waffle

Waffle is a weblog.
The author of Waffle, some guy in Sweden, also occasionally writes stmts.net.

Redefinition

A bit over three years ago, I saw the future. The iPhone was the future of smartphones. Nearly everything else was, according to its manufacturers, already established, gloriously superior in every way, shape or form (number of styluses: WinMo: 1; iPhone: 0; WINDOWS MOBILE CLEARLY WINS) and would outwit the iPhone in a heartbeat in every metric that mattered. These apparent champions are now mysteriously replaced by their full-screen multitouch successors. As a smartphone, iPhone has succeeded. However, I now find myself waiting for the future of the future.

Steve Jobs compares the currently sprouting wave of “post-PC devices” to cars, and PCs such as they exist today to trucks. For most things most people do, you don’t need a truck, and it’s a lot of dead weight to haul around with you. I can see his point to a certain extent, but right now, the argument doesn’t hold.

You could argue that what, say, the iPad provides is pretty much the useful parts of a computer, and you’d be right for many definitions of “useful”. But iOS isn’t well suited to take on every computer task. Although the subject is often used as a shield, this isn’t really about programmer obsessions like direct global file system access per se.

Some people will say “real people don’t care about the file system”. Well, okay, sure. Real people still group materials together from multiple applications into one folder for whatever project it is they’re working on. Real people still aren’t satisfied with a row of chronologically ordered, scrollable thumbnails where the only recourse for secrecy is removal, and the only recourse for reordering is frail and time-consuming and involves re-saving everything.

Real people still don’t know quite how to sync over documents to their iPad apps, nor how to sync their iPads at all or why they would, nor what those Apple people were smoking when they invented this scheme. Real people do use Dropbox to work around these problems — sometimes. Sporadically. For certain things. This is audacious on the level of asking people to connect a garlic press before they save — you probably don’t understand what it does for you and you just don’t realize why you’d need something else.

This has turned into a debate of whether the iPad is about content creation or not, or, passive-aggressively, whether we should listen to the programmers that got us into this mess, posited as a false choice between the new failure or the previous failure, hoping that pointing out which platform has Angry Birds will make every problem go away. That’s focusing on everyone’s hot buttons instead of the problem at hand.

The iPad may not need a visible, all-dictating file system as we know it, but it damn well needs a filing system. This “post-PC device” depends on a PC, or on nasty workarounds like emailing or cloud services, to do what it’s supposedly replacing. (Unless literally all you do is read mail and browse. I’m pretty sure most of those people would like to write a document and file it away every once in a while too.) I get that this might be a lot to solve without repeating the failures of past systems, but it’s badly needed. If this isn’t addressed in iOS 5, one wonders what the priorities are in Cupertino.

That’s one example. Usually, this argument can be made for other aspects. “Real people don’t care about concurrently running programs”. Nope, but they care about multi-tasking and moving information around, which is why the multitasking bar and copy-paste were such welcome features and why even confused extended family members of mine use them even when they call and ask for help about doing the same thing in Windows.

“Real people don’t care about Bluetooth profiles”. No, but they care about being able to send things to their computers, or to other phones — they don’t sit up all night and think about it, but they rightfully get mad when it’s not there. “Real people don’t care about lack of hardware access”. No, but if they bought a book app and they can’t turn down the backlight in it, they blame whoever made the app.

I’ve been using iPhone as a lead-in and iPad for the meat of this post. The iPhone is ultimately “still just a phone”. Most of the things it can do, like act as a bubble level or flute, is pure gravy; if your business didn’t buy it for you, you probably carry it to be reached and to mess around in Cut The Rope once in a while. The iPad is positioned directly as something that mostly replaces a laptop and is more powerful than the iPhone. The iPad is simply where the justified criticisms in the same iOS because of positioning really turn inconvenient.

“What do you mean I can’t organize my documents in a uniform way? I might not like exactly how computers work, but that’s what they do for me. It’s why I use them.”

Screw the debate about Flash, 7″ screens or device heft. The best thing Apple can do to take it beyond today’s PC is to bring it closer to today’s PC. They already have the innovative parts. The successor to flawed organization isn’t no organization. It’s time to salvage from PCs what still works so well.

