Options

Operad-based software

edited January 2016 in Chat

A few weeks ago I was approached by a company who is applying for a government grant to design more reliable systems of systems software using new mathematical techniques, especially operads. This company already makes systems-of-systems software, and they were approached by the US government to apply for this grant. They have some mathematicians who work for them, but they needed to bring an expert on operads on board to make their proposal credible. I understand operads pretty well. My job would be to come up with some new mathematics, while they would be doing some programming to modify the software they already make.

I decided to give it a try, since it seems like an interesting challenge, the company seems to have a reasonable chance of getting the grant, the people who work there seem smart, and the grant would help fund a grad student as well as giving me some more money. I hope working on this will give me some new ideas about complex networked systems. The grant not only allows but requires that my work be published - that's good.

If it gets funded, the project will start in August and last 4 years. I think I won't say who the company unless the project goes forward. I'll let you know when I find out more!

I've been pretty busy helping them write their grant proposal, but right now I'm done with that, at least for a while.

Comments

  • 1.

    I would be curious to know what makes software operad based.

    Comment Source:I would be curious to know what makes software operad based.
  • 2.
    edited January 2016

    The idea is that it'll use the theory of operads to describe how smaller systems are stuck together to form bigger ones. For a quick explanation of operads, try this:

    • John Baez, Review of Operads in Algebra, Topology and Physics by Martin Markl, Steve Schnider and Jim Stasheff.

    For their application to systems-of-systems, try this:

    • David I. Spivak and Joshua Z. Tan, Nesting of dynamic systems and mode-dependent networks.

    Needless to say, if I get paid to work on this for 4 years I'll have a lot more to say about it!

    Comment Source:The idea is that it'll use the theory of operads to describe how smaller systems are stuck together to form bigger ones. For a quick explanation of operads, try this: • John Baez, [Review of _Operads in Algebra, Topology and Physics_ by Martin Markl, Steve Schnider and Jim Stasheff](http://math.ucr.edu/home/baez/operad.pdf). For their application to systems-of-systems, try this: • David I. Spivak and Joshua Z. Tan, [Nesting of dynamic systems and mode-dependent networks](http://arxiv.org/abs/1502.07380). Needless to say, if I get paid to work on this for 4 years I'll have a lot more to say about it!
  • 3.

    I think I won't say who the company unless the project goes forward. I'll let you know when I find out more!

    OH and for those who want to guess - Wikipedia says:

    Corporations active in this research such as The MITRE Corporation, BAE Systems, Northrop Grumman, Boeing, Raytheon, Thales Group, CAE, Saber Astronautics, and Lockheed Martin.

    It seems there are also european collaborations, like T-Area-SoS:

    T-AREA-SoS (Trans-Atlantic Research and Education Agenda on Systems of Systems) aims to increase European competitiveness in, and improve the societal impact of, the development and management of large complex systems in a range of sectors through the creation of a commonly agreed EU-US Systems of Systems (SoS) research agenda.

    Would be for example interesting to hear their solution proposals to these kind of european large scale problems.

    Comment Source:>I think I won't say who the company unless the project goes forward. I'll let you know when I find out more! OH and for those who want to guess - <a href="https://en.wikipedia.org/wiki/System_of_systems">Wikipedia</a> says: >Corporations active in this research such as The MITRE Corporation, BAE Systems, Northrop Grumman, Boeing, Raytheon, Thales Group, CAE, Saber Astronautics, and Lockheed Martin. It seems there are also european collaborations, like <a href="https://www.tareasos.eu/">T-Area-SoS</a>: >T-AREA-SoS (Trans-Atlantic Research and Education Agenda on Systems of Systems) aims to increase European competitiveness in, and improve the societal impact of, the development and management of large complex systems in a range of sectors through the creation of a commonly agreed EU-US Systems of Systems (SoS) research agenda. Would be for example interesting to hear their solution proposals to these kind of <a href="http://www.theguardian.com/world/2016/jan/30/greece-threatened-closure-eu-border-dystopia">european large scale problems.</a>
  • 4.

