* Collection-oriented rather than cell-oriented. Eve is loosely based on datalog which is closer to SQL or prolog than to a spreadsheet. The relational query model is much more powerful than vlookup. I believe we can make it as approachable.
* Networked / multiuser. Excel deals poorly with concurrent access to data and reacting to events. Most of the users we talk to have problems with sharing spreadsheets and moving data through pipelines. We aim to provide a shared database so that one users computations can easily depend on another users data and can respond to changes or new data.
* Web-native. The web is unfortunately the only distribution platform with any reach. That's one of the things that killed VB6.
That said, the forms/3 HCI research is excellent and their recent work on preventing errors (http://eusesconsortium.org/wysiwyt.php) has been especially educational.
Thanks for the response. All of these pain points are definitely true, in my experience.
I'm working on a database-backed spreadsheet product too, and it sounds like we have very similar ideas about the weaknesses of the current spreadsheet model.
Our output product is still 'rich' spreadsheets rather than apps, though, and while we have a programmable VM and data processing language our language looks much more Excel than prolog or SQL. We also were really interested in the forms/3 research during our early development which is what prompted me to post.
* Networked / multiuser. Excel deals poorly with concurrent access to data and reacting to events. Most of the users we talk to have problems with sharing spreadsheets and moving data through pipelines. We aim to provide a shared database so that one users computations can easily depend on another users data and can respond to changes or new data.
* Web-native. The web is unfortunately the only distribution platform with any reach. That's one of the things that killed VB6.
* Not ugly. It may be shallow, but people don't like to use ugly software and will perceive it as being harder to use (eg http://www.ergonomicsclass.com/wp-content/uploads/2011/11/Tr...)
That said, the forms/3 HCI research is excellent and their recent work on preventing errors (http://eusesconsortium.org/wysiwyt.php) has been especially educational.