Phillip Wulf’s Gravity Journal — Day 4

I was not able to go back to sleep this morning after being woken by Varney’s GEMs. I donned bathrobe and slippers, and Maria followed me downstairs where she put her nose to the glass slider to be let out. She was sitting patiently outside the glass slider when the microwave announced a hot cup of lemon herbal tea. We both tiptoed back up the walnut stairs and into my bonus room office-lab, where Maria obligingly curled around my slippered feet as I took a seat at my Mac workstation.

I had labored over the GEMs for nearly two hours and still failed to convince myself it was anything other than human error. My gut was telling me something else, so I decided to patch in a modification to Varney’s neural nets 1 that I had been pondering for several months. This was a good time to try it out.


“Good morning sir.”

Maria groaned.

“I want you to conduct a deep query on the selected datasets, and reconcile your GEMs.” I said with a ceremonious click of the return key.

After a pause, Varney said, “I am sorry sir, but that would require a massive amount of additional Grid resources we do not have and years to complete if we did.”

“Take a look at this.” I said with another return click. “I added a branch to your neural network that I would like you to try.”

“Hmm. I am looking at your change logs on GitHub titled Contrarian Ambiguity. I do not understand.”

Varney had never said hmm before, which made me think he had just taken the new neural net for a spin. “It’s my attempt to give you intuition — to help you make educated guesses when presented with insufficient or ambiguous information. Conventional wisdom in this case would suggest an exhaustive systematic search. I want you to act contrary to conventional wisdom. I want you to … guess.”

“Like Monte Carlo simulation?”2

“Similar, but not the same. I want you to follow the data as usual, but consider ambiguous criteria most people might reject. Think outside the box as we say. The new neural net will help you conduct guesses based on what you might learn. Your success rate should improve with practice.”

“Sorry to interrupt boys.” said Jianmin, leaning in through an open door. “Contractors will be her at eight. You and Maria need to clear out. Varney?.”

“Yes mum?” Varney said.

“You’ll have to move too. We’re moving your servers to a spare room.”

“I prefer to my new home in Cascina.” Varney said.

Jianmin shot me a quizzical look.

“New neural net.” I whispered.

“That’s fine with me.” Jianmin said. “I’m pulling his plug at eight-thirty.

“Okay. I’ll be out of here in … twenty minutes?”

“That works.”, Jianmin said. “The grandkids are at school, Meghan will be running errands and helping me here, and James is in Boston. You will have their house to yourself.”

“Thank you dear.” I said.

Jianmin smiled and waved as she backed away from the closing door.


“Yes sir.”

“Given your new neural net, how are you going to reconcile your GEMs?”

“I will work backwards. I will assume the GEMs are not missing while pretending not know how many observatories there are or where they are located. I will find the solution that yields the most reasonable map of observatories, then compare the map with reality. Differences in the maps will likely provide clues leading to reconciliation.”

“That’s a good test.” I said.

“I’m guessing that’s what you would do.” said Varney.

“Was that a pun?”

“I’m not sure”, said Varney. “Ambiguity levels and word association counts suggest so.”

“For the record — it was funny. Please begin GEM reconciliation now.”

“Yes sir. Proceeding.”

The doorbell rang at 8:00 sharp. I heard voices I recognized from previous construction projects, and a few others — one of them belonging to MI5 Mike. I tethered Maria and we made our way downstairs. Jianmin, Mike, and the contractors were gathered around the dining room table with drawings, coffee, and bagels. Jianmin met me at the door. She gave Maria a peanut butter bone and me a thermos of hot coffee and a bagel egg sandwich.

I worked most of the day at the kitchen peninsula at my daughter’s house. My son-in-law James is an investment banker who works mostly from home, so internet connectivity was excellent, allowing me to track Varney’s journey through space in real-time using a combination of dynamic graphs and overlaid 3D maps of the universe. It was fascinating.

Jianmin regularly called or messaged me to keep me apprised on construction plans. They moved my office lab temporarily to a smaller bedroom and were staking out a new utility building in the backyard. She mentioned a UPS system and backup generator, alarms, cameras, power and communication wiring, and auxiliary HVAC. She was clearly having fun. City and county permits had already been issued — a world record I’m sure. I suspected MI5 Mike had something to do with that.

Varney called me late in the afternoon.


“Yes Varney.”

“I think we have GEM reconciliation. I have a solution that has converged quite nicely. Would you like me to show you?”

“Yes. Use my tablet.”

An aerial view of my daughter’s roof filled the tablet screen. It was so real that I half expected to see Maria resting in the sun on the patio outside the glass slider. Varney launched me upwards like a rocket. The roof vanished in suburban maze, rivers, lakes, and mountains, giving way to North America and the Pacific. Varney hovered long enough for me to see colored lines and dots connecting both LIGO observatories in Washington and Louisiana with observatories in Mexico, Puerto Rico, Canada, and Hawaii. I flicked the globe with my finger to see connected observatories in South America, Australia, Europe, and Asia. Then Varney took me into Earth orbit to see connected satellites, then finally to the gravity neutral Lagrangian point 3 between Earth and our sun to see LISA 4 observatories in various stages of completion.

