On Thu, May 27, 2010 at 11:50 PM, Oz Linden (Scott Lawrence) <o...@lindenlab.com> wrote: > It's fairly clear that Linden Lab doesn't have the resources to devote > to active work on both Snowglobe 1.x and 2.x, and it's not efficient for > the community as a whole to be splitting effort.
Snowglobe is an opensource project. Sure Linden Lab can decide not to do anything about 1.x (not that it has done much about it in the past year or so anyway), but you cannot tell the outside contributors what they want to focus on. And most of them still want to improve on SG 1.x. > I'd like to fairly quickly get to the point where all our new work is > happening on the 2.x branch. That said, I understand that might leave > behind things that the Snowglobe user/dev base wants and that some > people are not happy with some elements of 2.x. What I'd like to know > is... what needs to happen to make that choice that most people can be > happy with? I'm afraid this is not going to be very easy Oz. It's not likely that a lot of new development is going to go to 2.0 since a lot of people find the UI much more difficult to use. And Linden Lab has in the past few months since the introduction of the new UI only be willing to do minor tweaks to it, nothing of substance. See the JIRA's and mailing list threads about chat bar focus where the response from the head of the viewer team was basically "no can do", complain all you want but that's not changing. With attitude like that from LL the response is not surprising, OK, no we cannot move to 2.0 with such an inflexible approach to fixing most glaring UI issues. > One of my goals is to increase the rate and volume at which Linden Lab > can (and _does_) take changes from the open source base into the > internal code, but unless we can keep everyone on the same branch, that > will be much more difficult. This would be good indeed. It was very sad to see that almost none of bugs that were fixed in SG 1.x tree were taken upstream. Viewer 2.0 was released choke full of bugs (including crashers that could lead to potential exploits) that were fixed moths and months before in the public SG repository. > Please respond to this thread with your favorite reasons not to move > development to 2.x. We will review the list at the 6 June open source > meeting with the goal of setting some priorities. It's UI, the UI and more of UI. While the 2.0 at the first glance looks nice and modern compared to 1.x it's usability is far worse. 1.x UI could be skinned to look much more modern and many TPV already do this. >From what I have seen building from viewer-external branch so far, 2.1 is going to bring only tiny cosmetic changes without addressing the real problems of the UI's usability. SL is a 3d virtual world. 2.0 UI with it's non transparent floaters, inability to dock local chat history with other ongoing conversations, obscures much of the world, making it into a facebook page with a glimpse of something 3d behind it. Local chat changes from 1.x to 2.x also show this facebook fixation of 2.0 design, moved from functional, usable display to something resembling facebook or twitter updates. Try an experiment for yourself, go to a busy club with a lot of chat in 1.x and 2.0 and see the difference. If you get a couple of IM conversation goings + busy local chat, there is hardly any world left to see. Local chat input field is too small even for facebook or twitter style updates. 2.1 makes this even worse by trying to jam more buttons on the same line, so you get chat input, buttons, IM conversation items and notification counts all on one line. Chat is a clear loser there, but guess what, a lot of people come to SL to socialize and talk to each other, so de-emphasizing chat like that isn't going to improve the user experience. At the same time, we have this huge and mostly empty address bar, so the message seem to be, let's make it look like a web browser and heck with usability. Also. the changes of focus handling for local chat between the the two versions show that the designers of 2.0 have never actually used SL for socializing, otherwise they would notice that people start jumping around and doing other weird things because they thought input focus was on the chat line. Inability to open more than one user profile, or to look at a user and group profile at the same time is also very limiting, and makes some common tasks much more difficult. Just compare the list of user groups and the task of opening them and discovering what they're about between the two versions. This is how people discover things about Second Life, open people's profiles, look at their groups, their picks, and go visit those locations. The modal nature of 2.0 interface makes this painful.-> Compare how to do common things, "how do i upload picture of my cat" and share it with friends. In 1.0, it's file -> upload -> image. In 2.0 you have to figure out how to open inventory and that there is a menu in there that allows you to upload. Current trunk makes this even worse, you need to know that you have to click on a suitcase icon first, then click on "+" icon on the bottom which opens a menu two levels inside of which you find the upload. Just about as obscure and as non-discoverable as you can get, For a very good review of 2.0 UI I would recommend reading http://zhaewry.wordpress.com/2010/03/15/viewer2-0-beta-the-good-the-bad-the-ugly-and-the-odd/ In summary, getting experienced SL users (and most devs are) to invest their effort in 2.0 will take more radical improvements to the UI usability. Perhaps forward porting of the 1.x interface elements (chat, profiles, group profiles, etc) is the way to go. I'm hoping that your efforts on making LL more responsive and inclusive about community contributions will help reignite the interest of OS devs for the project. Latif _______________________________________________ Policies and (un)subscribe information available here: http://wiki.secondlife.com/wiki/OpenSource-Dev Please read the policies before posting to keep unmoderated posting privileges
