Slashdot Mirror
Blake Ross Working on Parakey Web OS
prostoalex writes "IEEE Spectrum is running an article on Blake Ross, creator of Firefox, and his new project called Parakey, which will bridge the gap between Web and desktop operating system. From the article: 'As he describes it, from a user's point of view, Parakey is "a Web operating system that can do everything an OS can do." Translation: it makes it really easy to store your stuff and share it with the world. Most or all of Parakey will be open source, under a license similar to Firefox's. There are differences between the two projects, however. Although Ross plans to incorporate the talents and passions of the free-software community, he's building Parakey around a for-profit business model. And he's leading the charge with a simple battle cry: "One interface, not two!"'"
How is this an OS? An OS manages the hardware and software resources for a computer. Is this just a virtual filesystem?
"Translation: it makes it really easy to store your stuff and share it with the world."
A P2P OS.
I've always thought you should be able to write interrupt handlers in Javascript.
Why must we have tools that try to do everything?
I remember hearing about some guys named Brian and Dennis and uh I forget the third guy's name - it was back in the 60's - trying to write an operating system based on the idea that each part should do one distinct thing, and do it well. I don't know if anything ever came of it, but I thought that it sounded like a good idea.
There is a major distinction between MY computer and the rest of the world. One is mine; the rest belongs to others. I treat them differently. I want my desktop to reflect it.
There are already too many people who seem to forget that my stuff is mine - spammers, politicians, cold callers, door-to-door salesmen, etc - and that I might want it separate from the rest of the world. I don't want my OS forgetting this too.
The first OS made entirely out of DONGS!
they want their lame idea back.
How we know is more important than what we know.
Nothing is said about he is planning on monetizing this. Any ideas?
Anything that makes it "really easy" for me to move/save/delete files while online from any computer means that unless you're amazingly careful, you're also making it that much easier for someone else to do it for you.
Maybe I'm just paranoid, but I have yet to see *any* vendor, be it closed source or open source take enough time and care with their code to write something that doesn't have gaping security holes in it.
What's going to happen when what was a simple browser problem becomes a file system problem? Drive by downloads that wipe your machine.
I say butter.
He is writing iPhoto/iWeb for windows?
a simple battle cry: 'One interface, not two!'
Of course, when MS - also seeing a change in the traditional boundaries - wants to embed a browser in their own OS, and make poking around the local file system feel similar to poking around web sites... that's the battle cry of... Teh Evil!
*sigh*
Don't disappoint your bird dog. Go to the range.
This sounds great for the vast majority of web users ... people who want to create blogs, picture pages, keep notes, network with their friends, use e-mail, chat, etc. Calling it an O/S is a bit of a stretch since it doesn't perform any hardware/software control on the computer or the server, it simply comes with an application to facilitate file manipulation/moving/sharing/tagging/etc. Sounds like MySpace meets e-mail meets Flickr, on steroids.
BTW - if you'd like to get more information on this product when it launches, you can get on their mailing list or just bookmark their site at www.parakey.com
Crack - Free with every butt and set of boobs
Including running like molasses and then BSOD!
But seriously, is this just another one of those "desktop in javascript" things? They've been done a million times, and they all suck.
I wonder how this project will manage to give a "web os" more power over hardware while not simultaneously throwing our security in the "web recycle bin".
IE had too much power over the OS and it caused problems. Firefox and IE7 do more to put some distance between the os and the web for good reason.
...but I just don't get the idea of what a "Web OS" is. It's still going to be a bunch of code running on a physical webserver running a traditional operating system, right? I mean, doesn't that make it just one big web application?
Why would you want to run an OS from within a browser, in your own OS, and then surf the net from the browser in an os which is in your browser on your OS!!. The idea is ridiculous. How about just making a "functional website" rather then glorify it by calling it an OS. I'm sure if you had enough time you could do the whole thing in flash. It stinks - as much as the idea of Office live.
Am I the only person appalled by these web interfaces, or even web desktops, being referred to as operating systems? It is technically wrong by a large margin. An operating system is the interface between hardware and software that manages the resources of hardware. Web "operating systems" do not manage any hardware.
