Apple doesn't believe in the ecosystem anymore?
I'm not sure the current Apple management "gets" the idea of an ecosystem. They've worked out that they can get a lot of orders this quarter by releasing a thinner, lighter, less functional version of last year's product... and if that drives away loyal customers then, so what, they can raise the prices to keep the revenue up.
Apple's past designs (allowing them the occasional hockey-puck mouse lapse) were great examples of balancing form with function - and when they decided to dump a legacy feature, though it never pleased everybody, they were usually on the money*.
Now, we get, whammo, bye-bye Magsafe, bye-bye HDMI, bye-bye SD, bye-bye USB-A, bye-bye SPDIF
far before these things can reasonably called "legacy", pretty obviously just to meet the Prime Directive of "smaller is always better"... and in return we get, well, a slightly improved display***, this year's version of last year's CPU and GPU (minor incremental performance improvements that you won't notice without a stopwatch). Oh, yes, there's the incredible potential i/o bandwidth and multi-monitor support of 4xTB3, coupled with even faster SSDs, which will be totally irrelevant to most people leaving, well, exactly the sort of pro-graphics/audio/video users who
also need 32GB RAM and a more bleeding-edge CPU & GPU (and who could possibly cope with finding a mains socket, unless all their other gear is running on sunlight). There's no
balance to this design - its exactly what you get if someone designs the box first then the engineers struggle to fit in what features they can.
Not to worry, though - if enough people will pay $3000 for a handbag to keep fashion houses in business, they'll pay $3000 for an ultrabook with a emoji-bar. That'll work for a few quarters, then they'll wonder why nobody is developing pro graphics apps, decent games or developer tools... but then, hey, all they need to do is knock out an XCode for Windows or Linux so that people can write iPhone apps (the Swift compiler is already done) and they can walk away from the Mac. Tim started the last launch by saying how important the Mac was to Apple** - that's the sort of thing that high-flying business types only say when they mean the exact opposite.
I differ, slightly, on things like the networking kit and displays. In terms of networking, the Mac has come a long way in terms of interoperating with third-party kit in the last decade, and most other networking/NAS kit now "just works" with Mac, it doesn't even have to be AFP or HFS+-based (and Apple are actually depreciating AFP). Where I live, most people get a free or heavily-subsidised modem/router thrown in with their home broadband and the best strategy if you want support from your ISP is to stick with that unless you need something more tweakable (in which case you're probably not looking at Apple's routers, either). They've never really been "professional" products (beyond very-small-home office) and they lack what
would have been the Unique Selling Point of acting as an always-on server for your iTunes media.
As for displays... the display market is in a right mess at the moment thanks to Intel, AMD and NVIDIA messing around - Neither Intel's TB3, USB-C implementations or their iGPUs support DisplayPort 1.3 which is really needed for 5k+ and 4k@60Hz displays over single cables. There's also a snarl-up over rival standards for variable-sync-rate displays. Good time not to be making displays. I'll also be interested to see what the practicalities of "docking" anything more than a mouse and keyboard via a 5k display are: In theory, there's 10Gbps of free bandwidth left out of 40Gbps... but I've yet to see an interface that can deliver anything like its full theoretical bandwidth under a "mixed load".
What the mythical "pro" user really wants, surely, is to be able to choose the size/style/aspect ratio/gamut/features that their application needs from a range of compatible third-party displays.
(What I find interesting, though, is how the LG UltraFine displays show all the signs of being designed by Apple right up to the point where it was time to sign off on the nice machined aluminium case & fused front panel/display/webcam assembly... look at the lack of any other input beyond a single USB-C/TB3, the "only way is USB-C" output ports (when there is plenty of space for a couple of USB-As) - the placement of the ports (more or less where they are on the old TB Display/iMac) and the odd "forehead" (to house the webcam) that wouldn't show up so much - or could be much smaller - if the whole front was glass. C.F. with the 4k displays that LG already make. No proof there, but it is rather suggestive that, at one point, there was going to be an LG-made Apple display.)
...that said, we have to accept that the "maturing" of the Laptop/Desktop PC industry will have consequences. Not long ago, a manufacturer could sell you an "upgradeable" machine safe in the knowledge that 2 years down the line they'd have a new model that was twice as fast, had twice as much memory, twice as much storage, expansion slots with twice the bandwidth etc. to the extent that it made more sense to buy a brand new machine and keep the old one as a spare or hand-me-down. Now, progress is much, much slower and if Apple sold you a "proper" Mac Pro mini-tower with PCIe slots, upgradeable memory, swappable drives they'd be lucky to see you again this decade (unless
all the upgrades were Apple proprietary and we wouldn't like that). They must have been hit quite badly by the number of us who stuck $300 SSDs in our 2011 MBPs and found that we had what felt like a brand new machine. You can see
why Apple might want to increase built-in obsolescence or increase prices - however, doing both at the same time
without significantly increasing functionality is a bit much. The counter to this is that the only reason
I was prepared to lay out $3000 on a MBP 5-6 years ago was the reasonable expectation of getting 5 years out of it.
*E.g. Most people certainly didn't need an ethernet port or an optical drive
on the road by the time the first rMBP came out, and they compensated by adding a second Thunderbolt/DisplayPort
and HDMI - so the TB port was no longer the only display connection - not to mention retina displays and superfast SSDs to sweeten the pill. As I said, not great for everybody but
somebody had put some thought in, beyond "it has to be thinner!"
**Actually, I err - he started by announcing that a 5-year-old sandbox construction game that already runs on everything from a $30 Raspberry Pi upwards is now available for AppleTV. Woohoo!
*** Providing you want a video-optimised gamut rather than Adobe - my experience with wide-gamut displays is that unless they're the
right gamut for what you're doing, and unless you're running a full colour-calibrated workflow, they're just a recipe for that tacky "Kodachrome Gold" look.