How does this compare to retool? I've been using that to build UIs directly on top of
Postgres, and it's been nice to be able to design the screens a little; eg inserting new rows linked to a currently selected row by primary key. Curious if you're targeting primarily a spreadsheet-like view, or if the interface can be customized. How does the spreadsheet view work across multiple linked table? Can you present the tabular info in non tabular ways (e.g. a row details or editor view)
We don't have any ability to customize UIs yet, although we're certainly open to feature requests.
We do have a "Record Page" for each individual record that shows a tabular view of related records from other tables, here's a screenshot from our website: https://mathesar.org/_app/immutable/assets/step-2.CKT9S1j0.a... - you can create new linked records directly there.
I'm not OP but from my experience with both tools there are some main differences within the products:
1. Mathesar includes a lot of cool UI elements for CRUD operations on tables, schemas, and columns. Outerbase doesn't have any UI for this, but only a SQL editor.
2. Mathesar has a lot of support for "relationships" (more accurately described as references in Postgres, but I digress) that Outerbase does not. For example, if you're inserting a foreign key column into Outerbase you have to manually type the value of the column you're referencing. In Mathesar, you'll see a dropdown of all the "related" rows and be able to search through their columns to find the record you need.
3. Mathesar is postgres-exclusive, while Outerbase works with many other DB providers.
And more in terms of governance and usage:
1. Outerbase is paid, with a very restrictive free tier. The option to self host is only available to enterprise customers.
We're working on multiple options for sustainability, both the traditional nonprofit route (major donors, individual sponsors, etc.) and also looking into revenue from hosting or services. We're committed to always being 100% open source and self-hostable no matter what.
I'm not related to the author or developers, but this appears almost like an Airtable-like (or Nocodb) UI on top of postgres, so it's really quite a different experience than a full database manager tool/IDE like DBeaver or DataGrip.
I love it personally and it looks pretty good so far. Also, amazing that the team started it as an open source project under a 501c3 nonprofit as well.
Having a more "spreadsheet" (or I would say, MS Access-like) UI is quite cool for certain use cases (e.g. low/no-code but backed by the full power of regular old Postgres). Can allow pretty agile development, although - it can get out of control if not managed with care... but nice to see the access controls/permissions there to help a bit.
sixdimensional covered this pretty well, but just to reiterate, I believe DBeaver is primarily a desktop app aimed at a technical audience. Mathesar is web-based and designed to have an intuitive, collaborative, spreadsheet-like UI for non-technical users who know anything about SQL or DB concepts. Also, Mathesar has deep integration with Postgres features e.g. you can configure roles and privileges through the UI.
> ... spreadsheet-like UI for non-technical users who know anything about SQL or DB concepts.
All the language is DB: it's tables not spreadsheets, records not rows, etc... It also doesn't hide things like 'schema' button or metadata like table description.
Maybe this project could do things like integrations with web services, easy actions on groups (like paste a list of ids and do something to all those records), and publicly available web pages (surveys, donations, pretty event pages.)
Far too technical for the target audience, who has never heard of a "linux" and doesn't care about cost, software freedom, the petite tyranny happily imposed by corporations, etc, they just want it to "just work" on their smartphone browser.
Metabase is a BI tool focused on querying, reporting, and visualization, but it’s read-only. Mathesar is closer to Airtable, you get a spreadsheet-like UI, you can view and query data like Metabase, but you can also add and edit data or even modify the data model (if the permissions allow it). Unlike Metabase, Mathesar also lets you create a database from scratch rather than just connecting to an existing one.
Metabase adds an abstraction layer with concepts like questions, segments, and metrics, while Mathesar works directly with Postgres tables, schemas, and permissions. Mathesar is deeply integrated with Postgres, so it supports Postgres-native features like role and privilege based access control, which aren’t possible in products designed for multiple database backends.