I find this usage appalling, and I hope that this terminology doesn't spread and dumb down the use of technical terms.
It's hardly an operating system. It's more like trying to combine all your data into one browser window. It is a unique idea - no application I'm aware of centralizes this functionality. Why noone has thought of it before, who knows. It seems like the only reason I'd want to use this would be to share files with other people. It wouldn't seem like it would be a big deal to write an application that showed you the files on your computer, give you the option to post them online to any web server of your choosing, and format them nicely with a web portal. Even filtering the site so that different sets of people would see different data would hardly seem to be a problem, as you can email a 'key' to them that sets their level of access.
As others have said, the most important thing to worry about would be security. What will prevent malicious code from altering your local files, or uploading sensitive data?
Sounds suspiciously like WebDAV, only with less brain-damage than Microsoft's implementation of a client for it.
.Mac services use it.) It has a lot of potential.
Personally I think WebDAV should get the "Internet's Most Unappreciated Technology Award", in terms of having a lot of promise but being seldom used. (Although Apple does drag it out every once in a while; I think the
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
Check it out eyeOS, its open source also :)
And why are some web devlopers so obsessed with the OS model?
Because deep down, all web developers want to be OS programmers?
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
How does this differ from You OS? Previously reviewed on Slashdot
-- Brought to you by Carl's JR
It can be more useful than a brick when the network connection is down? No? Then it can't do everything an OS can do.
Bogtha Bogtha Bogtha
I'm well aware that a "web operating system" would not fulfill the same functions as a true web operating system, and I'm as tired of the "WebOS" rhetoric as anyone else. I did explain this to Spectrum, and it seems the magazine decided to leave the mention but explain that it's only an "operating system" from the average user's perspective--which is difficult to prove either way, since my mother probably thinks an "operating system" is some kind of surgical device.
As for the "how is this different from XXX?" comments, I understand that it may be difficult to differentiate Parakey based solely on the description provided in this early article. Rather than chase those sorts of questions here, I'd rather continue working towards putting the product in your hands so you can decide whether it's different and, ultimately, whether it's worth your time. Thanks everyone.
Yeah. The source is free (at least the client component), but the service to hold your files on the web will cost $$$.
.Mac
.Mac, both in terms of the business model (client is free -- in Apple's case it's included with every system -- but the server space is what you charge for) and some of the functions (document storage, integration with desktop applications, web services). Perhaps what he should be more interested in are the rumors I keep hearing about how Apple is going to pull the plug on .Mac any day now...dunno if it's true, but I heard the service was going to stop being for-profit and start being a free service coincident with the release of Leopard. Course, I can't find any substantiation of that now.
Seems like it shouldn't be hard, then, to reverse-engineer the code and figure out how to use somebody else's servers as the data repository. Unless he's planning on doing something sneaky/evil, like using encrypted binary lumps or something. Even then, if it's really that neat an idea, people will figure out a way to do it on their own servers.
Think:
Agreed; the whole thing reminds me of
There seem to be a dearth of good (by which I mean, tightly integrated) end-user client programs for accessing remote volumes over the Internet from Windows clients; if all this project ever amounts to is making a nice interface for Windows users to manipulate files on a remote FTP/HTTP/SMB/NFS/whatever server, then it might be a nice thing to have. I wish him well, I guess.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
I just checked it out. It's very nice. I've no use for it, but it's a nice implementation of the idea.
I think this is another solution to the same semantic problem. That is how do I integrate my real digital worlds into my real world. I love my Mac G5's interface but it's not terribly portable. I love the remote graphics of X11. I love the lightweight powerful interface I get through a CLI in ssh. The problem as I see it is there are way to many solution verticals implemented without enough development done to mesh them all together. It's all pretty fragile. For me I want the same information availible on my cell phone, a website, a login, a desktop, and to those ends I have a myspace page, a flikr account, a linux box running a website with my blog and photo gallery and NX Server, a Laptop, a G5 and a G4. The all have pretty much the same information stored redundantly. It's way inefficient, since I'm duplicating processing power, storage, and memory just for different access features, it tends to break and it's difficult to keep in sync. From what I can tell of this parakey.com it seems like a good idea but it sounds like another vertical framework. Until there are more horizontal interfaces designed and integrated into these personal computing verticals we're going to continue to reinvent a square wheel.
