[{"data":1,"prerenderedAt":311},["ShallowReactive",2],{"project-41907719":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":35,"aspects":36,"strip":44,"post_text":45,"what":46,"said":90,"maker_in_thread":180,"shape":192,"grid":195,"attempts":273,"same_week":274,"maker_shows":31,"maker_hidden":303,"reader_version":304,"card":305},"41907719","Data Formulator","Data Formulator – AI-powered data visualization from Microsoft Research","title","Show HN: Data Formulator – AI-powered data visualization from Microsoft Research","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fdata-formulator","github.com","chenglong-hn",1729539735,2024,212,34,369,3515,11,"notable","data","ai data visualization","web_app",[24,25,26,27],"data visualization","ai charts","natural language","python data analysis","roasted",-0.219,{"positive":31,"neutral":32,"mixed":33,"critical":34},3,5,0,9,17,{"execution":37,"design":38,"docs":39,"idea":41,"performance":42},{"positive":33,"critical":31},{"positive":33,"critical":33},{"positive":33,"critical":40},1,{"positive":33,"critical":40},{"positive":33,"critical":43},4,[33,40,33,31,31,33,33,33,33,33,40,33,40,33,31,40,33,40,31,33,31,33,32,40,33,40,40,33,33,40,33,32,33,32],"Creating data visualizations with AI nowadays often means chat, chat and more chats...and writing long prompts can be annoying while they are also not the most effective way to describe your visualization designs.\n\nData Formulator blends UI interaction with natural language so that you can create visualizations with AI much more effectively!\n\nYou can:\n\n* create rich visualizations beyond initial datasets, where AI helps transforming and visualizing data along the way\n\n* iterate your designs and dive deeper using data threads, a new way to manage your conversation with AI.\n\nHere is a demo video",{"problem":47,"audience":53,"pricing":59,"motivation":60,"time_to_build":59,"stack":64,"openness":87,"stage":88,"team_size":89},{"value":48,"source":49,"evidence":50},"Creating visualizations with AI via long chat prompts is cumbersome; UI plus natural language is more effective","maker",[51],{"id":52},"post",{"value":54,"source":49,"evidence":55},"Initially Tableau\u002FPowerBI users; later more useful to Python data analysts",[56],{"id":57,"by":11,"text":58},"41916086","1. when started out, we have Tableau\u002FPowerBI users as the main audience in mind, hoping to grant them the power of data analysts who program in python to create charts requiring…",null,{"value":61,"source":49,"evidence":62},"Make AI-assisted visualization more effective than chat prompts",[63],{"id":52},[65,74,81],{"value":66,"source":67,"evidence":68},"Python","inferred",[69,73],{"id":70,"by":71,"text":72},"41912585","zurfer","Here is the link to one of the prompts. It seems like all the LLM tasks are in the agents directory:…",{"id":57,"by":11,"text":58},{"value":75,"source":76,"evidence":77},"Plotly","commenter",[78],{"id":79,"by":71,"text":80},"41912530","Anthropic recently released something that looks more polished but follows the chat paradigm. [1] As a builder of something like that [2], I believe the future is a mix, where…",{"value":82,"source":49,"evidence":83},"GPT-4o",[84],{"id":85,"by":11,"text":86},"41916237","That's a cool example! You are right, GPT-4o is much more powerful than we allow it to perform in Data Formulator, and our current design is to restrict it to a point that the…","open_source","beta","team",{"praise":91,"objections":114,"requests":137,"questions":149,"alternatives":159},[92,103],{"value":93,"source":76,"evidence":94},"Useful for saving time on data tasks",[95,99],{"id":96,"by":97,"text":98},"41911045","marktl","Definitely looks like something that could save me, and others, allot of time. Thanks for sharing!",{"id":100,"by":101,"text":102},"41911774","donq1xote1","Thanks for sharing and provide open source version! This is great!",{"value":104,"source":76,"evidence":105},"Data threads and UI-plus-natural-language approach are interesting",[106,110],{"id":107,"by":108,"text":109},"41915994","d_watt","Some of the metaphors for interacting with the models, and visualizing as threads are interesting. Definitely does a good combination of ease of prompting + interogatability of…",{"id":111,"by":112,"text":113},"41916773","gourabmi","I just tried this out on a bunch of log files that were lying around. Excellent! I would love to connect this to a SQL database through ODBC\u002F JDBC and ask questions about the…",[115,130],{"value":116,"source":76,"evidence":117},"AI output may be wrong; hard to verify transformations",[118,122,126],{"id":119,"by":120,"text":121},"41911839","goose-","Since Data Formulator performs data transformation on your behalf to get the desired visualization, how can we verify those transformations are not contaminated by LLM…",{"id":123,"by":124,"text":125},"41913627","croes","The problem with AI is still, how confident can we be the result isn’t wrong?",{"id":127,"by":128,"text":129},"41914402","nerdjon","We still really should not be confident. We have seen this time and time again that the output cannot be relied on. People keep trying to downplay the problem as \"not a big deal\"…",{"value":131,"source":76,"evidence":132},"Less consistent results than prompting GPT-4o directly",[133],{"id":134,"by":135,"text":136},"41911928","matt3D","After