Because priorities.
Obviously it would be better if the Rein alpha had a graphical user interface and during the initial development phase it was tempting to try and bolt one on. After all, the vast majority of users won’t touch an app if there’s no GUI.
On the other hand, Rein is like a construction site with open trenches, loose planks, and dangling wires. The very existence of a GUI might set an unreasonably high expectation of “fitness for a particular purpose.”
Finally, the goal was to release something in January and a GUI wasn’t in the cards.
Thankfully, there seems to be a recent trend in application and protocol development that starting CLI-only is ok, if not preferred. Take a look at Drop Zone or 21 Inc’s Bitcoin Computer. Coders like the CLI and young projects are smart to court them.
The other concept I’d like to bring up is dogfooding, where a project or company uses its own product to squash bugs and build features that surface only through daily use. Rein has the potential to be dogfooded so why not get it building itself as early as possible GUI or no? Maybe the first (known) transaction will tackle an issue from the repo?!
If you’d like a better user experience, you can wait for others to build it or jump in and help, whether by directly hacking on the codebase or by posting a job with Rein itself.