Show HN: Simple test, can ChatGPT group chat? https://ift.tt/gvYHumD

Show HN: Simple test, can ChatGPT group chat? I just wanted to see if ChatGPT could interpret a group chat, and each individual message, and determine whether true or false to respond and, subsequently respond if true. It's running live right now (It's just a 10m script nothing too serious). https://ift.tt/kgmnzH7 March 31, 2023 at 12:43AM

The Municipal Railway Planning Division & The First 5-Year Plan

The Municipal Railway Planning Division & The First 5-Year Plan
By Kelley Trahan

The San Francisco Municipal Railway 5-Year Plan, 1979-1984 was the first comprehensive service plan created by the first San Francisco Municipal Railway transportation planners. The plan introduced a grid system to provide more efficient crosstown service with better neighborhood connections that would improve access and increase ridership, moving away from Muni’s prior service design focused on trips to and from downtown. It also provided service standards, including coverage, capacity and stop spacing, many of which continue to inform Muni planning efforts today. The San Francisco Municipal Railway saw many changes at this time, including the opening of the Muni Metro, the conversion of some lines from diesel to electric trolley bus, a simplified fare structure and increased fares and historic streetcar service on Market Street. 

Prior to the mid-1970s, the San Francisco Municipal Railway’s service development was determined by a mix of privately-operated transit systems it had acquired, rather than one master transit plan. This changed when the federal government required the agency to submit a 5-year plan to be eligible for funding.  

The entire process began when President Lyndon B. Johnson signed the Urban Mass Transportation Act of 1964 into law in response to the nation’s long-term transportation investment needs. The new law provided grant and loan money to local transportation systems and created an oversight agency, the Urban Mass Transportation Administration (UMTA), which would be renamed the Federal Transit Administration (FTA) in 1991. The Metropolitan Transportation Commission (MTC) then was put together by the California State Legislature in 1970 and tasked with reviewing transportation grant applications submitted by agencies in the state. 

 

Urban Mass Transportation Act of 1964, Public Law 88-365, 78 STAT 302

An in-depth study of Muni was undertaken in 1974 by Wilbur Smith & Associates, a private consultant and author of the UMTA’s Characteristics of Urban Transportation Demand: A Handbook for Transportation Planners. The Municipal Railway Planning, Operations and Marketing (POM) Study was completed in 1977 and analyzed the Muni system and the travel needs of its riders and made recommendations for a 5-Year Plan. 

 A new in-house Muni Planning Division that had been created with UMTA funding in 1974 presented the POM Study at more than 70 community meetings, and after intense review and several revisions, The San Francisco Municipal Railway 5-Year Plan, 1979-1984 was drafted. 

Muni Planning Division Employees in Presidio Trolley Coach Yard, May 17, 1979

Since that first plan was written, transportation planning at the SFMTA has expanded to include capital projects, street design, bike lanes, paratransit, parking, historic streetcars and much more, while promoting safety, equity and sustainability. The SFMTA continues to publish an operating and service plan approximately every 10 years, now called the Short-Range Transit Plan (SRTP), which is based on the needs, goals and agency priorities to serve the people of San Francisco. 

Published March 30, 2023 at 11:56PM
https://ift.tt/viy0xaI

Show HN: Kaskada – modern, open-source event processing https://ift.tt/CYfTEQm

Show HN: Kaskada – modern, open-source event processing We recently open-sourced Kaskada – an event-processing engine built on Rust and Apache Arrow. It features a high-level, declarative query language designed specifically for reasoning about events in bulk and in real time. Unlike SQL, an aggregation over events produces a timeline indicating the value at each point in time over the events that have occurred up to that point in time. The query language provides composable, expression oriented syntax, including the ability to nest aggregations. Due to the focus on events and temporal queries, it provides capabilities to shift events forward (not backward, to avoid "leaking" information about the future into past values), ticks that create new times (allowing you to observe the current value of an aggregation every day, for instance) and joins that operate "at the current time". Example query: # Purchases up to a point in time let purchases_now = count(purchase) # Shift that *forward* one day (so the value now is from 1 day ago) let purchases_yesterday = purchases_now | shift_by(days(1)) # Compute the difference (purchases since yesterday) in purchases_now - purchases_yesterday More on the history of Kaskada is available in this blog post https://ift.tt/pcmKbIX . More information on the project (and documentation) is available at https://kaskada.io . https://ift.tt/ItCkj9W March 30, 2023 at 11:57PM

Show HN: GitHub Annotations for flake8 with my plugin https://ift.tt/mSoCOfX

Show HN: GitHub Annotations for flake8 with my plugin I made a plugin for the Python linter flake8 that turns it's output into Github Annotations by invoking it with `flake8 --format github`. https://ift.tt/InRmgUX March 30, 2023 at 02:18AM

