Coffeehouse Thread

13 posts

Forum Read Only

This forum has been made read only by the site admins. No new threads or comments can be added.

Helios

Back to Forum: Coffeehouse
  • User profile image
    exoteric

    The Singularity research operating system has evolved into Helios which implements satellite kernels in support of heterogeneous hardware environments. It's been a long time since we heard about Singularity. It looks like it's alive and well - perhaps now we should just talk about Helios, as it appears to be the successor (nice name too).

     

    http://research.microsoft.com/apps/pubs/default.aspx?id=81154

     

    I bet a lot of niners would love to hear more about this thing. This is cutting edge.

  • User profile image
    Charles

    The Helios team is in Cambridge, England. Not sure I can get to them any time soon. But, this is on the radar.

    C

  • User profile image
    felix9

    I think Helios is just a research branch of Singularity, rather than a successor.

    The mysterious 'Midori' project should be a much more developed 'product' based on Singularity, but ms just wont talk anything about it, you dont even try.

    However, you can find some very interesting stuff her : http://www.sdtimes.com/link/32627 and here: http://blogs.zdnet.com/microsoft/?p=4149

  • User profile image
    Royal​Schrubber

    Cool play with the name - Helios is greek god of sun, and sun is a star which is a kind of solar object that that exists before it implodes into singularity of black hole. Still, I prefer Singularity - it has a bit of a threatening connotation - like MS wanted to say they finally have something that will finally be able to suck and destroy linux and every other OS so that there will be only one kernel in all processors. I guess then Helios won't suck its competition, only overshine it. Smiley

  • User profile image
    Charles

    felix9 said:

    I think Helios is just a research branch of Singularity, rather than a successor.

    The mysterious 'Midori' project should be a much more developed 'product' based on Singularity, but ms just wont talk anything about it, you dont even try.

    However, you can find some very interesting stuff her : http://www.sdtimes.com/link/32627 and here: http://blogs.zdnet.com/microsoft/?p=4149

    Yes. Helio is not the evolution of the Singularity project.

    Midori is a private incubation project. There's no point in talking publicly about it. Speculate all you want... Smiley

    C

  • User profile image
    turrican

    At first, I thought you were talking about http://www.textpad.com

    "Helios Software Solutions"

    Smiley

  • User profile image
    Charles

    Satellite kernels are a very compelling idea Smiley

    C

  • User profile image
    exoteric

    Charles said:
    felix9 said:
    *snip*

    Yes. Helio is not the evolution of the Singularity project.

    Midori is a private incubation project. There's no point in talking publicly about it. Speculate all you want... Smiley

    C

    It's allright, you can share information with a few trustees Wink

    Seriously, there's still a lot of public research, so we are not completely in the dark Smiley

  • User profile image
    JoshRoss

    exoteric said:
    Charles said:
    *snip*

    It's allright, you can share information with a few trustees Wink

    Seriously, there's still a lot of public research, so we are not completely in the dark Smiley

    As someone who has compiled and ran Singularity, I can tell you that most of the thrill is just getting the thing to run. After that, unless you want to put huge amounts of work in for minuscule results, I would rather just read the white papers. There are a lot of wild ideas left in system software and I'm sure that most of these will never obtain the critical mass required to be useful outside of a sand-boxed VM.

     

    EDIT:

     

    Another thing about research operation systems, they seem to have cool features and terrible UX. Case in point, Squeak.

     

    -Josh

  • User profile image
    exoteric

    JoshRoss said:
    exoteric said:
    *snip*

    As someone who has compiled and ran Singularity, I can tell you that most of the thrill is just getting the thing to run. After that, unless you want to put huge amounts of work in for minuscule results, I would rather just read the white papers. There are a lot of wild ideas left in system software and I'm sure that most of these will never obtain the critical mass required to be useful outside of a sand-boxed VM.

     

    EDIT:

     

    Another thing about research operation systems, they seem to have cool features and terrible UX. Case in point, Squeak.

