Forgot your password?
typodupeerror
Firefox Mozilla Open Source Operating Systems Wireless Networking

Nokia Researcher Puts Firefox OS On Raspberry Pi 75

Posted by timothy
from the obscure-sports-quarterly dept.
judgecorp writes "Mozilla's mobile phone operating system only exists in an early beta form, but Oleg Romashin, a researcher at Nokia, has already got it working on the Raspberry Pi and posted video to prove it. We don't think this indicates any alternate strategy for Nokia if Windows Phone doesn't pan out, but it does show that Firefox OS is portable, and the Pi is capable, and both can be played with — which will please both Mozilla and the Raspberry Pi Foundation. And the Firefox OS work in progress is available for download (direct tarball link)."
This discussion has been archived. No new comments can be posted.

Nokia Researcher Puts Firefox OS On Raspberry Pi

Comments Filter:
  • by Anonymous Coward

    Raspberry Pi, Nokia, Mozilla, Firefox all in one slashdot article - nice work.

    • Re: (Score:2, Funny)

      by Anonymous Coward

      Apple's going to use Bitcoins to patent the idea.

  • by KiloByte (825081) on Thursday August 16, 2012 @09:06AM (#41009353)

    Sorry, but with only 224MB memory, this is a bad idea. I've tried multiple browsers on Pi -- including full-blown like Firefox or Chromium, and minimalistic like Midori; the only one that's actually usable is elinks. Especially if pages as bloated as Slashdot are involved.

    Gooseberry, if it ever becomes something more than vapourware, might get into an usable range (512MB minus video memory).

    It's interesting how no graphical browser of today can cope with that little memory, when back in the day we could browse fine with 4MB.

    • by jcupitt65 (68879)

      The mobile browsers work OK in that much memory. My ipad1 can browse fairly well and "only" has 256mb of ram.

      • by KiloByte (825081)

        Looks like we have different definitions of "fairly well". After an aeon or two of swapping, it manages to render Slashdot after all, but I don't have that much patience. And, on many pages (sadly, Slashdot excluded), elinks is actually not that bad.

        • by oakgrove (845019)
          What's wrong with Slashdot in Elinks? I browse this site and many others in Elinks everyday (right now actually) and haven't had any problems at all. Maybe it's because I'm on the old comment system.
        • by Threni (635302)

          Slashdot doesn't render well on any mobile browser. I've tried Dolphin, Chrome, Firefox and the stock one on Android 4.0.4 (s3) and (most of those browsers on) 2.3.4 and 2.3.7 and it's horrible. I'm actually quite suprised at how little effort has been spent on this site for mobile. In particular, if you try and click on your user name at the top right of the screen you get the `options` and `accounts` links; it's impossible to click the username itself. Also, each subject doesn't wrap properly, forcing

          • by JDG1980 (2438906)

            From a technical POV, it's surprising just how much the Slashdot site sucks, given that it's written by and for geeks. Not only does it not work very well on mobile, but it also doesn't support Unicode. As a result, copying and pasting quotes will often result in garbage being inserted where there should be dashes, smart quotes, or other special characters. Come on, it's 2012; there's absolutely no excuse for this.

            • by rbrausse (1319883)

              it's surprising just how much the Slashdot site sucks

              this is /.'s USP - if the site would work flawlessly it would feel like a boring news aggregator...

            • by tlhIngan (30335)

              From a technical POV, it's surprising just how much the Slashdot site sucks, given that it's written by and for geeks. Not only does it not work very well on mobile, but it also doesn't support Unicode. As a result, copying and pasting quotes will often result in garbage being inserted where there should be dashes, smart quotes, or other special characters. Come on, it's 2012; there's absolutely no excuse for this.

              Do not confuse geeks with website designers. It's just the same as assuming that since you can

            • by unixisc (2429386)
              And you forgot that it's not even available on IPv6
          • From my experience, no site render well on mobile browsers except if there is a website's version written for mobile devices. Something few are paying attention to.
          • Slashdot renders great on the last ~3 versions of Firefox mobile on my Nokia N9, (which are provided via OTA updates). Ajax and all, it is very responsive, while the rendering quality seems very well done. (It was good before those versions also, but it seems zippier and more responsive with each version since). Firefox on the N9 is a joy.

            Disclaimer: I made websites for a living, so I figure I am fairly discerning.

        • by jcupitt65 (68879)

          Slashdot is horribly broken on mobile, but most other sites that aren't daft work well on an ipad1. It's certainly fine for casual use.

        • by jcupitt65 (68879)
          I timed the slashdot homepage on an ipad1: 1.5s for something to appear, 5s for the whole thing to finish. Seems fine to me. The BBC homepage takes less than 3s and appears fully-functional.
    • You might want to check out the VIA pico-ITX ARM board [viaembedded.com].
    • by Revotron (1115029)
      Rather than add more memory, why not use the architecture of the Pi to write a new browser that doesn't suffer from software bloat and scope creep? The RPi project was initially targeted at teaching children how to write programs that run on small, simple, and affordable systems with no fancy toolkits or bloated libraries.