Yeah, I understand why that would be a concern. For what it's worth, our schemas are completely isolated and never touch any other data. We use them to store our DB functions and custom data types.
The functions are what lets us reflect the latest DB state in the UI in real-time without worrying about schema migrations or adding a bunch of reflection queries to each API call (which would make the UI much less responsive). We just need them installed _somewhere_ in the database.
Our custom data types (e.g. email, URL) are just available for convenience. If you don't have any columns that use those types, then you'll never need to deal with them.
It's a great looking interface but I have the same gripe with the tagline that I have with Grist's. It's really not spreadsheet like is it? You just don't have the row and column manipulation possibilites that defines a spreadsheet vs a table.
Thank you! And yeah, I understand your gripe. The tagline is meant to convey the general idea of "editing data in a table UI without a lot of clicks or keystrokes" in the fewest number in words; it does sacrifice precision (although that's why it's "spreadsheet-like" and not just "spreadsheet"). Do you have any ideas for language we could use instead?
I'll try to install this sometime. The permissions bit looks nice. It's nice to see a project that isn't reinventing the wheel. Is there a SQL editor? I don't see anything about one in the docs but that would be a bit of a dealbreaker for me.
Thanks! We do not have a SQL editor in the UI yet, but it's one of the things we're actively considering working on soon, it's mentioned in the "What's next?" section of our beta release announcement: https://mathesar.org/blog/2025/01/29/release-0-2-0
I'm not sure if this would fit your use case, but you can also use a separate SQL editor alongside Mathesar, we always reflect the latest state of your DB.
This looks exactly like something my finance team asked me for yesterday. Definitely will take a look. In my startup (a fintech) we tried to keep all the information about money in the DB, and have the finance team use that to compare with actual bank information
I can see this being very useful for many admin interfaces where some basic data must be managed by domain experts and UX and is not a priority. Many enterprise applications have such parts.
I wonder what the GPLv3 licensing means for such scenarios: Could people run Mathesar one microservice in an ensemble with proprietary services? Companies who don‘t want to open source their whole product might still be willing to upstream their fixes and improvements to the Mathesar component.
Yep, Mathesar is GPLv3, not AGPL, so there’s no issue running it alongside proprietary services. Companies can absolutely use Mathesar as a standalone service in their stack without open-sourcing their other components. Another example of this is WordPress, which is also GPL, and has a thriving hosting ecosystem.
GPLv3 only applies if you modify and distribute Mathesar itself, it doesn’t extend to services that simply interact with it. If a company makes changes to Mathesar and distributes that modified version, then those modifications would need to be open-sourced under GPLv3. But using Mathesar as a microservice in an enterprise stack? No problem.
We’d love to see companies upstream fixes and improvements, of course!
Does AGPL have trouble running alongside proprietary services? I always thought AGPL means that if you host the software, you have to make any changes you did available to the users. So if you host it without changes, there is no problem?
Yep, AGPL can run alongside proprietary services without issues, and if you host it without modifications, you don’t have to share anything. But if you modify it and make it available over a network, you have to provide the source to users.
Mathesar, however, is GPL, so you only need to share modifications if you actually distribute the software itself.
When I was at a big tech, the interpretation by their lawyers was that by running AGPL, we will have to open source everything in the network to users. The problem is in the definition of "Modification" - Per [AGPL](https://www.gnu.org/licenses/agpl-3.0.en.html#:~:text=To%20%...):
> "The "Corresponding Source" for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities."
This can be interpreted as even modifying any configuration to allow the software to run on your own infrastructure. Obviously this is a very aggressive interpretation but the lawyers didn't want us to test this phrase in court so all AGPL software had a blanket ban.
Mathesar doesn’t cache any database state, it always reflects the live structure of the database. If you run migrations through SQLAlchemy or Django, Mathesar will just show the database as it is before and after the migration, no extra steps needed.
