empty slots, option spaces, and the cowboy myth

What percentage of the expansion slots on PCs lie empty? 90%, 98%? That’s a lot of empty real estate! Who’s paying for that and why? Think of what all that empty real estate costs to create!

Slots and connectors create a set of options for the both the buyer and the market to play with. But why create those options if they are mostly not used? First let me work thru one explanation that I don’t find sufficient.

Options that don’t get used can be extremely valuable. For example, phones are almost never used call the fire department, but the option to make that call is still exceptionally highly valued. So, part of the answer is about risk. The empty slots lower the buyers’ perception of their risk. The slots as creating the option to solve problems; problems the buyer can not even name. That suggests that slots are more signal value and use value.

I don’t think that’s a strong enough value proposition to explain all this empty real estate. So let’s take another try; this time at the market dynamics rather than the individual buyer value.

Slots create options for the entire marketplace. In a sense they create the possibility that the market repair vendor errors. The slots a sign that the vendor is humble – i.e. willing to admit he can not do it all. But humble isn’t exactly right; the vendor is also signaling that he is generous. The slots say something about the abundance of value he is creating.

When the vendor creates a slot he’s relinquishing control over some amount of value creating energy implicit in his offering. At first glance that seems like an odd move on his part. Giving away the option to create value? Certainly the rational actor would horde those options. Presumably it is not just his humble nature that triggers this giving away.

That starts to get us closer to a plausible answer to the question of who’s paying for all this empty real estate and why?

Markets are not as simple as buyer v.s. seller! Most markets have a number of complementary players. It maybe we need to look these to find out who ordered the empty slots.

If the OS or CPU vendor is powerful he might force it. Both the OS vendor and the CPU vendor prefer an open hardware bus because it encourages a more commodity market in their complements. I suspect this is the key to puzzle. Microsoft/Intel ecology has so many empty slots. Other platforms have fewer. It’s no surprise then that the Apple keeps trying out product offerings with few or no slots.

In this model those slots are paid for by the hardware vendors to create a space of options who’s exercising creates value for the dominate market players, i.e. Microsoft and Intel. The landscape of empty slots is a substrate for a generative process, a process that adds to the network effect around their business model. Of course they are both so large that their business model is the industry’s business model.

Why do Macs have slots? I suspect that’s because the market expects it. That Apple garners little benefit from the slots, and it’s users don’t get much either. But it would be interesting to see statistics about how much of the slot landscape is filled in on both ecology. Of course Apple, like Linux does, benefits from the hardware vitality created in the Microsoft/Intel ecology so it’s a small price to pay.

The underlying idea in here is that the space of empty slots is like real estate. In an economic universe that implies that it serves an economic function. Typically that means somebody owns it. Somebody is maintaining it. Somebody it taking a tax off. Somebody benefits.

In the American story the huge swaths of unowned real estate that was left behind after European plagues wiped out the native american populations provided a very different dynamic from that seen in old Europe. All that land enabled made labor extremely valuable. The combination of ample options to move on, and high wages gave labor vastly increased freedom to experiment.

The same pattern exists on the landscape of empty slots. The space of empty slots lowers the barrier to experimenting. That output of that experimentation is valuable to those who create the landscape of empty slots. I.e. Microsoft and Intel.

Notice then what happens if the percentage slots begin to be filled in. If there is anything consistent about how they get filled in; for example they all start sporting graphic’s cards – that’s not stable. The industry will absorb any consistent slot usage into the core.

So then, is the space of empty slots a long tail? I.e. is their a generative process here who’s distribution is power-law which creates the occasional elite member, (in this case we could consider the graphic accelerator card as an example of a elite member). For this generative process to work it draws upon the power-law’s long tail of activity. In this case the long tail of activity actually has negative value. Since the majority of buyers don’t use the slots they are getting no value from them. The hardware makers are including them not because of buyer demand but because the dominate players in the market want them. The dominate players in the market want them because they tap into the the value created by the generative process that all that empty real estate creates.

Of course Moore’s law and his friends are all about creating huge swaths of empty real estate. So empty slots on motherboards are just one example.

0 thoughts on “empty slots, option spaces, and the cowboy myth

  1. David Coletta

    I think you’re right that slots are rarely used. But don’t forget that in the case of the Mac Mini and other small-form-factor machines, the slots have become virtual, in the form of USB and FireWire ports. At this point, both of these ports are really fast enough to function the way that slots do for the purposes of your argument.

Leave a Reply

Your email address will not be published. Required fields are marked *