Comments

  1. “The best thing Apple can do to take (iOS) beyond today’s PC is to bring it closer to today’s PC. They already have the innovative parts. The successor to flawed organization isn’t no organization. It’s time to salvage from PCs what still works so well.”

    See, I think this gets it wrong.

    I agree with many of your diagnoses, Jesper, but not your treatments.

    In the future, the market will bifurcate.

    iOS will hit the market for folks who really don’t need a file system. It’ll hit the market for folks who really want Cupertino to have the only admin account on the system.

    There’s nothing wrong with that. That particular market is huge.

    But at the same time, there will still be a market for folks who want general purpose computers as we currently understand them. That market will not disappear, even though it will be smaller than the iOS market.

    The best thing Apple can do to take iOS beyond today’s PC is to keep it far away from the design of today’s PC.

    But at the same time, Apple ought to continue to evolve the current PC. The market might be smaller, but making state of the art PC’s would be good for the company in very real, if somewhat intangible ways.

    The best thing Apple can do is to lay out a roadmap for parallel development of iOS and OS X as two very different things. Unfortunately, I don’t think they’re planning on doing that. Unfortunately, I think they want convergent development which will produce devices that geek ludites like me and you will be deeply unhappy with.

    Apple has clearly made the best general purpose computers in the world during the 2001-2010 decade. I have my doubts that will hold true during the 2011-2020 decade.

    By Chucky · 2010.11.18 17:57

  2. I have little doubt that Cupertino spends a lot of time and money trying to figure out how to fix this particular problem. As with most things Apple does, they’ll let us know when they find one they like. Until then, though, we’re SOL.

    By Phil Nelson · 2010.11.18 18:14

  3. Chucky: I get your general angle, but the whole point of this post was that normal people want folders and cross-app storage too. Seriously.

    I reject the notion that people are too stupid to appreciate those things and therefore don’t need them, as well as the related notion that organization has been antiqued. You know that they had file folders before computers, right?

    By Jesper · 2010.11.18 18:25

  4. “You know that they had file folders before computers, right”

    Right. But most people didn’t have file cabinets in their homes.

    And I think we’re going to return to that era.

    In the pre-computer era, I would’ve owned file folders, a file cabinet or two, and if I had some cash lying around, I would’ve owned a sweet IBM Selectric. But that’s because I have different needs and wants than the average consumer.

    Apple has been pretty clear that they see themselves as a CE company above all. And the prototypical CE product is a television. No file folders on a TV. Keep it simple. Apple is genuinely in the game console business going forward.

    In short, I think the median user doesn’t quite want the things you or I might want…

    By Chucky · 2010.11.18 18:50

  5. But most people didn’t have file cabinets in their homes.

    Of the households you know, how many of them do not own a ring binder?

    Apple has been pretty clear that they see themselves as a CE company above all. And the prototypical CE product is a television. No file folders on a TV. Keep it simple.

    Which is why the only view available in Photos is of every single pho- oh wait, that’s right, you can organize them until the cows come home. Same with the iPod app and playlists.

    And somehow, documents, of which you traditionally have many, work with a short time and want to keep, should escape the ability to be archived?

    Don’t “average consumer” me. The reason the average consumer doesn’t do a lot of bookkeeping on their iPads right now is because it’s such a pain in the ass. If they really want it to be able to replace the computer for those users, they’re going to have to solve it. Otherwise, the people with the potential to be most served will need to still own one iPad and one “real computer”.

    By Jesper · 2010.11.18 22:51

  6. great post. i agree with Phil Nelson – these are hard problems and apple aren’t about to crap all over the user experience by making the file system fully accessible (thankfully).

    definitely hoping for significant progress in iOS5 though. or buy/(clone and simplify) dropbox or something…

    By steveg · 2010.11.19 01:21

  7. steveg: I hope they are working on them. I hope the contrived verdict that everyone’s really well off now and that it’s not a real problem isn’t driving their development process.

    If they want to keep playing the security angle, they can’t just put a wrapper on the file system. I believe that there’s the ability to do it better. I also believe that the new solution will necessarily include two concurrently running applications being able to negotiate back and forth on a more direct level, without clipboards or URLs.

    By Jesper · 2010.11.19 02:12

  8. “And somehow, documents, of which you traditionally have many, work with a short time and want to keep, should escape the ability to be archived?”

    Jesper. I’m with you.

