Interesting conversation with a friend recently – it was one of those occasions when the sl experience was rather less satisfying than we would have liked it. A number of other friends had been unable to log in, we’d experienced problems with rezzing and bake-fail and, inevitably, the conversation turned to the shortcomings of sl.
You know the sort of thing… general whining and whinging about lag and such like, which led to criticism of viewers, the Lindens and the internet in general. Yet, despite our negativity, we were both in agreement that, although on one hand we could easily describe ourselves as sl’s greatest critics, on the other hand we’d be its staunchest defenders. Neither of us is about to leave, and despite all its faults, we’ll stick with it until the bitter end.
To be absolutely fair, when compared with the competition, sl doesn’t do terribly well – then again, it’s very much a victim of its own success and, unfortunately for us, it has a rather unique position which works both in our favour and against us. It’s very much a case of ‘damned if we do; damned if we don’t’ – the very nature of sl and the niche it operates in keeps it where it is – there’s no doubt at all that sl could do much better… it could be faster, with more efficient scripting, less bugs and could compete far more ably with other popular online pursuits but, as much as we like to think that’s what we want, the bottom line is that, in reality, that’s the last thing that we’d really want to happen.
The trouble with high performance, low lag and top-notch responsiveness is that it’s really incredibly demanding on hardware, and this where we hit our first hurdle. Linden Lab could easily build a viewer with blistering performance that utilised the very latest attributes of graphics cards and modern motherboards; and there will be those for whom that is the Holy Grail. “Please do it!”, they plead – but, for every one of those fortunate people with their 100GB optical internet, 8-core fluid-cooled super-processors, with 500 Jigglebytes of RAMBO and triple crossfire aquaflow Diamond graphic cards, (does it sound like i know what i’m on about?), there will be a thousand users plodding along with their Argos netbooks and a dongle connection, with a wire coathanger plugged into it to improve reception. Then there’s all the poor souls trying to run sl on an old PC they picked up from work when they upgraded so that spreadsheets wouldn’t crash the machines – these guys are still using V1.23 because that’s about the only thing that won’t cause their system to grind to a complete halt. Yes, the Lab could give us Super-SL™, but they’re not going to because to do so would mean that there’d only be 106 people logged into sl on a good day and 99.999% just wouldn’t be able to log in at all without fatally crashing and frying their graphics cards.
So much for client-side, what about server-side improvements? It’s another double-edged sword i’m afraid -bump up performance on the servers and you get bottlenecks at the receiving end: it’s a bit like the entire capacity of a rugby* stadium descending on the bar at half-time, with only one person behind the bar to serve them. (*English people please substitute ‘soccer’; Americans, ‘baseball’; Ancient Romans, ‘chariot race’).
OK, so we don’t want to overwhelm residents by smartening up performance, how about tightening up LSL, improving rendering, making cars run better across Sim borders and all that really annoying stuff that happens inworld that we have so much fun moaning about? Once again, sl becomes a victim of its own success and its rather unique take on the virtual world… ‘all content is resident-created’ has a sting in the tail!
A lot of resident created stuff is top-notch – cleanly coded, well put-together and with efficient physics and optimal integration with sl. However, an awful lot of resident created stuff is borky, inefficient, weird, wonderful, thrown-together, amateurish and barely works at all at the best of times – a lot of it is very old, in computerish terms, too… yes, there’s some content in sl that was created back in 2002 – those aren’t criticisms, by the way, it’s just the way things are, and it’s one of the best things about sl… anyone, with any level of skill can have a bash at doing their own thing, (a bit like blogging!). Here’s the catch: upgrade, modernise or improve how sl handles all these amazing creations and overnight you end up with a world full of things that no longer work, don’t look as they were intended or just go plain haywire – you’ll also get a lot of very peeved people. This doesn’t happen with slick games like Samurai Mushrooms of Delta 3 because all the content is created by gnomes in white coats to strict guidelines, using proprietary code that is designed to be readily updateable; the only creativity that participants get to explore with these virtual worlds is what to call their character and what flavour crisps should they have for in-game snacking.
This is the monster that we, in companionship with Linden Lab, have created. It can certainly be better – much better – but at the expense of leaving the vast majority of its stalwart supporters far behind, or requiring every one of us to accept that everything we ever create, buy or interact with is temporary and will only last as long as it takes until the next server update. i really don’t think that’s a price any of us would be willing to pay. So, that being the case, we’re stuck with sl as it is – warts and all – and perhaps that’s something we should be grateful for?
You’re perfectly flawed
You’re perfectly incomplete
Like cracks in the glass
And faded photographs
Otep – Perfectly Flawed