Linux In JavaScript, With Persistent Storage

An anonymous reader writes "Remember Fabrice bellard's [Linux-booting PC emulator in JavaScript] ? This modified version [Note: click on "emulator.html" in that directory to see it in action] allows the same emulator to boot the most recent linux kernel, 3.0.4, as well as providing the user with persistent storage. It is achieved by building a virtual block device, which stores data in HTML5 local storage. The block device can be partitioned and formatted as ext2, so it can be easily used."

Comments

uhhh

[Anonymous Coward]

why?

Re:uhhh

[Spad]

Because you can? Because nobody else has done it? Because it's cool? Because it's a challenge?

It depresses me that everyone always responds to these articles with "Why?" and "What's the point?" and "What a waste of time". The whole of human achievement is pretty much the story of people doing things just to see if they can, or because it's interesting to them, or because it's never been done before.

Re:uhhh

[wsxyz]

I and think it's undeniable that, looking back on this achievement, 100 years from now, future men will consider it one of the non-negotiable cornerstones of their civilization.

Re:uhhh

[knuthin]

Because this guy can.

Re:uhhh

[Anonymous Coward]

I wouldn't exactly say that javascript is cool.. Now doing it in Hypercard, on a Mac IIx, that would be cool.. and a much bigger challenge.

Re:uhhh

[DMiax]

why?

To show that we have gone way too far allowing remote scripting facilities in the browser, to the point that websites can run a complete OS of their choice on our machine just because one clicked a link. Or maybe I am misunderstanding the intention? This is NOT considered a good thing, right?

Re:uhhh

[goombah99]

why?

Isn't it obvious? He wants to have the linux in order to run a browser with javascript.

Re:uhhh

[somersault]

What is bad about it, if it's within the bounds of the browser's sandbox?

Re:uhhh

[SlashV]

Droste for the win.

Re:uhhh

[houghi]

You know the reason as to why they respond to these questions with "Why"? Because they can.
There is no reason to get depressed if people have not the same insight or opinion as you have. It will make your life a whole lot easier (and less depressing).
I am sure that some things these people do you will find a waste of time. Should they get depressed of that because you ask them "Why?" or "What's the point?".

The question of why is just as relevant as why not. The only thing it shows by your response is that you have different priorities. And that is a good thing. Remember: We are all individuals. ( ... I'm not.)

Re:uhhh

[Anonymous Coward]

It depresses me that some people think whacking away at a keyboard at some random futility is in any way comparable to an achievement along the lines of climbing Everest or inventing the transistor. If I decide to smear myself in feces and shove chopsticks up my ass and run screaming in the street, *just to see if I can*, that's worthy of attention?

Re:uhhh

[Allicorn]

Honestly, this should make you chuckle and smile and say "Wow!"

"Why" might be in there somewhere but if it's your first port of call, you're a lost cause - hand in your geek card.

Re:uhhh

[circletimessquare]

why not?

it'as easy to challenge something someone does, isn't it? now that you have discovered this, what have you done lately? it's hard to do something interesting yourself, huh?

there's people who dress up like characters from dead tv shows, people who try to grow giant pumpkins, people who bend giant steel beams and call it art, and people who spend years of their lives leveling MMORPG characters. why? why not?

and, i suppose, there are people who try very hard to comment on slashdot forums in the negative, regardless of intellectual or probative value

hmmm, maybe you understand the esotericity of human endeavour after all

Re:uhhh

[Kjella]

The whole of human achievement is pretty much the story of people doing things just to see if they can, or because it's interesting to them, or because it's never been done before.

Most of the achievements that actually push the world forward have been either to impress girls, make money or scientific curiosity. Hobbies don't tend to have the ambition to do any of those things, only to have personal value to you. It'd be a snowball's chance in hell if me watching TV or playing video games lead to anything like an achievement (or well, lately games have been giving me 100s of achievements for random crap). Sports or exercise might get me in better shape, but I'm not about to set any records at anything. I like to develop as a hobby, I think it's cool to be able to control a computer into doing things for me, like some people like to teach their dogs tricks only with less fur and far more obedient.

The only time I go "Why?" is when I feel you're making it extremely difficult for yourself for no particular reason I can think of, like can you scrub this floor with a toothbrush. Of course you can, but like... why? I fully understand when people want to do it themselves, like make their own pasta from wheat even though it's available in the store but not just arbitrarily limit myself to poor tools. Even within a hobby it's fun to do the best you can. Somehow using a javascript engine to run Linux sounds a bit like using the worst possible tool for the job. Yes, it's half "Wow, you can do that?" but the other half is "With a toothbrush? Seriously?"

Re:uhhh

[larry bagina]

That is worthy of attention. More so than climbing Mt Everest but less than inventing the transistor.

Re:uhhh

[Lord Kano]

A Mac IIx, you must have been one of those rich kids. I had a Mac Plus and I fucking loved it!

LK

Re:uhhh

[Lord Kano]

Because this guy's ePenis is bigger than yours and he wants the whole world to know about it.

LK

Re:uhhh

[Anonymous Coward]

OK, well then, keep an eye on Google News Montreal. Now where do I keep my chopsticks?

Re:uhhh

[Doc Ruby]

Most SW development isn't either to impress girls, make money, or even "scientific" curiosity (not eve CS). It's to enjoy doing something the doer has never themself done before, and/or have the results from your own labor. A pretty good amount of SW development motivated by that has pushed the world forward.

  That is a lot different from hobbies like watching TV or playing video games. The difference is between producing something, and consuming something. Very little consumption has ever pushed the world forward - probably none by a single person (and any is vastly outweighed by pushing us backwards).

This particular hobbyist effort might not push the world forward. So what? Its equivalent is "let's make a tiny S-100 bus 6502 CPU into a personal computer mimicking a $100K IBM machine". A few dozen hobbyists doing that pushed the world far forward. The same happened with radio and plenty of other electronic hobbies, and (among the rich) with astronomy, botany and other hobbies become sciences.

And even here the "Linux in JS" effort is by a hobbyist specializing in embedded script engine development. This particular stunt might not push anything forward but their own expertise - or passion. Which is where world pushes come from.

Scrubbing a floor with a toothbrush is hazing partly because it's been done before, which makes it even more pointless - and so more tedious. "Linux in JS" is new, and so less pointless. Even if it is a terrible tool for running Linux. It's a pretty good tool for exercising the imagination, and the skills and commitment to back them up.

Re:uhhh

[bratwiz]

It depresses me that some people think whacking away at a keyboard at some random futility is in any way comparable to an achievement along the lines of climbing Everest or inventing the transistor. If I decide to smear myself in feces and shove chopsticks up my ass and run screaming in the street, *just to see if I can*, that's worthy of attention?

Only if you set them on fire.

