[{"data":1,"prerenderedAt":580},["ShallowReactive",2],{"project-43896011":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":29,"sentiment":30,"stance_mix":31,"scored_comments":36,"aspects":37,"strip":44,"post_text":47,"what":48,"said":78,"maker_in_thread":173,"shape":183,"grid":188,"attempts":543,"same_week":544,"maker_shows":284,"maker_hidden":572,"reader_version":573,"card":574},"43896011","VectorVFS","VectorVFS, your filesystem as a vector database","title","Show HN: VectorVFS, your filesystem as a vector database","https:\u002F\u002Fvectorvfs.readthedocs.io\u002Fen\u002Flatest\u002F","vectorvfs.readthedocs.io","perone",1746458253,2025,279,138,202,3837,6,"legend","data","vector search","cli",[24,25,26,27,28],"vector database","semantic search","embeddings","xattr","filesystem","mixed",0.028,{"positive":32,"neutral":33,"mixed":34,"critical":35},7,16,3,10,36,{"idea":38,"performance":40,"design":42,"docs":43},{"positive":18,"critical":39},2,{"positive":41,"critical":18},0,{"positive":41,"critical":39},{"positive":41,"critical":41},[34,34,45,39,45,39,39,34,34,46,41,41,34,34,34,34,45,46,46,41,41,41,41,41,41,41,41,41,41,41,41,41,41,41,41,41,41,41,34,34],1,5,null,{"problem":49,"audience":47,"pricing":47,"motivation":47,"time_to_build":47,"stack":60,"openness":75,"stage":76,"team_size":77},{"value":50,"source":51,"evidence":52},"Searching files by meaning instead of name, with embeddings kept on the files themselves","inferred",[53,57],{"id":54,"by":55,"text":56},"43897147","jlhawn","If I understand correctly, this is attaching metadata to files in a format that LLMs (or any tool that can understand the semantic embedding vector) can leverage to understand…",{"id":58,"by":11,"text":59},"43897289","Hi, it is quite different, there is no LLM involved, we can certainly use it for a RAG for example, but what is currently implemented is basically a way to generate embeddings…",[61,68],{"value":62,"source":51,"evidence":63},"Python",[64],{"id":65,"by":66,"text":67},"43896170","malcolmgreaves","Fun idea storing embeddings in inodes! Very clever! I want to point out that this isn’t suitable for any kind of actual things you’d use a vector database for. There’s no notion…",{"value":69,"source":70,"evidence":71},"Meta Perception Encoder","maker",[72],{"id":73,"by":11,"text":74},"43897493","Hi, there are no LLMs involved, it is all local and an embedding (vector representation) of the data is created and then that is used for search later, nothing is sent to cloud…","open_source","prototype","unknown",{"praise":79,"objections":96,"requests":119,"questions":138,"alternatives":160},[80,89],{"value":81,"source":82,"evidence":83},"Clever idea of storing embeddings in inodes","commenter",[84,85],{"id":65,"by":66,"text":67},{"id":86,"by":87,"text":88},"43896254","natas","this is actually a great idea",{"value":90,"source":82,"evidence":91},"Nostalgia for BeFS and WinFS style metadata search",[92],{"id":93,"by":94,"text":95},"43897343","pseudosavant","This immediately made me nostalgic for BeOS's BeFS or Windows Longhorn's WinFS database filesystems, and how this kind of thing would have fit them perfect. So much cool stuff…",[97,105,112],{"value":98,"source":82,"evidence":99},"Search is linear over all files, not suited to real vector database use",[100,101],{"id":65,"by":66,"text":67},{"id":102,"by":103,"text":104},"43902592","PeterStuer","If there is no indexing, how will your search time not increase linear or worse with the number of files?",{"value":106,"source":82,"evidence":107},"Zero-overhead