Etherpad-lite: Spreadsheet version?

Created on 4 Jul 2018  ·  7Comments  ·  Source: ether/etherpad-lite

Would it be possible to create a fork of etherpad, modifying the text container to be a spreadsheet container, such as a jquery one? EtherCalc is a terrible alternative compared to etherpad, and would love to see a modified version of etherpad that would work as a spreadsheet.

Feature Request

Most helpful comment

Looks possible to me. But that would be a HUGE effort, there's a lot of things a spreadsheet would do that Etherpad currently does not implement.

All 7 comments

Looks possible to me. But that would be a HUGE effort, there's a lot of things a spreadsheet would do that Etherpad currently does not implement.

Not sure why this needs to be closed, especially if people would like to work on it and see the issue. Handsontable looks like it'd be a great candidate.

I don't see this as an issue of Etherpad _per se_, but I'm fine with letting this opened for others to discuss.

You've mentioned Handsontable, can you elaborate why we would need a fork of Etherpad instead of using it? And are there other options you've researched as alternatives? It would be good to have these points described to start the discussion...

I agree, ideally it wouldn't have to be a fork, it could just be an option when you make a new pad, to either choose between word / text pad or spreadsheet pad. I'll research into plugging it in but I'm not familiar with the codebase yet. The big concerns to me are if we'd need different websocket messages to correspond to the spreadsheet actions.

Who wouldn't love a spreadsheet version?

Given the current state of the project (security, stability, bug fixing, modernizing), I think it is severely out of scope, but I won't even self-limit where a motivated developer can reach.

Let's dream of a spreadsheet proof-of-concept:

  1. it should never impact the stability of the text version;
  2. the POC should start its existence as an out-of-tree plugin;
  3. new messages and hooks would probably be needed: these core modifications should be kept to a minimum, and never break compatibility;
  4. core modifications should be kept under a feature flag;
  5. if the experiment is successful the plugin could be mainlined in the official repo.

I'm thinking about something similar to the upcoming work on the new theme, which can be tutored immediately, but has to obey quality rules since the start.

Except for a spreadsheet everything is 100x harder, and I do not know if core contributors can give much support...

Would you like to try at these conditions, @dessalines?

For some reason I thought https://github.com/ethersheet-collective/EtherSheet was loosely more based on Etherpad. It's JS at least, though it looks to be dead and the site is gone.

My question is: Why not contribute to making EtherCalc less "terrible"? I suspect it would be vastly easier to improve an existing spreadsheet application than try to turn a word processor into a spreadsheet. Bear in mind, a good spreadsheet application is more than just "cells you can type text in", you want formulas and charts and the ability to reference data in one cell from another cell.

You could give a spreadsheet program a toolbar and UI that mimics Etherpad's (which would be cool), but it's a lot more than swapping one container for another, it'd be a wholly different application.

No activity, closing.

Was this page helpful?
0 / 5 - 0 ratings