iCloud: The Anti-Social Network
Lately I’ve been busy writing about the latest incarnation of Pages (see “Experimenting with “Take Control of Pages” Pre-book on Leanpub,” 3 February 2014). One of the defining characteristics I noticed early on about the new Pages is its iCloud-iness: with the newest Pages you can, via iCloud, shuttle your Pages document from your Mac to your iPhone to your iPad to Internet Explorer on Windows PCs in your office and then back to your Mac, and your document won’t lose a single drop cap or footnote. Sure, in order to do this Apple had to rip out a ton of features from Pages and redesign its document file structure, but Apple thought that was a good trade-off — iCloud
compatibility is that important to Apple.
This should not come as a surprise: two years ago, Apple CEO Tim Cook said that iCloud was “a strategy for the next decade.” In fact, however, it’s a strategy that has been percolating within Apple for almost two decades. What’s more, iCloud is a strategy that almost quintessentially reflects the personality of Steve Jobs, embodying a vision that he held and cherished since even before he returned to Apple.
Jobs delivered one of the earliest descriptions of iCloud (though it was many years before it would be called that) at Apple’s 1997 Worldwide Developers Conference during a ninety-minute Q&A session with developers. About fourteen minutes into that session, in response to a developer’s question, he said the following:
Let me describe the world I live in [at NeXT]: about eight years ago, we had high-speed networking connected to our now-obsolete NeXT hardware running NeXTSTEP at the time, and because we were using NFS, we were able to take all of our personal data, our home directories we call them, off of our local machines and put them on a server. And the software made that completely transparent, and, because the server had a lot of RAM on it, in some cases it was actually faster to get stuff from the server than it was to get stuff off of your local hard disk because in some cases it would be cached in the RAM of the server if it was in popular use. But what was really remarkable was that the organization could hire a professional person to
back up that server every night, and could afford to spend a little bit more on that server, so maybe it had redundant disk drives, redundant power supplies. And you know, in the last seven years, you know how many times I have lost any personal data? Zero. Do you know how many times I have backed up my computer? Zero.I have computers at Apple, at NeXT, at Pixar, and at home. I walk up to any of them and log in as myself, it goes over the network, finds my home directory on the server, and I’ve got my stuff wherever I am. And none of that is on a local hard disk.
[…]
But managing a network like this is a pain in the butt. Setting it up, getting it all to work is really complicated. One of my hopes is that Apple can do for this new type of network — not so new, but to the average person it’s new — with gigabit Ethernet technologies and some of the new server stuff that’s coming down the pike, and some thin, thinner hardware clients — hardware clients that are thinner, not necessarily software — that Apple could make that as plug-and-play for mere mortals as it made the user experience over a decade ago.
Replace “gigabit Ethernet technologies” with a wireless connection, a “thin client” with, say, a MacBook or iPad Air, and you’re looking at a description of iCloud: a service that allows you to get access to your data wherever you are. (Ironically, it’s also a perfect description of Google’s Chrome OS.) And that’s “you” as a singular pronoun: Jobs’s vision is person-centric, not team-centric. (Elsewhere in that session Jobs talks about using the network to be “in touch” with colleagues, but the vision of having his stuff wherever he happens to be is what really excited him that day in 1997.)
The names of the various services that Apple instituted over the years to realize Jobs’s vision are telling: iTools, iDisk, Mobile Me, iCloud. Not usTools, usDisk, MobileWe, or usCloud.
Related Apple services and their limitations mirror the me-ness of iCloud. Your iTunes account is associated with a single Apple ID, and makes no real provision for families or even for transferring ownership. If a couple breaks up, or a child goes away to school, the media obtained from one iTunes account can’t be divvied up the way books and videotapes and DVDs and Blu-rays and CDs and LPs can. It’s one owner, now and forever.
Some people look at the Internet and wax poetic over its potential for enabling collaboration and community. (And many of us here at TidBITS resemble that remark!) But the network vision that Jobs carried with him from NeXT to Apple was all about empowering the individual: community was, and is, an afterthought at best.
