We are looking at ways to improve the https://vizhub.com/create-viz page and are looking for feedback.
Some initial ideas:
Also tracked as GitHub Issue: Fix up Create Viz Page.
Minor suggestion:
When the user clicks one of the choices, it should be automatically forked so that the user directly starts editing his/her own fork.
It happened to me many times that I would choose a viz from that page and start editing right away, then I notice when I get the error that I didn’t fork it. So, I believe the automatic fork would make the process (at least slightly) more seamless.
Yes this is a really great point.
Related to the idea of an Interstitial forking rename dialog.
I think it makes sense to implement this auto-forking feature only once that is in place - so it’s clear that a fork will occur. Probably if we do it now, it may be a bit disorienting to click on one of the previews in the create-viz page and have it automatically fork.
Agree with Waseem.
As for me, then user is on create vizz page, one think of vizzes as templates rather than previews. So users want exactly what Waseem described.
Cool! (No, I’m not saying that because my template made it to the front of the page. 
)
This looks great! 
A couple of thoughts:
The first is whether examples could be labelled upfront as D3 v5, v6 etc. This would help ease one’s search and exploration. In fact, I wonder if the version number could be pulled automatically.
The same could apply for React but then again that remains mostly (as of the present moment) backwards-compatible. Indeed, the biggest jump for me has been more of a mental one (between classes and hooks).
Second, would it be useful to also have examples categorised by the kind of data stories they show? One of your videos touched very nicely on this important topic around different data types suggesting different data visualisations. Some basic categories could be distribution, composition, change over time, and relationships between data. Something like each of these could have an example of an appropriate visualisation for either categorical, ordinal, or numerical data. For example:
Obviously, the above is a terribly rough sketch and I defer to your knowledge about categorisations, and what best visualisations best match them.
I like the starters, was hard to find them among the others while going through videos.