Breaking compatibility took care of that problem for them - why is it different today?
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
What makes Parakey different from Desktop.com and WebOS, two companies that looked like they had a decent future during the internet bubble? Desktop.com had a full operating system inside Netscape 4.5 and IE 4. You could share file, folders, etc. with your friends, use online "AJAX" applications, etc. WebOS had a very slick interface, too.
At Desktop.com, we had a couple of problems making the product fly. The first was technical: it bloated the browser up to 32Megs and made it unstable. (Nowadays, I don't bat an eye when Firefox is hungrily consuming 250M.) The second was usability: the online experience just wasn't as smooth and easy as a local application. This was partly due to the lower connection speeds that people were using back then, but also due to the inability to seamlessly interact with local files. And the really big problem was the business model: you either have to charge the users, or figure out how to put ads somewhere where people aren't used to ads -- like in the application title bar. Ick.
So, "one interface, not two" is all fine and dandy, but I'll be interested in seeing if they actually make it work and worthwhile.
I've begun wondering lately why the processors themselves can't be extended to handle much of the functionality of the 'OS'- essentially migrate the basics into the CPU? If there was a way to get things up far enough to access some flash memory that would house some signed net/video/input/storage drivers to get things up and running. This is just a very basic idea and by no means is all-inclusive of every possible requirement..
I'm going to start with some obvious information to establish why I think there's a real development going on here.
Historically, computing improvement has been achieved by layering the technology, so that each layer operates with a high degree of autonomy from the layer below it. Depending on your perspective, there are anywhere from 6 to dozens of layers within the computer you're using to read this.
This layering, called "abstraction" by most, has minimized the amount of complexity that needs to be managed at any one point, allowing us dumb people to work together and improve the whole by incrementally improving each part.
Even with a single "layer", there may be multiple internal layers. For example, much of the software I write is managed through a file abstraction layer that takes care of the details of converting a memory object to a file on disk. I do something simple like ($obj->FWrite($object)) and all the rest is managed for me.
Now, on to the point. There are major abstractions in use today. EG:
Hardware ->
FirmWare ->
Operating System ->
Application ->
And there's a new, cross-system development now underway. Technologies like SOAP, XML, RPC, AJAX, and similar, related words describe a new layer of abstraction on top of the Application layer.
It's not a well-developed idea yet, and the foundational principles of this idea are now being explored. Yes, there are definitely specific implementations of this, but just like the Operating System developed after decades of exploration in designs, this next abstraction hasn't been well defined and/or commoditized yet. So far, any development in this area really requires a very specific implementation - code reuse is minimal.
IMHO, the best implementation of this new abstraction is probably XML/RPC. But it's honestly not much more than a transfer protocol.
So, yes. There will be a "Web O/S" - though we'll probably not call it such. It's closest cousin is called "Middleware" by IBM.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
All they have to do is get Google to buy em.
Windows RG Edition
"I bless every day that I continue to live, for every day is pure profit."
Ah...the classic "Get your product/service made for free and then sell it for profit" business model. Best of luck to people who work for this and don't get compensated for their time and efforts.
Buy Steampunk Clothing Online!
I think it might be comparable to using wordpress or some other CMS vs. updating your site's content in the html source without the benefit of a sql database.
I'd much rather my mom be trapped in a user friendly and productive content management system instead of nagging me every two minutes to do something for her.
For the average person all popular OSes are too complicated.
Does grandpa really need the contents of c:\windows\ showing up when he searches for one of the few files that he himself created?
I think a lot of people would feel comforted knowing that something like parakey is sandwiched between them and the 'real' OS and they don't have to worry about possibly destroying their computer some how.