It’s an afterthought that usually costs extra. For example, your iCloud Photo Stream makes 1,000 of your photos free for you to access from any of your devices, no matter how big they are. Your shared Photo Streams? Those count against your iCloud storage allocation, and only the first 5 GB of that allocation is free (your choices: use the 5 GB for backing up your iPad, or use it for sharing photos with others — for the prudent user, it’s Hobson’s choice).
Even when community or collaboration doesn’t cost extra, what iCloud offers tends to be markedly weak tea with no scones. Take the new Pages: it does offer free real-time collaboration among users, even among non-Apple users — but only via the Web version of Pages, an app that provides just the barest subset of the capabilities of the Mac Pages app, or even the iOS Pages app.
In iCloud, getting to your stuff so you can play with it anywhere is central; being able to share your stuff and let others play with it is an add-on, if it’s possible at all. And when Apple has in the past created such features for iTools or MobileMe, those features have generally suffered in comparison with the competition, and have been kicked ruthlessly to the curb in major transitions.
Granted, building a robust and reliable infrastructure for real-time data sharing and real-time collaboration (as Google has done) is a much harder engineering problem than building such an infrastructure for storing and syncing a single person’s data among several devices. Apple sells personal computing devices to individuals; for it to devote resources to solving the collaboration problem would not give it a big return on investment. Solving the storage and syncing problem, though, means being able to tell each customer, “You can have your tunes, and your docs, and your books, and your apps, and your mail, and your contacts, and your bookmarks available on
every single one of your devices” — that sells itself!
So don’t look to iCloud as it evolves for best of breed collaborative software, for media sharing flexibility, for services that foster and promote community. Apple has always been a personal computer company; group hugs are not in its DNA.
Steve Jobs was famously self-centered. It seems obvious in retrospect that his vision of networked computing, carried back to Apple from his years in the wilderness, would be similarly self-centered. Steve may be gone, but his singular vision lives on: it’s Apple’s strategy for the next decade.
It's not that Steve is I-centric. It is because it is VERY difficult to implement consistently collaborative algorithms in a stateless and real-time network. I believe that you have a professor in Ithaca who is renowned for his work on distributed algorithms and networks.
Just a faint example: your wife asks you to do a web banking transfer to pay an invoice. You say yes and then you are interrupted by a neighbor to move your car and leave the house. Your wife takes over and has to find out whether you did the transaction or not. Humans do that all the time, but try to have this done by machines over a network.
And, yet, that is what Google is doing with Google Drive and Google Plus, and Microsoft is working toward in SkyDrive. The challenges are, as you say, very difficult, but not impossible to overcome.
For Apple, collaboration and community are not, as Steve Jobs would say, "the high-order bit"—ubiquity of *personal* data is much more important in their network vision.
Yes, not to mention DropBox that is my favourite :-)
Apple took the simple and safe (for them) road, by making iCloud THE reference of data for each personal account. But only for one person, not for a workgroup.
In Apple forums, you can read horror stories of ladies who lost years of appointments on their iPhone, once their husbands/sons/daughters convinced themto adopt iCloud and sync their iPhone to the single "family account".
Even when you are alone to manage your data, when you foul your data, there is no way to restore a previous iCloud version, which makes it a VERY primitive tool in this respect.
On the other hand, the topic is very complex. Try and Google "atomic transaction in distributed operating system". It is a challenging field, even when not considering security aspects to protect the workgroup.
Nice contextual article. Wish there were more like this.
years ago I lost all my passwords syncing thru Mobile ME. since than ,I have lost a years digital photos because an iPhoto update needed to recode themas a result no power on earth will get me to trust anything important to iCloud or iPhoto. First time shame on you, second time shame onme
Great article. This explains so much. I always thought that the Apple people don't have families. Why make things so hard to share?
Very nice article in honor of Steve Jobs but missing one very important fact: The entire Apple Staff, included Tim Cook are unable to comprehend Steve’s thoughts and put it to work properly. Steve has known that and that made him looking obnoxious when he last his patience and came along as inpatient and intolerant. Steve lived on a level people even today cannot comprehend and therefore the results are sometimes disastrous.
