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from the fact that Android's share menu addresses many of these concerns, his point still stands: files are not an outdated, archaic concept.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.
Thread beginning with comment 524487
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[4]: Comment by tupp
by tupp on Sat 30th Jun 2012 09:44 UTC in reply to "RE[3]: Comment by tupp"
tupp
Member since:
2006-11-12

The first version of the Macintosh didn't have the concept of files at all.

Of course it had files. I used one. What do you think you put in the "trash can?"


As henderson said files aren't essential to how computers work.

Files are essential to any computer that uses a file system (which is 99.99999% of computers).


It is a design decision that was made many years ago to how computers work.

It isn't a "design" decision -- it was necessary to have files for the increased complexity and to conveniently save data.

Reply Parent Score: 2

RE[5]: Comment by tupp
by henderson101 on Sat 30th Jun 2012 10:03 in reply to "RE[4]: Comment by tupp"
henderson101 Member since:
2006-05-30

Files are essential to any computer that uses a file system....


EXACTLY!

But a Filesystem is an entirely human solution to the problem of data storage. Storing data does not *require* a file system. This is the leap you are making, and you are wrong. File systems only suit those systems that want to expose data storage to a human. Hierarchical file systems even more so. Look at the way Palm did it with Palm OS. There was no file system, all "files" we're actually databases of resources, these lived in memory and were directly edited, not cached. The original Mac had a file system, but what you saw was only half the picture. Again, a file was forked, had a data fork and a resource fork and OFS had no hierarchical notions at all.

The whole notion of files and storage comes from the original sequential storage on magnetic tape. Where there was no option but to store as a single block of data.

Also, to address your claims about fragmentation - nice try. No, look at the physical storage on a hard drive at an inode level. It's rare for an entire file to be a contiguous block of inodes. Depends on the file system dynamics, but a file could be split across multiple sectors quite easily. Especially something larger than a few hundred megabytes.

Go read a book. The Domonic Giampaolo one is a free download. Google it.

Reply Parent Score: 3

RE[6]: Comment by tupp
by tupp on Sat 30th Jun 2012 10:44 in reply to "RE[5]: Comment by tupp"
tupp Member since:
2006-11-12

But a Filesystem is an entirely human solution to the problem of data storage. Storing data does not *require* a file system. This is the leap you are making, and you are wrong. File systems only suit those systems that want to expose data storage to a human.

No. File systems primarily exist to allow both humans and computers to safely and effectively deal with large, complex groups of data and code.

You are correct, that storing data does not require a file system. Most hand calculators store computation data between each entry.

However, it wouldn't be wise to try storing a large amount of complex data without files, such as one needs when (for instance) digitally editing a feature film. One static electrical discharge (or other data corruption) and ZAPP! -- there goes your entire project and all the hours you worked down the drain. Plus, without files, how would you store (and transfer from the camera) all of the gigabytes of the many takes of footage?


Hierarchical file systems even more so. Look at the way Palm did it with Palm OS. There was no file system, all "files" we're actually databases of resources, these lived in memory and were directly edited, not cached.

My phone uses Palm OS, and I guarantee that it uses a file system, with actual files that can be stored on mini SD cards.


The original Mac had a file system, but what you saw was only half the picture. Again, a file was forked, had a data fork and a resource fork and OFS had no hierarchical notions at all.

No. The original Mac had folders and sub folders, and files. I used one, but you don't have to take my world for it -- merely look at a screenshot from the original Mac: http://toastytech.com/guis/mac11sortview.gif


The whole notion of files and storage comes from the original sequential storage on magnetic tape. Where there was no option but to store as a single block of data.

No. Punch cards and punch tape were storage, too.

Storage in files and directories became a necessity once data and programs got too big and complex.


Also, to address your claims about fragmentation - nice try. No, look at the physical storage on a hard drive at an inode level. It's rare for an entire file to be a contiguous block of inodes. Depends on the file system dynamics, but a file could be split across multiple sectors quite easily. Especially something larger than a few hundred megabytes.

Not that it matters to our discussion, but it certainly is not rare for an entire file to in a contiguous block of inodes. This contiguous state is often found when a drive first sees use.

Furthermore, give me a blank drive and when I copy into it the entire contents of another drive, almost all of the files will be contiguous.