indexing claim is misleading",[108],{"id":109,"by":110,"text":111},"43917047","yencabulator","> Zero-overhead indexing Embeddings are stored as extended attributes (xattrs) on each file, eliminating the need for external index files or services. Ain't no such thing as…",{"value":113,"source":82,"evidence":114},"Docs lack GPU backend and uninstall details",[115],{"id":116,"by":117,"text":118},"43896319","anotherpaul","Great idea indeed. The documentation needs a bit more information to be useful. What GPU backends are supported for example? How do I delete the embedding information after I…",[120,127,131],{"value":121,"source":82,"evidence":122},"Support for macOS",[123],{"id":124,"by":125,"text":126},"43896904","adenta","I wonder if I could use this locally on my macbook. The finder applications built-in search is kinda meh.",{"value":128,"source":82,"evidence":129},"Explain how to delete embeddings",[130],{"id":116,"by":117,"text":118},{"value":132,"source":82,"evidence":133},"Add an embedded vector index such as Weaviate",[134],{"id":135,"by":136,"text":137},"43899629","thirdtrigger","Might be interesting to add an optional embedded Weaviate [1] with a flat-index [2] to the project. It wouldn't use external services and is fully disk-based. Would allow you to…",[139,146,153],{"value":140,"source":82,"evidence":141},"Is the embedding for the whole file or per chunk?",[142],{"id":143,"by":144,"text":145},"43898397","asadawadia","is the embedding for the whole file? or each 1024\u002F512 byte chunk?",{"value":147,"source":82,"evidence":148},"How does it compare with LanceDB?",[149],{"id":150,"by":151,"text":152},"43896824","esafak","Files-as-vector stores is LanceDB's value proposition. How do you compare in performance, etc.?",{"value":154,"source":82,"evidence":155},"How can users debug why a file surfaced?",[156],{"id":157,"by":158,"text":159},"43896617","b0a04gl","If VectorVFS obscures retrieval logic behind opaque embeddings, how do users debug why a file surfaced—or worse, why one didn’t?",[161,165,169],{"value":162,"source":82,"evidence":163},"LanceDB",[164],{"id":150,"by":151,"text":152},{"value":166,"source":82,"evidence":167},"Finder search",[168],{"id":124,"by":125,"text":126},{"value":170,"source":82,"evidence":171},"BeFS and WinFS",[172],{"id":93,"by":94,"text":95},{"comments":174,"answered":46,"key":175},9,[176,177,180],{"id":58,"text":59},{"id":178,"text":179},"43896265","Thanks. There is a bit of a nuance there, for example: you can build an index in first pass which will indeed be linear, but then later keep it in an open prompt for subsequent…",{"id":181,"text":182},"43896356","Thanks, I'm working on implementing the commands to clean the embeddings (you can now do that with Linux xattr command-line tool). I'm supporting CPU or GPU (NVIDIA) for the…",{"commenters":184,"max_depth":185,"maker_replies":174,"first_comment_mins":186,"lifespan_hours":187},70,11,14,365,[189,191,195,198,201,204,207,210,213,214,217,220,224,227,231,232,234,238,242,245,247,250,252,254,256,258,262,265,269,272,273,275,278,282,286,289,291,293,296,298,301,305,308,311,312,315,318,320,324,328,329,330,333,335,336,338,340,342,344,347,350,353,356,358,360,363,366,368,371,373,375,378,380,383,385,387,389,391,393,395,397,401,404,406,409,412,415,418,420,423,426,428,432,435,438,441,443,445,448,450,452,455,458,461,464,467,470,472,474,476,478,480,482,484,486,488,491,493,495,497,499,501,503,506,508,510,512,514,516,518,520,523,525,528,530,532,536,539],[54,41,34,190,55,41],"project\u002Fneutral\u002Fidea",[192,41,41,193,194,41],"43897169","commenter\u002Fneutral\u002Fnone","lstodd",[196,41,41,193,197,41],"43897188","mywittyname",[199,41,41,193,200,41],"43897264","danudey",[202,41,34,190,203,41],"43897267","yjftsjthsd-h",[205,41,41,193,206,41],"43897281","simcop2387",[208,41,41,193,209,41],"43897387","0x457",[211,41,41,212,194,41],"43899632","commenter\u002Fcritical\u002Fidea",[58,41,41,190,11,45],[215,41,45,216,55,41],"43897492","project\u002Fcritical\u002Fperformance",[218,41,45,216,219,41],"43904151","freeamz",[221,41,46,222,223,41],"43899101","project\u002Fpositive\u002Fidea","pilooch",[225,41,45,226,194,41],"43899718","project\u002Fcritical\u002Fidea",[228,41,45,229,230,41],"43910951","project\u002Fcritical\u002Fdesign","SoftTalker",[65,45,45,216,66,41],[178,45,41,233,11,45],"maker\u002Fneutral\u002Fperformance",[235,45,34,236,237,41],"43896346","project\u002Fmixed\u002Fperformance","binarymax",[239,45,34,240,241,41],"43896358","project\u002Fneutral\u002Fperformance","PaulHoule",[243,45,34,240,244,41],"43897226","ori_b",[246,45,45,216,151,41],"43896863",[248,45,34,190,249,41],"43897125","int_19h",[251,45,45,226,110,41],"43917074",[253,45,34,190,249,41],"43922040",[116,39,34,255,117,41],"project\u002Fmixed\u002Fdocs",[181,39,41,257,11,45],"project\u002Fneutral\u002Fnone",[259,39,34,260,261,41],"43897191","project\u002Fneutral\u002Fdesign","3abiton",[263,39,41,264,200,41],"43897309","commenter\u002Fneutral\u002Fdesign",[266,39,41,267,268,41],"43899182","commenter\u002Fcritical\u002Fdesign","hadlock",[270,39,41,271,11,45],"43897459","maker\u002Fneutral\u002Fdesign",[135,34,34,190,136,41],[274,34,34,190,237,41],"43899984",[276,34,41,193,277,41],"43900491","lysp",[279,34,41,280,281,41],"43902946","commenter\u002Fneutral\u002Fidea","bobvanluijt",[283,284,46,222,285,41],"43904114",4,"ndsipa_pomu",[287,284,41,280,288,41],"43904215","sneak",[290,284,41,280,285,41],"43905512",[292,284,41,280,288,41],"43906592",[294,284,41,295,285,41],"43906998","other\u002Fneutral\u002Fnone",[297,284,41,295,288,41],"43909217",[299,284,41,300,285,41],"43913189","other\u002Fcritical\u002Fnone",[302,46,41,303,304,41],"43899494","other\u002Fneutral\u002Fidea","quantadev",[306,46,41,280,307,41],"43902998","wfn",[309,46,41,310,304,41],"43908811","commenter\u002Fpositive\u002Fnone",[157,18,34,190,158,41],[313,18,41,193,314,41],"43897102","refulgentis",[316,18,41,317,11,45],"43897444","maker\u002Fneutral\u002Fidea",[319,18,34,190,55,41],"43900153",[321,32,34,322,323,41],"43900841","project\u002Fmixed\u002Fidea","PeterZaitsev",[325,326,41,193,327,41],"43897049",8,"bullen",[109,174,45,216,110,41],[86,35,46,222,87,41],[331,35,46,222,332,41],"43896622","iugtmkbdfil834",[73,35,41,334,11,45],"maker\u002Fneutral\u002Fnone",[102,185,45,216,103,41],[150,337,34,240,151,41],12,[339,337,41,260,11,45],"43897474",[341,337,45,229,151,41],"43897609",[143,343,34,257,144,41],13,[345,186,34,190,346,41],"43899095","javier2",[124,348,46,349,125,41],15,"project\u002Fpositive\u002Fnone",[351,348,41,352,11,45],"43897501","maker\u002Fneutral\u002Fexecution",[354,33,46,222,355,41],"43896392","badmonster",[93,357,46,222,94,41],17,[359,357,41,193,11,45],"43897517",[361,357,41,193,362,41],"43897782","didgetmaster",[364,357,41,193,365,41],"43898096","tugdual",[367,357,41,193,362,41],"43899138",[369,357,41,303,370,41],"43897581","p_ing",[372,357,41,295,362,41],"43897732",[374,357,41,295,94,41],"43898020",[376,357,41,295,377,41],"43900950","aforwardslash",[379,357,41,295,370,41],"43898328",[381,357,41,295,382,41],