A MYSTERY - Why Did Bio-Supercomputing Go Dark?

a reply to: Bedlam

Apparently we're posting at cross-purposes.

Apparently.

To dispel the assumptions on my part, are you:

1) Just posting about biocomputational vector processors because they're interesting and the mystery of why no one seems to be working on it was a literary device to increase the interest

No.

2) You're actually wondering if much of the entire field has gone dark (what I picked up from your op)

No. I was enquiring about the very specific working model that was created last year (news coverage and PNAS article referenced in the OP).

3) You're wondering about the fate of that specific instance in the OP

Yes.

Having been a not insignificant part of a design team that built and fielded one of the fastest vector processors in the world at that time, the general topic is of interest. And I have at least a few insights into where things go and how they end up. This one is a bit afield from my usual stomping grounds but that doesn't mean I can't find out, in time.

I am coming at this from the perspective of biology and a special interest in proteins and prions - not physics or computer design. Specifically, I have been monitoring progress in prion proteins, protein engineering, biocomputing for medical applications and devices, and transhumanism. Aside from the hard info and breakthroughs, I am interested in speculating about creating an (organic) brain-machine hybrid replete with evolutionary, cellular and conscious memories - and transplanting said hybrid into robots. For fictional purposes, obviously.

I am aware that popular thinking dismisses the use of organics for computers, and the fictional trend is to speculate about transplanting 'consciousness' into mechanical computers - thereby attaining transhumanists' goal of 'immortality.' However, I also know that such an approach will miss the mark pretty much completely. Hence my 'curiosity' about the parallel processing bio-supercomputer announced last year, as referenced in my op.



PS. Two of the research partners on the project are businesses: Molecular Sense Ltd and Philips Innovation Services. Both are involved with developing "technical solutions and devices for the markets of biomedical devices and biomaterials." Seems clear that the short-term use of the protein technology developed for the bio-supercomputer model will be used for biomedical devices and applications - but again, I'm most interested in the longer term applications and the status of the work.

a reply to: soficrow

Perhaps the title of the thread threw me off. From the title, it seemed you were interested in the general case and were using the OP as a specific example. You didn't ask 'why did THIS project go dark' you asked 'why did (all of) bio-supercomputing go dark' with 'I can't find any recent research on the topic' for sauce. So I was looking at more of an approach to look for where the people went and what they were doing that had once worked in the field and it seemed rather puzzling that not only were you not interested in that conversation but became snarky about me not reading the OP, which wasn't true at all.

So if you're more interested in that one instance, there are ways to pursue that one as well. Might be fun. I'm going to be on and off line here a bit - this computer I use away from work has become senile and undependable from age and being transported about one too many times, and tonight starts my journey in assembling a new machine and retiring the old, a task that seldom goes as tidily as one might want.

Whilst I'm waiting for the interminable and no doubt repetitious task of installing and re-installing the OS until I get it where I want it, I'll use the time I normally spend cursing Bill Gates' monstrosity to start poking about with my old Macbook, starting with pulling up any recent prior related work by the device's authors through my corporate account so I don't have to pay for it, and start by seeing what they did, what they did prior, and who they are. Once I get a better look at them and the project and its antecedents, it's off to poking and prying into their more recent activities. Hopefully at least some of them are domestic. It's tougher to get the dope on Europeans as I have to ask someone else to do the snoopy bits.

a reply to: Bedlam

Thank you - and sorry about the confusion. Did my best.

Anyway - the locals are from McGill (Canada) and Berkeley - otherwise you're looking at Germany, the Netherlands, Sweden and London. The 2 businesses are Swedish as I recall, so it seems that that's where the initiative started. Philips has a long history of success with innovative consumer products - but I can't speak to their success with specialized industry markets.

Thanks again.

~ sofi

a reply to: Bedlam

Thanks again. More info to help you on your way.

Father and son Dan Nicolau Sr. and Jr. started the project. The lead Nicolau is with Dept. of Bioengineering at McGill University.

Can’t believe I missed it, but DARPA is one of the funders.

Key point: Seems the researchers think protein-based computing is more readily scalable and practical than DNA computing and quantum computing.

This research was funded by: The European Union Seventh Framework Programme; Defense Advanced Research Projects Agency (DARPA); NanoLund; The Miller Foundation; The Swedish Research Council; The Carl Trygger Foundation; the German Research Foundation; and by Linnaeus University.