Reply Parent Score: 2

RE[6]: Comment by tupp
by Alfman on Sat 30th Jun 2012 19:38 in reply to "RE[5]: Comment by tupp"
Alfman Member since:
2011-01-28

henderson101,

"Storing data does not *require* a file system. This is the leap you are making, and you are wrong. File systems only suit those systems that want to expose data storage to a human. Hierarchical file systems even more so. Look at the way Palm did it with Palm OS. There was no file system, all 'files' we're actually databases of resources"

Why is everyone nitpicking the terminology? Frankly I don't care if you want to call it "files" or "content", "directories" or "folders" or "drawers". "File system" or "database" or "repository"... call it whatever you will, I don't care about that and I doubt tupp does either. It's not the terminology that's important, it's the ability to separate/organise things to keep them them from becoming a jumbled mess containing everything.

From a typical user perspective, why should one care if documents reside in a "file system" or in a "database" (pedantic note: a file system IS in fact a kind of database)... Users should be completely oblivious to the inner workings of the OS. You surely agree here, so what is it about hierarchical organisation that's putting you and others off? It's illogical to argue that hierarchies are bad for uses on account of the fact operating systems also use them internally.

Reply Parent Score: 4

RE[5]: Comment by tupp
by lucas_maximus on Sat 30th Jun 2012 12:35 in reply to "RE[4]: Comment by tupp"
lucas_maximus Member since:
2009-08-18

Sorry I wasn't clear, One of the early prototypes at Apple before Steve took control of it.

From the book itself.

even deeper locked-in idea is the notion of the file. Once upon a time, not too long ago, plenty of computer scientists thought the idea of the file was not so great.

World Wide Web, Ted Nelson’s Xanadu, conceived of one giant, global file, for instance. The first iteration of the Macintosh, which never shipped, didn’t have files. Instead, the whole of a user’s productivity accumulated in one big structure, sort of like a singular personal web page. Steve Jobs took the Mac project over from the fellow who started it, the late Jef Raskin, and soon files appeared. UNIX had files; the Mac as it shipped had files; Windows had files. Files are now part of life; we teach the idea of a file to computer science students as if it were part of nature. In fact, our conception of files may be more persistent than our ideas about nature. I can imagine that someday physicists might tell us that it is time to stop believing in photons, because they have discovered a better way to think about light—but the file will likely live on.

The file is a set of philosophical ideas made into eternal flesh. The ideas expressed by the file include the notion that human expression comes in severable chunks that can be organized as leaves on an abstract tree— and that the chunks have versions and need to be matched to compatible applications. What do files mean to the future of human expression? This is a harder question to answer than the question “How does the English language influence the thoughts of native English speakers?” At least you can compare English speakers to Chinese speakers, but files are universal. The idea of the file has become so big that we are unable to conceive of a frame large enough to fit around it in order to assess it empirically.

Lanier, Jaron (2010-01-18). You Are Not A Gadget: A Manifesto (p. 13). Penguin UK. Kindle Edition.


The second are I have highlighted is the exact mental trap that you are falling into.

Edited 2012-06-30 12:45 UTC

Reply Parent Score: 3

RE[6]: Comment by tupp
by tupp on Sat 30th Jun 2012 15:47 in reply to "RE[5]: Comment by tupp"
tupp Member since:
2006-11-12

Sorry I wasn't clear, One of the early prototypes at Apple before Steve took control of it.

This description sounds like prototype which had the sole purpose of merely demoing Apple's version of the Xerox/3-rivers GUI. The file and folder icons were there (just like the Alto and Perq), but there was no content, hence, it had no "files." The intention for the actual OS was to eventually have a file system.


From the book itself.

Sounds like a fundamentalist about to quote scripture.

The text that you quoted and highlighted is B.S.

Also, why don't you come up with some original thought, instead of quoting someone else's conjecture?

Reply Parent Score: 2

RE[5]: Comment by tupp
by robertson on Sun 1st Jul 2012 19:10 in reply to "RE[4]: Comment by tupp"
robertson Member since:
2010-04-30

"The first version of the Macintosh didn't have the concept of files at all.

Of course it had files. I used one. What do you think you put in the "trash can?"
"

I think he may be referring to early prototypes of the Macintosh when Jef Raskin was still on the team. I recall a story conveying that Jobs didn't like Raskin's interface ideas very much, including a "document-centric" approach (as opposed to file/application-centric) and easily learned keyboard commands in place of a mouse.

If one looks at the Canon Cat, one gets an idea of a system in which the file structure is made irrelevant to the user in favor of a different, document-centric approach. The Etoile project is also aiming at a document-centric environment, as I understand it, but will also have utilities for exporting your work as files.

All of this is really only tangentially related to the linked article, which seems to be more about apps locking in your data in annoying ways. A document-centric system needn't make it impossible to extract and move your data around different "apps."

All this said, I like the "Unix way" of "everything is a file" for the transparency it provides. I also like the document-centric design philosophy for its kindness to the mind when one just wants to work on documents, graphics. The great both/and principle...

Reply Parent Score: 1