      In other words, the way it used to be, when 256MB was a blessing and not a hindrance.
    • by WankersRevenge (452399) on Thursday August 16, 2012 @10:27AM (#41010685)

      It's interesting how no graphical browser of today can cope with that little memory, when back in the day we could browse fine with 4MB.

      Back in the day, a good practice was limiting your entire page to under 100kb. Now, you're lucky if a page clocks under 1mb even with all the caching going on. Don't get me started on sites like the Huffington Post or Destructiod.

    • by vmlemon (1203598)
      Hmm, what about trying to build Escort (http://code.google.com/p/es-operating-system/wiki/UsingEscort) for the Raspberry Pi?

      It seems to be very lightweight; already runs on Linux; and it's based upon a new architectural design, and a brand new rendering engine. It shouldn't be too much of a hassle to build, time/dependencies-wise.
    • Thats about par with low end Android phones and tablets. My phone and tablet (each $150 or so) have around that 256MB RAM mark. Both seem to run the latest Firefox Mobile Beta fine.

      Optimization is always necessary, especially when something is written for x86 and compiled to ARM.

    • by steelfood (895457)

      That's because everything's done by the browser nowadays. Whereas in the past, you'd have static pages that the server dishes up with each interaction, today, you have the browser rendering all the interactions, with the server supplying only the variable data.

      For example, if this was old Slashdot, if you clicked on a comment, it opened up in a new page. For new Slashdot, if you clicked on a comment, the comment now opens up in the current page, while the rest of the page is reformatted (lenghtened, shorten

    • by SlashDev (627697)
      I recall the old days of EMS where a piece of software would double your memory by compressing/decompressing on the fly, whatever happened to those?
      • by tepples (727027)
        Are you talking about something like RAM Doubler for classic Mac OS or zRam for Linux?
    • by Osgeld (1900440)

      1) turn off javascript, it will help a shitton
      2) go look at webpages from the 4 meg days, now go look at a modern site, see how that works?

      • by KiloByte (825081)

        0. install AdBlock, configure it to axe not just ads but also all trackers and similar sleaze. And especially all those Fecesbook/Google+/Twatter/whatever "likes".

  • Is this an attempt by Microsoft to fragment the Raspberry Pi platform?
    • by Anonymous Coward

      I would be utterly astonished if Microsoft was paying any attention to the Raspberry Pi "platform" at all. They can't even get the Slashdot commentards to take it seriously.

    • by gl4ss (559668)

      the article doesn't mention under what capacity oleg did the porting work, pretty much that they just pulled from linkedin that he worked on meego browser.

  • Better than the phones they been making the past few years.
  • by DrXym (126579) on Thursday August 16, 2012 @10:21AM (#41010603)
    The Pi is 700Mhz ARM processor with 256MB total memory (some reserved for GPU) and some hardware for OpenGL and MP4. The more you can push onto the GPU the better because the CPU is designed to power set top boxes and the like where the CPU should marshal the hardware and do as little as possible otherwise. So Raspbmc works okay for the most part because mp4 content is being powered by the hardware and there is only one main process running. But Raspbian demonstrates that a desktop performance is awful even with the lightest of configurations. I don't expect the performance of Firefox OS to be earthshattering. Even budget phones would have a faster CPU and more RAM than the Pi. It might run and be interesting for that, but I think performance will be poor especially on heavy content. So calling the Pi "capable" is reaching a bit.
  • I have a RPi, but the current hardware version has some major issues.

    Not enough memory - As some other commenters noted, the 256mb memory is not enough to run X. Forget web browsing, unless you want to wait minutes to load websites. With Scratch loaded, which is one of the advertised used cases, there's only 10-20mb memory left, with the 240-16 memory split.

    Not USB 2.0 compatible - This is a major issue. I tried about a dozen keyboards, most don't work at all, the one that is semi usable repeats keys
    • by Hatta (162192)

      As some other commenters noted, the 256mb memory is not enough to run X.

      That's quite plainly bullshit. I run X on a P200MMX with 128MB of RAM. X itself is quite efficient unless you start loading up kdelibs or some shit.

      • by csumpi (2258986)
        You are correct, I should've defined the use case better. I can run X on it fine. I can open multiple terminals under xmonad, ssh to my server and do some administration (although scrolling the terminal window atakes about a second to redraw in full screen). I can even browse the internet with elinks. I can control the GPIO from python.

        But it's not for running Midori, Scratch or gimp.
    • by Anonymous Coward

      Three types of USB-related faults have finally been accepted by the Foundation, after a long struggle during which they denied everything for months:

      A. Limited endpoints:
      It can handle only 8 USB endpoints, which means that you can run out of endpoints when plugging in just 2 or 3 devices (a hub just uses up more endpoints and makes this still worse), and then the entire USB system and your networking dies.

      B. Data loss:
      The USB driver requires realtime response from the Linux kernel when handling USB's split

  • "...as soon as Stephen Elop found out the researcher was thrown out of a window while Elop threw a chair and yelled "You don't disrespect Windows Phone like that!"."
  • Chris Weber, the CEO and former Microsoft executive, fired Oleg Romashin stating that "Resistance to our Win7 strategy is futile. You will be eliminated."

Never test for an error condition you don't know how to handle. -- Steinbach

Working...