Slashdot Mirror
Operating Systems Are Irrelevant
zincks writes "David Gelernter (Yale Professor of Computer Science, and Unabomber target) has a
story in the NY Times which states, (1) Operating systems are relics of the past, (2) We should be able to access data anytime/anywhere, by (3) seeing a stream of 3D documents(?), so (4) he's written such
software, and (5) that's all you should care about so it doesn't matter that it runs under windows.
This is a fantastic (definition: based on fantasy : not real (?)) vision of the future by a premier technologist."
Hmm, an interface that is completely independant from the underlying OS, network, etc, etc. I think I may have heard of that before. What's that? In 1986??? Oh yeah.
I don't have a sig...Do you??
You know I own a car. And cars have gotten to the point where when I buy it typically I do not have to consider the road. It is irrellevant...
....
OR, is it relevant after all? Lets see in Germany I would get a sports car, Switzerland big luxury, Canada SUV because of the snow, Southern France Convertible,
My point is that while we do not make a big deal of the road or conditions, it does influence our buying decisions. And saying that it is irrelevant is just a pipedream...
"You can't make a race horse of a pig"
"No," said Samuel, "but you can make very fast pig"
Okay, that article is just an advertisement. I'm surprised that some editor at the Times let that pass for a column.
Let me summarize for the impatient. "Operating systems are irrelevant, except for Windows, which we should be thankful to Bill for, because it made everything more accessible, and he's oh-so-visionary. Buy my stuff, which is an incarnation of the vision that Bill wants to realize in Longhorn. Also, Linux is irrelevant."
The OS isn't dead per se, but, like he said, irrelevant. Ask the average Joe what OS they run. First off, they won't know what OS means. Then, if you ask 'em what kind of Windows they have, most of 'em still won't know. People have long since using computers for the sake of the OS (well, except for OSS zealots). His point is to slap some real useable software on top of any OS and live there, not at the OS level with files, folders, permissions, etc.
The whole artical read as a huge advert. /. isn't the only one placing ads as stories...
Guess
III.IIVIVIXIIVIVIIIVVIIIIXVIIIXIIIIIIIIVIIIIVVIII
While I didn't read the article (I see enough NY Post commercials on TV). I assume this guy seems to think the O/S is "all the pretty windows". I think the common person's viewpoint of the O/S is also the same, but when we get down to brass tacks, the O/S is what holds those windows on your desktop. File system access, memory management, task/process management, interfacing drivers to hardware. THIS IS THE O/S!! I doubt seriously that these floating 3-D documents will do little good if you can't even drive signal to your video card.
Perhaps this person should exhalt a new outlook on user interface design (ex: extending Windows, or KDE or Gnome), and not dismiss the O/S.
And for those ready to flame on my inclusion of Windows, Gnome and KDE on the same sentence, realize that these are all essentially window/interface managers, and not operating systems. Yes, MS bundles their manager and O/S in an unpackageable package, but the interface you see in MSWindows is not the MSWindows O/S. That is like saying a BASH shell is the O/S of Linux/UNIX.
Whoa, there. Was this guy watching Johnny Mnemonic while drugged up beyond belief? His drivel about being able to "see a stream of 3D documents" reminds me of the virtual surreality user interfaces in that movie. I wouldn't be surprised if he started spouting off unintelligible mutterings about "hacking the Gibson" and "finding the garbage file", too.
In Soviet Russia, Jesus asks: "What Would You Do?"
Read the article. Nowhere did he say that an OS wasn't necessary.
Those of us who've grown up on Wintel might relate. I got to spend about an hour working on some Macs yesterday, a mix of OS 9 iMacs and new ones running OS X. For anyone who's grown up on Windows, this is a refreshing change of perspective. I never did find any sort of command line, even when I briefly needed one, but the machine just "worked". Everything was responsive and fast and gorgeous and simple. When I have my own Linux box really tweaked out the way I like it -- WM with the genmenu menu structure -- it just runs like a deer.
You reach a point in any well-designed system where you don't interact with the system itself anymore. For example, I've got a site I frequent with a login and the "submit" button drawn in JavaScript instead of as an HTML button. ie lets me just hit "Enter", but Mozilla requires that I mouse-click "Submit". That's a Mozilla problem. Windows XP allows you to burn cds and read zips right in the filesystem browser, which is a good thing. KDE used to have some five different apps under "Text Editors", which is just not useful. That WM menu I had was easy to customize and had only the one or two that I used. These are issues of system design, not program functionality.
