Get the full ebook with Bonus Features for a Better Reading Experience on ebookmeta.

com

Graph Neural Networks: Foundations, Frontiers, and

Applications First Edition Lingfei Wu

https://ebookmeta.com/product/graph-neural-networks-
foundations-frontiers-and-applications-first-edition-
lingfei-wu/

OR CLICK HERE

DOWLOAD NOW

Download more ebook instantly today at https://ebookmeta.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Graph Neural Networks in Action (MEAP Version 4) Keita

Broadwater

https://ebookmeta.com/product/graph-neural-networks-in-action-meap-
version-4-keita-broadwater/

ebookmeta.com

Cambridge IGCSE and O Level History Workbook 2C - Depth

Study: the United States, 1919-41 2nd Edition Benjamin
Harrison
https://ebookmeta.com/product/cambridge-igcse-and-o-level-history-
workbook-2c-depth-study-the-united-states-1919-41-2nd-edition-
benjamin-harrison/
ebookmeta.com

Artificial Neural Networks with Java: Tools for Building

Neural Network Applications, 2nd Edition Igor Livshin

https://ebookmeta.com/product/artificial-neural-networks-with-java-
tools-for-building-neural-network-applications-2nd-edition-igor-
livshin/
ebookmeta.com

Platonism and Poetry in the Twelfth Century The Literary

Influence of the School of Chartres 1st Edition Winthrop
Wetherbee
https://ebookmeta.com/product/platonism-and-poetry-in-the-twelfth-
century-the-literary-influence-of-the-school-of-chartres-1st-edition-
winthrop-wetherbee/
ebookmeta.com
ISE EBook Online Access for Essentials of Investments 12th
Edition Bodie

https://ebookmeta.com/product/ise-ebook-online-access-for-essentials-
of-investments-12th-edition-bodie/

ebookmeta.com

Commando Club Curve 15 1st Edition Ja Lafrance

https://ebookmeta.com/product/commando-club-curve-15-1st-edition-ja-
lafrance-2/

ebookmeta.com

Juno Loves Legs Karl Geary

https://ebookmeta.com/product/juno-loves-legs-karl-geary/

ebookmeta.com

The One That Knows No Fear 1st Edition Steve Stred

https://ebookmeta.com/product/the-one-that-knows-no-fear-1st-edition-
steve-stred/

ebookmeta.com

Black Haze Violence Sacrifice and Manhood in Black Greek

Letter Fraternities Second Edition Ricky L Jones

https://ebookmeta.com/product/black-haze-violence-sacrifice-and-
manhood-in-black-greek-letter-fraternities-second-edition-ricky-l-
jones/
ebookmeta.com
Will There Be Tiers in Heaven Disability and the
Resurrection of the Body Nicola Santamaria

https://ebookmeta.com/product/will-there-be-tiers-in-heaven-
disability-and-the-resurrection-of-the-body-nicola-santamaria/

ebookmeta.com
Lingfei Wu · Peng Cui
Jian Pei · Liang Zhao Eds.

Graph Neural
Networks

Foundations,
Frontiers,
and Applications
Graph Neural Networks: Foundations,
Frontiers, and Applications
Lingfei Wu • Peng Cui • Jian Pei • Liang Zhao
Editors

Graph Neural Networks:

Foundations, Frontiers,
and Applications

123
Editors
Lingfei Wu Peng Cui
JD Silicon Valley Research Center Tsinghua University
Mountain View, CA, USA Beijing, China

Jian Pei Liang Zhao

Simon Fraser University Emory University
Burnaby, Canada Atlanta, USA

ISBN 978-981-16-6053-5 ISBN 978-981-16-6054-2 (eBook)

https://doi.org/10.1007/978-981-16-6054-2

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore
Pte Ltd. 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore
Foreword

“The first comprehensive book covering the full spectrum of a young, fast-growing
research field, graph neural networks (GNNs), written by authoritative authors!”
Jiawei Han (Michael Aiken Chair Professor at University of Illinois at Urbana-
Champaign, ACM Fellow and IEEE Fellow)
“This book presents a comprehensive and timely survey on graph representation
learning. Edited and contributed by the best group of experts in this area, this book
is a must-read for students, researchers and pratictioners who want to learn anything
about Graph Neural Networks.”
Heung-Yeung ”Harry” Shum (Former Executive Vice President for Technology
and Research at Microsoft Research, ACM Fellow, IEEE Fellow, FREng)
“As the new frontier of deep learning, Graph Neural Networks offer great potential
to combine probabilistic learning and symbolic reasoning, and bridge knowledge-
driven and data-driven paradigms, nurturing the development of third-generation
AI. This book provides a comprehensive and insightful introduction to GNN, rang-
ing from foundations to frontiers, from algorithms to applications. It is a valuable
resource for any scientist, engineer and student who wants to get into this exciting
field.”
Bo Zhang (Member of Chinese Academy of Science, Professor at Tsinghua Uni-
versity)
“Graph Neural Networks are one of the hottest areas of machine learning and this
book is a wonderful in-depth resource covering a broad range of topics and applica-
tions of graph representation learning.”
Jure Leskovec (Associate Professor at Stanford University, and investigator at
Chan Zuckerberg Biohub).
“Graph Neural Networks are an emerging machine learning model that is already
taking the scientific and industrial world by storm. The time is perfect to get in on the
action – and this book is a great resource for newcomers and seasoned practitioners

