By no means finished but it's first time I've shipped a side project from start to finish so it feels good :)
At least for right now. If the goal is something bigger, then getting users and solving the problems related to use that come from edge cases in browser configuration; drawing size and style; and saving across the internet can be field tested by reducing the impediments to use.
Getting rid of accounts also means getting rid of account management chores...like the one here...that eat into the time needed to make the user experience better. To be clear, no matter how good you make the signup, it is a bad experience that stands between a user's enthusiasm for the product and their use.
Going further, there are approximately two user experience scenarios that go with signup.
1. The security conscientious user will spend time thinking about security and managing identity and assessing risk associated with adding another account because every account increases attack surface area.
2. The security naive user will just use an existing identity pattern and that makes your site a more attractive target for black hat hacking and therefore increases effort that must be exhibited on non-functional aspects of the site.
Or to put it another way, would you rather spend your time managing an account system or doing creating something interesting...and productive: I found your comment from this resubmission of your site (plus an HN search): https://news.ycombinator.com/item?id=15283017. It goes to the normal page instead of the new better experience, as would be expected. And so my experience was worse than you hoped because it is hard to control incoming links.
Adding signup and accounts are a task that frameworks make easier than the important functional work. But they are hard to get right because no user says, "I wish I had to sign up for this instead of doing the interesting thing I came here to do." This means accounts are hard to get right. It is only the ease with which frameworks make creating a user account system possible that makes adding user accounts "best industry practice."
tl;dr = Treat the need for user accounts as a possible edge case.
Good luck.
I guess I got carried away working on what's behind the login wall, but better onboarding/signup is the next thing I will work on.
So I'll work on that next, probably a random easy sketch you can draw without signup and as you submit it, you submit a signup form.
Maybe my views are outdated about this, but I haven't see this done without registration, sure I can store stuff locally in the browser/device but the second you switch, your data is lost.
I guess I could make it easier for the latter security-naive user by offering SSO with Facebook/Twitter/what-have-you but in the end it's just glorified signup, a better UX sure, but working around the same problem.
If you can give me examples of other apps that have done this well, I'm all ears, this is very much a learning experience for me and I'd like to take away as much as possible.
An example of a site that does not require signup is this one. The most important user function -- encountering intellectually interesting material -- does not require signup. And my suspicion is that most of the people who create an account, read Hacker News for a while before doing so and that many people read Hacker News for a long time without making an account (unless applying to YC). And it is worth noting that an HN account only requires selecting a user name and password: no email and verification nonsense.
And again it is worth emphasizing that HN offers substantial benefit without signing up. The same with Google because singing up is orthogonal to the core function.
Looking at SketchDaily.io, the reminders and email are orthogonal to the acts of sketching and learning to sketch. They apply as much to sketching with a pen on paper as to sketching on a web site.
Because a person can sketch daily using a pen and paper instead of a website, the value of the website has to be in providing a better user experience. The odds that providing a website with a valid email address will provide a good user experience are often close to zero.
Maybe there are great features behind the signup screen. Maybe there aren't. Putting the signup screen in front of those additional features means that the website misses the opportunity to sell those features to any potential user who does not want to bother with the friction of signing up.
How bad is it? Well I'm no your page and your signup process sends me away from your page to another application in order to verify my email address and I cannot proceed until I do. So I can get distracted or not bother or the email can go into my spam folder or just get stuck because SMTP is an unreliable protocol. And all of those things stand between me and whatever enthusiasm I had five minutes ago.
I mean it's good for Google if I check my Gmail account. It's not good for me. It's probably not really good for your site either because your site gave me a chore. And the chore is in lieu of what is really useful for both me and the site...my use of the site so that you can learn what works better and worse and make improvements.
I'd put it this way, two people have complained about signing up after actually landing on your page. This might be evidence that some fraction of potential users don't see an obvious benefit from signing up. If the signup was not there, you might have received more interesting and better feedback about the actual function of the site rather than its administrative policies.
I did make the confirmation optional previously, I think first step will be to go back to that and work from there towards a signup-as-an-edge-case approach.
Thank you, you've given me a lot of food for thought.