I'm looking forward to the day I don't have to worry about how the system runs and whether it will continue to run. I'm not far from that with a Linux+WindowMaker desktop of my own design, but even then I have to struggle with issues like printing and file format compatibility and fonts. I guess there are people in corporate, standardized environments that have Microsoft SMS running and the whole MS Office suite customized and installed who probably feel their work is pretty transparent. I haven't yet SEEN one personally, but they probably exist.
-j
And...
Each [of linux and Windows] is nonetheless still solid enough to be a good, steady platform for the next step in software.
This does not indicate a future in which operating systems are really irrelevant. In fact, it would appear to be the opposite. Now, the operating system may appear to be invisible to the end user, but that isn't the same thing. People like Alan Cooper have been pushing for this kind of computing interface for ages.
The underlying operating system must be transparent, and rock-solid, fast, correct, and efficient.
Again, from the article:
nearly universal platform - and for the software future, "universal" is nonnegotiable.
Why does the OS have to be universal? The operating system may become invisible, but a properly written interface will be portable. No one will have to know how to use the "operating system" that powers their hardware, but they may figure out that some are more reliable at running their Interface Of The Future (TM) than others.
bytesmythe
Hypocrisy is the resin that holds the plywood of society together.
-- Scott Meyer
Or perhaps . . . just perhaps . . . he's more concerned with making this product available to the biggest market share. Not really so much concerned with advancing computer science, as with making money? Maybe?
The New York Times: Free advertising space for anyone with a PhD.
This article is a joke. This guy seems to know very little about nothing.
THE end of the Microsoft trial is great news whatever you think of the defendant - because the trial was all about the past, and we in the technology world have no more time to waste on that topic.
The past? Idiot. Idiot! Fool. If we don't look at our past and learn from it, we are gonna repeat it, and make the same damn mistakes in the future. What MS did only affects everyone in computerdom out there. Ask Be Inc, or Netscape, OS/2 or Linux companies what they think of about this being something we should forget about? No, it was about our present, and future. XP wouldn't be the POS it is if there was more competition.
Meanwhile, operating systems are lapsing into senile irrelevance. An operating system connects the user (and the user's software) to the ensemble of machines we call a computer. But nowadays users no longer want to be connected to computers. They want to be connected to information, a claim that sounds vague but is clear and specific.
But wasn't that the goal of computers from the beginning? To enable a 'paperless world' where we could input and receive information from a centralized location. Um, mainframe, anyone? And how is the OS irrelevant? Maybe to him it is, and to the home user, but to developers, hardware makers, and administrators, the OS is very much the heart and soul of the computer. It determines whether the software will run- the software that obtains the information you demand.
This kind of information management is simpler, more powerful and more natural than the Steelcase-inspired software we've got today - the files, the folders, the desktops and all those other high-tech office accessories straight out of 1946.
You know, I still use a file cabinet. As far as I know, they are a great resource when the network goes down, or a hdd crashes. I support large companies that still use them. Just because it is old, does not mean it is no longer needed, wanted, or relevant.
We built our system on Microsoft Windows because Windows is a reliable, solid, reasonably priced, nearly universal platform....
Well, one out of 3 ain't bad. No comment on what everyone else will point out here.
Of course, another operating system, Linux, is also clamoring for attention. Linux and Windows are both children of the 70's: Linux grew out of Unix, invented by AT Windows is based on the revolutionary work of Xerox research. In technology years, these loyal and devoted operating systems are each approximately 4,820 years old. (Technology years are like dog years, only shorter.)
Anyone know what he is talking about here? So, Windows and Unix are almost 5000 dog years old. How is this little piece of info helping his argument. Can anyone help me out here. I don't see it. I think he is trying to make linux look like the old beast of the ancients, when it is actually newer than Windows is. I mean, Windows the OS didn't happen till 1993 with NT 3.1- linux was 'born' in 1990. Prior to 93, windows was an OE.
They stuck me in an institution, said it was the only solution, to...protect me from the enemy, myself
Microsoft, people say, has driven up prices and suppressed innovation. But this is a ticklish argument at best: after all, over the decade of Microsoft's hegemony, computing power has grown cheaper and cheaper. Innovation has thrived.
He's comparing apples to oranges...Microsoft is software based, they haven't designed any hardware (joysticks and future DRM technology not-withstanding). If you look at software, it HAS been stifled a bit...there are very few innovations in the OS market over the past decade. Windows has, just recently, incorporated functionality that Unix had 20 years ago.
