On 1, I disagree. It still only one bus, and having more ports on the same bus dilutes throughput, especially with FCP 5.

On 2, we agree, but Express/34 is a better protocol. As cards proliferate, it will be better in my view.

On 3, we disagree. I want a slower burner, not a faster one. I struggle to burn at 2x because so often it affects compatibility with many set tops.

But as you say, don't fix it if it ain't broke

USB2 vs Firewire 400
 

Just to clear up the USB2 vs Firewire 400 debate...

USB2 is 480mbps and Firewire 400 is, well 400 mbps. USB2 is slower in performance, however, because of the protocol. USB2 requires that data packets for each bit be sent in both directions, that means roughly twice the amount of info is being passed through at any given time, thus making USB2's speed roughly half the total 480mbps bandwidth it has to play with.

On most devices, this is not noticable because rarely does any single device take up the full bandwidth of either. Hard drives, however, can take up more than 240mbps easily (half the usb2 bandwidth) thus taking a performance hit when compared to Firewire 400.

I'm not anti-Apple in any way (at least I don't care for them any less than Intel or Microsoft or Starbucks...), but this is the kind of disinformation that really does not help anyone and serves no real purpose. It probably does more harm in the long run. I've never read that book and from what you have quoted, I wouldn't want to.

If you want the complete history of Firewire and the evolution of the standard, why not just head on over to standards.ieee.org and take a look for yourself? You might find that it's nowhere near as cut and dry or simple as that book makes it seem and no, Steve Jobs isn't God and no, Apple didn't invent Firewire. Apple, along with a handful of other like-minded companies, coordinated with the IEEE to form a standards committee -- other key members included Intel, Sony, Philips, Hitachi, etc.. Apple has never even bothered to explore or implement many of the IEEE-1394 variations such as 1394C (Firewire over UTP cable) or multichannel implementations, etc.. Many of these have found their way into industrial applications.

If you still don't want to believe me, that's fine. There are thousands of people out there that believe Apple invented Firewire or that Apple invented the graphical user interface with application windows and icons (sorry, that was Xerox)...

USB is a full-duplex protocol and data can be sent upstream simultaneously with downstream data. I'm not going to argue that I haven't seen a performance hit with USB2 devices, because I have. I have also used some USB2 devices that perform better in USB2 mode than they do with firewire or vs. the firewire offering from the same manufacturer. It all comes down to implementation. The truth of the matter is that USB2 has superior bandwidth and the capability to offer lower latency and better data/bandwidth management than Firewire. The sad part of all this is that for some reason many manufacturers don't exploit these benefits and they slack off on their USB2 implementation. Equal blame (and perhaps more) can be assigned to chipset manufacturers and mainboard vendors - on both the PC and Mac side of the fence. So many use only 1 or 2 USB2 controller channels on their mainboards, yet they drive 4 to 8 root hubs from these. Seems like a good idea to add more USB2 ports to a system, right? Well, if you have 4 ports on your system via 4 root hubs tied to a single controller channel, then all 4 of those root hubs/ports have to share that 480mbps bandwidth. And while devices sitting on the bus (things like software keys, memory readers, printers, scanners, etc..), even when sitting idle are eating up data packets in the form of a status pulse.

...Very, very few hard drives on the market can come close to 240mbps (60MBytes/sec) bandwidth you reference. In fact, it usually takes two HDDs in a stripe set or similar arrangement to sustain a rate like that. If you're seeing a performance fall-off on an external HDD going from Firewire to USB2 then either your USB2 bus is heavily saturated to begin with or the system and/or the drive has a poorly implemented USB2 interface.

240mbps is only 30MB/sec, which hard drives frequently surpass. It even affects the rate at which your songs get uploaded to your iPod. When I got a new video iPod in which I was forced to use USB2 to sync to my powerbook, I noticed and immediate and significant data transfer rate hit.

Heh. Ooops... I guess I should wait until I'm not so tired before reading/posting. 30MB/sec is correct, sir. :)

More Info on Express/34
 

http://www.hardmac.com/news/2006-01-13/#4996

Note the comment about FW800 adapters.

A FW800 adapter is a no-brainer in this case and I would be surprised if it's not available by the time the MacBook officially starts shipping or very soon after. I'm seriously confused as to why Apple didn't use the Express54 slot though. But an Express34 to external PCMCIA adapter should be an option at some point, although if card makers rush to release E34 versions of their cards, the need for the adapter would be minimal. It's probably more likely to see an E34 to external E54 adapter... Many peripherals like some of the current audio cards are already cramped in a standard PC card form factor and even protrude from systems a bit. It's going to be difficult and/or expensive for manufacturers to cram all that stuff into half the space they have it in now.

I think it's cause the Macbooks are an inch thick.

Yeah, I'm sure it's an issue of space. They only needed another 2cm of width though.

It's quite simple, the debate between USB2 and FW400. Video conference with a camera using both protocols. It's usually not even close. USB2 has gotten a bad rap, but better bandwidth than FW400?