Obsidian Bases

Here's a summary of the themes from the Hacker News discussion:

Dataview Replacement and Functionality

A significant portion of the discussion revolves around Obsidian's new "Bases" feature and its relationship to the popular Dataview plugin. Users are eager to understand if Bases is a direct replacement and how its functionality compares.

• "Is this essentially an official replacement for the Dataview plugin?" asked mudkipdev.
• LordDragonfang responded, "Yes and no. It's not meant to be as "kitchen sink" comprehensive yet, but it's a like 90% replacement and it's a ton faster and more responsive. (And I suspect once extension support comes in the dataview team will fully change over)"
• jcutrell, a Dataview user, stated, "I am curious about how this compares to dataview. As a dataview user, I'm not immediately seeing something bases does that dataview doesn't, but I am not a power user."
• bad_username elaborated on the difference: "Dataview can be used for queries that output tables, but its strength is letting you write essentially custom imperative Javascript code that renders stuff in notes dynamically (dataviewjs mode). Whereas "queries to tables" is more or less what Bases does, the dataviewjs mode will probably always be unique."
• wscott offered a comparison: "Dataview can generate all of the data from Bases (and a whole lot more), but Bases is a lot easier to use as you can build queries in the GUI and the data comes out in nicely formatted table where you can edit the fields directly in the table rather than needing to load each data item one by one to make changes."
• budududuroiu noted a limitation: "Thought this could be a full replacement for Dataview but it’s just not there yet. I love the dataview flexibility and multiple display options. ... However, for task lists, Dataview still works better imo"
• "I've been using the 'DB Folder' and 'Dataview' plugins, I'll definitely look into this new option. Does it work with Dataview at all?" asked remram.

Comparison to Notion and Airtable (Database Functionality)

Several users draw parallels between Obsidian's new database-like features and established tools like Notion and Airtable, sparking a debate about Obsidian's identity and scalability.

• datadrivenangel observed, "So obsidian is now notion is now airtable."
• haydenlingle countered, "Yes, but their kicker has always been that you own your data in a readable format even if you don’t use their app and can self sync it if you prefer. I imagine they’ll keep adding functionality of notion and airtable while still keeping this underlying premise."
• al_borland expressed frustration with a competing tool: "Notion always felt painfully slow and fiddly. I have convinced myself that they have manufactured their entire perceived popularity through YouTube sponsorship. It seems like there is a better tool for every job."
• dottjt defended Notion's complexity: "The complexity is what makes it valuable to some people. Put simply, it does more than Obsidian."
• shminge raised concerns about Notion: "Are you not concerned about notion a) being online only and b) not letting you be in control of your data? I'm a strong proponent of File over App: who knows how long Obsidian or notion will exist - at least I know I can work with my Obsidian notes as long as text editors exist."
• "I can’t believe my eyes. Is that the self hosted notion alternative many of us have been looking for?" questioned barbazoo.
• nxobject hoped for wider adoption of .base files: "I hope more programs use ".base" files for database views – it runs a lot of workflows in Notion that would benefit from a diversity of implementations."
• wkat4242 stated, "I can definitely replace the 'Projects' plugin for me, maybe the 'Dataview' plugin as well? Hope it can rival Notion Databases in the future, e.g. by adding a Kanban view."
• Ayin_avocado noted, "It’s a markdown editor, but they can’t modify the markdown standard, so their scope is limited. All they can do is build features around it. Having a database isn’t mutually exclusive with the core functionality. You can simply not use it."
• bryanhogan felt it was a step away from Notion: "It does not change the content of the markdown files themselves, so there's no lock-in or other potentials long-term problems. It allows me to move further away from Notion, which is a great thing, and I hope to see them be able to fully replace Notion Databases in the future."
• Jgoauh explained the functionality: "a base is a table where each row is a specific file in your vault, and each column is a proprety of this file. Using a base you can manage, sort and filter your files and their propreties (ex add a rating, price, or deadline proprety to your files or only show files from my movie folder where my proprety is set to [this])"

Data Structure and Performance (File-per-Row vs. External Data)

A key technical point of discussion is how Obsidian's new "Bases" handle data, specifically the "one file per row" approach versus the desire for direct integration with external data sources like CSV or JSON for potentially larger datasets.