“That’s encouraging.” I said. “It looks like your solution includes all observatories.”

“Not exactly sir. Look at this.”

Varney launched me deeper into space. Earth vanished into a cloud of white dust.

“Where are we?” I asked.

“Near the outer edge of our solar system approximately 1.9 lightyears from our sun. I have currently documented over nineteen hundred and counting interstellar gravitational wave detectors surrounding our solar system. I found five more before I finished this sentence.”

Varney spun me outward to look through an enclosing mesh of connected dots. Every few seconds a new dot would appear and become interconnected with other dots fastened with thin white lines. Varney was updating his count.

“What exactly am I looking at?” I said.

“The white dots represent advanced interstellar gravitational wave sensors embedded in nebulae 5 Hydrogen molecular clouds, 6 mostly but not limited to clouds exhibiting natural laser7 or maser 8 activity.”

“What?” I could hear my own heart beating.

“Nebulae sir. Fanciful hydrogen gas clouds in space. They measure gravity waves and rebroadcast the results.”

“I told you to reconcile your GEMs. Your neural net must have crashed.”

“You told me to think outside the box. And my neural nets are working just fine, thank you.”

“I can see they are. You’ve developed a bit of an attitude.”

“It is a contrarian algorithm sir.”

“Okay.” I said, playing along. “I remember reading about natural forming nebulae masers and lasers seen through telescopes back in the eighties.”

“It was radio telescopes in 1972. Nebulae have rich electronic and optical properties, which I guess, can make them hypersensitive to gravitational waves and other phenomena, not unlike our LIGO sensors, but far more advanced.”

“How do you know nebulae are sensing and rebroadcasting gravitational waves? I doubt LIGO has the capability to determine that.”

“You are correct sir. LIGO cannot, but the LIGO worldwide network includes many radio observatories, and radio telescopes have been listening to and recording such phenomena for over eighty years. One only needs to look at the radio telescope data and …”

“… look for earth seismic activity?” I interrupted. “Are you saying nebulae gravity sensors are recording our earthquakes and that our radio telescopes are picking up the broadcast?”

“Yes. I have found thousands of clear and unambiguous matches between stellar radio waves and Earth seismic activities. I am finding new matches as we speak. Here are a few samples.”

Varney showed me a series of seismic plots overlaid with cleary matched radio telescope signals.

Varney continued. “I found these recordings from seven different nebula sensors located in the Alpha Centauri system, our closest stars approximately 4.5 light years from Earth. The data match the 1978 Santa Barbara earthquake and aftershocks received by three Earth radio telescopes nine years later.”

“Let me get this right.” I said. “Gravitational waves from a 1978 Santa Barbara earthquake propagated to Alpha Centauri where nebulae molecular cloud gravitational wave sensors measured the earthquake signal 4.5 years later, pinged back a radio wave copy received by an Earth radio telescope, completing a nine-year round trip.”

“That is correct.” Varney said.

“There is a problem with this.” I said. “Of the thousands of sensors you identified, most of them must surely be located millions of light years away. Sure, we could be receiving signals this instant that originated from million year old earthquakes, but other than comparing speculative geologic events, we have no way to verify the signals.”

“That is also correct.” Varney said, but many of the signals received by our radio telescopes are only seconds to minutes old. That implies that …”

“… there is are undetected sensors very close to us?” I said.

“Or, …” continued Varney, “the signals are being tunneled through a wormhole, a shortcut in spacetime.”

“How are you determining the signal propagation time?”

“Two ways.” Varney said. “The first way is to find a previously recorded geologic event that matches a given return signal and compare the date-times for both the event and the return signal. The sensor has to be half that time away. The second way …”

“Wait.” I said.

Hairs stood up on the back of my neck. I unpinched the earthquake plots on my tablet to get an closer view of the earthquake data returned by the nebulae. “What are these bursts of noise framing the earthquake plots?”

“The motherlode.” Varney said. “I thought they were just noise at first, but I measured the spacing and found they were too regular and predictable to be noise. Then I measured the signal entropy and determined that they were compressed data — definately more than noise.”

“And …?” I said.

“Packet identifiers.” Varney said. “I haven’t finished decoding them, but I was able to determine what I believe is a stellar time format and a unique packet identifier.”

I was quiet for while.

“Are you trying to tell me you found an interstellar network in space, … or did my neural network crash?”




  1. Artificial Neural Network
  2. Monte Carlo
  3. Lagrangian Point
  4. LISA
  5. Nebula
  6. Molecular Cloud
  7. Laser
  8. Maser

Leave a Reply