"43900183","WalterGR",[384,357,41,295,370,41],"43900315",[386,357,41,295,362,41],"43904947",[388,357,41,295,382,41],"43905097",[390,357,41,295,362,41],"43918046",[392,357,41,295,382,41],"43922494",[394,357,41,295,377,41],"43900889",[396,357,41,295,377,41],"43900858",[398,399,41,295,400,41],"43896983",18,"tzury",[402,399,41,193,403,41],"43897042","dominicq",[405,399,41,193,400,41],"43897347",[407,399,41,295,408,41],"43897077","cratermoon",[410,399,41,295,411,41],"43897089","ryanianian",[413,399,41,295,414,41],"43897487","sunshine-o",[416,399,41,193,417,41],"43897773","drob518",[419,399,41,193,414,41],"43898309",[421,399,41,295,422,41],"43898632","cryptonector",[424,399,41,193,425,41],"43901371","user3939382",[427,399,41,193,422,41],"43906374",[429,430,41,303,431,41],"43896453",19,"Ericson2314",[433,430,41,295,434,41],"43896531","7qW24A",[436,430,41,295,437,41],"43896591","packetlost",[439,430,41,193,440,41],"43897263","mrlongroots",[442,430,41,193,437,41],"43897762",[444,430,41,193,440,41],"43909200",[446,430,41,193,447,41],"43920554","dev-ns8",[449,430,41,193,440,41],"44045753",[451,430,41,295,431,41],"43897456",[453,430,41,193,454,41],"43896611","qwertox",[456,430,41,193,457,41],"43896874","hdevalence",[459,430,41,193,460,41],"43896905","90s_dev",[462,430,41,193,463,41],"43914114","fc417fc802",[465,430,41,295,466,41],"43896647","foobiekr",[468,430,41,295,469,41],"43896826","jonhohle",[471,430,41,295,249,41],"43897144",[473,430,41,295,466,41],"43898325",[475,430,41,193,431,41],"43899535",[477,430,41,295,466,41],"43900245",[479,430,41,193,431,41],"43900881",[481,430,41,193,466,41],"43901220",[483,430,41,193,431,41],"43905617",[485,430,41,193,466,41],"43905994",[487,430,41,193,431,41],"43906638",[489,430,41,193,490,41],"43896839","adolph",[492,430,41,193,431,41],"43897480",[494,430,41,193,466,41],"43898331",[496,430,41,193,431,41],"43899278",[498,430,41,193,466,41],"43900250",[500,430,41,280,431,41],"43900933",[502,430,41,280,377,41],"43901044",[504,430,41,505,440,41],"43897522","other\u002Fneutral\u002Fperformance",[507,430,41,303,431,41],"43900916",[509,430,41,505,440,41],"43901747",[511,430,41,505,377,41],"43901038",[513,430,41,303,440,41],"43901810",[515,430,41,303,377,41],"43911062",[517,430,41,303,437,41],"43897503",[519,430,41,280,431,41],"43906646",[521,430,41,303,522,41],"43896779","01HNNWZ0MV43FF",[524,430,41,303,431,41],"43897413",[526,430,34,190,527,41],"43897056","runlaszlorun",[529,430,41,303,431,41],"43899343",[531,430,34,190,440,41],"43897061",[533,534,41,295,535,41],"43900111",20,"gitroom",[537,534,41,295,538,41],"43900168","j45",[540,541,41,295,542,41],"43897318",21,"colordrops",[],[545,550,555,559,564,568],{"id":546,"name":547,"points":548,"category":549},"43905942","Clippy",1122,"ai",{"id":551,"name":552,"points":553,"category":554},"43880962","BreezePDF",791,"productivity",{"id":556,"name":557,"points":558,"category":549},"43899028","RealtimeVoiceChat",524,{"id":560,"name":561,"points":562,"category":563},"43873074","Anukari",512,"creative",{"id":565,"name":566,"points":567,"category":554},"43891245","AI Native Resume",301,{"id":569,"name":570,"points":571,"category":563},"43916956","Kindle Comic Converter (KCC)",240,false,"v2",{"id":575,"name":5,"line":6,"ai":572,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":576,"top":18,"kw":577,"maker":11,"domain":10,"tier":19,"attempts":41,"strip":578,"url":9,"mix":579,"praise":81,"objection":98,"sub":21},43896011,0.03,[24,25,26,27,28],"3312122335003333155000000000000000000033",[32,33,34,35],1794036674313]