• About us…

  • The archives

  • RSS The Gaming Session

  •  Better and faster with IPv6

  • ipv6 ready

As described on the tin

Max Kleiber dropped a comment just a few minutes ago that really gave me pause for some thought.

”[I]t would make marketing’s life a lot easier if they were promoting a product that works as intended all the time, every time, with no surprises except the ones people are looking for,” he said.

And it occurred to me that – while that’s very obviously the sort of ideal we hope for in a product – how infrequently it is that products actually live up to that.

For example, I’ve got a Pioneer DV-410V DVD player here. It’s got a bundle of nice little features, including a USB port, that you can use to play media from a USB stick or external hard drive. Mind you, swapping between disk-mode and USB mode crashes the firmware 9 times out of ten (you’ve then got to pull the power-cord out to reset it). As a bonus, whichever kind of media you’re playing slowly goes out of synch, until the audio and the video have about a second or so between them after 90 minutes.

Likewise, Optus have provided me with a little set-top box. Setting reminders for programmes from the electronic guide frequently crashes and resets the unit. When the electronic programme guide actually loads properly at all.

The list goes on; from Microsoft Office to boxes of matches, how much of the stuff we buy really performs exactly the way it is supposed to?

Seems like not very much of it. It’s frustrating, yes.

I put up with the DVD player, because what’s my alternative exactly? Pay extra for another model with a whole different set of quirks? I’ve at least learned to work around the current one’s problems to some degree. There’s no firmware update. The set-top box? It’s not really like there’s an array of choices.

Clearly, though, there’s quite a spectrum extending between what some might call “a piece of shite” and something that works exactly as described on the tin.

My preference, of course is that more things were closer to the top of that spectrum, and I suspect strongly that yours is also.

In a roundabout way, that brings me to Second Life, which was the original topic on which Max made his comment.

Second Life certainly doesn’t seem to be very high up there.

Now it is sort of understandable. Linden Lab back in the beginning had a largely incomplete product, and certainly couldn’t afford to spring for the QA resources to really stabilise the product properly. Instead, the Lab raced towards completion instead, leaving a lot of inadequately designed and tested code in its wake.

Obviously the Lab is paying for that now, as are we all.

There’s pieces of software there that have never been looked at since the day they were written, and that nobody really still knows how they were supposed to have worked, since nobody apparently wrote any documentation for them so that what they do could be checked against what they were supposed to do and how they were supposed to do it.

Those parts can’t be said to do what it says on the tin, because the tin is unlabelled.

Now, I’m the first to admit that I’ve actually got some serious misgivings about the new Scrum model that the Lab is using. Why? Because it feels a whole lot like the development model that the Lab started with, back in the day.

A lot of grabbing at the low-hanging fruit and less thinking about things that might have to stretch across a half-dozen or more sprint-cycles to get to any functional stage.

That said, I’m impressed somewhat with how the Snowstorm team has been doing. While my misgivings are still strong, the Snowstorm team seems to be showing some impressive performances. Things are happening, and they appear to be happening swiftly and efficiently.

I don’t know yet if that’s going to be enough – the mire is a deep one to try to climb out of – but maybe this could work.

I know, that’s hardly a ringing endorsement, but I am genuinely impressed by what is being achieved. I’m not sure if that level of achievement is sufficient.

Perhaps my most serious misgiving is a lack of visible goals. Lots of fruit-picking going on, but no apparent destination for it, and if there’s one thing that worries me about a software project, it’s when there’s no clearly defined goals – and I’m just not seeing them.

Got a news tip or a press-release? Send it to news@taterunino.net.
Read previous post: