Hacker News Reader

Comment by abtinf | Hacker News Reader

I'm sometimes tempted to use and recommend Airtable. There is a lot to like.

But their "records-per-base" limit is a complete non-starter. The highest tier of service (before Enterprise) has 50k record cap, which disqualifies it for many reasonable use cases.

Apologists for Airtable hand-wave away this concern. They say Airtable has integrations with other systems and databases; that Airtable is meant for current data only; that its not meant to store complete histories; that larger databases would be too resource-consuming.

None of it is persuasive. The limit seems arbitrary and seems obviously intended to drive businesses into the Enterprise pricing.

I've even tried to contact them about it in the past, to get a larger limit. I was starting work on a non-fiction book and thought it would make a great writing support tool for gathering facts and evidence. But I never heard back.

View on HN

In a few months you might be interested in Baserow. It's an open source alternative to Airtable that we're developing. Right now we have a very early test version at https://baserow.io that everyone can try, but it's still limited. This month we're adding some essential features and next months will be in light of the open source release. After that you can self host it without any limits.

I love the look of this. Just make it so it's easy to make full offline backups and I'll be happy to leave hosting and server maintenance to you.

I'll make sure this feature is going to be implemented in the future. Might take a while though, there are lots of other features that have more priority :).

This looks very cool. I'll be taking a closer look.

One minor nitpick - you might want to adjust your comments and site copy to say "it will be open source" rather than "it is open source". Cheers.

It might be under an open source license. In which case anyone who has access to the source can exercise their rights under that license. Being open source does not necessarily mean that it's available for download on the web, though eventually it will be.

Really encouraging to see an OS option.

Looks great, hope you'll succeed with this! Having an affordable on-prem alternative with native database support are killer features for me.

Very interested in this! Could you please incorporate rss support in your blog so I can keep track of it?

I've added the issue to the backlog!

Hey, looks like your auth / permissions are messed up. Looks like the app in general isn't functional at all?

I'm seeing DBs/groups from what I believe are other accounts.

You shouldn't be able to see groups and databases of others. Can you elaborate a little bit about what you are seeing that is not supposed to be there? Or maybe you can send me a screenshot at bram [at] baserow [dot] io?

Wow, I'm having the same issue right now. It seems to be related to having multiple replicas in the Kubernetes cluster. I've scaled it down to a single replica and that seems to fix it for now. As you can see this is clearly an early alpha version. Thanks for notifying me about this issue and my apologies for the inconvenience.

Hard API limits on 100 rows returned at a time also make it completely infeasible for some use cases. Last time I checked, this couldn't even be lifted for enterprise customers too.

I used fieldbook prior to Airtable, but unfortunately they closed down and I am still sad about it. They had a better pricing model (well, for me, obviously not for them since they couldn't sustain a business), and a magical search function that would find different bits of your search string in different fields. I have one client I use airtable with, but their pricing makes almost everything else not feasible for me.

I really liked the idea of airtable and use it for toy projects. But the record limit, version retention limits, and expensive per use pricing made it so I can’t use it on real projects.

Even simple stuff like asset tracking will go past the record limit and if I have to build something to dump and archive and then query both, I’d rather not bother as most systems I archive to have some UI and workflow option. So I’d rather have one mediocre system for the entire data than an awesome UI for part, and a mediocre system anyway.

The team seems smart, but I assume they have some technology architectural problems that make them not get big.

I would rather pay by data stored or I/o or something but charging the same for a user who adds 100 items per month as one who only reads or only adds one is hard to pay for.

Hey, I am so happy to see Baserow. I am building https://dwata.com, which is primarily an admin for SQL and API (Stripe, Mailchimp) data. So it expects an existing DB with tables and stuff but I have plans to create tables with custom schema, on demand (migrations take care of).

Airtable is a great inspiration from the UI point of view but I am have a strong opinion of self-hosted. So Dwata will try best to stay self-hosted. It is open source (https://github.com/brainless/dwata) but I am surely planning to make money by selling license/support - have not reached there yet.

Someone suggested I go to Airtable after getting frustrated with the query options in Excel on a personal project, in lieu of using a full database.

It was a spectacularly bad suggestion; Airtable seemed to require the same sorts of weird support columns and formulas as Excel in order to support multivariate queries. I came away thinking it was just a pretty browser-based spreadsheet toy, and not a thing useful for actual work.

I'm still baffled by the love.

(And I went on to use Pandas/Jupyter Notebooks/Python for my project.)

Making it per-base rather than per-user is odd.

Suppose I'm thinking of using airtable to build something shared between all our staff. If I have 10 staff on the pro tier they can add 5000 items each.

But if I pay ten times as much for 100 staff they can only add 500 items each.

My guess is that it's more of a technical limit, and they had difficulty making it perform well for larger bases.

Records-per-staff is certainly an issue, but in fact the problem is much more severe.

A great deal of the value of the Airtable comes from having multiple sheets in a single base. Put another way, the value comes from the ability to maintain relationships of data.

So those 10 or 100 staff aren't just entering individual items. They are, say, adding project information--ideas, bugs, tasks, managing the next sprint. Every additional bit of information you add to Airtable compounds the problem.

Airtable is the only data management tool I've ever seen that punishes users for storing more data.

The records-per-base issue is super annoying. I've actually been in touch with their enterprise group and I was told the enterprise limit was 100,000 records. That's a total show-stopper. So I created cloudternal.com, without any of those limits.

I am a fan of Airtable, but the 50k record per base cap is annoying. Perhaps this will prompt Airtable to relax the cap.

Reply on news.ycombinator.com