    I’m in favor of general purpose computers. I’ll still be doing my work on one in ten years. I just think Apple is slowly exiting that particular business line.

    “Don’t “average consumer” me. The reason the average consumer doesn’t do a lot of bookkeeping on their iPads right now is because it’s such a pain in the ass. If they really want it to be able to replace the computer for those users, they’re going to have to solve it”

    Does the median consumer spend a lot of time doing bookkeeping on their current general purpose computer? I’d argue that they don’t. And I’d further argue that they’d prefer to do even less bookkeeping than they already do.

    Not to mention that things like the lack of a transparent file system help a device maker achieve consumer lock-in.

    Many of the things that you and I see as bugs in iOS are actually features from the standpoint of a CE device company looking forward.

    The “Simple things should be simple and complex things should be possible” design philosophy is not intended to make the leap from OS X to iOS.

    Lamborghini began life making tractors. People still need tractors. Lamborghini just doesn’t serve that market anymore.

    By Chucky · 2010.11.19 15:34

  9. I’d argue that they don’t.

    Not today they don’t. Do you think we’ll still be doing this sort of stuff overwhelmingly on paper in ten years? (I doubt we’re doing it “overwhelmingly” today, even.)

    Your opinion about the customer electronics angle makes sense except that Apple itself is driving business to adopt iPads. Your TV may not have folders, but it doesn’t have VPN either. iPhones and iPads have had VPN since the beginning, and they’ve been bending over backwards for corporate features. That’s pretty funny for a piece of customer electronics.

    By Jesper · 2010.11.19 19:10

  10. “Your opinion about the customer electronics angle makes sense except that Apple itself is driving business to adopt iPads. Your TV may not have folders, but it doesn’t have VPN either. iPhones and iPads have had VPN since the beginning, and they’ve been bending over backwards for corporate features. That’s pretty funny for a piece of customer electronics.”

    Sure. You can pull out your iPhone on the street and magically use VPN to reawaken your childlike sense of wonderment. I dig. It’s CE paradise.

    But what are you VPN connecting to? Your general purpose computer.

    Think of it this way, Jesper:

    My significant other doesn’t have an admin account on her OS X laptop. She is smart, and makes good use of her OS X laptop, but she has no interest in being a system administrator. As long as I admin her machine, she’s a happy user.

    I want a general purpose computer that I can admin. She doesn’t.

    And the median consumer is more like her than it is like me.

    iOS is all about machines where the user never faces a message like “With great power comes great responsibility.” And that’s what Apple is going to be preinstalling into its laptop form factor boxes in a few years time, no?

    “The iPad may not need a visible, all-dictating file system as we know it, but it damn well needs a filing system.”

    Sure. iOS will actually be usable in a few years. And somewhere during that interval, it’ll acquire some kind of filing system.

    But it won’t have a filing system that has the power of a true transparent hierarchical file system.

    That’ll be an acceptable tradeoff for the median consumer. But it’s going to force a consumer like me to find a different way to source my general purpose computers, no?

    By Chucky · 2010.11.20 18:04

  11. Sure. iOS will actually be usable in a few years. And somewhere during that interval, it’ll acquire some kind of filing system.

    But it won’t have a filing system that has the power of a true transparent hierarchical file system.

    You probably won’t be able to, say, do symlinks, no. You will be able to do folders, or maybe tags, but probably folders. And that’s what I asked for in the post. I didn’t ask for symlinks – you don’t have the kind of power where you could make the most of them anyway! So it’s very strange of you to use something I didn’t wish for as an argument against the post. (Unless you think they’re going to, say, limit hierarchical structures.)

    I think the filing system will be more powerful from the perspective of actually using it to do stuff with than would a filing system that’s feature-for-feature compliant with transplanting the file system itself there. That’s my hope, anyway; redone from the beginning, not dumbed down.

    By Jesper · 2010.11.21 00:17

  12. “That’s my hope, anyway; redone from the beginning, not dumbed down.”

    You are indeed a hopeful person.

    I think iOS is going to be dramatically dumbed down for reasonably sane design reasons.

    I just wish Cupertino would continue to develop a non-dumbed down system for folks like me, and I’m not particularly hopeful on that count.