You can also modify the data model directly in Mathesar, but if you’re using an ORM-managed database, you’ll need to update the ORM layer yourself to reflect any schema changes. If your ORM expects strict control over the schema, it’s a good idea to configure our permissions to prevent users from making data model changes in Mathesar's UI. Otherwise, it should coexist just fine.
Every single time I see an electron app, I wish there was a way to deelectronify it and just run it as a server.
Mainly I am bitter because I run an openbsd desktop and would have a pretty good chance at getting it to run if it were a normal web app. cross platform my elbow.
A quick perusing of the repo I can see it's heavily optimized for commit count. Large number of commits are spelling/padding changes. You got to do what you got to do to attract investors but I would not waste any time on this.
Mathesar project lead here, it's a nice surprise to see this on the front page! I'm happy to answer questions if anyone has any.
How does this compare to retool? I've been using that to build UIs directly on top of Postgres, and it's been nice to be able to design the screens a little; eg inserting new rows linked to a currently selected row by primary key. Curious if you're targeting primarily a spreadsheet-like view, or if the interface can be customized. How does the spreadsheet view work across multiple linked table? Can you present the tabular info in non tabular ways (e.g. a row details or editor view)
We don't have any ability to customize UIs yet, although we're certainly open to feature requests.
We do have a "Record Page" for each individual record that shows a tabular view of related records from other tables, here's a screenshot from our website: https://mathesar.org/_app/immutable/assets/step-2.CKT9S1j0.a... - you can create new linked records directly there.
As for working with multiple linked tables, while editing data, you can search through nested links easily, here's an screenshot illustrating the record selector: https://mathesar.org/_app/immutable/assets/step-3.SKu3rwM2.a...
We also have a visual query builder that can join data across tables: https://github.com/user-attachments/assets/7ff100fe-cd26-422... and https://github.com/user-attachments/assets/0df680d3-621d-466...
I hope that helps!
This would probably work for us but being able to
* do unit conversions by a multiplier - showing converted values in ui and converting on read and store
* hiding some (empty and irrelevant for specific situation) fields conditionally on values of other fields
would make this a much better fit.
I'll be trying this out for my needs, it seems closer to what I've wanted.
Along the way other nice projects for slightly different but also overlapping use cases include teable.io, budibase and supabase among others.
Any comments on Outerbase?
That poped up on HN too a few weeks ago.
I'm not OP but from my experience with both tools there are some main differences within the products:
1. Mathesar includes a lot of cool UI elements for CRUD operations on tables, schemas, and columns. Outerbase doesn't have any UI for this, but only a SQL editor.
2. Mathesar has a lot of support for "relationships" (more accurately described as references in Postgres, but I digress) that Outerbase does not. For example, if you're inserting a foreign key column into Outerbase you have to manually type the value of the column you're referencing. In Mathesar, you'll see a dropdown of all the "related" rows and be able to search through their columns to find the record you need.
3. Mathesar is postgres-exclusive, while Outerbase works with many other DB providers.
And more in terms of governance and usage:
1. Outerbase is paid, with a very restrictive free tier. The option to self host is only available to enterprise customers.
2. Mathesar is entirely free and self-hosted.
Your website says you're a non-profit. Do you have any monetization plans down the line, otherwise how are you planning to maintain funding?
Here's some background info on our current funding: https://mathesar.org/blog/2024/03/28/mathesar-foundation-ann...
We're working on multiple options for sustainability, both the traditional nonprofit route (major donors, individual sponsors, etc.) and also looking into revenue from hosting or services. We're committed to always being 100% open source and self-hostable no matter what.
How do you differentiate from, say, dbeaver, or any similar database management software? What's the main reason to get Mathesar instead of DBeaver?
I'm not related to the author or developers, but this appears almost like an Airtable-like (or Nocodb) UI on top of postgres, so it's really quite a different experience than a full database manager tool/IDE like DBeaver or DataGrip.
I love it personally and it looks pretty good so far. Also, amazing that the team started it as an open source project under a 501c3 nonprofit as well.
