Please read to the end to understand how Liquidart started
22nd April 2021 - 11:10 PM IST
The initial idea I had, before beginning Liquidart was to create a server framework from scratch. At that point, even the name Liquidart didn't exist. The idea was simple. The idea was to provide a minimalistic framework which was fresh and modular, so that people can use it to expand their server applications.
When I found out that Aqueduct was discontinued, I decided to migrate the package and revive it. I actually liked Aqueduct and it was the reason that I never wanted to learn nodejs just for server side programming. There were great tutorials explaining the use of Aqueduct for Server Applications, from programmers like Nick Manning @seenickcode ... The Aqueduct documentation was also helpful in explaining the structure of the framework and the proper way to use it.
The entire Aqueduct package was a nightmare to migrate to Null Safety with more than 6000 errors showing up during migration. It took me 4 consecutive days of work to resolve all of those. I rescheduled all of my other projects to accommodate this. I barely had proper sleep for those 4 days. I hoped that the migration to null safety will prepare the package for publishing. Little did I know that the entire Aqueduct codebase was absolutely messed up. I have been trying to solve some list-related errors for the past 2 days, only to find out that those errors have been present in the project for more than 8 or so months. It was not the migration that caused all these problems. The codebase was messed up even before migration. That's why it is almost impossible to bring this package to a functioning state. And I am all alone working in this project, which makes it even worse.
At this point, after going through so much complexities and troubles trying to revive the project all on my own, I am forced to go back to my initial decision. I will be creating a server framework from scratch and changing the documentation and examples to support the new codebase. I believe that is the right decision to make at this point, especially since the Liquidart package hasn't been adopted by many programmers. It is only a month old. It's better to provide a working alternative than to shove a broken package on people's faces pretending that it is the best thing in the world.
enhancement critical