Sure, this isn't a good fit for most people like me with our 1800 php files, 3000 html files, countless psd gif jpg, mb, 3ds, wav, ogg, etc...
but we're not the "average joe". One of the main reasons IE is so damned popular is because a large portion of computer users ONLY use their computer to update their profile page on some community site, check their mail (ever had to help someone set up outlook over the phone, people hate all the steps and foreign terms),
and search for information.
When last did you see somone buy and encyclopedia on cd/dvd? stuff like Wikipedia is faster and easier and like so many things on the web it's straight to the point.
Sure other people have attempted something like this in the past maybe. But maybe their implementation sucked. Maybe this won't suck so bad. With so many examples online of just how useful a good CMS can be, I'm sure they have a good chance to get it right and make something good.
I work with Blake on Firefox. As one of the few people who's actually seen and used Parakey, I can tell you that the assumptions being made here are misguided. It's a unique product that surpasses anything similar I can find out there today.
The article referenced does a poor job of explaining what Parakey is about and an even worse job of describing how it works. It won't be long before you all can see for yourself.
- A
I don't know how different it is from YourOS. I kind of like the idea, though, even if it's not the unified solution many of us want.
You are welcome on my lawn.
The idea of a "web operating system" sounds very 1990s to me. It doesn't really sounds the least bit interesting these days. On the other hand, I don't understand why linux distros don't take more advantage of things like wikis. I'd really like to see linux distros become more integrated with wikis - error messages could have links to wikis or forum posts, control panel applets could contain links to editable howtos, etc. there are some security issues but nothing impossible.
Buddy, you will soon get more women (and sex) than you will know what to do with. Do your best! You are fucking on behalf of well over a million Slashdotters, the vast majority of whom can only dream getting jiggy with the ladies.
It's just like an intelligent but illiterate professional dancer talking about how he has worked out there are 35 "senses" because he hasn't listened to anyone long enough to find out that the word perception exists. It appears that many are spinning different definitions of existing terms to profit from confusion or due to simple ignorance or lazyness. My instant reaction to this usage is to treat anyone who uses it as ignorant and gently correct them like those who are misled into thinking that linux is part of a gnu operating system - while the gnu operating system is the hurd.
An apologist could argue that it is something that operates a system of web links or something and not a computer system - but of course my instant predjudiced opinion of people who radically change meanings of existing terms and loudly proclaim them is that they are some sort of not paticularly funny clown to be ignored or somebody up to no good.
...go right ahead.
In the article, you say you would like to open source all of it, but...Well, why not stick to your guns and keep wading through the vulture capitalists until you get what you want? Hold out for an angel investor instead.
I am inferring from the article you are primarily a windows user, true? And follow up, although you say it will be browser agnostic, will windows OS be the primary dev platform?
And money, this will be subscription based or ad driven? That appears to be the only two viable ways other than burning up startup capital, so which is it? And when will your alpha (or beta-whatever) code be available to people?
First of all, SOAP _is_ XML RPC. And AJAX is SOAP launched from a web client by javascript (as opposed to a piece of middleware).
Everything you're talking about is using XML and the web CGI model for transactions. This technology exists in MULTITUDE other forms. Let me see, uh, RMI, CORBA, ringing any bells? How about jabber? If I scripted interaction with a jabber client using Lua for driving voice/text prompt tree navigation or something, would we make some new acronyms for that?
What you're talking about is still client/server, but dealing with application platforms and interpreted code (specifically using web servers for arbitrartion, since everyone seems to like those, and they can be the basis for some nice frameworks).
I don't think it needs a new name. We've been doing it all along, but we were programming to APIs. Now we create platform independant artifacts that depend on other software to act as an interface, providing a runtime (although at times I wish javascript had a JIT compiler too). You are simplfying the distribution aspect of your software product... if you get can to the website, then you can start using it, oh and please enter your credit card number here.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
... that the internet works and you can get to a browser.
Big assumption to make in an OS. Better to have local documentation that is thorough.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
Ok fellas, lets migrate all our systems from Windows, Mac and Linux to this new uber OS.....hmmmm... where are the install CD images?...... what are the hardware requirements?.... are drivers available???.....oh wait!
