Projects/kde.org/Planning

From KDE TechBase
< Projects‎ | kde.org
Revision as of 23:14, 23 September 2008 by Tampakrap (talk | contribs)


Introduction

Purpose

This document outlines a plan for moving the KDE.org family of websites to a content management system (CMS) and explores future improvements to KDE websites.

Points for consideration:

  • Where is the current state of the KDE family of websites?
  • Where is the direction of the KDE family of websites?
  • What obstacles are there that would make improving KDE websites difficult?
  • Porting of kde.org websites to a CMS system such as Drupal.
  • Designing of systems to accommodate open collaboration services.
  • Re-word the content to further a healthy perception of KDE (see: http://aseigo.blogspot.com/2008/09/what-does-kde-app-mean.html)
  • Centralizing the management of users, information and services for all parts of the KDE community.
  • The potential of having a physical meeting of all people working on KDE websites or strongly interested in working on KDE websites.


The current state and purpose of the KDE.org websites

Current sites

Current State

The current state of the kde family of websites is quite poor. There are numerous pages and subdomain websites which are out of date, inaccurate or both. Compounding this problem is the lack of any specific maintainers for pages and for subdomain sites.

The KDE.org Family

As well as having a rather large website at kde.org, there are also over 20 sub-domains (each with an independent website), hosted beneath kde.org

List of KDE.org main sections

  • About KDE
  • Download
  • Community
  • Develop
  • Explore


List of KDE.org sub domains

I've taken this list from http://www.kde.org/family/. Also, the current status of the kde.org sub domains can be viewed here: Projects/kde.org/Site Status. Please note that these lists may be out of date and incomplete.


Purpose of current sites

Future requirements

As the scope of the KDE project grows, the ability of the KDE family of websites to scale and meet the demands of an increasingly large userbase will become increasingly important. These websites will need to make it efficient to find news, documentation and other content. These websites will also have to foster a community of collaboration among developers and users.

To accomplish these goals, a platform which can support thousands of users will need to be adopted.

Porting of KDE.org websites to a CMS

CMS Options

Drupal

Why Drupal?

  • Positives
    • Well supported application
    • Good community
    • Scalable
    • Proven - used by large corporations
  • Negatives
    • Less flexibility than custom app
  • Scalability
    • An interesting article on some issues with Drupal (2007 = a bit old)
  • Extensibility
  • Modularity - Drupal was designed to be modular
  • Future proof
    • Drupal 6 - current stable release
    • Drupal 7 - on the horizon with big changes
    • Should we wait for 7?


Converting current website pages to Drupal pages

  • Methods
    • Manual copy/paste
      • Positives
        • Ensures everything is copied and works
        • Gives an opportunity to review the content
      • Negatives
        • Slow
        • Time consuming
    • Automated script
      • Positives
        • Speed
      • Negatives
        • Accuracy
        • Obsolete Data ported too
  • Research for Automated Script
    • How does Drupal store page data?
    • How do we determine needed data?
    • How do we handle corrupted data?
    • Shell script or PHP for conversion? Gimme egrep :D
  • Theme
    • Re-create Oxygen theme for Drupal 6.x
    • Changes to current theme? Visual refresh?


Designing for Open Collaboration Services

One of the main things that came out of Akademy 2008 was 'brining the KDE community to the desktop'. What does this mean? Ideas like 'seeing KDE users nearby' is a good example.

We all know that KDE has a large community that is going through a new phase of growth. Because of this, coordinating, designing and maintaining open collaboration services given the existing scattered state of different sections of the community would be almost impossible. And even if it was possible, maintaining the resulting mess would be an effective step backwards.

If we were able to centrally manage communities and services such as kde-look.org, kde.org and its sub-websites, svn, mailing lists, Amarok, Konversation, etc. then Open Collaboration services would be doable and far more powerful than previously possible.

If you think about the idea behind Akonadi (http://pim.kde.org/akonadi/), why don't we apply this to the KDE web community? Also, once we've done that, why don't we build our own web based 'Nepomuk'? Why not build the KDE community on these same 'pillars' as the desktop? This would make Open Collaboration services one of the most advanced out there for any community and allow the users of KDE unprecedented ease of use and integration with each other. Of course, due appreciation of privacy and security need to be taken into account. And once again, having centralized management would make maintaining privacy and security a lot easier.

Method:

  • Feasibility study
    • Can we unite management of programs like SVN, kde.org and kde-look.org?
      • Who are we trying to unite?
        • Amarok
        • KDE-Look.org
        • all KDE.org sub-domains
        • Konversation
      • Physical location
      • Leadership/Direction
      • Technical Problems
        • Scattered Databases
        • Different CMS's
      • Contact relevant communities regarding their thoughts
    • 'Akonadi' for the web
      • Centralisation of all data and users
      • Too much scattered information to go in one big database structure?
      • Performance Issues
  • Design phase (assuming feasibility study succeeds)
    • 'Akonadi' for the web
      • Database design
      • API design
      • Open Collaboration compliant
      • Separate application or Drupal module?
Retrieved from "https://techbase.kde.org/index.php?title=Projects/kde.org/Planning&oldid=33789"