Project Open Letter

Project Open Letter was a year old on April 27, and frankly I don't see much (if any) improvement in the way SL issues are being dealt with by the Lindens.

There are things that were going on at that time that seem to have been marginally resolved, but the stuff that's gone away has been replaced by new and, in some cases, more serious issues.

Let's look at some of the issues that were highlighted:

- Inventory loss - this is still frequently reported by residents posting on the official SL forums and elsewhere.

- Problems with Find and Friends List - Friends List outages have become relatively rare, but Find functions being unavailable is still fairly common.

- Grid stability and performance - Teleports failing, attachments failing to attach correctly and persistent greyness/failure of textures to rez are still a daily occurrence.

- Build tool problems - I still see posts from people saying that there are issues with this, though I don't build enough personally to have much of an opinion on it.

- Transaction problems - inventory deliveries still fail often, as do L$ transactions.

All this despite the much-ballyhooed rollout of Havok4. Overall, the state of the grid is still somewhat poorer than a year ago, IMHO.

Comments?

P2

Just that when you think something can't get worse..it can.


I don't know P2. It is hard for me to comment on POL because when Cristiano came up with the idea and began working so hard on it we all thought it would make a huge impact. Other then the couple quick town halls it never really did. I was in world the other day and visited a shop that the owners obviously have not been around for a long time (Old residents too). There was a POL kiosk in there displayed prominently. It was kind of sad to think of all the work Cristiano did and all the hope we put in it just to be sitting here a year later with out marked improvements.

I don't bad mouth LL or SL much at all now. I have come to accept that this is just the way it is going to be. I can be onboard for the fun but the serious side of taking it as a "virtual world" lost it's luster when I began to see more human error then virtual living. :/

OK, back to trying to see the positives in SL over the negatives for me.

The only thing that has been addressed is the physics system, and Havok4. Physics crashes have taken a dive. But the problem is the asset server. 3 of the 4 things on that list all are database issues. The fourth is a problem with version control and the thing getting so big that no one knows what exactly does what.

Inventory loss - couldn't agree with you more. Just guessing, but I think we are still maybe a year away from a more robust asset serving system. Right now I think they are in triage mode. How to speed things up? Not sure I would divert the course - seems we are stuck on an old architecture until a deep, difficult, complete transformation is done. Expect a bumpy ride - it's like changing out an engine while driving.


Friend / Find / Search - I suspect a significant failure mode is due to strain on the system overall. If so, some creative throttling may make some difference - perhaps throttle anything hammering the database the way a bot might. Another failure mode: third party subcontracted services failing. Some things may need to roll downhill, here.

Grid stability/performance: most of this (April for instance) was due to failures of third party subcontracted services. I hope the new CEO does some subcontractor shopping. Fast.

Build tools - this one comes and goes - I think there has been a lot of whack-a-mole bug fixing, and as of right now, the good guys are winning. Stay tuned though.

Transaction problems - worse than ever, is my sense of it. Deep concern here, though I personally don't know how this works. I fear it has to do with the asset servers timing out. Meaning it may kinda be around until the asset serving is revamped.

Havok4 - brilliant rollout that while still a bit flawed, did wonders for region performance. Crashes and other nonsense demonstrably reduced. But all of that can't be enjoyed unless you can actually rez anything to interact with (asset serving). Far better than a poke in the eye with a sharp stick, though! The Havok4 rollout is 'how it should be done' by a class-act professional team, in my view.

* * * * *

Project Open Wallet.

Project Open Wallet thoroughly trounced Project Open Letter, providing something like 70 million reason$ for 'business as usual' in 2007 - thoroughly pwning some thousands of signatures.

Strangely, some of the very signatories to Project Open Letter lent even more support to Project Open Wallet, giving 295 vote$ of confidence monthly as compared to one paltry signature. Just what was Company to think?

The intent of Project Open Letter was quite sensible - but predictable, when up against the hurricane-force gales of cash blowing the Company's way. Wouldn't surprise me if 2009 dawned on a 100 million dollar Company. It's 2010 - 2012 that will matter, I think, for the wind can suddenly blow a different way. However, before it does, we may end up with a MicroSoft.


* * * * *

I've heard a lot of the 'better faster cheaper - can't have them all' arguments, and there's some truth to that.

But there's also truth in that a good CEO can turn "this or that" into "this and that" - it's the only reason you pay someone the big bucks, because ostensibly they don't really *do* anything else than turn or's into and's. Because just anybody could give you or's.

Right now it's the 'holiday' period for the new CEO guy. It may take 6 - 12 months for stuff in the pipeline that can't be changed to flush through - but slowly, incrementally, we'll see if he's got the right stuff.