Come on guys - we are not all in the first year of high school and no good at anything but football here. Silly lies to children to dumb things down about computers are not necessary - statements like the above are more likely to intially confuse people into thinking a qemu window is running in a browser than what I assume you really mean from the rest of the context.
"Personally I think WebDAV should get the "Internet's Most Unappreciated Technology Award"
Too late, IMAP already won that one.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
The only thing I've got to go by is the Spectrum article, so I'm going to read into it *a lot*. So if my assumptions are correct you should be very excited by Parakey for the following reasons:
For profit or not, this is a great project.
At least the guy is upfront.
Not like numerous other characters who release their stuff under GPL, all the while reserving themselves (and only themselves) the right to sell it commercially to people under other, less restrictive licenses.
Let the flames pour down on me and my petrol-impregnated shell suit but this is of course why BSD is the TRULY open source license - once you're released it under BSD you really can't forcibly make money off it, except by providing stand-out support.
[x] auto-moderate all posts by this user as insightful
Yeah, calling it an OS is a bad idea, but these do have a purpose. With the whole Web 2.0 thing, we're basically seeing a move back towards thin clients. All you really need on a machine now is access to a modern web browser. From there, you've got email, IM, FTP, SSH, you name it. This basically looks like they're just rolling everything together into one bundle of Ajax apps. I'd imagine it'd be handy in an enterprise setting.
"Knowledge is power. Power corrupts. Study hard. Take over the world."
I really dislike the use of the term "webOS", especially for a product aimed at the average user. Mom and Dad probably don't know what an operating system is or what it really does for them. Geeks, on the other hand, will just shake their head at the misuse of th term. In short, the term does nothing to describe the product.
That said, I think Parakey does seem interesting. Even though I'm a geek and more than capable of figuring out how to publish things online, it would be very convenient to have one place to store various types of media. I think the concept of sending virtual keys to your friends is a good one as well. There have been many times when I've wanted to share a file with a friend or two, but not the public at large. Despite knowing how to use many tools, it is still very difficult to do this. I think that with the right feature set, Parakey could take off with the geek crowd despite being targeted for average users, much like Firefox has done today.
SIGFAULT
Is that the Internet isn't fast or reliable enough for the most part to act as part of a computer. A major distinction between local and remote is that all local data I have immediate access to, and I don't worry about losing it. That's a major difference from any remote data. Though my link to the net is somewhat fast in relative terms, 6mbps, it doesn't even approach the slowest local links. My mouse has a faster connection than that. So things I get remotely I have to wait longer for. Also, I find that the net is rife with unreliability. My connection can die, or the connection to the server, or a link in between, etc. However the data on my system is quite secure. For the most part, if the system is operational I have access to my data.
Thus to try and blur the distinction is not useful because there is a real distinction brought on by the physical limits of what the Internet is. I suppose I could play make believe like it's part of my local system, but I'll quickly notice the difference.
Thus I find it useful to have it as a distinction, and to have different kinds and level of access to things. I do not treat access to files on a remote site the same way I do on my hard drives simply because of speed. Thus it wouldn't be useful to present them in the same interface to me.
I thought that last sentence of GGP was kinda elegant. Ignore GP; he clearly has no poetry in his soul.
Instead of flaming, I have to agree with you.
BSD is the only one really open.
We are Turing O-Machines. The Oracle is out there.
... what about plain old privacy? Would anyone really want to do this with their files? For a minute let's not get into how it's stored online, and what kinds can you store.
But the simple fact that you can store all kinds of files online (and potentially access them - something similar to google docs or such??) brings int he question of privacy too. Security implies deleting, overwriting, and accessing (ofcourse), but would anyone really want to put files online which they truly hold dear? What if we have someone, or a bot, trawling through our files for information, numbers, etc?
I guess this would make big business for anti-virus/internet security products, who now not only need to protect home PCs, but also online "OS"s of sorts! I guess that's where the revenue sharing/for-profit model partially comes from.