Hardware has been where innovation has taken place. More transistors on a wafer, faster memory seek times, faster hard drive rotation, larger hard drive capacity. These are the big changes in computers, not the software.
--trb
The thing about his style is that he seems to believe that the way to get people to listen to him is to say something radical that can be wildly misinterpreted, and then get on the soapbox. He's also well known for saying things like the entire educational system in this country sucks and has to be rebuilt from scratch. So it's no surprise to me that he says the OS is irrelevant. In theory it's his way of getting people to at least look at their assumptions and question them. I mean, come on, how many people do you run into every day that tell you "Yes, I agree, Windows sucks, but why fight it?" WHY FIGHT IT? Because it sucks. Gelernter's point is that you should always be on the quest for the "powerful yet simple" solution to the problem.
In a rather interesting chapter of the book, he offers a variety of drawings for new desks. After all, who said that the traditional setup is the best one? So he creates a variety of stacks, slants, and other combinations that might work better for people.
I think the OS *should* be irrelevant. Awareness of it makes things complicated. Imagine if the rules of a Turing machine were different depending on what computer you ran it on, and on some computers its rules just didn't hold at all. Computers will be simpler when somebody can just say "Email" and not have to worry about Outlook, or POP, or any of that nonsense. That's my two bits.
duane, listening to old dr. dobbs mp3's he found referenced on slashdot last week
www.HearMySoulSpeak.com
Yes, from one perspective, you could say operating systems are irrelevent.
That's presupposing that whatever operating system is in place provides the needed infrastructure for managing processes and scheduling hardware access in a reasonable way. Doesn't matter whether it's Windows, MacOS, Linux or OS/2.
By the same token, my travels from home to work depend on my car, not on the roads.
And it's true that what my body contacts is the car, not the road (motorcyclists sometimes have it rougher, I suppose).
If the roads are well-maintained, plowed in the winter and other cars obey the traffic laws, I'd almost begin to believe that cars were more important and roads were irrelevant.
But if my super highway developed a large pothole, that illusion would disappear quickly.
Likewise, if the owner of the road decided to erect a toll booth and exact some money from me for use of the road, I'd begin to appreciate the importance of roads.
Operating systems: only when they work right do you not notice them.
"Provided by the management for your protection."
An interface that is completely independant from the underlying OS?
HTML? VRML? PDF? XML? JAVA?
These all seem to be what this guy is calling for.
You say things that offend me and I can deal with it. Can you?
then along came the micros like the altair8800. With these there was no operating system per se. Usually you just loaded interpeted basic from a bootloader and then managed your disk or tape using program you wrote in basic. For example PIP was the name of a program for reading files on a floppy. you loaded pip off tape and then you could access the floppy. when you were done you got rid of PIP. there was no OS only indiviual ad hoc programs. You could say the programming envionment was the OS but givine that your programs were doing peek and poke instructions instead of using an API i'd say there was no OS.
then with the rise of more memory and disks, apple, commodor, trs-80, and IBM started to emulate the mini-computers which had those VMS/CPM/DOS operating systems. then we graduated to Windows and macOS with real APIs.
then what, well along came the Browser, and the idea that one could replace the OS and APIs with a new sort of middleware that would be platform independent. Mosiac and netscape could open text files, and even do many operatiung system inteaction functions like telnet. (KDE extends the metafore to launching programs from a browser inteface).
the other fork was java, which combined a programming language with all the high level functions needed to act like an OS. in other words it was both a language and an API rolled into one. Sun began to talk about how JAVA was really a new kind of OS. you no longer needed to worry about what the computer or OS running underneat was. JAVA was you environment.
so now someone is saying OSs are dead. whoopee. I've seen it before.
Some drink at the fountain of knowledge. Others just gargle.
It's been something like 8 years since OpenDoc's component based applications it the net and I think it is the concept this guy speaks of( can't get to the article ). The idea was that you had a file format which allowed application data to be stored together with many applications all in one file. The applications were small, specific apps which were more like plug-ins for a "container" application.
The whole idea behind OpenDoc was that your data was what was important and YOU, the user, could mix and match small lightweight applications to create your own "super" document. The application or "Parts" developers would have to provide a free viewer for their data format so that you could email your "super" document to someone else and they could read it's contents by downloading the viewers.