    •••••••

    “You will be able to do folders, or maybe tags, but probably folders”

    Yup. But you definitely won’t be able to do folders and tags. At least not for the next decade or two. And no third party will be able to add features to the filing system. And there will be partial lock-in to whatever filing system is implemented.

    “You probably won’t be able to, say, do symlinks, no”

    Yup. And therein lies a problem. I don’t need symlinks often. But in the cases where I do need symlinks, well, then I need symlinks.

    And when Apple starts selling its laptop form factor boxes with iOS preinstalled in a few years, what happens to the edge cases like me? Windows? Ubuntu? Ugh. I’m not a happy camper about what is happening inside The Fruit Company.

    I just want a computer that follows the basic “simple things should be simple and complex things should be possible” design philosophy, and I don’t know where to get those computers looking forward a few years.

    “So it’s very strange of you to use something I didn’t wish for as an argument against the post.”

    I’m not arguing against your post. I find it an interesting post, and agree with large parts of it. I’m just riffing off of it to my particular concerns.

    I actually think iOS is already the best game console ever invented, and it will only become a better game console as it matures. But I’m just not really into Pokeman.

    By Chucky · 2010.11.22 01:08

  13. I have Angry Birds on my HTC Desire HD and it rocks.

    There should be layers (Like on Android): File system “Views” of parts of the file system.

    “Views” are application specific views into a segment of the file system and should show only what a typical user cares about (“Documents”, “Music”, “Pictures”, “Movies”, etc.). A power user should still be able to save an mp3 into “Pictures” or a jpg into “Movies”. The view should provide a neat abstraction over the file system that is simple for common use, but easily expands to a full representation on request. For example the down arrow on “Save As” dialogs on Mac OS X.

    By Chris · 2010.11.22 03:38

  14. Chucky: What if some of the cases where you “need” symlinks reveal gaping holes in the structure or usage of the file system? You need a file to “exist” in two places. What if it could really exist in two places in the Hypothetical iOS Filing System (HiFS?)?

    (And now you’ll say, well that’s a symlink. And I will claim that it isn’t, based on a) symlinks only existing in non-console OSes, and apparently iOS is entirely a console OS, and b) the non-symlinks being necessarily actually manipulable on the Finder/Open/Save-equivalent level, instead of in the Terminal.)

    I may not have complete faith in their every design decision going forward. But I am confident that at least some part, in being reinvented, will come out saner and more capable in the process. Just as we don’t need to ignore everything that has happened previous to iOS, we don’t need to clone every design decision, quirk or mistake of those technologies.

    Chris: I’m not wild about “type buckets” (like Windows 7 libraries or any such contraption). To begin with, they do absolutely nothing for “projects”, and to wrap up, if your solution for that is let’s override them at the drop of a hat, they won’t even be type buckets anymore.

    They solve the problem of “let’s neatly collect everything of one kind”. They don’t solve the entire problem of organization in all its splendor. Neither does search, tags, folders, keywords and so on. Replacing an entirely static system (file systems) with an entirely dynamic system (search or type buckets) won’t advance anything.

    On Mac OS X, type buckets exist where they should; in the open/choose panel, alongside every other tool at your disposal. It is pure there because you can’t drop pictures in the music bucket. And it is limited to where it makes sense because there’s only buckets for discriminate types, and none for “oh, generic documents”. That’s how it becomes useful.

    By Jesper · 2010.11.22 20:05

  15. “I may not have complete faith in their every design decision going forward. But I am confident that at least some part, in being reinvented, will come out actually partially saner and more capable in the process.”

    Again, you are indeed a hopeful person.

    I think iOS will be a less powerful system than OS X is today for at least a decade of development.

    “Just as we don’t need to ignore everything that has happened previous to iOS, we don’t need to clone every design decision, quirk or mistake of those technologies.”

    But we don’t need to recreate the wheel. Especially when some of the new “features” of the neo-wheel include a magical new square form factor.

    I don’t think the goal with iOS is to advance the “simple things should be simple and complex things should be possible” computer.

    I think the goal with iOS is to create a revolutionarily game console.

    Diff’rent stroke for diff’rent folks. I’m just not really into Pokeman.

    By Chucky · 2010.11.23 01:02

  16. But we don’t need to recreate the wheel. Especially when some of the new “features” of the neo-wheel include a magical new square form factor.

