Antti: I would suggest that the kind of Rexviewer fixes that rise to the top of the list are those that involve improved compatibility with the current OpenSim code, developer, and user base. This strategy would allow people running OpenSim grids to recommend the rex viewer as a valid alternative to LLviewer. For this to happen, the viewer has to:
(1) be compatible with viewing and interacting with existing OpenSim regions (2) provide a pathway for improvement over (1) Item (1) is consistent with the development of modrex, and will lead naturally to RexNG adoption. The following is a partial list of high priority items (some from http://modrex.wordpress.com), at least from my perspective: 1. Recommended performance settings need to be determined for existing OpenSim worlds, since the opensim user will experience reduced FPS rates in denser areas as compared with LLviewer 2. Primitives and objects that can be seen in LLviewer, but not in Rexviwer, need to be implemented (trees, flexiprims, particle systems...via LSL invocation). 3. Shadows need to be made a Preferences->Graphics option rather than having to manually set a flag in each prim. Otherwise, opensim regions will be unable to take advantage of shadows. 4. Terrain texture mapping at different height levels 5. ..Anything that OpenSim users expect to see when they go in and interact with their regions using LLviewer. Some of these features may already work with your server, but are not passed through to modrex? For (2), you have "significant value added" - improved lighting, shadows, meshes. But thsi value added will not be realized practically, without (1) Regarding your first statement on "improved authentication/storage" services, I hope that you are working closely with the OpenSim dev community so that everyone can benefit and that RexNG remains a human interface controller. Let OpenSim handle the asset and authentication services--if they need improvement, work with their devs. Just my 2c. -p Antti Ilomäki wrote: > We're currently considering our options. As you all probably know, > we're focusing on the new architecture, including an entirely new > viewer and improved authentication / storage services. > > Our internal resources are naturally limited and we are looking to the > community to join in and assist in the production. It has been heart > warming to see all the great conversation on our mailing lists, it > appears that many brilliant people are genuinely interested in what we > are doing. It is also very important to note that we are working for > you guys, one of the important measurements of success for us is how > many interested developers and users we have, so all the feedback and > contributions are vital for us. Thank you for everything you have done > so far. > > I realize that the present realXtend viewer is something relatively > few people are interested in working on for licensing issues. It's > also important to note that fixing certain issues is a lot of work; > one of the very reasons we chose to start building a new viewer was > that creating the features we really wanted was becoming excessively > difficult and time consuming on the current codebase. > > I have many things I would like to see fixed in the present viewer as > well and the NG architecture is still some time away. I would > appreciate your help in deciding what kind of fixes we should focus on > right now, so write down your ideas and let's discuss them in this > thread. Let's see what the community can come up with, find out the > crucial problems and see who, how and when they could be fixed. > Although the progress on the new viewer is quite promising, we don't > intend to just abandon the present one. Then again, this is an open > source project, so if you know people who could help us fix the issues > we find, it would greatly benefit us all. > > So let's hear your thoughts! > > Here's one to begin with: > - The default click & drag function for users with building rights > seems to cause more problems than it's worth. Moving objects could be > done exclusively in a slightly more complicated (push button to grab & > move for example) but far less accident-prone way. > > > > > 2009/3/23 Jani Pirkola <[email protected]>: > >> Dear realXtend, Antti, >> >> Modrex seems to advance fast and I am truly excited to see it starting to >> work! Thank you! >> >> As Paul Fishwick noted in his new modrex blog, there are new issues that >> need work to be done at the viewer side. I hope it is not DIY, but you guys >> have some plans to make basic fixes. If so, do you have plans or schedules >> already for 0.5 release of the viewer? >> >> >> Best regards, >> Jani >> >> >> > > > > -- Paul Fishwick, PhD Professor and Director, Digital Arts and Sciences Programs University of Florida Computer & Information Science and Eng. Dept. Bldg. CSE, Room 301 P.O. Box 116120 Gainesville, FL 32611 Email: [email protected] Phone: (352) 392-1414 Fax: (352) 392-1220 Web: http://www.cise.ufl.edu/~fishwick --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/realxtend http://www.realxtend.org -~----------~----~----~----~------~----~------~--~---