The concept of document-centric, also called data-centric computing isn't new it's just the one very large monopolist must protect their operating system and make sure the "application" remains relevant. Document-centric computing abstracts the applications, greatly reduces the application size since they are now made of many smaller plug-ins, and most importantly, it reduces the barrier to entry into the market. Two or three coders could whip up a pretty good spell-checker Part or html editor Part as opposed to a full blown application containing the spell-checker, graphics editor, text editor, etc, etc.
This kind of stuff won't show up until Microsoft is gone or irrelevant. IMHO.
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
That should keep him quiet for a few decades...
Nice article, but the guy has a bit of a people-centric view. OS's are not only important to people, but machines are communicating as well. Think of computers communicating, but also networking hardware, machinery in processing plants, equipment in planes, trains & automobiles, ships, telephone systems etc.. There things are dependant on timing, realtime processes etc. A nice 3-d data view is completely irrelevant. The OS becomes even more important
As I take it, his basic point seems to be that both Windows and Linux are based on OS concepts developed at least 30 years ago:
-- Slashdot: When Public Access TV Says "No"
After reading the NYT article, I feel like having a good, loud BURP from the Pepsi-like nutritional value this guy's ideas are worth. What a load of pie-in-the-sky thinking. OK, yes, it would be very nice to work with one interface that allows me to store and retrieve all the assorted bits and bytes of my life. But no matter what that eventual piece of hardware looks like, it's still going to require some sort of OS to manage how it connects, how the HMI (Human Machine Interface) communicates with the entire information infrastructure behind it.
I've worked in IT for 10 years now, and this guy is as close to the "Useless White Guy" I see playing CEO's and CTOs on HP, IBM, Sun and M$oft commercials everyday. For a compsci guy, this guy is clueless about the infrastructure and hardware and how to get it to work properly to make his 'ideas' work in the real world.
I guess my 'realist' side is crying out to be heard today.
I can no longer read Dilbert. It's too depressing, because it is too real. -- Hyperhaplo
About 18 years ago Jeff Raskin and Canon Inc. built a computer called the Canon Cat based on some of these principles. The Cat memory was basically one massive flat file you could search, order, compute, or edit with a few simple universal commands. (I thought it was too massive for my limited brain.) The architect Jeff Raskin has a footnote in P.C. history as the guy and computer project who Steve Jobs took over when Steve was trying to regain power at Apple after the Apple III and Lisa diappointments.
Isn't it amazing how "Unabomber target" has come to imply a certain level of prestige?
I object to that article, and to the next reply.
With all due respect to the good (and smart!) professor, people have been singing this tune for years now. Oracle was gonna make OS's irrelevant. Then Sun was. Blah blah blah.
.WAV files played when email arrives, many (not all, but many) people project into their computers their own vision of what it should look like, sound like, behave like.
What the nerdish community (I am including myself in this) fails to realize is that a lot of people (and not just geeks!) like they do their cars and their office cubicles: they personalize the hell out of them. From soul-crushing cute-and-fuzzy-bunny stickers to desktop pictures to skins to gut-wrenching
The OS-agnostic, computer-as-information-portal idea is a grand one and might work for business travellers hitting kiosks and library users, etc. However, until a rich, customizable environment gives this "personalization community" the same ability to turn their desktop into a fairy princess or a NASCAR cockpit or what have you, OS's will continue to rule.
My two cents: keep the change.
"Don't matter how New Age you get, old age is gonna kick your ass." - Utah Phillips
over the decade of Microsoft's hegemony, computing power has grown cheaper and cheaper.
I've heard this before, and I will agree. See In the beginning was the command line. The point is, MS made it possible for computers to become cheap and commonplace. But now that they've done that, there is no further benefit from them maintaining a monopoly. The idiot who wrote this article doesn't seem to understand that much, and is still narrow-mindedly believing that they can do no wrong.
It boils down to ``OS doesn't matter - you need windows'' - in other words, a blatant bit of technically inaccurate flamebait. (And very good flamebait too - I've bitten...) Unfortunately, there's still idiots out there who believe what he's saying, and will think ``if even the experts say the OS is irrelevant and we should all buy Windows, then I will''.
We need Microsoft itself to be the universal stepladder that lets us climb out of our hole and smell the roses. ... euch. Troll just isn't a good enough word for it. Pass the 2x4x24.
We rely on the courts and antitrust laws to keep Microsoft from abusing its enormous power. We need Microsoft itself to be the universal stepladder that lets us climb out of our hole and smell the roses.
So everything will be grand if only we put Microsoft between us and our hardware? Please. Making a Microsoft operating system into a 'universal stepladder' will only tempt them to abuse this power further. Let's not turn this guy's vision into an us-against-them issue. This vision (and its implementation) is long overdue.
It's interesting that this guy is flag-waving for Microsoft in the first place. After all, he is competing against Microsoft's own 'window manager', which will become moot if his vision comes to fruition. If what he says is true, in two years he will then compete directly with Longhorn's UI.
Each is nonetheless still solid enough to be a good, steady platform for the next step in software. But Windows is the marketplace victor and has now won a decisive legal imprimatur. There is no technical reason for us to move to Linux; why should we switch?
What is really needed is a nice OS layer that gives support to these new user interfaces (that replace windows managers). Linux is a nice open solution to this problem. What happens to this guy when Microsoft comes out with their own new-and-improved GUI for Longhorn? Microsoft closes their OS (not window manager) API and its game over.
Who knows, maybe he's just trying to get a job on the Longhorn project. But if he's going to try to compete directly against Microsoft, I don't see how he can possibly win. He'll find out first hand how powerful Microsoft really is. He would be wiser to develop for an open platform, and beat Microsoft from the side instead of from the top.
----- rL
The main point is not that there is no need for OS's, only that his UI will be abstract enough to where you don't need to know the desktop metaphor to use the computer. He believes that organizing information according to the 3 dimensions of time (past/present/future) are the most "intuitive". Where I disagree is that people don't always look at the world in terms of time's dimensions, but in terms of essences. For example, we all supposedly believe that we are the same particular person throughout time (I'm still me, just like I was yesterday), so time is irrelevant in that context/sense. If I'm typing a paper, I don't want to have to look at the "story" of how it was 2 days ago, I want it to be permanently changed by saving my modifications. If I want to know how it was 2 days ago, I would back it up. The desktop metaphor doesn't have to be dead just because it's not a brand-spanking new fad. Most problems with the metaphor come from not being consistent with the metaphor: since when do I have to throw a book into the trash before I can take it off my desktop? (hint, hint, MacOS!)
P.S. Linux rules ya mutha! (sorry, couldn't help myself...)
What other innovations are we missing until Microsoft deems it 'time'?
I'm no Microsoft apologist, but I really disagree with that statement. It was the market that decided that pen computing wasn't ready when the Apple Newton was around. Later the smaller and cheaper Palm Pilot took the market by storm.
My point is: the Newton was not unsucessful because Microsoft did anything to kill it. Palm was sucessful without any help from Microsoft. Microsoft is not the technological super-being who dictates who survives and who fails in the market. Some of Microsoft's own "innovations" have failed quite spectacularly, Microsoft Bob anyone?
Addmitedly, Microsoft has the significant advantage of having an awful lot of money and commanding quite a bit of media attention whenever it does anything. Will this be enough to make notebook sized pen computing sucessful? Only time will tell. But blaming the failure of a particular item, this "Pen for Windows" which you lack any details about, on Microsoft, instead of on the people who failed to develop it well, or market it properly is just senseless MS-bashing.
-Spyky
If you tear the pages out of that book and plaster Ben Franklin all over it with green ink, I can assure you, the smell it changes to is much more comforting.
slashdot: where everyone yells sarcastic metaphors to themselves to understand the issue
Bill Gates from the article, "Why are my document files stored one way, my contacts another way and my e-mail and instant-messaging buddy list still another, and why aren't they related to my calendar or to one another, and easy to search en masse?"
...
Hmm, let's see
Because your documents are different than your contacts which are in turn fundamentally different than your email which are in turn fundamentally different than your buddy lists. Yes, they are all data. So are books, CD's, audio-tapes, which are all quite different. Both in terms of media and interaction. This is such a silly question. Why on earth would you want a 'Universal' viewer for your data...?! This would necessarily lower it to the common denominator, but then it is the differences that make these kinds of data useful.
The computer he seem to describe would be able to pull up the information based on what you wanted based on a request, not on some method of searching for a file.
Or perhaps this can be rephrased. What are files, and how do you interact with them? This is what makes the Reiserfs filesystem so interesting. They are thinking about these kinds of issues in unconventional ways. Reiserfs is not just another journaling filesystem. Moreover, Hans Reiser is subjecting his ideas to the test of reality by producing tangeable testable results.
--Lawrence Lessig for Congress!
Are computers really going to change that much in the future? Any technology goes through an s-curve of innovation.. Then it levels off. How different is you car now than from 50 years ago? Although there are changes/improvements in ergonomics/pollution/reliability, the format is the same.