Re:uhhh

[MichaelSmith]

Its a great way to run Linux in places where installing Linux is strictly forbidden.

Re:uhhh

[Anonymous Coward]

Maybe you could make a movie about lazy zombies. That would be great.

Re:uhhh

[Mordocai]

It depresses me that some people think running around with a ball, throwing it, and smashing into other people is in any way comparable to writing javascript that can run a linux kernel. Calling programming -anything- "whacking away at a keyboard at some random futility" vastly underestimates the knowledge required to do anything cool with programming. Yeah, I wrote a calculator when I first started programming. There were thousands of calculators out there that were better and more efficient than mine. However, whacking away at a keyboard at that random futility taught me what I needed to be able to move on to creating bigger and better things. I think it would be damn cool to be able to hit a web site and run linux when I'm stuck at work on a crappy windows computer. Get this to the point I have a fully functional bash shell and a package manager, and to where it can compile small programs, and I will use the crap out of it.

Re:uhhh

[segin]

Except that this provides practical future value, especially in a world where "web apps" seem to the the Next Big Thing. It may be cost-efficient to run some legacy client application in a browser using JavaScript virtual hardware to translate, e.g. serial port I/O, into some AJAX or similar request towards an equally-legacy (emulated?) mainframe.

This similarly would allow companies to squeeze additional blood from stones by offering old DOS games running in a JavaScript emulator using HTML5 Local Storage for save games, while keeping the main game executable, bit-for-bit, to that they shipped two or more decades ago on floppy disks (or CD-ROMs).

Rewriting said games for the modern web is not a port. That is a remake, unless you are using LLVM + Emscripten to port the original C/C++ to the web. And it has to be the original game code, not a brand new codebase from scratch. There are some games whose experience would actually be changed by a remake, due to subtle programmer errors in the original code, which for accuracy's sake, must be maintained.

Great!

[Anonymous Coward]

So when the web browser becomes the OS we will still be able to issue cryptic command-line incantations to do things that everyone else has to point-and-click to do!

Re:Great!

[Anonymous Coward]

Mod +1 funny.

Re:Great!

[rubycodez]

I thought emacs was the wonderous OS that does that. It's a crappy editor, though

Whiskey Tango Foxtrot!

[Anonymous Coward]

That's neat and it's clear that this guy has MAD skills. But, why the HELL would anybody spend so much time doing something so very technical and difficult with no value whatsoever? It isn't art, it isn't a literary piece it isn't useful. It is just a monumental waste of very valuable time.

WTF?

Re:Whiskey Tango Foxtrot!

[cryoman23]

aww but it is art. Beautiful wonder full art.

Re:Whiskey Tango Foxtrot!

[Slashdot Assistant]

From the author's site:

I happen to be interested by the implementation of Javascript engines these days - but I don't know yet if I will write my own any time soon ! Anyway, this emulator was a way to learn how to write optimized code for recent Javascript engines, in particular Jaeger Monkey (for Firefox 4) and V8 (for Chrome).

Re:Whiskey Tango Foxtrot!

[niftydude]

Three awesome reasons:

1) Because you can,
2) Because no-one else ever has,
3) Because there are useful lessons that can be learned by performing an exercise like this.

and I'll go ahead and speculate on the fourth and possibly best reason:
4) Because the developer enjoyed solving the problems involved in doing it


The time he spent doing this is probably equivalent to the time you spent watching all 5 seasons of the Battlestar Galactica. I'll leave it to you to decide which was the more monumental waste of time.

Re:Whiskey Tango Foxtrot!

[Svippy]

He got to the front page of Slashdot. I don't think I need to say more.

Re:Whiskey Tango Foxtrot!

[somersault]

I think it's very cool. Technical, yes. Difficult? Probably not that difficult seeing as Linux is designed to be amazingly portable. It doesn't even need the C standard IO library to compile, so really you'd just need to emulate a few low level interfaces for things like memory access, keyboard input and a terminal display. There are a few different ways to implement it, but if he has it running quickly then that would be quite impressive :)

Re:Whiskey Tango Foxtrot!

[somersault]

Having read TFA (well, the README), it's as the summary says and the only really new thing here is the HTML5 storage driver. So my last comment would have been better made on the previous article about JS/Linux. This storage driver is very useful if you want to mess around in JS/Linux, so it was a brilliant idea. It would be pretty funny to see X on this, maybe even with WebGL 3D acceleration.. I certainly wouldn't want to wait around for it to compile natively though!

Re:Whiskey Tango Foxtrot!

[Anonymous Coward]

This isn't Linux ported to the browser (e.g. C to LLVM to JavaScript, which has been done in other instances). This is a reasonably "normal" build of x86 Linux running on an x86 emulator written in JavaScript, with a few special-cased virtual devices and drivers.

Re:Whiskey Tango Foxtrot!

[Doc Ruby]

Because they're a geek.

And you're not. Why are you reading Slashdot, let alone posting to it? Poser.

Re:Whiskey Tango Foxtrot!

[somersault]

I know it is. It's still linux running on JavaScript in a browser, even if it's not as efficient as it could be.

Re:Whiskey Tango Foxtrot!

[somersault]

(I guess when I wrote my original comment, I didn't know that though - but then I read TFA)

I hope this JavaScript fad blows over soon.

[Anonymous Coward]

I really hope that this JavaScript fad blows over soon. While it was just mildly annoying back in 2008 when it started picking up steam, now it's just getting stupid.

Look, JavaScript was a mistake. It was poorly designed from the beginning, and it took 15 years before implementations of it weren't complete shit. Hell, it's basically a mistake of history and circumstance that it's so widely available.

As a language, there's nothing special about it. It's not "Scheme-with-a-C-like-syntax" like some foolish people claim. It's doesn't even do prototyped-based OO sensibly (refer to Self to see it done correctly). It's just a half-assed scripting language.

And don't go blaming the DOM for JavaScript's bad reputation. It's a poor language in any context. Lua is a better scripting language, and Python is a better prototyping language, and Java, C++ and C# are better languages for building large networked applications.

Get over it, guys. There's nothing special about JavaScript. It's merely a limited version of essentially every other programming language ever implemented. Just because you can use it to do stuff that other languages can easily do it does not mean that JavaScript is "unique" in any way.

The Ruby fad has already ended. I hope the JavaScript one ends just as swiftly.

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

LOL, so those strings of characters you mention which produce an array of bytes which are interpreted mechanically by the processor are soooo much better than the string of characters in javascript which produce an array of bytes which are interpreted mechanically by the processor. Yup, you're right.

Re:I hope this JavaScript fad blows over soon.

[Svippy]

Do you want us to use Flash instead? O! Enlighten us, wise one, about the numerous other languages that are available for web browsers!

Re:I hope this JavaScript fad blows over soon.

[bmuon]

News flash: it won't and it will only get better.