    The idea is that it'll use the theory of operads to describe how smaller systems are stuck together to form bigger ones

    I remember sheaves being popular for this many years ago, particularly the work of Joseph Goguen and Grant Malcom. Uday Reddy and Yellamraju Venkata Srinivas also did interesting work in this area. I wonder what has happened in that area since Goguen's passing and how that work relates to the operad approach.

    How free would you be to discuss this work with the outside world?

    Comment Source:> The idea is that it'll use the theory of operads to describe how smaller systems are stuck together to form bigger ones I remember sheaves being popular for this many years ago, particularly the work of Joseph Goguen and Grant Malcom. Uday Reddy and Yellamraju Venkata Srinivas also did interesting work in this area. I wonder what has happened in that area since Goguen's passing and how that work relates to the operad approach. How free would you be to discuss this work with the outside world?
  • 5.

    Bet it's related to this one: https://govtribe.com/project/design-of-dynamically-composed-systems-of-systems-request-for-information-rfi

    Been working simulation tools for the big SoS projects over the past 15 years, another is pending shortly for me

    Comment Source:Bet it's related to this one: https://govtribe.com/project/design-of-dynamically-composed-systems-of-systems-request-for-information-rfi Been working simulation tools for the big SoS projects over the past 15 years, another is pending shortly for me
  • 6.
    edited March 2016

    Bet it's related to this one...

    Nope, it's not related to the the DARPA "Design of Dynamically Composed Systems of Systems". But thanks for pointing that out, Paul!

    How free would you be to discuss this work with the outside world?

    There is some proprietary software, and I'd be signing a nondisclosure agreement, mainly to keep me from telling the competitors about that software. On the other hand, I won't be doing any programming. My actual job will be to come up with new mathematical ideas on how to use operads to describe systems made of subsystems that are communicating along channels that may or may not work, and I'm encouraged to publish papers about this - in fact, I think this is required by the grant proposal.

    So, I believe that I'll have a lot of freedom to talk about the issues I'm actually interested in. But needless to say, I'm a bit worried about this. If I don't like this sort of arrangement, I'll avoid doing something like this again. But I'm hoping that a bit of contact with the so-called "real world" will make me come up with a lot of interesting math.

    Comment Source:> Bet it's related to this one... Nope, it's not related to the the DARPA "Design of Dynamically Composed Systems of Systems". But thanks for pointing that out, Paul! > How free would you be to discuss this work with the outside world? There is some proprietary software, and I'd be signing a nondisclosure agreement, mainly to keep me from telling the competitors about that software. On the other hand, I won't be doing any programming. My actual _job_ will be to come up with new mathematical ideas on how to use operads to describe systems made of subsystems that are communicating along channels that may or may not work, and I'm encouraged to publish papers about this - in fact, I think this is required by the grant proposal. So, I believe that I'll have a lot of freedom to talk about the issues I'm actually interested in. But needless to say, I'm a bit worried about this. If I don't like this sort of arrangement, I'll avoid doing something like this again. But I'm hoping that a bit of contact with the so-called "real world" will make me come up with a lot of interesting math.
  • 7.

    It looks like the company got the grant! Apparently some details need to be negotiated... but it seems to be happening.

    Comment Source:It looks like the company got the grant! Apparently some details need to be negotiated... but it seems to be happening.
  • 8.

    If it gets funded, the project will start in August and last 4 years. I think I won't say who the company unless the project goes forward.

    ... but it seems to be happening.

    So as the project goes forward, which company is it?

    Comment Source:>If it gets funded, the project will start in August and last 4 years. I think I won't say who the company unless the project goes forward. >... but it seems to be happening. So as the project goes forward, which company is it?
  • 9.
    edited March 2016

    I'll announce it iff the project actually goes forward---that is, iff the company actually gets the grant.

    Comment Source:I'll announce it iff the project actually goes forward---that is, iff the company actually gets the grant.
Sign In or Register to comment.