Linked by Thom Holwerda on Fri 29th Jun 2012 22:55 UTC
OSNews, Generic OSes "Whenever there is a conversation about the future of computing, is discussion inevitably turns to the notion of a 'File'. After all, most tablets and phones don't show the user anything that resembles a file, only Apps that contain their own content, tucked away inside their own opaque storage structure. This is wrong. Files are abstraction layers around content that are necessary for interoperability. Without the notion of a File or other similar shared content abstraction, the ability to use different applications with the same information grinds to a halt, which hampers innovation and user experience." Aside from the fact that a file manager for Android is just a click away, and aside from the fact that Android's share menu addresses many of these concerns, his point still stands: files are not an outdated, archaic concept. One of my biggest gripes with iOS is just how user-hostile the operating system it when it comes to getting stuff - whatever stuff - to and from the device.
Permalink for comment 524383
To read all comments associated with this story, please click here.
RE: Comment by tupp
by Doc Pain on Sat 30th Jun 2012 00:49 UTC in reply to "Comment by tupp"
Doc Pain
Member since:

Files and directories are fundamental to computers, [...]

No. They are fundamental to the user in the first place. Files are an abstraction of organization criteria (such as directories), carrying meanings such as carrying information or containing a hierarchy.

But you validly can do more with files than just see them as a means to access stored data. Files are also means to access hardware infrastructures, such as the device files in UNIX: Writing to a port takes nothing more than writing to a file. Using this abstraction method, even "mounting a network" is possible as shown in Plan 9. That way, files can be seen as an abstraction of hardware, a means of the OS to access resources and address functionalities.

The concept of files is present on nearly all kinds of computers, even though the practical implementation at file system level is quite different and usually not portable (as the content of the files, generated by programs, is often also not portable). Still this concept of files and directories, even if they have other names (e. g. datasets and libraries) can be found everywhere. Put a UNIX guy infront of a z/OS mainframe or an AS/400 interface - he will quickly see the similarities, even though there might be a "database file system" responsible for actually storing and retrieving the data. Then take a z/OS hacker and put him infront of a KDE session. Even though he sees a graphical representation, he will conclude what files and directories are. No big deal you say? Try the same with any output of our "modern" educational system. :-)

[...] and hiding such a basic organizational model from the end users only makes them utterly helpless and ignorant.

What you describe here is a resonance effect: Today's users are hardly able to figure out the concept of files and directories, as they do not use them. They even don't use programs, let alone an operating system. They "do stuff" - and that's it. OS, programs and files are just means to get stuff done, but "the authorities" (advertising, education, commerce) tell them that it's not needed to know anything to utilize a computer to get stuff done.

Of course, those who first learn those basics are highly superior to those who rely on a pre-chewed widget wizard with magical charms that finds their stuff when they need it. You can see that in reality where "professional" secretaries "organize" the company's correspondence in files like "Letter 1.doc", "Letter1.doc", "letter 1(a).doc", " letter2.doc", "letter 3 .doc" and so on. Of course, finding information within such a mess is more complicated and more time consuming than utilizing the concept of files and directories in the first place. Those candidates also store the companies e-mail in the Inbox (which grows to over 100,000 messages locally stored, without any backup). Trust me, I've seen it. It hurts.

Files confuse the average user. Directories confuse them even more. Don't try to change (or improve) things, this also confuses them.

Di I sound impolite? I hope I don't, because it's not meant to be a harsh unjustified statement, it's just based on my individual observations. That's why I say "average users", because novice users have the chance to learn and leverage the concept of files for their benefit. But that learning has to be done at the beginning of an "IT career" (and because IT is ubiquituous, for nearly everyone's career).

As you said, files basically are a model to the user. Learning a model and not confusing it with reality takes some prequisites: the ability to see the difference between them, on many levels (language level, visual representation level): the icon is not the file, the file is not the actual content of the file. Dealing with a model requires the presence of the concept of abstraction, something representing something else. The same way this concept is used on OS level (file /dev/console represents the system console), the user can utilize the concepts to increase comfortability, productivity, speed and security when using a computer.

Depending on file systems and their features, "plain files" may even be limited. Document management systems allow "tagging" of data (not represented as files anymore - they might be files, but they can also be binary blobs within a large database file). Even though it sounds complicated, it can be implemented easily by any half-baked Linux hacker using "plain files" and a few tricks.

You are surely familiar with "enterprise" document manangement systems. They use files "in a hidden way", and claim to care for everything. As long as it works, it's a real benefit for those who cannot understand the concept of files and directories. The "fun" starts when it stops working. Maybe you end up with a 2 TB proprietary binary blob database, with no chance to get your data out of it.

Such a complete reliance on software to find and organize everything on a computer really wastes more time in the long run and causes much more frustration when any problems occur.

True, but the average user has been taught to think: "The computer will do it." Maybe you can remember the times when DOS had file names in the 8.3 manner, and other operating systems (much older, see mainframe era) also had restrictions in what you could put into a file name. At this time, people were able to organize their stuff even within that very limited environment. Some of them even knew their files "by name".

But today, dealing with "bare metal abilities" is highly discouraged (by "the authorities") and more and more made impossible by the software manufacturers. The more advanced, yet simplified the interfaces for interacting become, the more the file concept steps into the background, even though it might still be present at the bottom level, hidden and locked.

And, really, it is needless to hide the file hierarchy system, because there is hardly any "learning curve' -- any child or elderly person can grasp the concept of files and directories within about five minutes.

That might have been correct 20 years ago, but sadly, this is not the truth anymore. While "unspoiled" young people actually have the chance to learn this, the majority seems to be learning-resistent and memory-freed.

Moving within this concept requires a kind of "language", to name files and directories. There's typically a grammar based on the file system way of doing things, and vocabulary to be supplied by the user (the actual file names and directory organisation to be created). Expecting the "point & grunt" generations to learn that language is something hard to consider. And if those knowledge is not present, the whole concept of files cannot be adopted as a tool to do something (because it is "only" that - it can be a powerful tool if applied properly, and a pain in the ass if not). Educated decisions are often helpful, again something that is hard to consider for the discussed audience.

I want to say: There's nothing wrong with files per se. They are intended to be used by a certain audience with specific needs and present knowledge. But they aren't for everyone. Those who cannot use them will have to deal with whatever kind of "desktop search" (extend it to remote services, social networks and the cloud) to find what they're searching for.

In order to gain market share (which is the main purpose of innovation, as it seems), you need to hide all "unneccessary" elements. If you don't do it, your product will be regarded "too complicated", because it deals with files and directories and transitions their use to the user, instead of "protecting" him from all that "annoying" stuff.

In keeping with Job's notions, perhaps Apple should actually make "The Wheel":

We're exactly heading into that direction, just be patient! :-)

Reply Parent Score: 6