Re:I hope this JavaScript fad blows over soon.

[somersault]

The special thing about it, is that it's available on almost any modern computing device, down to phones. It's become the Windows of programming languages in a way - not the best option, but it's just so damn common that you should at least learn to use it. I don't know why we don't have any alternative scripting languages for browsers yet. CoffeeScript looks nice, but it just compiles down to JavaScript anyway.. so not the most efficient way to do things if you need to ship your framework as libraries on your website rather than it being built into the browser.

I did suspect that Ruby was a bit of a fad, but it is a nice language all the same.

Re:I hope this JavaScript fad blows over soon.

[Junta]

Hell, it's basically a mistake of history and circumstance that it's so widely available.

Herein lies the reason why javascript is not 'just a fad'. No matter your opinion of the DOM and javascript syntax, it is *capable* of being used to get the job done and it is everywhere. Other than making tasks absolutely impossible, it's hard to offset in difficulty the benefit of being everywhere. No other language will be everywhere so long as javascript is 'good enough'. Any browser attempting to bring their own favored child in will not meet with adoption because Javascript will work too and on other browsers. Short of getting Microsoft, Google, Apple, and Firefox to adopt the language with *zero* footprint to start with, nothing will change.

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

Well, obviously - that's the only way it can go.

Re:I hope this JavaScript fad blows over soon.

[dingen]

Lua is a better scripting language, and Python is a better prototyping language

So you're saying we should build a Lua/Python interpreter in Javascript so we can use these superior languages in our browsers?

Re:I hope this JavaScript fad blows over soon.

[tepples]

and Java, C++ and C# are better languages for building large networked applications.

As for Java: A browser maker wants to blacklist the Java plug-in because of how it interacts with SSL.

As for C++ and C#: You have to be an administrator to install a C++ or C# program on a PC, and a lot of people don't know the administrative password of the machine they use. This could be because A. they aren't the head of the household or B. they aren't in the IT department. It's even worse on a set-top box:, you have to be a major corporation to get a C++ or C# program digitally signed for installation.

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

... Lua is a better scripting language...

Hahahaha. Oh wait, you were serious? Let me laugh even harder.

And you think there are things special with Lua? Yeah, maybe special about being awful.
Nobody thinks there is anything special with JavaScript, it was the one that won the browser scripting wars.
Where was Lua then? It wasn't even in the running.

Your awful opinion of JavaScript won't do anything to it.
It is a very, very capable language. The only terrible thing about working with JavaScript is working with the terrible DOM that it is built around. The HTML DOM is what is wrong with JavaScript.

It will still be here as long as browsers still exist, and will outlive everyone here probably, unless Native Execution Inside Browser Sandbox takes off. (NEIBS, the new buzzword coming to you in a near future!)

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

Umm, you don't have to be an administrator to install a C++ or C# program on a PC. I compile and install C, C++ and Java software to my unprivileged user's home directory all the time on Linux, Solaris and Mac OS X systems. I run Windows programs written in C++ and C# from my unprivileged account's desktop just slightly less often. I've never run into problems.

Heck, my mother (she's 76) has her own PC running Windows 7, and even she knows how to download and install software on her own, while using a non-administrator account.

Why do you JavaScript morons and Web 2.0 weenies always make the false claim that "you have to be an administrator to install a program on a PC"? Do you realize how badly it undermines your credibility when you so publicly show that you don't even know how to install software?

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

No, encourage browsers to use the script tag for languages other than JS. That was the original intent.

Re:I hope this JavaScript fad blows over soon.

[MightyMartian]

It's been present on pretty much all browsers of any note since the late 1990s. Love it or hate it, Javascript stopped being a fad a decade ago, and has so much momentum now that unless you want to give up web development, you'd better just man up about it.

Re:I hope this JavaScript fad blows over soon.

[santosh.k83]

You can only afford to nest virtual machines up to a point, after which the drag on performance becomes unaffordable, even on the fastest machines. A JavaScript x86 emulator is really cool, and it might be interesting to those who want to run (or serve) system code over the net, or push the limits of JS and it's implementation, but that's about the extent of it's potential as far as I can see.

Re:I hope this JavaScript fad blows over soon.

[santosh.k83]

"(NEIBS, the new buzzword coming to you in a near future!)" Awful... I much prefer NaCl...

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

How is Ruby more or less a fad than Python? How is Python a better language? Sure, they follow different language philosophies, but they offer many of the same features and occupy the same niche. And readability depends on one's aesthetic. Is white space more readable than begin end blocks (or curly braces)?

Re:I hope this JavaScript fad blows over soon.

[mrmeval]

Ruby and others in a browser, not so revolutionary now but interesting.
http://ejohn.org/blog/the-browser-scripting-revolution/

TCL in a browser, if it were integrated it would be more relevant.
http://www.tcl.tk/software/plugin/:Python

I doubt gawk, perl, bash scripting or others will run in a browser soon.

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

Not sure why this was modded offtopic. This type of "demonstration" is a direct result of the ubiquity, not the extensibility of the language.

Re:I hope this JavaScript fad blows over soon.

[Anonymous Coward]

You should think twice before suggesting that people like the GP "at least learn to use JavaScript". The people who are the most critical of JavaScript are often the ones who have used it the most, who know it the best, and who have likely used many other programming languages. They know in-depth just how awful JavaScript truly is.

Re:I hope this JavaScript fad blows over soon.

[somersault]

I wasn't meaning "you" as in the GP. I was making an analogy that like it or not, you're going to need it in future if you're in certain lnes of work

Re:I hope this JavaScript fad blows over soon.

[somersault]

I really like the look of Pyjamas, another poster mentioned it. I was planning on trying out Python soon anyway, and Pyjamas seems to be a very, very nice way to create portable apps that can be run normally as Python apps under Linux and Windows, or compile to JavaScript for running in browsers (as long as you use no C libraries). Very cool.

Educational sandbox?

[Irick]

This could be a great thing to embed into online how-tos and the like for teaching basic or even advanced linux. ... heh, embedded linux on the rise again :P

Re:Educational sandbox?

[Guylhem]

Believe me or not but for a 101 bioinformatics course I taught this year, I picked up jslinux to teach grep, pipe and other shell scripting basics.

This was just the simplest option to get a linux prompt on every machine of the computer lab - and studenta who had brought their own laptops could work on the very same examples as the rest of the class, in an identical environment.

Busybox grep is good enough for teaching the basics to students, and they can access the very same setup from home if they want to rehearse or try to solve the problems in a different way. .

What is truly missing is not persistent storage but basic network support, to be able to wget example files. Not anyone can cook an initrd (loop mounted ext2 filesystem)

New Java VM/Script?

[Doc Ruby]

Oracle is making announcements about Java 7 and 8 this week. Supposedly the new stuff is better integration between Java and HTML5, and between Javascript and the JVM.