Show HN: DataLegendAI – Amazon Sales Tracker – The Most Accurate https://ift.tt/vpUu9DG

Show HN: DataLegendAI – Amazon Sales Tracker – The Most Accurate DataLegendAI, unlike any other website, provides Amazon sellers with the actual (not estimate) data from the Amazon marketplace using the power of AI, providing unparalleled accuracy for essential data like historical sales, current stock, competitors count.I launched this with my friend a month ago, would appreciate any feedback on it! Thanks. https://ift.tt/lYtbOKj March 29, 2023 at 11:24PM

Show HN: Nextflick.io – Watch a random movie trailer https://ift.tt/SxyOTPE

Show HN: Nextflick.io – Watch a random movie trailer I want to introduce Nextflick to you, a website I made showing you a random movie trailer, so you can find out what to watch next. Technically, Nextflick is a rather simple website. Everything is rendered on the server-side. Movies are scraped from the TMDB API and dumped into a NoSQL database. When you request a page, a random movie is selected and it shows you a YouTube embed with a respective trailer. Even though this may seem like such a simple tool, I am immensely proud of it. It is my first project I ever actually shipped, and I had this idea in my head for over 5 years. Sadly I am suffering from a bad load of perfectionism, which always got in the way of finishing anything, until today. I deliberately chose to create something that is small and straightforward. I chose Go as a language because it helps me enormously to "get sh*t done". I don't have to think about too many language constructs or idiomatic ways of how to achieve something. I can simply write down my thoughts. I love the error handling as well, it makes me worry less about unexpected errors because the API pretty much tells me exactly where they happen and I can handle them right then and there. NoSQL helped me even more to keep things simple. Since there are no write operations happening on the website itself, it can be used in a readonly mode. If I want to download a new set of movies from TMDB, I simply write everything into a new NoSQL file and then swap the old against the new. I consciously did not implement any sort of filters. I wanted the experience to mimic that of going to your local cinema: you see a set of random trailers, some of them you won't like, but maybe one will stick. You might end up watching a movie you never thought to look for in the first place. There are currently 5.546 movies in the database. This number seems a bit low to me, given that TMDB has over 1.000.000 movies stored. However, I applied some criteria to the movies I imported in order to achieve a certain amount of quality: * They needed to have a trailer * They needed not to be flagged as "adult" * They needed to have an average TMDB user rating of over 6.5 to filter out low quality results * I realise that there are people who are interested in "crappy" B-movies. Pending feedback, I might be willing to reconsider this. * I excluded certain genres like live music or documentaries * They needed to be released, so you can actually go and watch them * They needed to have a runtime of at least 80m, I didn't want to include short films for now * They needed to be released after 1965 * I know that there are a lot of gems from before that time, however, I felt that most of the older ones didn't have proper trailers, or at least not of the sort as we would expect them today, so it felt a bit off to include them * They needed to be in English (for now, if people like it I might add more languages later on) There are also some outstanding issues I have not addressed yet: * YouTube shows the full title of the video name in the embed, this kinda gives away the surprise of watching the trailer until the end. However, since they disabled the option to hide video information, I was not able to find an easy fix for this * Some YouTube videos are marked as "private" or are not available in certain countries, so you will see an error every now and then. I might need to start using YouTube's API to filter these out * Trailers do not autoplay in Safari and Firefox * This one is unfortunate, as it negatively affects the experience of browsing through the trailers. I haven't figured out if there is a way to "fix" this * I tried my best to reach very good Lighthouse results. Oddly enough, the YouTube embed itself performs really poorly by loading megabytes of JavaScript and is flagged by it * Some movies are flagged incorrectly by TMDB, you might end up seeing some movies that are actually not precisely following the criteria I mentioned above https://nextflick.io/ March 30, 2023 at 12:01AM

Show HN: Hacker News Summarizer (Chrome Extension) https://ift.tt/9DlYyWV

Show HN: Hacker News Summarizer (Chrome Extension) Hey Hacker News community, We're excited to announce the launch of our new Chrome extension, "Hacker News Summarizer," which uses OpenAI API to summarize articles on Hacker News. It helps you quickly understand the key takeaways without reading the whole article. Plus, no login or confirmation is required to use it! Simply install the extension and click the "summary" link next to "comments" when you want to read an article. Disclaimer: Not all articles are summarised, or supported but we will change and try to support them later. Additionally, this extension is not affiliated with OpenAI or Hacker News. Chrome Extension: https://ift.tt/KfHZ5C3 ... Try it out today and let us know what you think in the comments below! Thanks, The Hacker News Summarizer team https://ift.tt/jVFr3Sh March 29, 2023 at 04:18PM