Making making apps easier

[Pascal Bergmann]

https://airtable.com/ - raised $52M to help businesses to build apps. By utilising a spreadsheet-model, Airtable makes itself very approachable to their target group.

There is definitely a big market out there for things like Eve. Maybe someone has fun in narrowing down Eve's vision and build an "Apple-y-easy" GUI for a specific target group (e.g. knowledge workers, managers, marketing...)

[magicmo...@gmail.com]

Airtable is not a language at all. It is a set of prebuilt modules, that are not actually user programmable. They have programmed 9 commonly used functions, like Event Marketing, and created a fixed schema for this database. they then allow you to add/modify/update records into those modules. You use a Restful HTTPS interface to do all your database work, which means this data is stored on their servers. So AirTable is not a general solution but a half-assed clone of SalesForce's very successful model. 

The heavy lifting of drawing the data structures is their value added part; you merely update the data and your calendar draws. Calendars exist all over the place in the business universe, so this is a very good module to offer. The universe of coding is evolving towards re-usable parts, and the 9 components they are offering are reasonably commonly used. But how general is AirTable? Not at all. They are doing the programming, the schemas are static and under their control, and if they are doing the drawing you can't modify their components at all. And they are closed source. So is this any different than a VB6 calendar plug in from 20 years ago? Well slightly, because it is using a remote server, and the web as the foundation instead of the Win32 API. But let's get serious, some of their modules like the Digital Video module is a toy. Video asset tracking is a commonly sold shrink wrap product, and there are over 20 products already in the digital asset management universe that range from simple to elaborate; why would you build one when this is such a complex area? And the second you want to change their data structures you are screwed. 

AirTable at 12$/month is cheap enough to get some traction, and from that early traction they have clearly succeeded in raising a huge chunk of capital for round 2, but their fundamental model is deeply flawed, because they have no way of representing and updating state in their system, and the idea that the server has to hold the sole source of truth means you will have to constantly pump data back and forth to their server, and since they are transmitting all information in textual form, this is not efficient at all. If you haven't seen all of Joe Armstrong's (the inventor of Erlang) lectures, please check them out on youtube, i love the one where he rants about how ericsson didn't sweat night and day to save a few bits in the packets to see it all wasted on JSON strings going back and forth... ;-)

Since their approach is so rigid, AirTable is not an entrant in the next gen language race, and if a next gen language takes hold, AirTable will have to be rebuilt entirely. They don't actually disclose their underlying technology stack, but maybe it is the conventional dogpile of HTML/CSS/JavaScript/MySQL/and some frameworks thrown in for good measure. That tower of babel is what Eve and the other projects are trying to eliminate, and merely putting a pretty wrapper around a giant pile of JavaScript with some Web API's doesn't impress.

[Pascal Bergmann]

Thanks for your reply :) You're absolutely right. Technically, Eve is way beyond anything Airtable can and wants to be. However, Airtable is something the normal business guy understands. Eve isn't. Airtable is also something that could've been easily built with Eve. Wouldn't that have been a use case for Eve? Wouldn't it have brought some money to develop Eve as a language? The concept behind Eve is genius. Chris & team just didn't have the time to get the interface right. Maybe someone one day takes Eve, builds a business product on top and funds Eve language development. Please. ;)

[William Taysom]

I'd say Airtable is a product meant to be used right now. Eve seemed more of a research project reexamining basic ideas. Genuinely wish we were much further down the Eve path, but here we are.