Will that revised tech be good support for an interactive user shell with a Javascript commandline calling Java objects, reporting back HTML5 in a DOM? Interactive HTML5 GUI objects that can take GUI events back into Javascript logic or just Javascript glue to Java objects in the JVM?

Will Android's Dalvik JVM follow that route, or take its own route to that architecture?

Re:New Java VM/Script?

[Anonymous Coward]

dunno, you can already glue the webkit-view on android to dalvik code. if you want.

most often there's no point.

Re:New Java VM/Script?

[g4b]

I dont think Java will play a big role anytime soon in most of the web. Especially this demo shows, that js itself has become so powerful, that powerful conversion a compilation apis will evolve, completely removing java from interest in the web. and I fear the people who might consider your path of using integration to a VM in the background will use adobe products more for these tasks, where interaction is already given - even if the flash vm sucks.

Your point is however interesting, because it shows what you think of, something which crosses barriers of mobile (no adobe!) and desktop browsers - there the only problem is, the number of applications for this model being used both in browsers and mobile devices are too few; the development cost of apps building either on complete web based logic or application based logic for mophos are smaller; and at last: while java is a common thing between android and the desktop, there is already the separation between oracle and dalvik and the separation between ios, symbian, winmob, and others, not all supporting java or flash as common backbone, again throwing up the question if targeted dedicated apps using common apis are not cheaper.

So my answer is: does not matter what oracle does there. It's a market which already falls from its zenith.

Java plays a role in conversion languages however at google. I still prefer pyjamas.

Re:New Java VM/Script?

[g4b]

ah yeah, and there is the werewolf in the vampire: .net with silverlight and browsers would compete here too on windows devices.

but vampires get fewer after generations of desperate although somehow successful geeky witchhunts.

Re:New Java VM/Script?

[Doc Ruby]

I don't see a GUI shell on a JVM being a Web technology. Except that the GUI and JVM could be remote from each other, transparently. Mainly I'm interested in a GUI shell to an OS and a complete machine, distributed or not. That's not really the "Web", though it could be the "Internet".

I'm especially interested in a mobile platform that works this way, so Android - which is an OS dedicated to running a JVM.

I'd like to use a fully dynamic GUI markup format like HTML5 for presentation, Java classes for computation and data management, and all the existing platforms, libraries, code in repos and tools supporting them for application development. I think an environment that's Javascript for interactive programming, Java for compiled heavy lifting, HTML5 for presentation would be very flexible yet highly specific in application. It would be the modern revision to ksh/C/printf, inherently networked, on JVM instead of C/Unix VM.

I suppose the missing piece would be Sqlite, if it were better integrated to server-side full SQL (explicit JOINs, Java/script stored procedures, etc), corresponding to ksh/C/printf's dbm. Just as Javascript is the dynamic/interactive control language for interfacing HTML5 to Java, Sqlite could be the lightweight data metamodel gluing full SQL to HTML5 local storage.

With that platform (HTML5/Javascript/JVM/Java/Sqlite), there'd be as little difference between "the Web", a local app, and "the Internet" as there is in C/Unix between domain sockets and remote sockets, or as there is on (properly used) X between distributed client/server and both on localhost. So "the Web" would go away as a distinction. Which, especially for mobile use, is essential given the continuing unreliability and diversity of networks and other hardware in use during a single session or across multiple ones by the same user for a continuing purpose.

So it does matter what Oracle does. Because it matters how Oracle's new JVM implements Javascript support, since it's the most common JVM in use. And because Oracle's JVM sets the context for what Google does with Dalvik for all the Android devices rapidly eclipsing (pun not intended, but realized and approved :). the rest of the JVM instances. Especially in the context of Oracle's existential lawsuits against Google's Android. The support in each (and, preferably, the same in both) of those defining JVMs will determine whether "Web 3.0" has a platform like the one I described.

But, probably at least as important to me, is whether I get that platform. I've been waiting for it instead of a VT100 XTerm for over 10 years, ever since a "Netscape GUI CLI" plugin appeared briefly and then disappeared.

Re:New Java VM/Script?

[g4b]

Now I understand. There is however in this scenario also the possibility that Google will Go another way.

Re:New Java VM/Script?

[Velex]

Java and Javascript are two completely different technologies. The only thing they have in common is a C-like syntax and four letters.

It's the same thing to glue a Javascript object to a Java object as to glue a C# or Ruby or PHP or Perl or CGI or whatever you're running server-side object to a Javascript or VBScript or Ruby or whatever you're running client-side object.

I find languages like Javascript and Ruby very interesting, but just don't get it mixed up that Java has anything to do with Javascript. Java is like C++ with training wheels affixed to a rail track (not saying C++ is good, either). It's a very "correct" language, and it's Java's way or the highway. Javascript, on the other hand, gives you more than enough rope to hang yourself, along with several ways to shoot yourself in the foot and a blindfold if you want.

Very different technologies, and it's a pet peeve of mine when people assume they have anything to do with each other or that they must needs have anything to do with each other. Javascript won't go away in the browser, true, but I'd rather have C#, PHP, or anything else server-side any day of the week. (Javascript server-side with something like Ruby's DRb over AJAX? Might be interesting.)

Re:New Java VM/Script?

[Doc Ruby]

You are wrong. As Oracle has been announcing, Javascript is now supported by (announced) technology in the Oracle JVM.

You were correct for a long time. When introduced, the only connection was a superficial syntax similarity that is shared with many other OOP languages. But, as I pointed out, the "Sun" JVM now has Javascript support in it. FWIW, I wasn't talking about the browser, either.

Re:New Java VM/Script?

[JasterBobaMereel]

..you can use Javascript as a scripting language in a JVM - no still no connection, you can script JVM with Groovy already does this mean it is Java ... no

JVM is a virtual machine, the main language for programming it is Java
You can also program in other languages and script it in several languages - One of this will soon be JavaScript

9.80 BogoMIPS

[Anonymous Coward]

Pentium MMX. Is that what everyone gets as well?

Re:9.80 BogoMIPS

[Anonymous Coward]

Yup, Pentium MMX and 9.80 BogoMIPS here too

Sequence of last story and this one...

[Junta]

"Windows 8 is going to use less memory"
"Oh yeah? Well Linux can run in javascript, ha!"

Re:Sequence of last story and this one...

[Anonymous Coward]

Does that mean they could make a javascript linux Windows 8 Metro App?

Its all nice and stuff

[dev658]

Except that the author of original emulator is an asshole.
This is because he didn't release the source of it (its obscured javascript) and he nether licensed it under a free licence.

And author of this mod, who say that he obtained a permission to _host_ these obscured files just supports that bad guy.
Both should be ashamed of this. They didn't do a rocket science, nobody would want to buy this stuff.

