Looks really interesting, and we're currently scouting for good table solutions, but this is a killer:<p>> Onomics is currently free for beta users, though we’ll eventually start charging for some premium use cases (hosting high volume charts, heavy usage, advanced charts, branded charts for example).<p>You've got to make money, that's for sure. But this open ended future tax (what is heavy usage/high volume anyway?) means that I can't invest the time in checking out your offer.<p>Much better would be some bold statement like "features x,y and z will always be free".
I was intrigued by the title of this post as I was looking for some nice django+d3 combinations for data viz but ... A hosted table with a bar chart? That's new.<p>I don't really see the audience for this. In the business world, no one will be willing to host their precious, sensitive data on some unknown server. For private use, I'd argue that one either creates a table programmatically in html or simply does it in excel if it's just something small. While we're on the topic of excel, I reckon one could recreate this app with VBA to look identical or near identical within less than 24h (possibly 3h-8h?).<p>Then comes the description itself - Stephen Few might get a small headache reading it:<p>> tables present all the actual data so readers feel more confident they’re seeing the source information.<p>As opposed to what, charts? What else do charts show if not "actual" data?<p>> Tables also allow you to rank items and show the relative magnitude of any item versus the rest.<p>Do charts not have exactly the same property? The relative magnitude of one item is easier seen in a well made chart than a table.<p>> Lastly, tables have a high information density: they allow you display a large amount of data in a limited space. Imagine the above table as a bar chart and how unwieldy that would be<p>The above table does in fact contain a bar chart. Doesn't seem so unwieldy to me. Charts, in my opinion, can have a much higher information density which is still comprehensible than tables - a scatter plot with 2 series each made of 1.000 items might be eye-opening if done right (and data has some trends etc.) - the same data displayed as a table, however, would take minutes/hours to digest.
Journalists will love it!<p>It's not easy for journalist to do a simple table like yours.
The problem is very time consuming for them to do this kind of work.<p>If you want to get traction in newsrooms, you'll need to add the option to tweak the size of the chart to be able to fit their CMS screen state, and twitter/facebook/etc post sizes.
Why would I ever want to split up that table into two columns? Maybe this works in a blog where everyone is an amateur, but any professional worth their salt would stay away from that in the business world.