     

    -Josh

    I have read a couple of the papers but have also downloaded the code on Codeplex - not to compile and run it - that doesn't really interest me at this stage - but to look at the actual code. It looks like some very exciting work is taking place in Singularity, Helios and Midori.

     

    It will probably take years before there is something public for us to "see". -- There's two kinds of people, those who focus on the visuals, those who focus on the fundamentals and then of course the ones who can't count as well as the ones who are interested in both - but it's first things first and UI is certainly way down on the list; composition must have strong fundamentals.

     

    The things that excite me a lot are the languages and compilers aspects: Sing#, MSIL+, TAL, iTalX, Bartok, Phoenix, and so on.

  • User profile image
    vesuvius

    exoteric said:
    JoshRoss said:
    *snip*

    I have read a couple of the papers but have also downloaded the code on Codeplex - not to compile and run it - that doesn't really interest me at this stage - but to look at the actual code. It looks like some very exciting work is taking place in Singularity, Helios and Midori.

     

    It will probably take years before there is something public for us to "see". -- There's two kinds of people, those who focus on the visuals, those who focus on the fundamentals and then of course the ones who can't count as well as the ones who are interested in both - but it's first things first and UI is certainly way down on the list; composition must have strong fundamentals.

     

    The things that excite me a lot are the languages and compilers aspects: Sing#, MSIL+, TAL, iTalX, Bartok, Phoenix, and so on.

    There are also those of us that like to make money. Language theory is not the most lucrative of areas. I am currently working in research at a sh** hot University, and the principal requirements are my ability to create systems that work that assist in their objectives. Ultimately, this results in products that that university can sell, and generate revenue for future research. This is a good thing.

     

    For the most part a lot of the projects you list are interesting, but there is no immediate benefit (apart from intellectual edification of course) and in order to reap any benefit it would take a company like Microsoft with the investment capabilities to make it work. Research work in general isn't the best paid work, and if you are going to be involved in it, it must be beneficial so you can launch the damn thing.

  • User profile image
    exoteric

    vesuvius said:
    exoteric said:
    *snip*

    There are also those of us that like to make money. Language theory is not the most lucrative of areas. I am currently working in research at a sh** hot University, and the principal requirements are my ability to create systems that work that assist in their objectives. Ultimately, this results in products that that university can sell, and generate revenue for future research. This is a good thing.

     

    For the most part a lot of the projects you list are interesting, but there is no immediate benefit (apart from intellectual edification of course) and in order to reap any benefit it would take a company like Microsoft with the investment capabilities to make it work. Research work in general isn't the best paid work, and if you are going to be involved in it, it must be beneficial so you can launch the damn thing.

    One does not have to be a researcher to appreciate excellent research and ideas. This is just out of interest it has no or little practical impact on my day to day work where the tools and primary language I work with are extremely undernourished even compared to Visual Studio and C# (with some exceptions); battling these constraints is an interesting challenge by itself but a daily fix of research is stimulating and provides perspectives to the world we currently live and what we could live in some years time.

  • User profile image
    Bass

    exoteric said:
    JoshRoss said:
    *snip*

    I have read a couple of the papers but have also downloaded the code on Codeplex - not to compile and run it - that doesn't really interest me at this stage - but to look at the actual code. It looks like some very exciting work is taking place in Singularity, Helios and Midori.

     

    It will probably take years before there is something public for us to "see". -- There's two kinds of people, those who focus on the visuals, those who focus on the fundamentals and then of course the ones who can't count as well as the ones who are interested in both - but it's first things first and UI is certainly way down on the list; composition must have strong fundamentals.

     

    The things that excite me a lot are the languages and compilers aspects: Sing#, MSIL+, TAL, iTalX, Bartok, Phoenix, and so on.

    I think what's cool about Singularity is the possibility of running such an OS on a machine with no MMU, and still having strong multiprocessing ability.

Conversation locked

This conversation has been locked by the site admins. No new comments can be made.