YaST/Research/Ideas

From openSUSE

AutoYaST documentation

The docs about profile is maintained by Uwe and thus is constantly obsolete. The worst offender seems to be X11.

IMO we should push the documentation to the module maintainers and provide infrastructure for that. Uwe cannot scale.

Everything must be available in autoyast

All functionality needs to be configurable via autoyast, but for us autoyast is commonly just the 2nd thought.

Heartbeat module enhancements

The module is very rudimentary and it is a question for PM how they want to position it. But it's clear customers need autoyast support for heartbeat configuration.

How to integrate customer-own changes

The example was progress dialog, but a generic framework might be of interest here.

LDAP server

The module is SMB-only, e.g. does not support departments.

storage API brainstorm

(proposal code, map abuse, proposal UI, nil returns)

builtin cleanup

Can be moved to Research projects when there is a list of what to remove and what to add.

module loading error message improvement

Can be moved to Research projects as soon as there is minimal specification.

scr parsing error on stderr (not y2log only)

?????

deprecated functions

Can be moved to Research projects when there is a list of what to remove and what to add.

Move YaST to git

Can be moved to Research projects as soon as there is minimal specification on how to tackle the problems of such a migration. ZYpp migration done.

package bindings testsuite

Can be moved to Research projects as soon as there is minimal specification.

package bindings api cleanup

Can be moved to Research projects when there is a list of what to remove and what to add.

package bindings use namespaces (AddRepository -> Repository::Add)

Can be moved to Research projects when there is a list of what to change.

DBus:: in ycp

Can be moved to Research projects as soon as there is minimal specification.

Command line internal API, introspection, bash completion

Can be moved to Research projects as soon as there is minimal specification.

Integrate YaST into KDE 4 SystemSettings

Integrate YaST modules into KDE 4 SystemSettings

• Instead of having two separate applications to configure one's system, integrate the YaST modules into SystemSettings and also eliminate duplicated functionality.

Required knowledge: C++, Qt, YaST2 (?)

Skill level: low or medium (?)

Willing mentors:

There is already ongoing effort on a new version of the control center based on System settings concept, but not to integrate with it. http://svn.opensuse.org/svn/yast/branches/tmp/tgoettlicher/yast2cc_rewrite/

Totally disagree with this note, openSUSE's tools (Yast modules) must be into KDE SystemSettings (as is already doing a known distro) because the things must be most standard as possible (as now with the packagekit project for install software) at the GNU/Linux world (where there are already many differences). If each distro develops its own Control Center, is a chaos; like install software was before packagekit.

SAN to speed up the bottleneck disk

Integrate YaST modules SAN SystemSettings

• Some smart algorithm allows to predict the disk read and write commands. Some part of the RAM get used to buffer the read and write requests. As options of each disk you can choose if you like to have on/off the read-buffer and extra the write buffer. In the settings you can choose the total amount of buffering (e.g. 12GB if you run 16 GB and need for the applications and operating system only 4GB). Or you can choose the buffer space dynamic via some percentage x%. As an alternitive you use 90% of the statistically unused RAM of the last week. The SAN-middleware software should be scaleable to use different kind of disks (USB, IDE, SATA and Fiberchannel.) and be able for hotpluging new disks.

Required knowledge: C++, YAST (?)

Skill level: high (?)

Willing mentors:

This will support ressource intencive applications which access a lot the disks, like databases, CAD, video and image manipulation, virtualisation and indexing - searching - organizing your diskspace

Videoconferencing

Save Videoconferencing in a professional 64bit environment

• Some intelligent alogrithm use the peernetwork to distribute telephonecalls and videoconferences. Realtime intelligent allogrithm allow to adapt to the bandwith (priorities like audio first, etc.) and to deliver hight quality communication in full HD and Hifi Surround Audio. Echo get suppressed and the webcam can follow the participants (automatic zoom in & out) and the visualisation get adaped to the best image results with the actual light. It could handel 25 participants including a parallel chatroom, presentationmode and sketch via tablets digiboards.

Required knowledge: Realtime Applications, Voice over IP, C++, YAST (?)

Skill level: high (?)

Willing mentors:

See also some state of the art webcams: [1]

Minimalistic User Friendly Yast Interface/Plug&Play Background Facilities

Produce a minimalistic interface for the *regular-non-technical* user. And also a backend which helps the user automatically when needed.

One of the long running problem with non-technical users is the perpeption that "Linux" is difficult and painful to configure and use. Part of the problem is where YaST offer too much, so the regular users are overwhelmed by the posibilities and the choices. Other problem is the way that regular users "think" when they want to use their computers.

So a minimalistic approach would look to make easier and pain free the most common task such as application install, peripheral install: printer/scanner/webcam , Folders sharing/Network printing and drivers installation for recently connected hardware.

More info: [2]

Required knowledge:

Skill level:

Willing mentors: