User Tools

Site Tools


wiki:development:roadmap

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
wiki:development:roadmap [2015/01/27 11:23]
stefanbaur
wiki:development:roadmap [2017/05/11 09:45] (current)
stefanbaur New section: NX-Libs (missing) features table
Line 1: Line 1:
-====== ​Development Planning ​====== +====== ​ΞX2Go Developer Meetings ​====== 
-  * [[wiki:​development:​planning:​whiteboard]] +{{:wiki:​development:​x2go-mascot-ccc.png?​nolink&​100 |}} 
-  * [[wiki:​development:​planning:​schedule]] +  * [[wiki:​development:​planning:​start|Planning the next meeting]]. 
-  * [[wiki:​development:​planning:​start]]+  * [[wiki:​development:​planning:​meetings_list|Past ΞX2Go Developer Meetings]] 
 + 
 + 
 +---- 
 + 
  
 ====== X2Go Roadmap ====== ====== X2Go Roadmap ======
Line 20: Line 25:
  
 === Prio #1 (not started yet) === === Prio #1 (not started yet) ===
-**Run multiple instances of X2GoClient**,​ or multiple sessions from one instance, all simultaneously (kind of how pyhoca-gui works) ​must work on+**Fix Lockup of OS X client when running multiple sessions //​sequentially//​** 
 + 
 +At present, X2GoClient on OS X goes into an endless loop and has to be force-terminated after the second or third connection attempt. 
 + 
 +Task assigned to: Mihai 
 + 
 +**Run multiple instances of X2GoClient**,​ or multiple sessions from one instance, all simultaneously (kind of how pyhoca-gui works) 
 + 
 +must work on
   * Windows   * Windows
   * OS X   * OS X
   * Linux (probably already does)   * Linux (probably already does)
  
-would love to see X2GoClient gain the multi-session feature from PyHoCa. ​ In fact, the GUI design of the taskbar icon's context menu hasn't made sense from the start, if X2GoClient is and will always be+Stefan ​would love to see X2GoClient gain the multi-session feature from PyHoCa. ​ In fact, the GUI design of the taskbar icon's context menu hasn't made sense from the start, if X2GoClient is and will always be
 limited to one session.  ​ limited to one session.  ​
  
Line 37: Line 50:
 h1 has a design in mind that addresses multi-sesson support in one X2Go Client instance (also x2goclient2.git on git.x2go.org is capable of that, but at the moment it is half-baked and probably temporarily discontinued). h1 has a design in mind that addresses multi-sesson support in one X2Go Client instance (also x2goclient2.git on git.x2go.org is capable of that, but at the moment it is half-baked and probably temporarily discontinued).
  
-Task assigned to: Mihai, Mihai, and again, Mihai+Steps: 
 +  - Fix Multi-Instance on Windows 
 +  - Write wrapper to turn Multi-Instance into Multi-Session mode (especially important for OS X, where Multi-Instance isn't supported) 
 + 
 +Task assigned to: Mihai, Mihai, and again, Mihai - with some help by Mike#2 for the Windows client
  
 === Prio #2 (not started yet) ===  === Prio #2 (not started yet) === 
Line 60: Line 77:
 Task assigned to: Stefan, aided by Mike#1 (hard to say, Mike#1 will coach Stefan for an equivalent of 4h to get things straight, no matter how long it takes) Task assigned to: Stefan, aided by Mike#1 (hard to say, Mike#1 will coach Stefan for an equivalent of 4h to get things straight, no matter how long it takes)
  
-<​del>​**HTML5 Client**</​del>​tagged as unrealistic given current budget+<​del>​**HTML5 Client**</​del>​ tagged as unrealistic given current budget
  
 If budget remains, a HTML5 client would be nice. If budget remains, a HTML5 client would be nice.
 +
 //There seems to be a disagreement amongst the developers regarding how such a client should be implemented.//​ //There seems to be a disagreement amongst the developers regarding how such a client should be implemented.//​
 +  - One option would be to copy what other remote solutions are using - which means sub-par performance compared to our native NX/X2Go speeds.
 +  - A from-scratch approach would bring native speeds, but is a time and money sinkhole.
  
 Task assigned to: <none yet> - could be Alex, Mihai, ...? Task assigned to: <none yet> - could be Alex, Mihai, ...?
  
  
-<​del>​**Alternative to NX-Rewrite**</​del>​tagged as unrealistic given current budget+<​del>​**Alternative to NX-Rewrite**</​del>​ tagged as unrealistic given current budget
  
 h1 has some ideas what other F/LOSS codebases might be suitable as a time- and money-saving alternative to doing the NX-Rewrite. h1 has some ideas what other F/LOSS codebases might be suitable as a time- and money-saving alternative to doing the NX-Rewrite.
Line 179: Line 199:
  
  
 +==== first attempt at documenting missing NX-libs features for certain applications and desktop environments ====
 +^Feature/​Application or DE^NX-libs Version^V. req'd for Gnome 3.14^V. req'd for KDE5 ^
 +|XServer: RANDR| | | |
 +|XServer: RENDER| | | |
 +|XServer: Composite| | | |
 +|XServer: GLX| | | |
 +|XServer: DRI| | | |
 +|XServer: BIGREQUESTS| | | |
 +|Client: libXrandr| | | |
 +|Client: libXrender| | | |
 +|Xinerama| | | |
 +|MESA| | | |
 +|VirtualGL| | | |
  
wiki/development/roadmap.1422357787.txt.gz · Last modified: 2015/01/27 11:23 by stefanbaur