    I’m speaking out of my ass too — there’s no such thing as what I’m describing today — but at least I’m not passing judgement based on that. I’m glad you’ve got everything so figured out that you’re already pre-derided every future action. Let’s come back and debate this when it really exists. (And since that won’t be a full on file system with every historically important ability, and that’s the only thing you’ll be placated by, except not really, you’re right that I’m a hopeful person.)

    By Jesper · 2010.11.23 01:23

  17. I’m just not really into Pokeman.

    stop saying this!!!

    By zem · 2010.11.23 09:21

  18. “stop saying this!!!”

    Blame it on Jesper. He lodged the thing in my brain in the first place.

    By Chucky · 2010.11.26 22:30

  19. “I’m glad you’ve got everything so figured out that you’re already pre-derided every future action.”

    I don’t have everything figured out, but I do think I have a grasp on the likely.

    Do you really think you’ll be able to have an admin account on the laptop form factor box you buy from Cupertino in 2014? I think it pretty likely that you won’t, but I’m certainly not certain on that.

    In short, I don’t have it figured out, but I think I can see which way the wind is currently blowing. And knowing which way the wind is blowing tells you something about tomorrow’s weather, but it doesn’t tell you everything.

    “Let’s come back and debate this when it really exists.”

    Of course, 2013 or 2014 will be too late for this debate. If anyone is interested in actual outcomes, the debate has to happen now. And it is. See some of the “Geek Ludite” and “Freedom 3” posts Gruber has been approvingly linking to of late. The debate is happening, and the “lockdown” faction are the only ones on the playing field.

    Microsoft has repeatedly delayed full-on implementation of Paladium due to public outrage. If you’d like to continue to get computers from Apple in the future that aren’t locked down, now would be a time to start voicing some outrage.

    •••••

    “And since that won’t be a full on file system with every historically important ability, and that’s the only thing you’ll be placated by…”

    No. as I keep repeating, I’ll be placated by an OS that manages to well implement the “simple things should be simple and complex things should be possible” design philosophy. The details used to implement that design philosophy are essentially unimportant to me.

    But “simple things should be simple and complex things should be possible” is not what iOS is fundamentally all about. iOS is about “simple things should be simple, and a very limited subset of complex things should be semi-simple”. Great. It’s a smart design philosophy for mobile phones. Unfortunately, Apple seems to see the iOS design philosophy as their mid-term path forward for their entire product line.

    The Alan Kay design philosophy is the wheel. It works nicely, although it can always use refinement and imaginative extension. But iOS is really a fundamentally different philosophy that can never replicate the functionality of the wheel for users like me. I’m just not really into game consoles, even if they can create excellent experiences on hardware with limited resources, like mobile phones.

    •••••

    We’ve been living in a pretty special era for the last decade. Cupertino has been selling us boxes with UNIX For The Masses. There was nothing like it before, and I think it likely that soon there will nothing like it again. That seems a reasonably worrisome outcome to me that is worth discussing in advance.

    By Chucky · 2010.11.26 23:10

  20. That seems a reasonably worrisome outcome to me that is worth discussing in advance.

    Which is what I’ve been doing for the past several years. That doesn’t preclude me from noticing things that are working well in the “new world order”. I’d be remiss if I didn’t. I think that the best way to fight the more locked down model of computing is trying to figure out how problems that are easily solved with more restrictions can be solved more easily without requiring them.

    For example, most of the benefits psychologically tied to the App Store are advancements in installers, distribution and good hardware. Why aren’t those advancements getting across? More importantly, why isn’t, say, the Linux package management kicking Apple’s ass in public mindshare? It does more things more robustly without requiring centralization and with a longer track record!

    Steve Jobs warns of “being trapped by dogma”. As much of a shame it is that we find ourselves limited by the poverty of his imagination, I don’t see the attraction of taking the foregone conclusion that the only thing that will satisfy your demands is the current solution. That’s just having faith in the other end of the spectrum.

    And yes, I do think hindsight is the only way to go for judgements; that way, you only judge what actually happens. If hindsight is the only way in which you derive your ideas, you’re in big trouble. I’m actually also worried about knee-jerk reactions to anything that brings the same advancements that I talked about as being the harbingers of the closed model. If that’s the case, the only successful carrier of those advancements will be the ones with the closed model, and I don’t see that strengthening the case against it.