Re:Its all nice and stuff

[wzzzzrd]

don't click, it's goatse.

Re:Its all nice and stuff

[Anonymous Coward]

Some of us like goatse, you insensitive clod!

New malware vector?

[Anonymous Coward]

Rather than using javascript to load infected files that use *.pdf and other infectible formats, they can now run a botnet just using an infected ad straight inside your web browser.

Re:New malware vector?

[GuldKalle]

But where is the advantage? I would assume that the linux VM would face the same restrictions as any other javascript in a browser.

Re:New malware vector?

[ksandom]

As soon as someone figures out a more useful way to access the internet than simply tunneling through http/s, this will effectively provide rooted-box-in-a-box. Ie stuff-all effort to get a huge number of hosts at your disposal. Persistent storage provides an easy way to resume where you left off.

Re:New malware vector?

[mysidia]

As soon as someone figures out a more useful way to access the internet than simply tunneling through http/s

How about a kernel mode driver inside the browser-based Linux box for a network stack tunnelled over HTTP/s to a dedicated application running on the same host as the browser to decapsulate tunnelled traffic and dump it onto a local virtual network bridge, kind of like the one used by VMware Workstation or Xen ?

Or even just an application utilizing a Java extension API to allow the Linux-in-browser js application to map a little chunk of host memory for raw network I/O

Re:New malware vector?

[Anonymous Coward]

Annoyed user: "Why the fuck the browser got so slow?" (closes tab)
Wanna-be hacker: "Damn! Foiled again!"

Mining bitcoins via parasitic computing this way would be more profitable than this kind of "botnet".

Also, no, I don't think there will ever be connecting to arbitrary services with javascript. It would be too easy to do "var sock=new Socket('blah.com', 25); sock.send('HELO ... MAIL FROM ... RCPT TO ... DATA ... I've got an urgent business proposal ...')"

So it would be extremely useless botnet.

Re:New malware vector?

[GuldKalle]

As soon as someone figures out a more useful way to access the internet than simply tunneling through http/s [...]

As soon as that happens, an attacker could just write his program in javascript instead of stuffing it into a VM running on the same JS engine, with the same access restrictions.

[...] to a dedicated application running on the same host as the browser [...]

What's with all the hoops? if the attacker can run an application with net access, there is no need for a Linux VM.

Re:New malware vector?

[DragonWriter]

How about a kernel mode driver inside the browser-based Linux box for a network stack tunnelled over HTTP/s to a dedicated application running on the same host as the browser to decapsulate tunnelled traffic and dump it onto a local virtual network bridge

So, to break the Linux-in-a-browser out of the browser security sandbox the malware distributor simply has to gain the ability to run a HTTPS server on the same client machine outside that the Linux-in-a-browser can talk to and that will then issue the actual arbitrary network connections.

But, if the attacker can run arbitrary code that can make network connections outside of the browser on the same client, why do they even need the compromised Linux-in-a-browser at all? You've essentially postulated a "browser based" attack that, as a prerequisite, requires the ability to run arbitrary code outside of the browser on the same machine, rendering the browser-based portion of the attack superfluous.

Re:New malware vector?

[mysidia]

What's with all the hoops? if the attacker can run an application with net access, there is no need for a Linux VM.

I'm suggesting the application would be something 'legitimate' with legitimate uses eg users who WANT their Linux-in-browser to have network connectivity, to help with training or whatever function this Linux VM is being used for.

Re:New malware vector?

[mysidia]

Who says this has to be arbitrary code?

I suppose it depends what your definition of 'malware' is; there are application loads that don't require arbitrary connnections either (e.g. masmsively distributed brute force Password cracking). I would consider anything that has access to a users' cookie database through JS and the ability to make any sort of outgoing connection a potential danger.These sort of connections are done every day using Ajax.

Javascript code is fine. HTML5 provides a technology called Websockets. There was recently an article about about a Javascript based SSH client

Next step forward in "An OS in a browser in an OS"

[knuthin]

Look at everyone going meta.
Now I want to run X on this virtual machine and fire up a browser. (Maybe just install Lynx on it if I can't install X on it.)
That'd be a browser in an OS in a browser in an OS. :D

Questions

[bigsexyjoe]

What's the login? If I download it, will my js Linux remember things from session to session? Could I install a webserver on it?

Re:Questions

[felipekk]

I logged in with root and no password.

I guess you could install a webserver on it, then maybe even host those same files (including "emulator.html")...

Then on another session open a web browser, point to your webserver and open the emulator...

That way you'd have an emulated web server serving an emulator that is being run on an emulated web browser emulating linux.

Re:Questions

[sho-gun]

standard linux fresh-install login.
username root
no password.

Re:Questions

[Anonymous Coward]

root with no password. I don't know. Very unlikely.

What I want to know is how difficult would it be to install other things on it like the gcc compiler. It would be perfect for the Linux programming intro course offered at my school.

Re:Questions

[nstlgc]

I N C E P T I O N

Re:Questions

[Cyphax]

The first link in the article above should point to a different version of this emulator, which actually did include gcc and a small hello world program. I remember playing with that shortly and being hugely impressed by the fact that works really, really well. :)
This whole emulator is massively impressive imo. :)

Re:Questions

[Anonymous Coward]

no login, just enter root as user

no network though. would be handy if it had network.

Re:Questions

[Anonymous Coward]

You mean standard like that idiotic distribution that gives all users unfettered root access via sudo, and is therefore no better than Windows for security?

Login Name and Password?

[Anonymous Coward]

Emm, what's d login name and password?

5ryn

Re:Login Name and Password?

[dev658]

You really aren't a hacker. Try 'root' man....

Re:Login Name and Password?

[ksandom]

type in root. It has no password.

Makes me wish

[sgt scrub]

While jsLinux is cool, and this is a cool addition, it just makes me wish JavaScript wasn't the only languageVM embedded in the browser. The thought of what could be done if one could take advantage of what the various scripting languages do best instead of trying to fit JavaScript to everything makes me sad.

Re:Makes me wish

[catbutt]

Well Dart is coming out in a couple days.

Although I'm not sold on the "use the best language for the job" mentality, I have better things to do than learn new languages and port my code from one to another. I see no reason why a single language can't do everything from user scripts to systems development, while maintaining elegance and expressiveness. Not saying that language has been invented yet (or will be any time soon), but still.

Re:Makes me wish

[xdor]

I still think those who complain about JavaScript being the only option don't really understand how awesome JavaScript is.

Heck, if you don't like coding JavaScript for the browser: port your language of choice to JavaScript! Then you can happily code in SmallTalk, ADAScript, MEL Script or whatever other codec you find more appealing.

Yes, its that flexible

Re:Makes me wish

[Anonymous Coward]

I still think those who complain about [insert favorite language] being the only option don't really understand how awesome [insert favorite language] is. Heck, if you don't like coding [insert favorite language] for the [platform in question]; port your language of choice to [insert favorite language]!

