/wasm-demo</a><p>Thanks, <a href="https://github.com/ggerganov/ggml">https://github.com/ggerganov/ggml</a>,
Show HN: TxtNet Browser – Browse the Web over SMS, No Wi-Fi/Mobile Data Needed
Hello all,<p>This is my second year of working on a project[1] with the goal of browsing the web, on an Android smartphone, without reliance on Wi-Fi or mobile data. While this concept might seem aimless, my goal was to provide a way for people in areas with limited, expensive, or censored cellular internet access a way to view the web in a basic format. I finished work on a basic client-server model last year[2], and this year, I implemented a new pseudo-distributed peer-to-peer model that allows any TxtNet Browser user to use their own smartphone to run a background server service that communicates via the user's own primary mobile number. The main advantage to this model over last year's use of the Twilio API is the fact that with an unlimited SMS plan from a consumer carrier, you will likely end up paying significantly less than the amount you would pay for Twilio credits (averaging about ~$0.50 per website). There's a lot going on with the stateless nature of SMS, GSM-7 encoding, and Brotli compression, so please ask any questions you might have!<p>I've also started up a test server instance running on a +1 country code phone number, so feel free to test out the app with your own smartphone. Like mentioned in the GitHub repo, please be aware that I (necessarily) have access to every phone number and associated request that is sent. Of course, anyone can host their own server instance, and if you would like to share it, feel free to get in touch so I can add the number to the repo! Also, there are likely many bugs still lurking, so feel free to report those.<p>[1] <a href="https://github.com/lukeaschenbrenner/TxtNet-Browser/">https://github.com/lukeaschenbrenner/TxtNet-Browser/</a><p>[2] <a href="https://news.ycombinator.com/item?id=32905496" rel="nofollow">https://news.ycombinator.com/item?id=32905496</a>
Show HN: TxtNet Browser – Browse the Web over SMS, No Wi-Fi/Mobile Data Needed
Hello all,<p>This is my second year of working on a project[1] with the goal of browsing the web, on an Android smartphone, without reliance on Wi-Fi or mobile data. While this concept might seem aimless, my goal was to provide a way for people in areas with limited, expensive, or censored cellular internet access a way to view the web in a basic format. I finished work on a basic client-server model last year[2], and this year, I implemented a new pseudo-distributed peer-to-peer model that allows any TxtNet Browser user to use their own smartphone to run a background server service that communicates via the user's own primary mobile number. The main advantage to this model over last year's use of the Twilio API is the fact that with an unlimited SMS plan from a consumer carrier, you will likely end up paying significantly less than the amount you would pay for Twilio credits (averaging about ~$0.50 per website). There's a lot going on with the stateless nature of SMS, GSM-7 encoding, and Brotli compression, so please ask any questions you might have!<p>I've also started up a test server instance running on a +1 country code phone number, so feel free to test out the app with your own smartphone. Like mentioned in the GitHub repo, please be aware that I (necessarily) have access to every phone number and associated request that is sent. Of course, anyone can host their own server instance, and if you would like to share it, feel free to get in touch so I can add the number to the repo! Also, there are likely many bugs still lurking, so feel free to report those.<p>[1] <a href="https://github.com/lukeaschenbrenner/TxtNet-Browser/">https://github.com/lukeaschenbrenner/TxtNet-Browser/</a><p>[2] <a href="https://news.ycombinator.com/item?id=32905496" rel="nofollow">https://news.ycombinator.com/item?id=32905496</a>
Show HN: TxtNet Browser – Browse the Web over SMS, No Wi-Fi/Mobile Data Needed
Hello all,<p>This is my second year of working on a project[1] with the goal of browsing the web, on an Android smartphone, without reliance on Wi-Fi or mobile data. While this concept might seem aimless, my goal was to provide a way for people in areas with limited, expensive, or censored cellular internet access a way to view the web in a basic format. I finished work on a basic client-server model last year[2], and this year, I implemented a new pseudo-distributed peer-to-peer model that allows any TxtNet Browser user to use their own smartphone to run a background server service that communicates via the user's own primary mobile number. The main advantage to this model over last year's use of the Twilio API is the fact that with an unlimited SMS plan from a consumer carrier, you will likely end up paying significantly less than the amount you would pay for Twilio credits (averaging about ~$0.50 per website). There's a lot going on with the stateless nature of SMS, GSM-7 encoding, and Brotli compression, so please ask any questions you might have!