Do we really need a change in the basic desktop format? Why would I want a 3-d desktop? The fact of the matter is what is now available enables 99% of the users to do what they want to get done. The problem with computers is finding problems to solve that fit the computing parigdigm.
love is just extroverted narcissism
Not only is that line a joke, but the whole article is a joke!
The headline for the article does not make sense because the body of the article clearly states that the OS *does* matter and that OS should be Windows! Did we miss that?
His argument is that Microsoft has won. Give them your money and allow them to do what they wish to your PC, just use this guy's document management system to view content -- which, if you did not notice, requires that you run *all* MS software (Outlook, Word, Excel, etc.)!
What a crock! Who the hell needs this *individual's* document management system if we are all supposed to use MS software? We would already have a universal system by virtue of *everyone* running the same software!
This guy is just trying to make money the easiest way he can.
*sigh* No, NOT GREP. You'd have sets of relations on things and you'd say, i want all these types of items that involve a certain relation. And I don't mean a filesystem relation but, all things related to project X, or all things related to my entertainment centre.
Think beyond files and cli. Think about having a bunch of things that have relations, ideas and what not that can be search on. Yes, grep might come into play on the lowlevel, but i'm sorry, my mom won't ever know grep.
-
ping -f 255.255.255.255 # if only
From the article:
"We built our system on Microsoft Windows because Windows is a reliable, solid, reasonably priced, nearly universal platform - and for the software future, "universal" is nonnegotiable. We need to run the system on as many computers as possible and manage the maximum range of electronic documents.
Each is nonetheless still solid enough to be a good, steady platform for the next step in software. But Windows is the marketplace victor and has now won a decisive legal imprimatur. There is no technical reason for us to move to Linux; why should we switch? Why should our customers?"
I think it is amusing that David Gelernter, the author, is clamoring for a new paradigm in the way that we look at information yet buttresses his argument for using Windows as a platform by saying Windows is the old paradigm. I guess my best answer to Mr. Gelernter as to why he should port to Linux is "don't." If it is a good idea, we will. And if the information is really entirely removed from the operating system with which it is stored, then this is merely a matter of implementation. If it isn't, and we can't, then you really haven't done what you've set out to do, have you?
My
Limekiller
I'm pretty sure that files will eventually go the way of punchcards. A file is just a spot on disk you can write to. When it comes to real data, nobody really keeps data in files anymore. It's all in databases. A file doesn't give you transactions, concurrency protection, or easy backup through replication. A database gives all this and more.
One of the things that we can curse microsoft for is giving a bad name to an operating system (Windows) with an integrated database (the Registry). The Registry is a horrid implementation of a rather good idea.
Maybe the state's highest function is to grind out insoluble problems. (Zelazny, Hall of Mirrors)
True, "transparent" is a better adjective. A little clearer would be, "the choice of OS is irrelevant."
However, in the case of the email program, I would argue that the filesystem is irrelevant. The exact same program could be implemented using hierarchical directories/files, a database, or a raw disk. No filesystem is necessary.
In the case of OS, it's difficult to say that an OS is not needed, since an OS is, by definition, an abstraction layer that makes the hardware accessible. Any abstraction layer accomplishing that job can be called an OS. So even if a program is written directly in ASM for a computer, has its own bootloader and never touches a "conventional" OS, chances ar that there's some layer of the program that serves as such.
But even so, on a single-purpose machine, the OS might be considered to be irrelevant - consider the case of Palm/Palm OS.
Solution to blink tags: wrap them in another blink tag, with a javascript delay loop, so they cancel each other out
What he really wants is ultimate coordination of information via the super human intellect. The OS is just and OS. It is typically always transparent. I never see my OS actually writing files, managing memory, other than through the basic application management interface that allows me to install/remove applications. The applications themselves are what matter. (Otherwise the copying of files continually from one directory to another would be fun and using more memory is fun... yay.... i... *DONT* think so).
Our current society is based upon the capitalistic market forces to solve supply and demand decisions because the socialistic models do not work. (Well, thats debatable. The socialistic models under a massive oppresive soviet system doesnt work, but I believe in some european countries it works?).
Typically, this capitalism vs socialism brings up many fears of the coldwar when one tries to migrate towards socialism. There is inherintly nothing wrong with the socialism paradigm. The fundermental flaw lies in the inability for the human mind to organise and coordinate on such a massively perfect scale the logistics behind a truely utopian socialistic society.