Yes, Turing-complete language is that flexible!

Re:Makes me wish

[xdor]

Touché

Re:Makes me wish

[BZ]

If you want to embed multiple language VMs in the browser, you immediately start having issues with cross-VM reference cycles causing leaks. The infrastructure needed for breaking those is .... nontrivial. You also get very complicated interactions performance characteristics as the VMs interact. Note that historically browsers have had issues just solving these problems for the JS VM and C++ DOM, without adding more VMs into the mix.

A better bet may be having a single VM that multiple languages can compile to (a la JVM, which all sorts of things other than Java can target). But at that point, perhaps that VM should simply be the JS VM. For example, you can already compile C code to JS. See http://syntensity.com/static/jsconf_eu_Emscripten_lo.pdf for the recent state of that. Now you do get a slowdown over, say, gcc -O3; the slides list this slowdown as somewhere between 2x and 8x depending on the workload. But JS VMs are pretty continuously improving, and it's not clear how fast running C code on the JVM, say, would be... Would be interesting to have those numbers.

Re:Makes me wish

[Anonymous Coward]

"What could be done..."

Running an entire OS isn't enough for you?

Re:Makes me wish

[sgt scrub]

I have better things to do than learn new languages and port my code from one to another

That sounds a lot like what a person that writes Python, Perl, Squeak, etc... would say when asked to re-write everything in JavaScript for the web. :)

IMHO. Computer languages are tools. The more tools you know how to use the easier life will be.

Re:Makes me wish

[sgt scrub]

Can I get a link from you of an example please. Something small and easy to dig through would be nice.

Re:Makes me wish

[sgt scrub]

I was referring to your second example, JVM, though it isn't a very good example. All the languages need to be "JIT'd" before use. Parrot is the better example. If they would nail down the embedding API we might have what I think people would like. I certainly would.

Re:Makes me wish

[BZ]

Yes, but why is Parrot a better approach than just compiling to JavaScript and using the existing JS VM?

(And note that Parrot is much like ActiveX: an unspecified single-vendor kinda thing.)

Re:Makes me wish

[sgt scrub]

1) You loose a lot in translating code. No optimizations.
2) You can not use libraries native to the language.
3) You end up having to write/correct the translators with "if (typeof myFunction=='undefined') {"

unspecified single-vendor? ActiveX is a response to NPAPI with gapping holes in system security. Adding a VM that can interpret native code eliminates ActiveX, NPAPI, and PPAPI while, if done correctly, creates a jail/sandbox for web content.

Re:Makes me wish

[BZ]

> 1) You loose a lot in translating code. No
> optimizations.

This is true of any intermediate representation not designed with a particular language in mind, to some extent. The emscripten compiler does in fact do a fair amount of optimization, and it's operating on LLVM bitcode that's already had some optimization passes applied.

> 2) You can not use libraries native to the language.

In the specific case of emscripten, again, you just compile the libraries,

> 3) You end up having to write/correct the
> translators with "if (typeof
> myFunction=='undefined') {"

Why correct? You do have to write the translator; you do it once.

> ActiveX is a response to NPAPI with gapping holes
> in system security.

Yes, and an attempt at single-vendor lockin of the web. Pepper is missing the security holes, with any luck, but has the lockin thing in spades.

If all the UAs could agree on a single VM and if that VM were not controlled by any one company, then hitching the web to that VM might be OK. But unless you want to give a single company control over development of the web, this is not that likely to happen in the near future...

Re:Makes me wish

[sgt scrub]

Parrot is GPL'd. Being locked into Open Source for code visible to any user isn't much of a lock in.

Re:Makes me wish

[BZ]

1) Many project can't use GPL'd code. Opera comes to mind, say.

2) Being GPL'd in this case just means you can fork it, not that you can affect its development or prevent changes that break you. Google has complete power to evolve Parrot in any way it wants, and if it happens to break other browsers that happen to be using Parrot or breaks websites they have no choice but to deal. It's not quite as bad as being locked into something you _can't_ fork, of course. But don't pretend like it's "not much of a lock in".

Re:Makes me wish

[BZ]

Er, I just realized there's some confusion here (on my end) between Pepper and Parrot. So ignore what I said here while I go and read up on Parrot details. ;)

Re:Makes me wish

[sgt scrub]

:) and while your at it think about why google didn't spend some effort advancing parrot instead of dumping yet another scripting language on us. it boggles the mind.

Re:Makes me wish

[JasterBobaMereel]

The answer is Curl http://en.wikipedia.org/wiki/Curl_(programming_language)

this is a full programming language that also does markup, scripting, CSS, all in one language ...

Re:Makes me wish

[sgt scrub]

Curl allows one to have my_python_code.py, my_perl_code.pl, and my_javascript_code.js files on the server and content type tags for perl, python, and javascript and the browser will interpret the three scripting languages and apply the code to the content? That is the behavior I, obviously poorly, tried to describe.

sorry this isn't what you people make it out to be

[Anonymous Coward]

This isn't linux in javascript. This is more of an output to screen from a linux server.

1. All the important code has been "compiled" with some javascript "compiler" that basically takes javascript code and fucks with it to where you can't read it
2. Does not follow the GNU license.
3. There isn't enough code to make it a full linux in javascript.
4. You won't be able to copy this and run it off your own computer

nothing to see here people

Java

[Anonymous Coward]

I suppose at some point someone might get a JVM running in Javascript...

Re:Java

[Lennie]

Yes, let's port the JVM to nodejs. So you can run Java on the server.. ;-)

Sandbox is all you have

[tepples]

Some post-PC pessimists are under the impression that this educational sandbox will soon end up being all you have because the operating system publisher or hardware maker won't give you the cryptographic keys to boot anything else unless you represent a major corporation.

Re:Sandbox is all you have

[sourcerror]

"unless you represent a major corporation"
So we'll have RedHat, Ubuntu, and Oracle Enterprise Linux and Suse. And I guess the Ubuntu folks get a key for debian as well.
Of course it would negatively affact minor distributions, and I'm totally against it, but don't pretend that Linux would just suddenly disappear.

Re:Sandbox is all you have

[bartonlp]

The "secure boot" if mandatory will make it almost impossible to recompile source to optimize your system. Also, good by Gentoo. I think mandatory "secure boot" is a big mistake and Microsoft should be ashamed of itself, but monopolies seldom are ashamed.

Re:Sandbox is all you have

[Anonymous Coward]

Only the bootloader needsto be signed for secure boot, so as long as grub is signed, you can boot whatever yoou want. Worst case, upstream grub is signed and a downstream distro has less oppurtunity to ship a beta or forked version of grub.