<p>I've also started up a test server instance running on a +1 country code phone number, so feel free to test out the app with your own smartphone. Like mentioned in the GitHub repo, please be aware that I (necessarily) have access to every phone number and associated request that is sent. Of course, anyone can host their own server instance, and if you would like to share it, feel free to get in touch so I can add the number to the repo! Also, there are likely many bugs still lurking, so feel free to report those.<p>[1] <a href="https://github.com/lukeaschenbrenner/TxtNet-Browser/">https://github.com/lukeaschenbrenner/TxtNet-Browser/</a><p>[2] <a href="https://news.ycombinator.com/item?id=32905496" rel="nofollow">https://news.ycombinator.com/item?id=32905496</a>
Show HN: TxtNet Browser – Browse the Web over SMS, No Wi-Fi/Mobile Data Needed
Hello all,<p>This is my second year of working on a project[1] with the goal of browsing the web, on an Android smartphone, without reliance on Wi-Fi or mobile data. While this concept might seem aimless, my goal was to provide a way for people in areas with limited, expensive, or censored cellular internet access a way to view the web in a basic format. I finished work on a basic client-server model last year[2], and this year, I implemented a new pseudo-distributed peer-to-peer model that allows any TxtNet Browser user to use their own smartphone to run a background server service that communicates via the user's own primary mobile number. The main advantage to this model over last year's use of the Twilio API is the fact that with an unlimited SMS plan from a consumer carrier, you will likely end up paying significantly less than the amount you would pay for Twilio credits (averaging about ~$0.50 per website). There's a lot going on with the stateless nature of SMS, GSM-7 encoding, and Brotli compression, so please ask any questions you might have!<p>I've also started up a test server instance running on a +1 country code phone number, so feel free to test out the app with your own smartphone. Like mentioned in the GitHub repo, please be aware that I (necessarily) have access to every phone number and associated request that is sent. Of course, anyone can host their own server instance, and if you would like to share it, feel free to get in touch so I can add the number to the repo! Also, there are likely many bugs still lurking, so feel free to report those.<p>[1] <a href="https://github.com/lukeaschenbrenner/TxtNet-Browser/">https://github.com/lukeaschenbrenner/TxtNet-Browser/</a><p>[2] <a href="https://news.ycombinator.com/item?id=32905496" rel="nofollow">https://news.ycombinator.com/item?id=32905496</a>
Show HN: We built a ClickHouse-based logging service
Hey hn! I'm one of the co-founders of highlight.io, an open source monitoring tool.<p>Today we're sharing a ClickHouse-based logging solution we've been working on. We wanted to showcase how we built it and share how you could try it out to give feedback.
Since we started working on highlight.io, we've been hyper-focused on "cohesion", or ensuring that when you install your monitoring stack, all of the resources in that stack (user interactions, requests, traces, logs, etc.) are connected in a consumable way. We've written up more about our philosophy on this here [1].<p>We started building towards this by connecting your client-side app and your server-side exceptions with session replay and exception monitoring; i.e. if an error happened in a server-side app, we would make it easy (with session replay) to trace all the steps that a user took leading up to it.<p>Especially for larger companies using highlight.io, the request to tie in logs came up repeatedly, and we wanted to build this with the same philosophy in mind. Now, you'll see client-side and server-side logs all in one place, brought together in the context of a user session, as well as logs in the context of an error.<p>Like the rest of our stack, this project is written in Go and Typescript, and for log ingestion/querying, we're using ClickHouse [2]. Before deciding on ClickHouse, we were planning to use OpenSearch (an aws fork of elasticsearch [3]) for this part of our product, but as our traffic has increased, we encountered quite a few pains with write throughput for OpenSearch. After evaluating a few options, we eventually landed with ClickHouse (their cloud offering was icing on the cake), which has also proven to be much more cost-effective so far.<p>Building with ClickHouse from scratch has been an exciting journey. Eric (the mastermind behind this project) wrote a blog post [4] on a handful of ClickHouse learnings we've gathered since starting the project.<p>For those wanting to try out the product locally, you can run the following commands [5]:<p>git clone --recurse-submodules <a href="https://github.com/highlight/highlight">https://github.com/highlight/highlight</a>