Steve may had very precise ideas and concepts how iCloud could and should work but his staff is still not able to put that into the right perspective. That is the reason why others like Google, Dropbox etc. are now the biggest competitors in Cloud services, and they understand Steve’s vision and adapted it to their own concepts and understanding successfully.
Very sad to say but Apple is doomed without Steve or someone on the top who has real understanding of Steve’s ideas and concepts. Tim is a very good manager but a terrible leader for this once so inventive and thoughtful company.
Lets face it, iCloud has so many performance issues, that the current staff can only been seen as a bunch of guys who pretend to know what they are doing but actually running around like chicken without a head.
I am with the "Dropbox" faction! With use of encrypted DMGs for the super-private stuff, and the use of symbolic links for the user library stuff such as keychain and such, I have been happily using Dropbox for everything except iTunes and other Apple apps for two years. And of course, I use MS-Office, not iWork gorp.
I do still use 'SuperDuper' for the weekly clone, but every Mac I own and from anywhere I am, It all works -- and syncing is unbelievably fast. And, like the original post, I have not lost one bit of data -- ever. Of course part of that credit goes to all the other things that make the Mac and Apple world so great: quality hardware, OS, and support. Thank you, Apple, but forget iCloud.
Oh, yes -- I forgot to mention that Dropbox sharing and linking makes everything work across wife and family as well as collaborative associates. Yes, my Dropbox is really big... but worth every penny.
In the closing paragraph I sense an implication that Apple can not – or will not – envisage any meaningful improvement to features for collaboration until after 2024.
If that's what's implied: sorry, I find it unbelievable. Where's the evidence for that decade-long assertion?
In patent grants and elsewhere, there's strong evidence to suggest that Apple is well prepared to deliver innovations in the iCloud area and beyond.
I can't imagine that 2014 is the beginning of a ten-year period of stagnation …
In my post above -- yesterday -- I was so determined to make a strong statement about Dropbox that I neglected to tie my comment to Cohen's original article.
I am well-schooled in cloud tech, but not a "world-class expert" -- and still I am convinced that Apple could make iCloud into a spectacular service if they really had it as a priority. I am still surprised by the crippled and inept product.
I like my Macs because they do what I want, but it gets more difficult as the "Jobs Plan" for lockdown gets more intense, more complicated to keep hacking.
I have 50GB storage at Box--free. I can do anything I want with it. iCloud isn't so generous, and has a lot of restrictions, especially for those of us with multiple devices and platforms. iCloud doesn't do Windows. Apple doesn't want me to get/acquire my software outside their walls, but I prefer it that way. I can encrypt data and give friends a key to retrieve it in my multiple online services. I don't know if I can do that with iCloud, and don't care.
I hate the 'cloud', starting with the name. I've been using remote servers for over 30 years [imagine transferring video on a 1200 baud modem, overnight], and the best thing about it is that when it's [often] unavailable, I get free 'break time'.
I think you are pretty far off in your comment about iTunes and AppleIDs. It is the media companies that have dictated to Apple that they can't allow transfer of ownership to ensure they don't lose out on another potential sale of the song, etc. I can imagine Apple had to accept this to get these companies to finally agree to the concept of iTunes and to get the cost down to where it would be viable. If unrestricted and transferrable, Apple would probably have had to pay the record companies $10/song since they would assume fewer additional sales would occur in the future... And if so, we'd all still be without a legal means of acquiring our media because iTunes would have immediately failed. So better that we have to make some redundant purchases at affordable prices in those situations than the alternative.
Interesting article, but the criticism of shared Photo Streams is unfair.
> Your shared Photo Streams? Those count against your iCloud storage allocation
Actually, they don't count against your iCloud quota. See “Do Shared Photo Streams use my iCloud storage?” in the Shared Photo Stream FAQs:
Interesting article, but the criticism of shared Photo Streams is unfair.
> Your shared Photo Streams? Those count against your iCloud storage allocation
Actually, they don't count against your iCloud quota. See “Do Shared Photo Streams use my iCloud storage?” in the Shared Photo Stream FAQs:
http://support.apple.com/kb/HT5903