    By Jesper · 2010.11.28 02:19

  21. I dont know you, but I see lots of trucks around…

    By juan · 2010.11.29 18:06

  22. “More importantly, why isn’t, say, the Linux package management kicking Apple’s ass in public mindshare? It does more things more robustly without requiring centralization and with a longer track record!”

    Sure. But it ain’t for the masses.

    If OS X didn’t exist, I’d run Windows, not Linux.

    iOS fails the “complex things should be possible” part of the formula. But Linux fails the “simple things should be simple” part of the formula. (And I essentially need a mass market OS for the third-party devs.)

    •••••

    More broadly, better app installation procedures than exist at current are welcome. And I’ve been wanting “saved state” as normal app behavior for years.

    I’m all in favor of improving the OS. I don’t want OS X to be preserved in amber as is.

    I’m just not willing to run an OS where I’m not allowed to have an admin account.

    “Steve Jobs warns of “being trapped by dogma”. As much of a shame it is that we find ourselves limited by the poverty of his imagination, I don’t see the attraction of taking the foregone conclusion that the only thing that will satisfy your demands is the current solution. That’s just having faith in the other end of the spectrum.”

    The problem is that we’re not trapped in the poverty of Steve-o’s imagination. We’re trapped in the poverty of Steve-o’s desire to die with the most cash.

    I don’t think I have an unfounded faith-based pessimism about Cupertino’s future path. It’s think that all the signals Cupertino have been sending are that they think they can best maximize their share price by organizing around a lockdown model into the future.

    Just upthread, juan notes that there are still a lot of trucks around. That is true, and that will continue to be true. I’m just don’t think Apple thinks it in its own financial interests to continue in that market much longer.

    You speak of what will “satisfy my demands”. And again, I just want a box that well implements the “simple things should be simple and complex things should be possible” model in a mass market OS. I want an easy to operate truck like I’ve got right now. I’m just doubtful about being able to procure those boxes from Cupertino in a few years.

    So comfort me, Jesper. Tell me where my math is wrong:

    • I don’t think iOS will be ready to satisfy those demands in a few years.

    • I don’t think an “non-lockdown” OS X is likely to be still under active developement in a few years.

    I’d certainly love for my math to wrong on one of those.

    By Chucky · 2010.12.02 17:36

  23. I don’t think an “non-lockdown” OS X is likely to be still under active developement in a few years.

    I do. If we go twenty years out, things might start getting a bit fuzzy, but that’s because the original Mac OS didn’t survive twenty years, and because we don’t really know anything 20 years from now. Who saw the iPhone in 1987?

    I don’t think iOS will be ready to satisfy those demands in a few years.

    That’s a little more uncertain. I believe that iOS will eventually be gutted of the unnecessary divisions and leave behind one of the first OSes that balances good separation and the desire of applications to be useful. (I was going to say ‘freedom’, but I don’t mean that freedom is or is not compromised. I mean as in the case where you start with all freedoms intact, and then, because you want to, build a system with containment as a primary tool of choice. Theoretically, you can’t do anything without depriving yourself of the freedom to do something else. In reality, you can do freedom 0 and “freedom 3” simultaneously.)

    I realize that this prediction is a very hard pill to swallow. Some people believe in the market. I believe in time, reasoning and humanity. In time, what is not essential will whittle away, even though the chart of that timeline does not involve a straight line of universal improvement. It might not be tomorrow or within three years, but eventually, there will be an alternative solution which does what iOS claims to do without the onerous overtones, while being as compelling to use. (Android isn’t it in this incarnation.) Once that happens, either iOS adapts or it falters.

    It helps that to Apple, all this unmotivated locking and overzealous constraining, all this control, is a means to an end. I don’t fundamentally believe that every developer and manager and Apple wake up every morning, eager to constrain everyone, but I think that large parts of the organisation, including many of the top brass, believes that their current approach is what must be done to achieve the benefits of what they do.

    All that is needed to break their current strategy is for the people in the right positions to be disenthralled from this concept, to instead catch the idea that you can do all these things the right way, instead of the current easy way, lose none of the benefits, and hand people back the keys. (That’s hard work, and it’s not an easy thing, but it is clear.)

    By Jesper · 2010.12.03 00:59

Sorry, the comment form is closed at this time.