cd highlight/docker;
./run-hobby.sh;<p>To send logs to highlight, you can use your own OpenTelemetry implementation [6] or use our SDKs [7] which provide lightweight wrappers over OTEL.<p>Like the rest of highlight.io, we plan to make money from this with our hosted cloud offering. For those interested in trying out the cloud-hosted version, you can get setup at app.highlight.io.<p>To open the floor for feedback, we would love to get some thoughts on what we've built so far. Beyond that, what are parts of a logging product you wish you had with your current setup? And are there any notable pain-points of using a hosted monitoring product? (We're toying with the idea of an enterprise deployment). Excited to hear from everyone.<p>[1]: <a href="https://highlight.io/docs/general/company/product-philosphy">https://highlight.io/docs/general/company/product-philosphy</a><p>[2]: <a href="https://clickhouse.com" rel="nofollow">https://clickhouse.com</a><p>[3]: <a href="https://news.ycombinator.com/item?id=26780848" rel="nofollow">https://news.ycombinator.com/item?id=26780848</a><p>[4]: <a href="https://www.highlight.io/blog/how-we-built-logging-with-clickhouse">https://www.highlight.io/blog/how-we-built-logging-with-clic...</a><p>[5]: <a href="https://www.highlight.io/docs/getting-started/self-host/self-hosted-hobby-guide">https://www.highlight.io/docs/getting-started/self-host/self...</a><p>[6]: <a href="https://www.highlight.io/docs/getting-started/backend-logging/http">https://www.highlight.io/docs/getting-started/backend-loggin...</a><p>[7]: <a href="https://www.highlight.io/docs/getting-started/overview#for-your-backend-logging">https://www.highlight.io/docs/getting-started/overview#for-y...</a>
Show HN: We built a ClickHouse-based logging service
Hey hn! I'm one of the co-founders of highlight.io, an open source monitoring tool.<p>Today we're sharing a ClickHouse-based logging solution we've been working on. We wanted to showcase how we built it and share how you could try it out to give feedback.
Since we started working on highlight.io, we've been hyper-focused on "cohesion", or ensuring that when you install your monitoring stack, all of the resources in that stack (user interactions, requests, traces, logs, etc.) are connected in a consumable way. We've written up more about our philosophy on this here [1].<p>We started building towards this by connecting your client-side app and your server-side exceptions with session replay and exception monitoring; i.e. if an error happened in a server-side app, we would make it easy (with session replay) to trace all the steps that a user took leading up to it.<p>Especially for larger companies using highlight.io, the request to tie in logs came up repeatedly, and we wanted to build this with the same philosophy in mind. Now, you'll see client-side and server-side logs all in one place, brought together in the context of a user session, as well as logs in the context of an error.<p>Like the rest of our stack, this project is written in Go and Typescript, and for log ingestion/querying, we're using ClickHouse [2]. Before deciding on ClickHouse, we were planning to use OpenSearch (an aws fork of elasticsearch [3]) for this part of our product, but as our traffic has increased, we encountered quite a few pains with write throughput for OpenSearch. After evaluating a few options, we eventually landed with ClickHouse (their cloud offering was icing on the cake), which has also proven to be much more cost-effective so far.<p>Building with ClickHouse from scratch has been an exciting journey. Eric (the mastermind behind this project) wrote a blog post [4] on a handful of ClickHouse learnings we've gathered since starting the project.<p>For those wanting to try out the product locally, you can run the following commands [5]:<p>git clone --recurse-submodules <a href="https://github.com/highlight/highlight">https://github.com/highlight/highlight</a>
cd highlight/docker;
