Act Voyager ---- long story
Port Core Act to DBIx::Class schema ( 1.000 USD )
- Make Act 'hackable; with "Act-out-of-the-Box"
- DBIx::Class schema
Dancer2 implementation ( 1.500 USD )
- Make Act::Object sub-classes into proxies for DBIx::Class::Result classes
- They have the same attributes and methods as the original ones
- Retaining the tests that currently exists
RESTapi ( 1.000 USD )
- move mod_perl handlers to Dancer2 routes
- have Dancer2 routes fill in the original template
Theme Based Templates ( 1.500 USD )
- define all useful resources
- write the POST, GET, PUT and DELETE methods (and more)
- test, test and test
- write documentation
- Find a proper frontend framework (Foundation)
- Define Layouts
- Define Pages
- Redo the entire ACT with new templates and themes
- Decide which SASS variable will be editable
- Run several default Themes and Colorschemes
- Write a admin/theme-selector
The next two milestones, porting Core Act and building a Dancer2 implementation, are the steps that will ensure that all the old conferences (and the current) can be viewed and visited by bots once the Jurassic Apache and mod_perl implementation are shut down. For these legacy instances it will be sufficient that the pages can be rendered and no changes in the schedule or attendees have to be made. Therefore not all the handlers need to be moved to Dancer2. These routes gather information from the database and are used to fill the variables for TemplateToolkit. Those templates do not need to be changed and thus we can keep the websites backward compatible. Using Act::Legacy::Objects as proxies to the DBIx::Class::Schema for Act gives the flexibility to change the underlying database structure, without breaking the code that handles the original Act::Objects.
And somewhere along the line, there should come a tool that can read the 175 .INI files, the httpd.conf and possibly write them back as well. Those files contain valuable information not stored in the database, but are actually the 'root' objects for DBIx::Class::Schema. Most objects are related to an instance of a conference, but since there are no Conference tables yet, there can not be any relationships to select talks given during a specific conference, or list all attendees.
And then, of course we had a wonderful hackathon, full of action, fun and drama and produced some really cool prototypes that actually might work "in production". Unfortunately BooK and I had not fully prepared the hackathon, and there was no real clear path of direction. For a moment I even was worried about the crew members of the Act-Voyager. But you can't expect things to go very smooth when there are such great minds together that have their own backgrounds, their own experiences ... and yet have to get to the point that they were honourable people that were willing to work on Act, despite the differences.
There were made several patches to Act, Act-out-of-the-Box, done some design at the RESTapi side and some great ideas about how to move forward. Enough 'materials' to work with for the next couple of weeks and get to the next galaxy-stop, the Port of Act Core to DBIx::Schema.
Many thanks for all that support for this project, those who support this financially, but also those crew-members I consider my friends. People that encourage me to continue this exploration and voyage, that will lead us to a Next Generation Act! Thanks for all that came to Lyon (where my MacBook was stolen from the hotel room - and thanks for the generous help to buy a new one while waiting for the insurance). Special thanks for BooK who helped clearing the mess after the mutiny on the command-deck - and Liz who made sure I kept my head focused on this community project.
it's going to be a very exciting Act 2015!
Theo van Hoesel