Usability Engineer, Master in Human Computer Interaction
For a really impressive web OS, try YouOS. It works in several browsers (I've tried Opera and Konqueror), looks and acts like a desktop OS with GUI, has its own API, and it's open source.
Please correct me if I got my facts wrong.
What you are doing is good, but here is an idea that is even better:
Instead of making a Web O/S, why don't we forget all about Javascript, HTML, XHTML, web forms etc, web services etc and make a distributed computing platform that allows us to easily program distributed applications? enterprises and even web users do not really need only a shared file system, but a way to create distributed applications as well. The current situation is disappointing at best: lots of 'standards', lots of bad implementations, and great difficulty to even make the most basic of distributed applications. Even a distributed 'hello world' takes a tone of coding, xml files, web server configuration, and ultimately user frustration.
The distributed application platform would consist of the following components:
-a programming language that is distributed, lazily downloaded, scripted and compiled, statically and dynamically typed, object-oriented and functional, able to modify itself and to express tree structure (ala LISP), that allows meta-programming so as that language and project configuration happen at the same level.
-a virtual machine is responsible for executing the language, managing linking with the libraries on the local computer and accessing the resources of it with security as the first concern.
-a 'browser' which acts as a host for the virtual machine mentioned above. The role of the browser is to provide a URL box and space for the downloaded programs to run their GUI.
-a set of libraries that include abstractions for a GUI, networking, distributed storage, remote method invocation, model-view-controller etc.
What would the above solve? well, it would make the 'web' a truly interactive place, where anything can be coded with a library. The real drawback of the current standards are that they are not Turing-complete, they are simple passive descriptions of predefined functionality, and we all know that no one can anticipate all the needs!
Your Web O/S effort then could be reduced to a set of applications running in this platform.
Although Blake is certainly out front in Firefox land, I didn't think he was the creator, at least the sole one anyway.
I thought that Dave Hyatt ( now with Apple ) had quite a bit to do with that.
Who are the creators of Firefox?
a man, a plan, a canal, panama
"This has nothing to do with fame or fortune; it's about improving the experience for things we do everyday."
Oh really?
IEE Spectrum Articles, Blog, Wired Articles == attention whoring (e.g. fame)
For-profit business model == money grubbing (e.g. fortune)
Nothing wrong with wanting recognition and some $$$ in pocket for your hard work. Not everyone's an introvert or RMS. Plenty wrong for trying to pretend otherwise.
You're doing plenty good work. Don't taint it by spewing out blatant lies.
There were these other guys in the 70s (curiously enough also named Brian and Dennis) who wrote an operating system that treated everything as a file, and files as streams of characters terminated by an end-of-file character. This let you do cool stuff like I/O redirection and pipes very easily. Their operating system was called "Eunuchs" or something like that, and it had some pretty damn good small focused apps in it. That wasn't the point of the system, though - the system was a reaction to the prevailing culture of separate drivers for everything instead of a simple generic file paradigm for all I/O.
Of course, eventually Emacs and tin were written, and that was the end of small focused apps for all practical purposes. Nowadays people think perl is elegant, and most operating systems are written by penguins.
Personally I really like the idea of Parakey. I've been thinking about how to do similar things for myself and my family for a while now. Being able to give myself, my friends and family and the public different levels of access to my data is something I want to do. I already do so via my own web pages, but in many ways it's clunky and inconvenient and I don't have a solid way of making sure the copies of my data in various locations are kept up to date which Parakey may be able to help with. Also, I can see my parents using it for photos that they want to share with the family.
However, I'm not interested in storing my data on external servers (Parakey's site). Some of my data is personal and sensitive and should remain only on servers I run myself, though I would still like to be able to access it at any time. I hope that Parakey offers the capability of running your own personal server which can be accessed from the outside.
Steve
Although I would hope that any local documentation store (offline copy of a wiki, for example) would be formatted in such a way as to be viewable from same lynx or elinks. This shouldn't be too hard, I know twiki can be using the print preview site template, I'm not sure about MediaWiki.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON