[opensource-dev] Open Viewer Development Announcement

[Ann Otoole]

I use SLv2 code base viewers unless I need to go into complex build 
environments. The SLv2 code has a lot of performance issues that don't exist in 
the 1.x code. And it has nothing to do with avatar attachments since the crashes 
occur in regions with only me in them. When I need to go to those "fabulous 
builds" (with ktris/fr measurements in the range of 3 million polys or more) I 
am forced to use snowglobe 1.4. I think there is a lot of efficiency to be 
gained from revisiting the menus/commands and considering a multi row user 
configurable hot bar for people to customize for the modes they operate in.

So anyway...

How about soliciting requirements from us customers and then letting us 
customers prioritize them and then you go through the process of explaining why 
you will or will not implement. Maybe there is a bastion of software methodology 
expertise amongst us customers and we could organize a requirements system for 
you. Then the OS developers can choose if they want to continue with the high 
demand features in a TPV for SL or just take what the customers are asking for 
to open sim. Oh and metrics accessible to in world devs to back up decisions 
regarding establishing limits would be good so we can run the tests ourselves 
and see if we come up with different results. Like how scriptless attachments 
don't slow me down on tp or crossings so to me the issue people have, as proven 
by having them get rid of the lousy scripts infesting hair and shoes and then 
them wearing the scriptless attachments while teleporting and they stop having 
issues is important. Just making assumptions all attachments are the cause of 
all issues is an error since the issue appears to solidly be associated with the 
scripts.

>> Linden Lab will absolutely have the final word about what goes         into the 
>>viewer. 
>>

Sure. Of course. I for one have no expectation of LL putting in the jiggle 
feature. I don't even care because I don't use it nor any viewer with it. But I 
am one person. From my technical perspective opinion it is a hack and does not 
take into account chest/neck/pectoral attachments that may be part of an 
expensive oufit. Therefore it is not what I would call a professional feature 
because it interferes with or fails dominant use cases and requirements even 
though it is a very much in demand capability that keeps vast numbers of people 
from using your viewer as they openly state jiggle is why they do not use LL 
viewers. 


>> We're a multi-million dollar business with hundreds of       thousands of 
>>customers,

Yes. As long as we customers, as an aggregate group, decide to keep giving you 
all those millions of dollars. 


>> some kind of product management and quality control

Really? Why do previously fixed defects keep coming back? There is one I have 
always wondered about. How many times has an LL dev fixed the broken high 
resolution snapshot feature the majority of businesses rely upon, or rather once 
upon a time relied upon, for product imagery to create effective marketing? That 
is just one defect I recall having been fixed 3 times and then obviously 
overwritten by someone's obsolete code from their hard drive. SQA, or apparent 
lack of it, is a frequent theme amongst LL bashing festivals in various forums I 
have seen. Since I am not in your building and privy to your actual processes I 
can only go by what is observed in your deliverables. As for product management? 
I can't really go into this topic without getting personal so I won't.

BTW large groups of customers screaming at you is a serious symptom the board 
needs to concern themselves with. IMHO anyway. I suppose you can stop holding 
office hours and eliminate that issue if people won't maintain their 
professional bearing and sense of self control in meetings.

Is LL even open to constructive feedback about all aspects of LL's product 
delivery? Or is this restricted only to what LL exposes for feedback? Since this 
list is OS dev then I am curious about where the feedback mechanisms will be for 
the other aspects of LL's service delivery.

Thank you for the opportunity of having a dialog.

-----------------------------

From: Oz Linden (Scott Lawrence) <oz at lindenlab.com>
To: Henri Beauchamp <sldev at free.fr>
Cc: opensource-dev at lists.secondlife.com
Sent: Mon, August 16, 2010 2:56:25 PM
Subject: Re: [opensource-dev] Open Viewer Development Announcement

I've said this before, and I'll repeat it again here:

Don't waste everyones time suggesting that we throw away Viewer 2,     or that 
we revert the UI to Viewer 1.   It is absolutely not going     to happen, and 
any suggestion to that effect will be ignored.

That does not mean that we don't recognize that some choices in V2     were not 
optimal, and that some probably need to be revisited, and     we're open to 
doing that.  But we will do it in the context of calm     discussions of what 
problems exist and creative ideas for how to     solve them.  We are not moving 
backwards, we are moving forwards.

Think about it for a minute - there are an infinite number of     possible 
solutions for how to build a UI for a virtual world viewer     - what are the 
odds that the first or second attempt produced the     best possible UI?  We 
need new and creative ideas focused on     specific problem descriptions.  

That is all described on our process documentation pages on the     wiki, but 
I'll hit the high points here:
-----------------

	* Linden Lab will absolutely have the final word about what goes         into 
the viewer.

>We're a multi-million dollar business with hundreds of       thousands of 
>customers, and we need to deploy a coherent and       reliable software product 
>to them.  If anyone thinks they can do       that without some kind of product 
>management and quality control,       they are welcome to go build a business 
>the size and complexity of       ours and demonstrate it.
>



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.secondlife.com/pipermail/opensource-dev/attachments/20100817/0b884aa5/attachment-0001.htm