How to fail a roguelike development (Guide from the Evil Wizard) #2

Hail to you!
It is me again, your best friend, the Evil Wizard!

I have second, very important advice for you:
never use any tools, algorithms or libraries!

There are a lot of them all over the Internet. Beware, the only purpose of their existence is to rob you! Steal your creativity progress.
Don't listen to weaklings, who say that using engines will save your time and will let you to start the gameplay itself.
Oh no, see through their intentions. They want to take a year of fun from you!

The only way to make roguelike is to write your own engine from scratch!
Because roguelike development is valuable itself and the final result isn't really important.


Your humble servant, the Evil Wizard.

System Architect

Do you remember, I have wrote about the 'waves of understanding'?
The milestones of development.

Okay,  additional role to this list: System Architect.

I played with realtime, showed Fallen at DevMeat meting in Saint-Petersburg and found that it needs few engine updated.
I wrote about it beforehand.

The result is simple, this upgrade to game engine will give us useful tools, remove overburden complexity and so on, so on.
I don't think it is very interesting for you to read.
The more important part, all this upgrades could be done long before!
They are oblivious and, as every oblivious things, need a lot of experience to implement. 

I read that system architect is the leveled up programmer.
I have to agree. Vast experience with Fall From Heaven the Roguelike and Fallen gave understanding that flexible project architecture are top important milestone for success.
Wish to learn it earlier, but it is cool that I got it a least now.
^_^

So, I work on widget mechanism and wrap over the old legacy code.
What's next?