Biological supercomputer model could change how we solve complex problems

Protein-powered chip makes biocomputer extremely energy efficient

Researchers at McGill University have helped to develop a model of a biological supercomputer that can solve complex mathematical problems using very little energy.

……the new model biocomputer — developed by a team of international scientists from Canada, the U.K., Germany, The Netherlands and Sweden — is energy-efficient, performs multiple calculations simultaneously and is roughly the size of a book.

"A biocomputer requires less than one per cent of the energy an electronic transistor needs to carry out one calculation step," study co-ordinator Heiner Linke, director of nanoscience at Lund University in Sweden, said in a press release.

….At the core of the model is a 1.5 square-centimetre microchip, which uses myosin, molecular motors that carry out mechanical tasks in living cells, to move protein filaments along artificial paths.

….This is not the first research to apply parallel computing to complex problem-solving. DNA computing and quantum computing, for example, take a similar approach.

"However, these approaches have not proven, so far, to be scalable and practical from a fabrication and operational perspective," the researchers note.

Perhaps we should go to first principles and ask what the hell this thing is expected to do.

originally posted by: Maxatoria
Perhaps we should go to first principles and ask what the hell this thing is expected to do.

Maybe. ...Do you have a DARPA contact?

a reply to: soficrow

Doubt it sitting here in the UK But the problem starts with someone describing it and from there on people can work to solve it.

Doubt i'd much clearance in the US for more than knowing what the base commanders lunch will be.

a reply to: Maxatoria

lol. ...Maybe Bedlam can help.

Still can't believe I missed the DARPA funding.

originally posted by: soficrow

originally posted by: Maxatoria
Perhaps we should go to first principles and ask what the hell this thing is expected to do.

Maybe. ...Do you have a DARPA contact?

Generally DARPA floats the specs and reasons for desirability around the community. I have been away from home base a long time and have quit looking, as I'm stuck here interminably. And I'd normally ignore biologicals anyway, unless it was especially creepy and/or fun looking. When I get this monster up and running all the way I'll poke around in the RFQ pile and see if they fielded it around.

a reply to: Bedlam

Generally DARPA floats the specs and reasons for desirability around the community. ...I'd normally ignore biologicals anyway, unless it was especially creepy and/or fun looking. When I get this monster up and running all the way I'll poke around in the RFQ pile and see if they fielded it around.

Reviewing DARPA’s calls for proposals here.

Some computer developers seem to think the competition is about building the next-gen supercomputer - quantum computing v/s biocomputing. Wrong.

The immediate key goals clearly have to do with enabling, connecting, expanding, and processing sensory perception. The focus is biological, and bridging the bio-electronic gap (eg., with hybrids).

a reply to: soficrow

DARPA are known as the bat s--t agency, generally they float out idea and see what comes back, its their job to be like you would think of some eccentric Victorian gentleman who wants to paddle across the Atlantic on a horse, most ideas don't work as such but do encourage people to have the ideas.

Biological is a lot more of a PITA than silicon generally, Intel or AMD aint going to be releasing a general purpose CPU thats 2 inch in size and just needs probably 100 watts per hour at most but you need a specialist to keep it running who turns up every Tuesday.

You are more looking at AI by the looks of it and thats a totally different area, theres emulations of bio neural systems that learn out there, they're not bad but like the MS one that ended up spewing out racist text its only as good as the input (like any kid)

a reply to: Maxatoria

Okay, but - it's happening. Meaning - enabling, connecting, expanding, and processing sensory perception. Going biological, and bridging the bio-electronic gap (eg., with hybrids). I think maybe the use of proteins gets around the PITA thing - super stable, etc.

originally posted by: soficrow

The immediate key goals clearly have to do with enabling, connecting, expanding, and processing sensory perception.

The target goal is likely to be design of an interfacer. They've wanted that for a long while.

originally posted by: Bedlam

originally posted by: soficrow

The immediate key goals clearly have to do with enabling, connecting, expanding, and processing sensory perception.

The target goal is likely to be design of an interfacer. They've wanted that for a long while.

Uh huh. Looks like it.

But to what end?

originally posted by: soficrow

But to what end?

Oh, there are just LOTS of uses for that, depending on how it works.

a reply to: Bedlam

Yes. But. Endgame?

originally posted by: soficrow
a reply to: Bedlam

Yes. But. Endgame?