Having a more "spreadsheet" (or I would say, MS Access-like) UI is quite cool for certain use cases (e.g. low/no-code but backed by the full power of regular old Postgres). Can allow pretty agile development, although - it can get out of control if not managed with care... but nice to see the access controls/permissions there to help a bit.
All interesting choices, looks cool!
Yes this definitely feels like MS access but open source and based on postgres which is awesome!
sixdimensional covered this pretty well, but just to reiterate, I believe DBeaver is primarily a desktop app aimed at a technical audience. Mathesar is web-based and designed to have an intuitive, collaborative, spreadsheet-like UI for non-technical users who know anything about SQL or DB concepts. Also, Mathesar has deep integration with Postgres features e.g. you can configure roles and privileges through the UI.
> ... spreadsheet-like UI for non-technical users who know anything about SQL or DB concepts.
All the language is DB: it's tables not spreadsheets, records not rows, etc... It also doesn't hide things like 'schema' button or metadata like table description.
Maybe this project could do things like integrations with web services, easy actions on groups (like paste a list of ids and do something to all those records), and publicly available web pages (surveys, donations, pretty event pages.)
What about tools like LibreOffice Base and KDEs Kexi?
Far too technical for the target audience, who has never heard of a "linux" and doesn't care about cost, software freedom, the petite tyranny happily imposed by corporations, etc, they just want it to "just work" on their smartphone browser.
I'd be more curious about how it differs to Metabase?
Metabase is a BI tool focused on querying, reporting, and visualization, but it’s read-only. Mathesar is closer to Airtable, you get a spreadsheet-like UI, you can view and query data like Metabase, but you can also add and edit data or even modify the data model (if the permissions allow it). Unlike Metabase, Mathesar also lets you create a database from scratch rather than just connecting to an existing one.
Metabase adds an abstraction layer with concepts like questions, segments, and metrics, while Mathesar works directly with Postgres tables, schemas, and permissions. Mathesar is deeply integrated with Postgres, so it supports Postgres-native features like role and privilege based access control, which aren’t possible in products designed for multiple database backends.
I hope that helps!
I tried to propose this to our team, But the option that it installs some schemas was a blocker. I do like using this tool on local pgdump.
Yeah, I understand why that would be a concern. For what it's worth, our schemas are completely isolated and never touch any other data. We use them to store our DB functions and custom data types.
The functions are what lets us reflect the latest DB state in the UI in real-time without worrying about schema migrations or adding a bunch of reflection queries to each API call (which would make the UI much less responsive). We just need them installed _somewhere_ in the database.
Our custom data types (e.g. email, URL) are just available for convenience. If you don't have any columns that use those types, then you'll never need to deal with them.
Ah. Thermians, from the Klaatu Nebula. Or was it Dalmations?
I think it was termites. But they built a huge spaceship.
It's a great looking interface but I have the same gripe with the tagline that I have with Grist's. It's really not spreadsheet like is it? You just don't have the row and column manipulation possibilites that defines a spreadsheet vs a table.
Thank you! And yeah, I understand your gripe. The tagline is meant to convey the general idea of "editing data in a table UI without a lot of clicks or keystrokes" in the fewest number in words; it does sacrifice precision (although that's why it's "spreadsheet-like" and not just "spreadsheet"). Do you have any ideas for language we could use instead?
Any chance it’s going to support PostGIS, with a map to show geometry data?
This is one of the things we're actively considering working on soon! It's mentioned in the "What's next?" section of our beta release announcement: https://mathesar.org/blog/2025/01/29/release-0-2-0
Does anyone know if there's something similar for mysql?
Yes, you can try Visual DB: https://visualdb.com
I'll try to install this sometime. The permissions bit looks nice. It's nice to see a project that isn't reinventing the wheel. Is there a SQL editor? I don't see anything about one in the docs but that would be a bit of a dealbreaker for me.