Hell, Windows has boot.ini on Win2k3 and below and whatever Vista and above you... you can boot linux from the NTloader just fine if you have a system tah only has MS signing keys and no way to disable secure boot.

Re:Sandbox is all you have

[santosh.k83]

Either the PC manufacturer has to provide an easy and standardized mechanism for the user to add his own signing keys, OR, provide an easy and standardized mechanism to disable 'secure' booting. If not, the PC will be no more than a glorified MS version of iPad, and I'll be boycotting any such manufacturer. So will a LOT of people I'm sure, and public and legal pressure will certainly be brought to bear on them.

Re:Sandbox is all you have

[tepples]

If not, the PC will be no more than a glorified MS version of iPad, and I'll be boycotting any such manufacturer.

Until all well-known laptop manufacturers choose to start making "a glorified MS version of iPad". When faced with a choice between a $350 mainstream laptop with only the Windows key and a System76 laptop for twice the price, what will the general public choose?

Re:Sandbox is all you have

[santosh.k83]

Fortunately I believe people (and even most governments, perhaps not the US one though) won't accepted the dumbing down of the PC to an appliance. I certainly hope so. Besides being taken to court (which will definitely happen if Windows is imposed), the system is also certain to be cracked in time. IOW, we're heading for a mess at best, a protracted battle at worst. The fact that people haven't accepted lockdown even on smartphones and consoles gives me hope that this will turn out to be only yet another temporary roadblock towards universal and open computing.

Re:Sandbox is all you have

[Velex]

You know damned well it doesn't work that way. Ask your local soccer mom. Linux has a scary black background with a cryptic text interface and is either a.) a tool for copyright infringement to steal from hard-working, job-creating content owners b.) only used by subversive teenagers c.) used to commit cyber crimes (just like in The Lawnmower Man and isn't it wonderful that writer in Maine can explain computers to us simple folken, say thankya) d.) a tool of the devil with its black background. Nobody who does Real Work has a need to run Linux. Real Work is done in Word and Excel.

There are people who honestly think things like this. Yes, there are people who honestly do not realize that linux-based system power incredibly useful software like Apache. There are people who use MyTwitFace every day who think that being able to even pull up the HTML source of a webpage makes someone a potentially dangerous hacker.

You should be afraid and always strive to prevent those people from gaining political clout. They are not ignorant or stupid. Never think they are ignorant or stupid. I will never understand what they really think. I only understand what they say and do.

I guess that being said, I'm going to go order my Android tablet now. Powered by a Linux-derived kernel and open-source software. Yet I still need to "root" it to be able to install the free software I want to use and write. I'm not sure what to think of that. I just don't know what to think, and I'm afraid of what the future might hold when this tablet breaks and I want to buy the next latest-and-greatest device.

Re:Sandbox is all you have

[tepples]

Fortunately I believe people (and even most governments, perhaps not the US one though) won't accepted the dumbing down of the PC to an appliance.

People already accepted the dumbing down of TV-connected video game computers like the C64 to appliances like the NES, and they accepted the dumbing down of the smartphone from Windows Mobile 6 to iOS 1.

Re:Sandbox is all you have

[Anonymous Coward]

they accepted the dumbing down of the smartphone from Windows Mobile 6 to iOS 1.

That wasn't much of a choice. Windows Mobile 6 is basically unusable garbage. You also seem to be forgetting about Android...

Re:Sandbox is all you have

[tepples]

You also seem to be forgetting about Android...

I didn't mention Android because it didn't exist in a product when the first iPhone came out.

Native Client

[tepples]

Google Chrome has "Native Client", a verifiably type-safe subset of native code. One might, for example, port DOSBox to Native Client.

Re:Native Client

[PwnzerDragoon]

One might have already done that.

Re:Native Client

[Anonymous Coward]

You may want to take a look at naclbox: http://www.naclbox.com/

Re:Native Client

[sgt scrub]

Salt and Pepper, PPAPI, are really cool if you write C/C++. I was talking more broad like Parrot.

Re:sorry this isn't what you people make it out to

[Anonymous Coward]

There should a "Post as Anonymous Troll" option.

buy slashdot

[Anonymous Coward]

mod me out without posting or proving this isn't real
nice job
never again
enjoy your bots and bullshitting people

Year of

[JustOK]

All hail the year of linux in the browser!

Re:Next step forward in "An OS in a browser in an

[Anonymous Coward]

Yes. And the nice thing about this is that you can make things as slow as you need to. For a while, there was a threat that hardware would outpace the demands of software, but now that we're building everything in bloated virtual languages running in virtual machines running on virtual hardware, we can NEST that shit and continue to drive the demand for ever faster hardware.

Because otherwise, you know, my 2002 desktop runs KDE 4.7, Win7, the latest firefox and thunderbird, and everything else I throw at it with no problems. There's serious danger in that I haven't needed to upgrade it for a decade. We can't have that now, can we?

Worthless

[Anonymous Coward]

Doesn't run in my Android browser.

Recursion

[garyebickford]

That'd be a browser in an OS in a browser in an OS. :D

... and that could be extended to an arbitrary depth!

Re:Recursion

[ksandom]

That would be an interesting competition.

reporter: With me now is Tom who has just booted a computer inside a computer, insiiiiide a computer sixTEEN times!
tom: Yes, it took about a year to boot the last layer of nesting and takes nearly 2 hours to run ell ess, but it was totally worth it. I'm really excited.
reporter: And what do you want to do with it now?
tom: In a couple of years, I hope to be up to 17 layers, and it's my life ambition to get up to 32.

Seriously though, if I was to do such a thing, I'd put it in a VM to begin with. So at least then I could pause and resume the VM on different hardware so that I could continually upgrade the hardware as newer stuff became available. Now what would be really cool, is to implement hibernate (or suspend to ram and dump the ram) so that the browser could be changed out without rebooting the JSVM. Now that persistent storage is sorted, perhaps that isn't too hard?

The next thing I'd do is try to automate as much as possible. Because once it starts slowing down, it would be pretty frustrating to interact with.

Re:Recursion

[garyebickford]

The next thing I'd do is try to automate as much as possible. Because once it starts slowing down, it would be pretty frustrating to interact with.

:D
Oblig.: Sand Universe

Re:Recursion

[ksandom]

LOL that was brilliant!

this is what I like least about slashdot

[Anonymous Coward]

blah blah blah SCO.... blah blah... now Linux is more relevant than EVER! blah blah blah.... built on JavaScript just because we CAN! blah blah

Christ, people, get a grip.

Re:this is what I like least about slashdot

[oakgrove]

Today, you is what i like least about Slashdot.

Re:sorry this isn't what you people make it out to

[somersault]

You have no clue what you're talking about.

One in each tab ...

[PPH]

... and I've got a Beowulf Cluster!

Re:One in each tab ...