The obvious goal is weapons control. Faster linkage to the plane you're flying. If you can manage in time to do realistic sensory feeds, then your mind might be convinced that you ARE the plane, its senses your senses, the controls your body. It would be kickass. And eventually you might be able to do that by remote, the lag might be a bit offputting. But it might work.

A non-obvious goal - enhanced interrogation. Why bother with torture when you can pipe it in to the noggin? Or maybe you can just jack in and remember it out of their heads, no torture needed, just some prep time.

Devious end goals - transfer of consciousness to another body, or a mechanism. Or the 'Vacuum Flowers' punchline - you hypercube n number of people together to form a superhuman collective mind. That didn't end well, though.

originally posted by: Bedlam

originally posted by: soficrow
a reply to: Bedlam

Yes. But. Endgame?

The obvious goal is weapons control. Faster linkage to the plane you're flying. If you can manage in time to do realistic sensory feeds, then your mind might be convinced that you ARE the plane, its senses your senses, the controls your body. It would be kickass. And eventually you might be able to do that by remote, the lag might be a bit offputting. But it might work.

A non-obvious goal - enhanced interrogation. Why bother with torture when you can pipe it in to the noggin? Or maybe you can just jack in and remember it out of their heads, no torture needed, just some prep time.

Devious end goals - transfer of consciousness to another body, or a mechanism. Or the 'Vacuum Flowers' punchline - you hypercube n number of people together to form a superhuman collective mind. That didn't end well, though.

Like almost all other industries, the military is phasing out human personnel - as in, Humans Need Not Apply. So your first two speculations don't really play. For example, the following was published in 2004, and many of the author's seemingly off-the-wall speculations are already a reality.

2004. Anticipating Military Nanotechnology

....Vehicles used for surveillance, reconnaissance, and target location could be miniaturized much further. NT and microsystems technology would permit vehicles and mobile robots of decimeter down to millimeter size, some using biomimetic forms of propulsion. One variant would be to use small animals (rats, insects) controlled by implanted electrodes. Although the munitions payload of small robots would be limited, they could attack at sensitive spots, or act in swarms to achieve a mass effect. Small satellites and launchers may significantly reduce space launch costs; swarms of them could act as large effective radar, communication, or electronic intelligence antennae. Small satellites could also be effective in attacking larger satellites – by direct hit or by manipulation after docking.

Your third speculation has obvious merit - "transfer of consciousness to another body, or a mechanism" - but begs the question, "Why?"

a reply to: soficrow

If we're talking of military use its got to be simple for the field teams to swap in/out when its needed and robust enough to be able to take all sorts of crap as a Bio operated tank will certainly come under stress at points just the same as a jet will.

The real problem is bringing it all together as its probably a decade or two from being something you can just buy.

You need to design the externals to meet the bio comp and given its a living organism theres going to be some ethical issues and also how the hell do you teach that jar of gloop what 2+2 is.

a reply to: Maxatoria


As I said earlier, I doubt biocomputing is being developed for use in battle by the military. So the question is, "What DO they want it for?"

I read something recently about live insects carrying micro-cams, and being controlled using a micro-pack on their backs. Sorry - can't find it, but here's something else entirely. Point being, soldiers are going obsolete already.

….the Samsung SGR-A1 sentry gun installed on the South Korean border. The SGR-A1 is capable of asking humans for a password and shooting them with either lethal or non-lethal rounds if it doesn’t hear the correct answer.

....responsibility has been transferred from human to machine.

FOUND something (not the latest tho):

Remote-controlled flying cyborg beetles could replace drones

A group of engineers and scientists has come up with a way to remotely control insects, which they believe could provide an "improved alternative to remote-controlled drones".

...Drones, or unmanned aerial vehicles, are becoming increasingly prolific. Their current and proposed uses include everything from photography and filming, to delivery and emergency services, as well as for military, transport and construction purposes.

..."This technology could prove to be an improved alternative to remote-controlled drones as it could go into areas which are not accessible before," he told the Telegraph.

...Experiments were carried out with Mecynorrhina torquata, or giant flower beetles, which measure and average six centimetres in length and eight grams in weight.

This particular species was chosen for its ability to lift relatively heavy loads, and all the beetles involved in the project went on to live for their usual lifespan of between five and six months.