• andyferris inquired, "So where does the data "live"? I was looking at the syntax, it defines predicates for filters and views and so on, but I don't see the "rows". There is this file.name and file.ext thing - but where do you set them? What type of file does it point to? CSV? JSON? Something else? The docs seem incomplete."
• kepano clarified, "The rows are individual Markdown files and the columns are YAML frontmatter properties in those files."
• andyferris followed up: "Hmm... so I can't use this to render and filter a table with 10k rows without having 10k markdown files? If I understand correctly, the intention seems to be "curated list of links" which the user can sort, filter, etc when viewing. I guess that's cool, if you use Obsidian lots and have many notes/links - but when I clicked the article and saw the table I was hoping for a "dataframe" plugin for .md (much like how mermaid works, defined in a codeblock) that references a nearby CSV/JSON/etc file."
• segphault agreed with the sentiment: "I am with you on this, I wish Obsidian would optionally allow you to use YAML or some other structured data directly in the fenced code block or base file. I really, really want something that kind of takes an Obsidian-like approach to local databases, sort of like Excel/Airtable but with flat, human-editable text files that live on your filesystem with a schema driven property editor. It’s kind of a bummer that this gets so tantalizingly close but doesn’t take it to the logical conclusion."
• aetherspawn provided perspective on scale: "If you have 10k rows this isn’t for you, but 99% of use cases have less than 100 rows... For example, book list, movie list, customer list, invoice list, asset register, key register… once you hit a certain point, obsidian probably isn’t the right tool anymore. But no reason to go to the monthly SaaS “right tool” at the POC stage."
• pbronez clarified the feature's nature: "The “base” name appears to be misleading. I assumed this feature would add a structured data format to Obsidian, but it does not. This is exclusively a query engine for your existing markdown files. An obsidian base is primarily defined as a code block in an existing markdown file. That code block defines a series of queries (filters, really) that produce a result set of markdown file names. So… more a way to work with a large collection of markdown files than a relational database."

Plugin Ecosystem, Extensibility, and Core Features

The discussion touches on Obsidian's extensibility through plugins, user preferences for core vs. community plugins, and the perceived bloat or necessary complexity.

• wjrb expressed hope for API support: "I hope the API has support to allow extensions---I see that it is on the Roadmap[0]."
• "Are there any good LLM plugins for Obsidian, beyond just throwing Claude Code / Codex at your markdown folder?" asked dadrian.
• gavmor praised a plugin: "There's no decent RAG functionality, AFAICT, but the Text-Generator plugin has been fantastic w/ larger contexts and a template that pulls either/both links and backlinks into the inference query window. Hands-down my most productive interface to LLMs for [years since GPT3.5] years running."
• "The whole Obsidian ecosystem feels really electric right now," commented dtkav.
• criddell questioned the decision for Bases to be a core plugin: "Why is this a core plugin? It’s interesting, but seems a bit niche. How do you decide if it’s a core plugin or a regular plugin?"
• abhinavk explained, "Core plugin is first-party closed-source plugins the Obsidian team created and maintains. Regular (community) plugins are third-party, open-source and vetted only at the submission time."
• criddell expressed a desire for stability: "I kind of wish they would slow down development and just polish what is already there. They are on this treadmill where they feel the need to keep adding new features, new maintenance burdens, bloating the product in every dimension."
• theappsecguy defended Obsidian's development: "Obsidian has done more than anyone else for the PKM community without lock in, it’s quite disheartening to see so many complaints."
• hifikuno shared their plugin philosophy: "I only reach for plugins when I feel something is missing, but honestly the only plugins I use are Style Settings which let's me customize the theme a bit more and Calendar so I can have... well... a calendar."
• muppetman cautioned against over-pluginization: "Don't install Obsidian and then spend hours adding addons. You'll get overwhelmed, confused and wondering why all the Influenzers are saying it's CHANGED THEIR LIFE. It hasn't."
• kid64 raised security concerns regarding community plugins: "The community plugin system is both an unacceptable security risk, and a necessary part of achieving even a baseline level of usability. Imagine the scale of theft that must already have taken place..."

User Experience and Personal Workflow Philosophies

A recurring theme is how users approach Obsidian, with a divide between those who prefer a minimalist, core feature set and those who leverage extensive plugins to build complex systems. There's also discussion about the perceived "wankery" and "cringeworthy" aspects of user-generated content about Obsidian.