Thanks! We do not have a SQL editor in the UI yet, but it's one of the things we're actively considering working on soon, it's mentioned in the "What's next?" section of our beta release announcement: https://mathesar.org/blog/2025/01/29/release-0-2-0
I'm not sure if this would fit your use case, but you can also use a separate SQL editor alongside Mathesar, we always reflect the latest state of your DB.
This looks exactly like something my finance team asked me for yesterday. Definitely will take a look. In my startup (a fintech) we tried to keep all the information about money in the DB, and have the finance team use that to compare with actual bank information
I can see this being very useful for many admin interfaces where some basic data must be managed by domain experts and UX and is not a priority. Many enterprise applications have such parts.
I wonder what the GPLv3 licensing means for such scenarios: Could people run Mathesar one microservice in an ensemble with proprietary services? Companies who don‘t want to open source their whole product might still be willing to upstream their fixes and improvements to the Mathesar component.
Yep, Mathesar is GPLv3, not AGPL, so there’s no issue running it alongside proprietary services. Companies can absolutely use Mathesar as a standalone service in their stack without open-sourcing their other components. Another example of this is WordPress, which is also GPL, and has a thriving hosting ecosystem.
GPLv3 only applies if you modify and distribute Mathesar itself, it doesn’t extend to services that simply interact with it. If a company makes changes to Mathesar and distributes that modified version, then those modifications would need to be open-sourced under GPLv3. But using Mathesar as a microservice in an enterprise stack? No problem.
We’d love to see companies upstream fixes and improvements, of course!
Does AGPL have trouble running alongside proprietary services? I always thought AGPL means that if you host the software, you have to make any changes you did available to the users. So if you host it without changes, there is no problem?
Yep, AGPL can run alongside proprietary services without issues, and if you host it without modifications, you don’t have to share anything. But if you modify it and make it available over a network, you have to provide the source to users.
Mathesar, however, is GPL, so you only need to share modifications if you actually distribute the software itself.
When I was at a big tech, the interpretation by their lawyers was that by running AGPL, we will have to open source everything in the network to users. The problem is in the definition of "Modification" - Per [AGPL](https://www.gnu.org/licenses/agpl-3.0.en.html#:~:text=To%20%...):
> "The "Corresponding Source" for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities."
This can be interpreted as even modifying any configuration to allow the software to run on your own infrastructure. Obviously this is a very aggressive interpretation but the lawyers didn't want us to test this phrase in court so all AGPL software had a blanket ban.
Does / how does this handle migrations? Is it happy co-existing with an SQLAlchemy or Django ORM- managed PG instance?
Mathesar doesn’t cache any database state, it always reflects the live structure of the database. If you run migrations through SQLAlchemy or Django, Mathesar will just show the database as it is before and after the migration, no extra steps needed.
You can also modify the data model directly in Mathesar, but if you’re using an ORM-managed database, you’ll need to update the ORM layer yourself to reflect any schema changes. If your ORM expects strict control over the schema, it’s a good idea to configure our permissions to prevent users from making data model changes in Mathesar's UI. Otherwise, it should coexist just fine.
I love the name. A reference to the head Thermian from Galaxy Quest.
Never give up, never surrender!
By Grabthar's Hammer, you shall be avenged!
Love this. Will be trying it out.
It would be great if this was a desktop app. That way I wouldn't need to spin up another server
At least it's not an electron app.
Every single time I see an electron app, I wish there was a way to deelectronify it and just run it as a server.
Mainly I am bitter because I run an openbsd desktop and would have a pretty good chance at getting it to run if it were a normal web app. cross platform my elbow.
There is a docker image at least. Would be cool as a standalone tool though, I kind of agree.
see also https://mathesar.org/ for screenshots
A quick perusing of the repo I can see it's heavily optimized for commit count. Large number of commits are spelling/padding changes. You got to do what you got to do to attract investors but I would not waste any time on this.