giving it a whirl I'm a little underwhelmed, but maybe I'm using it wrong. I'm getting less consistent results than if I prompted GPT4-o for a Vega graph after providing it…",[138,142],{"value":139,"source":76,"evidence":140},"Connect to SQL databases via ODBC\u002FJDBC",[141],{"id":111,"by":112,"text":113},{"value":143,"source":76,"evidence":144},"Jupyter extension",[145],{"id":146,"by":147,"text":148},"41914078","paddy_m","This looks good, especially the data ingest. A couple of questions: What type of data person are you aiming this tool to be useful for, data analyst, data scientist, or data…",[150,155],{"value":151,"source":76,"evidence":152},"How to verify transformations and avoid hallucinations",[153,154],{"id":119,"by":120,"text":121},{"id":146,"by":147,"text":148},{"value":156,"source":76,"evidence":157},"Handling missing Python libraries like ARIMA",[158],{"id":107,"by":108,"text":109},[160,166,173],{"value":161,"source":76,"evidence":162},"Buckaroo",[163],{"id":164,"by":147,"text":165},"41914233","I have a tool [1] that is tackling some of the same problems in a different way. I had some core views that shaped what I built. 1. When doing data manipulation, especially…",{"value":167,"source":76,"evidence":168},"Chat-based LLMs such as GPT-4o and Anthropic's tool",[169,170],{"id":79,"by":71,"text":80},{"id":171,"by":11,"text":172},"41915749","potentially! I myself as an experienced programmer finds directly chatting with GPT-4o much easier, since I have full control. But Data Formulator only transforms data in the…",{"value":174,"source":76,"evidence":175},"Hex's Vega-lite DSL approach",[176],{"id":177,"by":178,"text":179},"41910405","data_ders","way cool! I hope to take it for a spin tomorrow! Q: Does your team see potential value in a DSL for succinctly describing visualizations to an LLM as Hex did with their DSL for…",{"comments":18,"answered":181,"key":182},7,[183,186,189],{"id":184,"text":185},"41916351","you find it! you can't imagine how often I'm annoyed when see April being ranked before March due to alphabetic order... Thus the sorting agent, and now running by default in the…",{"id":187,"text":188},"41916606","yes! chat is still a necessary component, since it's sometimes the only way for us to communicate unstructured information to the system.",{"id":190,"text":191},"41915830","this is really really cool! directly working with table is sometimes the only way to clean the data as well :) I wish multiple ways of interacting with data can co-exist…",{"commenters":35,"max_depth":43,"maker_replies":18,"first_comment_mins":193,"lifespan_hours":194},371,23,[196,198,200,202,206,208,210,213,214,216,219,221,222,223,227,230,233,236,240,242,243,246,249,251,254,255,256,259,261,263,264,267,269,272],[79,33,33,197,71,33],"other\u002Fneutral\u002Fidea",[70,33,40,199,71,33],"project\u002Fcritical\u002Fexecution",[184,33,33,201,11,40],"maker\u002Fpositive\u002Fexecution",[203,33,31,204,205,33],"41912882","project\u002Fneutral\u002Fdesign","DeathArrow",[207,33,31,204,71,33],"41913099",[187,33,33,209,11,40],"maker\u002Fneutral\u002Fdesign",[211,33,33,197,212,33],"41914011","flessner",[164,40,33,197,147,33],[190,40,33,215,11,40],"maker\u002Fpositive\u002Fidea",[217,40,33,218,147,33],"41915900","commenter\u002Fneutral\u002Fdesign",[107,220,40,199,108,33],2,[85,220,33,209,11,40],[119,31,40,199,120,33],[224,31,33,225,226,33],"41911905","commenter\u002Fcritical\u002Fidea","larodi",[228,31,31,229,205,33],"41912889","project\u002Fneutral\u002Fexecution",[231,31,40,232,124,33],"41913616","project\u002Fcritical\u002Fdocs",[234,31,33,235,11,40],"41915812","maker\u002Fneutral\u002Fexecution",[237,43,40,238,239,33],"41912232","project\u002Fcritical\u002Fidea","hggigg",[146,32,31,241,147,33],"maker\u002Fneutral\u002Fnone",[57,32,33,235,11,40],[177,244,31,245,178,33],6,"maker\u002Fneutral\u002Fidea",[247,244,33,248,11,40],"41910793","project\u002Fneutral\u002Fidea",[96,181,32,250,97,33],"project\u002Fpositive\u002Fnone",[134,252,40,253,135,33],8,"project\u002Fcritical\u002Fperformance",[171,252,33,209,11,40],[123,34,40,253,124,33],[257,34,40,253,258,33],"41914244","monkeydust",[260,34,33,197,147,33],"41914341",[262,34,33,235,11,40],"41916112",[127,34,40,253,128,33],[265,34,33,266,11,40],"41915795","maker\u002Fneutral\u002Fperformance",[111,268,32,250,112,33],10,[270,268,33,271,11,40],"41917233","maker\u002Fneutral\u002Fprivacy",[100,18,32,250,101,33],[],[275,280,285,289,294,298],{"id":276,"name":277,"points":278,"category":279},"41893524","HN Update",644,"ai",{"id":281,"name":282,"points":283,"category":284},"41914544","Rust Web Framework",488,"languages",{"id":286,"name":287,"points":288,"category":279},"41929804","Pablonet",472,{"id":290,"name":291,"points":292,"category":293},"41915958","Counter-Strike Football",420,"games",{"id":295,"name":296,"points":297,"category":279},"41926770","Agent.exe",406,{"id":299,"name":300,"points":301,"category":302},"41903864","Epublifier",290,"productivity",false,"v2",{"id":306,"name":5,"line":6,"ai":303,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":307,"top":18,"kw":308,"maker":11,"domain":10,"tier":19,"attempts":33,"strip":309,"url":9,"mix":310,"praise":93,"objection":116,"sub":21},41907719,-0.22,[24,25,26,27],"0103300000101031013030510110010505",[31,32,33,34],1793991103185]