[sldev] What is the point of firstlook and giving feedback to LL

[Henri Beauchamp]

On Thu, 10 Apr 2008 10:47:07 +0000, Matthew Dowd wrote:

> Sorry, this post is harsh, negative, and will offend someone in the
> UI team, I've no doubt but it needs to be said!

Welcome to the club of the "rude people" saying what must be said...
and (hopefully) listened to.

I'm afraid I'm going to howl with you and be rather rude and blunt
once again with LL... :-/
Believe me, I would prefer to congratulate everyone, but it is simply
impossible in the current state of the affair.
  
> I see that Dazzle is going the same way as the Communicator window
> fiasco. Namely that LL release a firstlook, and then completely ignore
> the feedback, release the feature almost exactly as in the firstlook
> client and are then surprised by the negative feedback, and spend the
> next six months drip feeding small tweaks in an attempt to improve the
> mistake.

I wholeheartedly second this, but really... What did you expect ?

LL as a whole proved countless times that it is totally autistic to
its _customers_' feedback and needs. Sometimes, one or two Lindens
will do a good job explaining, and even replying (yes, they do,
sometimes) to the residents either on the blog or on the mailing
lists like this one. But 90% of the time, your pleas will fall into
deaf ears. Plus, in the remaining 10%, the Linden who will be kind
enough to reply and/or listen to you is not the one who will be
able to make things move in the way you wish (and on a side note,
I feel very sorry for the Lindens in question, because they are
most likely the ones who will be hurted by my criticims: the others
simply won't care).

The developpers obviously don't work following the requests of the
residents but only do what they themselves find "cool and dandy".
There is no such thing as a clear roadmap that the residents could
at least influence a little (what about making _serious_ and
_exhaustive_ polls for the new features, or polling about the
feedback to a release candidate ?).

The project management is either absent or screwed, and instead of
concentrating the power of the devel teams on fixing the bugs which
daily ruin the residents' experience, they just let the various devel
teams do what they wish and push new "features" which prove bad
mistakes in the end, bloating further the code, and adding more
bugs to the bunch of well known but still unfixed (for months) bugs.
And I have been managing for over 20 years, so I know what I am
speaking about...

The QA is incompetent, delaying uselessly the patches submitted by
the open source community while they proved they were working
perfectly well and were not breaking anything, and on the other hand
they let huge, obvious bugs slip through in new releases (such bugs
that you stumble onto just 5 minutes after you logged in with the
new viewer...).

I can already make a prediction: if LL does not change really quickly
the way they handle the development and the relations with their
customers, they will "soon" (I give it 2 years at most to happen) find
out that other companies can do better and swip the lot.
The only reason why it did not already happened is that LL lived on
a monopoly. This is currently changing and the only chance left to
LL will be that their customers created a huge amount of contents
that new games will take a few years to equal.

I for one, am not far from either:

- forking the viewer for good to make my own and show all the script
  kiddies at LL how a single person (who is not even a genius),
  working only during part of his free time, can do a better job
  than what they are doing.
- giving up everything SL related and moving on to other, more
  gratifying things (Nicholaz is showing the way right now...
  saddly !).

I already gave up submitting bug reports to the JIRA as this proved
to be a total loss of time and energy.

> OK, so with both the communicate window and dazzle there are a number
> of "i hate it please switch back" comments, but there were and are a
> number of comments which quite clearly set out some of the real
> issues, and even suggest solutions, and yet are totally ignored.

Rule #1: LL is always right regarding SL development.
Rule #2: when LL is wrong, refer to rule #1.
Rule #3: LL never commits mistakes.
Rule #4: when a mistake happens, it's the ISP's fault, the Greemlins,
         or because of the full moon (the gradient in the gravitation
         changes tends to influence the database transactions...).

But let's get serious again...

> There is another similarity between Dazzle and the Communicate window
> - whoever is behind the design of these seems to forget the raison
> d'etre of SL. The focus of attention in SL should be on the 3D world,
> not on the UI.

Matthew... I think I'm falling in love with you... Err.. OK, I said
"serious", didn't I ?...

Yes, of course you are right. But LL script kiddies just _love_
playmobil-like UIs. They want the residents to drool at how pretty
the UI is. They don't care the least about how _useful_, _practical_
or _accessible_ is their UI: all what counts is that it looks good
damnit !

> Both the communicate and dazzle distract the attention from the 3D
> world onto the UI. This may hurt the egos of the UI team but a
> successful UI for an SL client should be one which the user
> doesn't notice, and just uses!

Matthew... would you marry me ?

> If LL had actually bothered to listen to the feedback on the
> firstlook client from users they would have noticed two major
> points:
>  
> i) that what was not needed was a new colour theme but more some
> level of easy (i.e. not hacking XML files) control over the
> colours. This is captured in jira as https://jira.secondlife.com/
> browse/VWR-5059 namely the ability to switch skins from within the
> client, and include the current colourisation as one of the
> default skins included with the client

Frankly, I don't think more than 5% of the residents really care
about UI skinning for the viewer. If you would let them the choice
in a poll between getting the current viewer bugs fixed and moving
to Dazzle, I bet than 95% of the votes would go to the bug fixes.

This is simple logic... But LL is not a cartesian company.

Rule #5: LL's logic is flawed, but flawed logic rocks !

> ii) that the new scheme far from being "more accessible and
> pleasing", actually causes eye strain. This is also captured
> in jira, had anyone from the UI team bothered to check, as
> https://jira.secondlife.com/browse/VWR-5080 , and in there one
> of the comments provides medical and documentary evidence why
> the scheme chosen by LL is a potential health risk and
> accessibility problem! 

Frankly, I am curious to see any "evidence" about the health
issue, but I agree wholeheartedly that beside being distracting
to the eye, Dazzle's UI is definitely going against accessibility
basics.

> Dazzle should not have got into an RC without at least one of
> these *major* issues being addressed. Had both been addressed,
> you might have managed a blog post which generated overwhelming
> postive responses on the blog comments and forums, whereas by
> ignoring both of these,  LL has once again successfully snatched
> disaster from the jaws of victory!
>  
> Matthew

Amen !

Henri.