./run-hobby.sh;<p>To send logs to highlight, you can use your own OpenTelemetry implementation [6] or use our SDKs [7] which provide lightweight wrappers over OTEL.<p>Like the rest of highlight.io, we plan to make money from this with our hosted cloud offering. For those interested in trying out the cloud-hosted version, you can get setup at app.highlight.io.<p>To open the floor for feedback, we would love to get some thoughts on what we've built so far. Beyond that, what are parts of a logging product you wish you had with your current setup? And are there any notable pain-points of using a hosted monitoring product? (We're toying with the idea of an enterprise deployment). Excited to hear from everyone.<p>[1]: <a href="https://highlight.io/docs/general/company/product-philosphy">https://highlight.io/docs/general/company/product-philosphy</a><p>[2]: <a href="https://clickhouse.com" rel="nofollow">https://clickhouse.com</a><p>[3]: <a href="https://news.ycombinator.com/item?id=26780848" rel="nofollow">https://news.ycombinator.com/item?id=26780848</a><p>[4]: <a href="https://www.highlight.io/blog/how-we-built-logging-with-clickhouse">https://www.highlight.io/blog/how-we-built-logging-with-clic...</a><p>[5]: <a href="https://www.highlight.io/docs/getting-started/self-host/self-hosted-hobby-guide">https://www.highlight.io/docs/getting-started/self-host/self...</a><p>[6]: <a href="https://www.highlight.io/docs/getting-started/backend-logging/http">https://www.highlight.io/docs/getting-started/backend-loggin...</a><p>[7]: <a href="https://www.highlight.io/docs/getting-started/overview#for-your-backend-logging">https://www.highlight.io/docs/getting-started/overview#for-y...</a>
Show HN: We built a ClickHouse-based logging service
Hey hn! I'm one of the co-founders of highlight.io, an open source monitoring tool.<p>Today we're sharing a ClickHouse-based logging solution we've been working on. We wanted to showcase how we built it and share how you could try it out to give feedback.
Since we started working on highlight.io, we've been hyper-focused on "cohesion", or ensuring that when you install your monitoring stack, all of the resources in that stack (user interactions, requests, traces, logs, etc.) are connected in a consumable way. We've written up more about our philosophy on this here [1].<p>We started building towards this by connecting your client-side app and your server-side exceptions with session replay and exception monitoring; i.e. if an error happened in a server-side app, we would make it easy (with session replay) to trace all the steps that a user took leading up to it.<p>Especially for larger companies using highlight.io, the request to tie in logs came up repeatedly, and we wanted to build this with the same philosophy in mind. Now, you'll see client-side and server-side logs all in one place, brought together in the context of a user session, as well as logs in the context of an error.<p>Like the rest of our stack, this project is written in Go and Typescript, and for log ingestion/querying, we're using ClickHouse [2]. Before deciding on ClickHouse, we were planning to use OpenSearch (an aws fork of elasticsearch [3]) for this part of our product, but as our traffic has increased, we encountered quite a few pains with write throughput for OpenSearch. After evaluating a few options, we eventually landed with ClickHouse (their cloud offering was icing on the cake), which has also proven to be much more cost-effective so far.<p>Building with ClickHouse from scratch has been an exciting journey. Eric (the mastermind behind this project) wrote a blog post [4] on a handful of ClickHouse learnings we've gathered since starting the project.<p>For those wanting to try out the product locally, you can run the following commands [5]:<p>git clone --recurse-submodules <a href="https://github.com/highlight/highlight">https://github.com/highlight/highlight</a>
cd highlight/docker;
./run-hobby.sh;<p>To send logs to highlight, you can use your own OpenTelemetry implementation [6] or use our SDKs [7] which provide lightweight wrappers over OTEL.<p>Like the rest of highlight.io, we plan to make money from this with our hosted cloud offering. For those interested in trying out the cloud-hosted version, you can get setup at app.highlight.io.<p>To open the floor for feedback, we would love to get some thoughts on what we've built so far. Beyond that, what are parts of a logging product you wish you had with your current setup? And are there any notable pain-points of using a hosted monitoring product? (We're toying with the idea of an enterprise deployment). Excited to hear from everyone.<p>[1]: <a href="https://highlight.io/docs/general/company/product-philosphy">https://highlight.io/docs/general/company/product-philosphy</a><p>[2]: <a href="https://clickhouse.com" rel="nofollow">https://clickhouse.com</a><p>[3]: <a href="https://news.ycombinator.com/item?id=26780848" rel="nofollow">https://news.ycombinator.com/item?id=26780848</a><p>[4]: <a href="https://www.highlight.io/blog/how-we-built-logging-with-clickhouse">https://www.highlight.io/blog/how-we-built-logging-with-clic...</a><p>[5]: <a href="https://www.highlight.io/docs/getting-started/self-host/self-hosted-hobby-guide">https://www.highlight.io/docs/getting-started/self-host/self...</a><p>[6]: <a href="https://www.highlight.io/docs/getting-started/backend-logging/http">https://www.highlight.io/docs/getting-started/backend-loggin...</a><p>[7]: <a href="https://www.highlight.io/docs/getting-started/overview#for-your-backend-logging">https://www.highlight.io/docs/getting-started/overview#for-y...</a>
