Metaset

It's done.
Added better feature videos. Rewrote some copy.
I wanted to get feedback from only those who are interested. So provided an Early Access form instead of a public download link.

Will be posting on social media and other spots shortly.
Now the fun starts!

I am now ready with the website: https://metaset.io.

• After looking at various templates, I decided to code the website from scratch
• Recorded a quick walkthrough video
• Listed all the relevant features
• Added an early access email submission, instead of a direct download link for the binary. I would like to get feedback from a small group of early users.

After about 6 months of trial and error, deleting whole codebases and starting over, talking to friends about ideas, designing and redesigning… I felt that I was finally ready. It did what it needed to in my initial vision, performed well on the sample datasets, and didn't have any obvious bugs. I was ready to show it to other people.
But first I needed a website.

Getting started was just a bit challenging:

• I had done a bit of Mac programming, but had to learn a lot of new topics. Thankfully I am proficient in Swift and many ideas from iOS transfer over easily.
• Had a false start when I began building a wizard-like interface for composing a query. Although a lot of fun, this proved to be a lot more challenging than I'd like a side-project to be.

But I had a very rough and limited prototype up and running in a couple of weeks.

Had the idea after I needed to visualize some data I had gotten in CSV. Usually I would write some quick Jupyter notebook code or put it in Excel. But writing code is too much work when I just want a couple of charts. And Excel is both frustrating and not powerful enough. I could use something like Tableau but I find it a little overwhelming, and I am wary of handing over my data.

I figured I must not be the only one with the same frustrations and preferences.

At some point I found a post by Jakob (Mac developer of database apps at http://eggerapps.at) where he was looking for something similar. I got in touch with him, and he was quick to validate the idea.

That was it! I had enough to get started.