[Anonymous Coward]

You might just be on to something here, and I'm imagining the Vagrant guys are already on this. Imagine a one-click staging environment--feed the main tab a branch of puppet/chef & your own app to use, and launch.

Remember comrade....

[theendlessnow]

With Linux, the browser runs you!

A Persistent Block Device "How To" would be nice.

[Anonymous Coward]

Where are instructions about how to create the advertised 'persistent block device'? Perhaps I missed it but the readme says that "This project allows the virtual machine to access a persistent block device. This means a user can format and partition this device inside it's virtual machine." BUT I do not see the instructions anywhere obvious. Even booting up this virtual machine ... I had to guess to login ... I eventually found user 'root' and an empty password would work to login. A quick 'df -h' does not reveal any persistent block device present. So I ask myself ... after going to all this trouble to host this demonstration ... why aren't there a few more instructions other than "For more details, see the source code." offered?

Signed perplexed.

Will it run VMWare vSphere

[Anonymous Coward]

Start it up in VMWare instance. Start up VMWare. Hot move the instance running the web browser to itself. It would so rock.

Obligatory "Yo Dawg"...

[Anonymous Coward]

We heard you like Linux... so we put some Linux in your Linux so you can Linux while you Linux.

Self contained thin client applications, anyone?

[Tastecicles]

This'll be handy for those who've bought into the Google Chromebook and discovered that all they get is a browser... no, seriously, don't you guys think you were even the slightest bit done over, when you can get a webbook for £60 with free lifetime data allowance?

The license of GRUB, for one thing

[tepples]

Only the bootloader needsto be signed for secure boot, so as long as grub is signed, you can boot whatever yoou want.

The license of GRUB requires that it be distributed with "Installation Information", which includes private keys for signing. Operating systems would need to be shipped with a non-GPL bootloader, and makers of home PCs would have no incentive to include keys for this bootloader because boot-time malware could work by installing it and setting it as default.

you can boot linux from the NTloader just fine if you have a system tah only has MS signing keys and no way to disable secure boot.

The NT loader of Windows 8 will check the signature on any kernel it loads, and (I'm guessing) so will the NT loader of Windows 7 and Windows Vista after a service pack to allow their use on machines that require UEFI secure boot.

security problem

[Anonymous Coward]

and still no one sees the security problems and thinks maybe just maybe html5 encompasses far too much for web a browser.

increase features to hell with the consequences!

Software Restriction Policies

[tepples]

Umm, you don't have to be an administrator to install a C++ or C# program on a PC. I compile and install C, C++ and Java software to my unprivileged user's home directory all the time on Linux, Solaris and Mac OS X systems.

It appears you've never used a thoroughly locked down machine with /home mounted noexec.

I run Windows programs written in C++ and C# from my unprivileged account's desktop just slightly less often. I've never run into problems.

It appears you've never used a thoroughly locked down machine with Software Restriction Policies.

Why do you JavaScript morons and Web 2.0 weenies always make the false claim that "you have to be an administrator to install a program on a PC"?

For one thing, PCs have the measures I described above. For another, please address video game consoles and other set-top boxes, which won't run anything unsigned or self-signed.

Re:A Persistent Block Device "How To" would be nic

[deimios666]

cat README This will solve your problem.

Re:A Persistent Block Device "How To" would be nic

[Anonymous Coward]

cat README

Re:sorry this isn't what you people make it out to

[Doc Ruby]

Liar.

Education purposes..

[eexaa]

Did anyone manage to run stuff like gcc in this? It would make a _GREAT_ education tool.

(network access would be great too, but I guess that would be pretty hard with javascript...)

Re:Education purposes..

[leighklotz]

Fabrice's original came with tcc and qemacs, both of which he wrote.

iPad

[prograde]

Ok, that was, by far, the easiest way to get Linux running on my iPad.

JavaScript a la PowerShell?

[Doc Ruby]

Can I get a JavaScript engine that copies the best features of MS PowerShell script (as PS has copied the best features of JS, Perl, csh, and Java/C++)?

Mainly I'm looking for a typed object pipeline with reflection the shell can access. Reflection that exposes APIs of all the classes (bundled in apps and in the OS) installed in the system. Which, as Javascript, should mean "installed in the Internet". Javascript that wraps reflection via CORBA or some other webservices registry/server would be extraordinary - a transformative technology. But just an object pipeline with language iterators and collections operators against localhost resources would be awesome.

Sync Browser Linux With Server Linux

[Doc Ruby]

This web app would be even cooler if the local Linux state could be synced with a server's state. If I could run commands locally, generating a history file that I could send to a server to execute over again there. Or vice versa, where I create state in the local Linux by rerunning history commands downloaded from the server. Or sync either direction, line by line. A kind of "VMWeb".

As it is I don't see any way to install any app in it, either by downloading a binary or by compiling typed-in source locally. In fact I don't see any network operations exposed to that shell. Which is too bad, since it's running in a complete network browser - to say nothing of reaching the network OS the browser is running in. The busybox installed in this instance doesn't seem to have any network commands enabled.

This Is Not Linux in Javascript

[Doc Ruby]

This cool Web app is not "Linux in Javascript". It is in fact a "Javascript PC Emulator", just as the app says in the app's page title. It's a bootloader and a virtual PC implemented in Javascript running in the browser JS engine. Which loads a stripped-down Linux binary into itself and runs it, as if it were running on the PC. The Linux was written in C, compiled into PC (x86) machine instructions like any PC Linux, and then runs on the Javascript PC emulator.

I suppose it might be possible to run a Windows binary on it, if that bloatware would fit in the browser. Maybe DOS, or even Novell Netware (though this Linux demo has its networking stripped, and in any case the browser enforces the originating-server-only network access).

Very admirable project. Truly journalistic bad headline and summary.

JSLinux

[Anonymous Coward]

No lspci

I'm holding out for the Brainfuck version ...

[Kaz Kylheku]

which dynamically compiles x86 code into Excel macros.

*duck*

Re:9.80 BogoMIPS - f00f?

[Anonymous Coward]

Cpuinfo has some interesting tidbits as well:

[root@JS/Linux /root]# cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 5
model : 4
model name : Pentium MMX
stepping : 3
cache size : 0 KB
fdiv_bug : yes
hlt_bug : no
f00f_bug : yes
coma_bug : no
fpu : no
fpu_exception : no
cpuid level : 1
wp : yes
flags :
bogomips : 9.80
clflush size : 32
cache_alignment : 32
address sizes : 32 bits physical, 32 bits virtual
power management:

Like it claiming to have the f00f bug - has anyone verified that this virtual cpu has it and that Linux fix for it works properly?

Latest kernel

[loxosceles]

...is actually 3.0.6. kernel.org is back up but it's not updating properly.

But let's not forget...

[randomir]

.. the (html5) local storage capacity limit in most browsers is 5MiB (to 10MiB).