Bringing this into relevance, one must consider the perfectly centralized vs haphazardly distributed nature of information.
It is obvious that if all human information stored by companies, governments, scientific, personel information, etc, were to be stored in a single storage system, then it could be perfectly sorted, arranged and the searching of the information would be maximal and perfect.
This provides two problems:
Who is going to search, store and coordinate said massive volumes of information? (Future AI)
Ethicly, do you want your entire personal details and history to be stored in one accessable place. This opens up abuse by authorities (Q:Why would authorities want to abuse others in utopia?). There is a fundamental loss of freedom in the contiual tracking of every move of individuals within society. How do you know your location is being constantly tracked.
Information is currently decentralized across the many industries. The information revolution started to occur with the internet and search engines. The next information revolution will occur with the perfect coordination of information but will society accept it?
-Tim
It seems pretty clear after reading the article that he is talking about a file system (perhaps directory system aware of content formats) that would be interesting if it could work. Want to find a document? Type in some key words and there it is. This is certainly an improvement over Explorer, and a definite improvement over *nix search methods.
Bringing some automagic searching capabilities to my desktop would be a good thing. Trouble is, to get his article published he had to color it with grounded opinions on the anti-trust trial and Linux, which it appears some on
Because of this, his ideas have to be attacked, but they seem worthy of an attempt and relevent.
Ed Barbar, President and General Manager, Furnit USA
One of the secrets to Microsoft's success was to de-commoditize the OS; remember the hoopla surrounding Windows 95, the songs, the celebrities, the massive lines outside of Fry's? They turned the win32 api into a gala multimedia event. It was so successful, that it took this Gelernter chap to remind us that all an OS is is a hardware abstraction layer.
;)
Point being, the OS is *supposed* to be invisible to the user, and nearly irrelevant.
This is why Netscape had to be crushed -- they wanted to make the browser the platform.
This is why Java had to be crushed -- they wanted to implement "write-once-run-anywhere". (There is a whole career-field of experts dedicated to figuring out why a HARDWARE VENDOR like Sun would push this, but that's a different post
I'll go out on a limb here and suggest that the Be OS idea was right... store everything in a database, potentially allowing any number of front-ends. Let's separate data from the display layer, and let people run their "Windows skin" or "Unix skin". Why not?
I'd like to see an ANSI standard Operating System. Hmm...
Final disclaimer: These are my silly ideas. Please treat them gently, as they are only half-baked.
This guy is just trying to make money the easiest way he can
Yup. Lets look at the story submission text:
* Yale Professor of Computer Science
Yup, Yale sure is CS central
* Unabomber target
This is his selling point for himself?
* Operating systems are relics of the past
subtext: No, I didn't just rip off Gosling's old speeches that failed to come true...
* We should be able to access data anytime/anywhere
This is such a fricking overused and pointless buzzphrase in the tech world that it's ridiculous. It's like a politician's "what about the children".
* seeing a stream of 3D documents
Yeah, everything gets better when it's 3d. Uh, huh. Very 90s.
The man is obviously an ass. Given the "irrelevant OS" and "3D documents" crap, it sounds like he harvested a bunch of random mid 90s research papers.
May we never see th
David Gelertner doesn't know what the hell he is talking about.
I had him for a class at Yale (got an A, so I'm not bitter). He was going off about his journaling os or whatever the hell it was (sorry, it's been several years). It was SUCH bullshit. Everybody in the class basically signed up to see this semi-celebrity professor, and everybody more or less had the same impression. The guy is a complete fraud.
Look, I don't mean to sound insensitive, but the Unabomber thing was probably the best thing that happened to his career. I mean, he sits there making pie in the sky os predictions not more complicated than any first year cs student could make and acts like the sun shines out of his ass for it.
I don't dislike the guy personally. He's a bit of a slob, but he's ok. it's his really bad academic work that I take issue with.
OK, finally we have had someone to post who has actually used and can competently describe this concept (damn /. effect).
:).
Basically, what I think most of the meta-physical, quickly tangential arguments that are going on here regarding this theory is leading to an unasked question: What exactly does this system index in each file? Dose it simply index the modify/creation times? The owner/group? The actual data depending on file type? What if my mp3's have no id3 tags? Does it then have some database schema that get queried somehow? How does a query work, exactly?
These are all questions that have not been really asked (I haven't read through the noise of all the other declarative posts).
This seems similar, IMO, to how sports casters can get the most obscure facts and related them in obscure ways. It's like:
"Well, this is the second time in history that a [basketball] player has missed his first free-throw and made the second after scoring more than x points in a row. The first player to do this was..."
OK, this example is a bit contrived but I've heard them relate even weirder connections.
It seems like the idea is to gather as much data (stats) as possible, to have a very efficient (read: normalized) database schema, and to be able to have a powerful query language (SQL, whatever) to relate or combine that data in any concevable way.
If this is the doctors idea then I don't think it's that "new" (he's aparently been a proponent of this for a while now), but maybe he could learn something from Monday Night Football (thats the oblong pigskin game like rugby, for all you people outside the US
why run from Vincenzo?
You really shouldn't judge a professor by his New York Times article. It's not a technical article, and it's not intended for the tech-savvy Slashdot audience. It's supposed to 'wow' your average computer user or perhaps drum up some buzz so that he can get better funding. Just because he's a dreamer doesn't mean he doesn't have any idea what he's talking about.
Also, your sarcastic comment about Yale being "CS central" is way off mark. It's not a top-tier school (I think they're ranked in the 20s or so for graduate school), but it is a strong program and they have several really good people there.
I don't agree with what he's saying, either, but that doesn't make him an ass. If you want to make a fair analysis of his research, you should maybe check out one of his research papers at citeseer.
> "The man is obviously an ass. Given the "irrelevant
> OS" and "3D documents" crap, it sounds like he
> harvested a bunch of random mid 90s research papers.
I agree completely. This is just more "the network is the computer" snake oil, repackaged in a "you're not cool unless you agree with me, if you're cool you'll 'get' what I'm saying" kind of rap. If I remember correctly, Sun tried this line of BS for years, and everyone -- I mean everyone -- basically laughed them back to their senses.
Here are my problems with this Gelertner guy's ideas:
1. First, he pins a large part of his argument on the idea that hierarchial data storage (i.e. directories and files) is somehow antiquated, silly, and inefficient. However, a couple of hundreds of years of information processing using increasingly refined methods (first using paper, pen and ink and then, in this century, increasingly automated means) has demonstrated the power and effectiveness of the paradigm. One could point out that the WIMP interface and the filesystem/directory/file paradigm represent the highest expression of human organization of information. One might also point out that SIX THOUSAND YEARS AGO the Sumerians were using a similar system to record transactions using cuneiform and clay tablets. Virtually every single human information processing system since then has incorporated the basic idea of a document (scroll, folio, parchment, clay slab) stored in a bin (basket, crate, locker, book, file cabinet). And, all of these methods were in some way hierarchial.
One can even make a case that the "file cabinet" paradigm duplicates internal, instinctual human methods for organizing ideas. Consider: we think of things in sets and subsets, don't we? A baseball is a type of ball, which is a type of sphere, which is a three dimensional shape, and so on... But you see what I mean.
2. Let's consider his basic idea: organizing documents and files on a timeline rather than a hierarchial system. So, you'll go into his 3-D viewer, and zip backwards and forwards in "time" to see documents and files you've got stored. Now, you have to remember WHEN you've created a document rather than what it's about. How is this supposed to help you locate information? There's no way around it; he is going to HAVE to provide some sort of hierarchial organization or his system will be completely useless.
Again, humans think about information in terms of sets and subsets, and we understand new information in terms of information we already have -- placing it in directories in our heads, in a way. Building a system that works differently will make it unpleasant for humans to use, and eventually, human-type modifications will be made leading back to the current paradigm.
He could incorporate a database which would let us look for related information, but this still wouldn't be as good as a hierarchial structure, because it wouldn't let us get a "top-down" view of our data, drilling down by layer. We'd be limited to the searches we're capable of formulating. See what I mean? If he tried to beat this by creating broad category based searches, he would eventually end up with something very similar to the dreaded directory and file approach, albeit in a virtual, database-ish form. In my view it's inescapable. It keeps coming back to the way in which we think.
3. FINALLY, getting back to the "the network is the computer" problem, people just don't want to have their data on some remote server where they can potentially get locked out of it. NO ONE is going to buy into this sort of thing. People want strong, general-purpose computers which STILL WORK if the cable goes out or they forgot to pay their bill that month. They're just not going to enjoy the dependence on remote data stores Mr. Gelertner thinks they will.
Anyway, this is just my opinion. But I think his ideas are a bunch of hooey.
Farewell! It's been a fine buncha years!