Show HN: Thoughts on Flash in 2023, in Flash, in 2023
Spent the past few days making this - something halfway between a demoscene program and an experimental film. I wanted to celebrate the unique computer-y aesthetics of flash, while showing off some weird and obscure tricks I've picked up over the years since it's been deprecated. (Also, some maybe-not-subtle commentary about AI-art and the tools of the future)
Show HN: Thoughts on Flash in 2023, in Flash, in 2023
Spent the past few days making this - something halfway between a demoscene program and an experimental film. I wanted to celebrate the unique computer-y aesthetics of flash, while showing off some weird and obscure tricks I've picked up over the years since it's been deprecated. (Also, some maybe-not-subtle commentary about AI-art and the tools of the future)
Show HN: Thoughts on Flash in 2023, in Flash, in 2023
Spent the past few days making this - something halfway between a demoscene program and an experimental film. I wanted to celebrate the unique computer-y aesthetics of flash, while showing off some weird and obscure tricks I've picked up over the years since it's been deprecated. (Also, some maybe-not-subtle commentary about AI-art and the tools of the future)
Show HN: Thoughts on Flash in 2023, in Flash, in 2023
Spent the past few days making this - something halfway between a demoscene program and an experimental film. I wanted to celebrate the unique computer-y aesthetics of flash, while showing off some weird and obscure tricks I've picked up over the years since it's been deprecated. (Also, some maybe-not-subtle commentary about AI-art and the tools of the future)
Show HN: Create Comics Using Stable Diffusion
Show HN: How to find the best bank for your needs
Hi everyone,<p>I just launched Bank List, an online banking directory to help people find the perfect bank for their needs.<p>It currently has 170+ banks/EMIs and it's mostly focused on individuals and businesses from Europe, but I plan on adding new regions soon.<p>The business model is affiliate marketing, I earn commission when someone signs up on any of these banks through my link. Maybe I will also add featured listings on the site, still need to think about this though.<p>I would love to receive feedback about the website and business idea in general.<p>Thank you
Show HN: Play the front page of Hacker News as interactive text games
Hey HN, I hacked together this fun project that lets you play any article on the front page of Hacker News as an interactive text game.<p>You can create a game about anything by going to infinityarcade.com/<anything you want goes here><p>You can also use a URL, which is what this page uses.<p>The whole thing is open source: <a href="https://github.com/themaximal1st/InfinityArcade">https://github.com/themaximal1st/InfinityArcade</a>
Show HN: Play the front page of Hacker News as interactive text games
Hey HN, I hacked together this fun project that lets you play any article on the front page of Hacker News as an interactive text game.<p>You can create a game about anything by going to infinityarcade.com/<anything you want goes here><p>You can also use a URL, which is what this page uses.<p>The whole thing is open source: <a href="https://github.com/themaximal1st/InfinityArcade">https://github.com/themaximal1st/InfinityArcade</a>
Show HN: CozoDB, Hybrid Relational-Graph-Vector Database
Hi HN! We're thrilled to share CozoDB v0.6, a monumental update to our FOSS database, which already unifies relational and graph features. With the addition of vector search, CozoDB becomes an even better companion for LLMs like ChatGPT.<p>This release introduces vector search using HNSW indices within Datalog, enabling seamless integration with powerful features such as ad-hoc joins, recursive Datalog, and classical whole-graph algorithms. This update significantly broadens CozoDB's capabilities.<p>Check out the linked release note for an in-depth look at the new features, comparisons to other systems, and intriguing AI development possibilities. We'd love for you to take a look! I'll be here to answer any questions you might have.<p>Looking forward to your feedback!