To me, this is what it's really designed for.
It may not be. I'm somewhat assuming will be able to run multiple SATA channels over wire. If only get one of each protocol it multiplexes then it will
not be a good candidate to increase direct attached storage bandwidth.
It will be simpler to implement something that just does one of each handled protocol rather than something that does multiple. If they are trying to minimize controller costs you may not get this.
In that context the real purpose is more aligned with addressing "docking port" / "Apple Display Connnector" problem. ( reducing connectivity to just one "wire" ). The multiprotocol handling feature primarily attacks that issue.
If Apple can bring ADC back in a industry standard I doubt their adoption rate will be slow. Likewise, it is goofy that docking ports for laptops isn't a industry standard.
The direct attached storage application is likely more secondary to those.
But if designers toss other ports not included in the chipset and go solely with LP, it might (add up the cost of added FW, SATA/eSATA, and additional USB), I should think so.
You can't toss out ports when there is a HUGE number of legacy devices that folks have to connect. Additonally, I think it is very unlikely that Intel will support FW over LP. Intel is out to kill FW. They have been for a long time. Including FW will only prolog its lifetime. Even if Apple cajoled them into it, would more likely be FW400.
Video , PCI , SATA , USB seema more likely combo. Firwire/eSATA can sit on the remote PCI connectivity without them having to muck with the protocol.
Users aren't going to want a new box to plug into but certainly adapt more quickly to an old external box that had multiple wires going back to the computer: docking station , monitor (with sound and USB ) , and to much less volume multi connector storage.
All of the old stuff still plugs into legacy connectors. Just have faster and fewer "wires" going back to the computer.
But users will want existing port tech as to avoid bridge devices necessary to function with existing devices they already own IMO. So for the most part, the change-over would occur over time rather than all at once (save perhaps Apple).
The issue in some part though is that LP isn't really a replacement, but an aggregator. It isn't replacing; it is primarily consolidating. On the computer the interacting is still with the old protocols so you need something to "talk" that.
For example international phone traffic is consolidated onto single fibers. That doesn't mean that the individual phones on each side of the connection disappear. You still have standard POTS wiring in the homes.
Granted there is "distant future" talk as though will only need Light Peak. However, that is about as practical as will have direct connect to fiber phones in the home. I think that talk is just to help drum up support (get folks yelping about Leap Peak to computer and peripheral vendors). It is not where the large value add comes from.
That's my point. Most users won't be able to tell the difference,
If you label one as a 250Mbs wire and one 10000Mbs wire. I think the value proposition is quite obvious to even naive users. Folks pay more for Cat 6 than Cat 4 cable. This one is several orders magnitude bigger. Bigger ==> higher cost .... not that huge of a leap. Plastic / glass isn't as strong as a differentiator as that.