• "The complexity is what makes it valuable to some people. Put simply, it does more than Obsidian," noted dottjt.
• "The idea is that you can store properties, or metadata about the current file, in Obsidian notes... What Bases allows you to do is visualize a certain kind of note as a table or eventually as a Kanban view or another type of view. So, it’s like a visualization layer on top of the data that you already have." said kepano, quoted from The Verge.
• "There is so much wankery around Obsidian, it's so cringleworthty. Obsidian is a nice/fancy editor for markdown files. That's all it is really," stated muppetman.
• "It’s a nice UI to edit a collection of markdown files. Every Obsidian YT video is about mind maps, how to organize your files, using relative links and weird plugins that break the premise of having universal markdown files," commented dimitri-vs.
• "Start with the problem you're trying to solve and use the features to solve it. Don't just try to cram its features into your life," advised AstroBen.
• "Just use it how you want to use it. For me, a bit of structure is enough - too much and it becomes unwieldy. But don’t let anyone tell you it’ll either save your life or become your “second brain”, that’s all bullshit," said dmje.
• "The graph is often the most overhyped and underused thing there. It looks complicated and that's the selling point of all that ecosystem around productivity systems and all that. The appearance of deep complexity and work," commented cloud_watching.
• "Maybe if the documentation would have an example of how you would create a collection of documents and then have a view on that with a base it would be clearer," suggested nedt.
• "It’s a markdown editor, but they can’t modify the markdown standard, so their scope is limited. All they can do is build features around it," said isege.
• "For me it's the opposite and I highly disagree. Valuable features such as this make working with markdown files much better. It's overall a huge plus for working with Obsidian," countered bryanhogan.
• "This is obviously for people for whom notes are a primary knowledge storage mechanism, and for whom that mechanism is obsidian," said theshackleford.
• "The default search is not great and the syntax of the dataview plugin is not amazingly well designed," asserted azeirah.
• One user lamented, "I writing this as some massive sql vquery on a couple billion records churns away. I'm not great (i'm much less impressive than that previous comment sounds in fact), but im way above beginner. I'm shocked at how hard this seems to be."

AI and RAG Functionality in Obsidian

A smaller but present theme is the integration of AI and Retrieval Augmented Generation (RAG) capabilities within Obsidian, with users discussing existing plugins and the potential for improved functionality.

• "Are there any good LLM plugins for Obsidian, beyond just throwing Claude Code / Codex at your markdown folder?" asked dadrian.
• "There's no decent RAG functionality, AFAICT, but the Text-Generator plugin has been fantastic w/ larger contexts and a template that pulls either/both links and backlinks into the inference query window," stated gavmor.
• "The fact there is no app to do "RAG on localhost for my own notes" really makes me wonder whether the tech works at all," mused Noumenon72.
• obsidianbases1 mentioned popular AI plugins: "Copilot, Smart Connections and Text Generator are all popular LLM plugins in Obsidian."
• pixelbro recommended a specific plugin: "I really enjoy Obsidian Copilot (by Logan Yang)."
• mrtsepelev suggested an alternative: "While not a plugin, I've had a great experience using the free tier of Gemini CLI over my md-repository."
• hal Jordan commented on the evolution of AI: "Good is a little more proscriptive than that. Its not hard to see the difference between the, we can call them gen1, ai functions and gen2/3 functions we have now. The difference between gpt-3.5 and NotebookLM."

Git Integration and Data Synchronization

Concerns and discussions also emerged regarding Obsidian's Git plugin and general data synchronization methods.

• "My last request of the Obsidian team is a better git plugin. Their official built-in sync product is fine, but I'd still like to manage my own versioning so I can use automations. The currently available git plugin is extremely dangerous (!!!) if set up incorrectly," warned echelon.
• "I've never had it wipe anything before[1], but I do have a stretch of 200+ commits in my personal vault where my laptop and desktop were fighting back and forth on the contents of one setting file," responded LordDragonfang regarding potential Git plugin issues.
• "The Git plugin is great for single-device backup IMO, but not great for device sync or collaboration. I've been working on making Obsidian "work for work" with a real-time collaboration plugin called Relay [0]," related dtkav.
• safety1st asked about Obsidian Sync: "I just have a notes folder in Syncthing and it seems to accomplish the same thing. Or is there more to Obsidian Sync that I don't know about?"
• pgorczak elaborated, "It also syncs config and plugins."
• omnibrain shared past issues: "In the beginning I had my Obsidian folder in my iCloud, but had not so happy experiences regarding the conflict resolution."
• bryanhogan detailed various sync options: "In general you can use many cloud provider, SyncThing Fork or GitHub. ... The easiest way to sync would be with the official Obsidian subscription, which costs the most amount of money from the options mentioned here."