v
vi Foreword

alike! Its chapters are very carefully written by many of the thought leaders at the
forefront of the area.”
Petar Veličković (Senior Research Scientist, DeepMind)
Preface

The field of graph neural networks (GNNs) has seen rapid and incredible strides over
the recent years. Graph neural networks, also known as deep learning on graphs,
graph representation learning, or geometric deep learning, have become one of the
fastest-growing research topics in machine learning, especially deep learning. This
wave of research at the intersection of graph theory and deep learning has also influ-
enced other fields of science, including recommendation systems, computer vision,
natural language processing, inductive logic programming, program synthesis, soft-
ware mining, automated planning, cybersecurity, and intelligent transportation.
Although graph neural networks have achieved remarkable attention, it still faces
many challenges when applying them into other domains, from the theoretical un-
derstanding of methods to the scalability and interpretability in a real system, and
from the soundness of the methodology to the empirical performance in an applica-
tion. However, as the field rapidly grows, it has been extremely challenging to gain
a global perspective of the developments of GNNs. Therefore, we feel the urgency
to bridge the above gap and have a comprehensive book on this fast-growing yet
challenging topic, which can benefit a broad audience including advanced under-
graduate and graduate students, postdoctoral researchers, lecturers, and industrial
practitioners.
This book is intended to cover a broad range of topics in graph neural networks,
from the foundations to the frontiers, and from the methodologies to the applica-
tions. Our book is dedicated to introducing the fundamental concepts and algorithms
of GNNs, new research frontiers of GNNs, and broad and emerging applications
with GNNs.

Book Website and Resources

The website and further resources of this book can be found at: https://
graph-neural-networks.github.io/. The website provides online preprints
and lecture slides of all the chapters. It also provides pointers to useful material and
resources that are publicly available and relevant to graph neural networks.

vii
viii Preface

To the Instructors

The book can be used for a one-semester graduate course for graduate students.
Though it is mainly written for students with a background in computer science,
students with a basic understanding of probability, statistics, graph theory, linear
algebra, and machine learning techniques such as deep learning will find it easily
accessible. Some chapters can be skipped or assigned as homework assignments for
reviewing purposes if students have knowledge of a chapter. For example, if students
have taken a deep learning course, they can skip Chapter 1. The instructors can also
choose to combine Chapters 1, 2, and 3 together as a background introduction course
at the very beginning.
When the course focuses more on the foundation and theories of graph neural net-
works, the instructor can choose to focus more on Chapters 4-8 while using Chapters
19-27 to showcase the applications, motivations, and limitations. Please refer to the
Editors’ Notes at the end of each chapter on how Chapters 4-8 and Chapters 19-27
are correlated. When the course focuses more on the research frontiers, Chapters
9-18 can be the pivot to organize the course. For example, an instructor can make
it an advanced graduate course where the students are asked to search and present
the most recent research papers in each different research frontier. They can also
be asked to establish their course projects based on the applications described in
Chapters 19-27 as well as the materials provided on our website.

To the Readers

This book was designed to cover a wide range of topics in the field of graph neu-
ral network field, including background, theoretical foundations, methodologies, re-
search frontiers, and applications. Therefore, it can be treated as a comprehensive
handbook for a wide variety of readers such as students, researchers, and profession-
als. You should have some knowledge of the concepts and terminology associated
with statistics, machine learning, and graph theory. Some backgrounds of the basics
have been provided and referenced in the first eight chapters. You should better also
have knowledge of deep learning and some programming experience for easily ac-
cessing the most of chapters of this book. In particular, you should be able to read
pseudocode and understand graph structures.
The book is well modularized and each chapter can be learned in a standalone
manner based on the individual interests and needs. For those readers who want
to have a solid understanding of various techniques and theories of graph neural
networks, you can start from Chapters 4-9. For those who further want to perform
in-depth research and advance related fields, please read those chapters of interest
among Chapters 9-18, which provide comprehensive knowledge in the most recent
research issues, open problems, and research frontiers. For those who want to ap-
ply graph neural networks to benefit specific domains, or aim at finding interesting
applications to validate specific graph neural networks techniques, please refer to
Chapters 19-27.
Acknowledgements

Graph machine learning has attracted many gifted researchers to make their seminal
contributions over the last few years. We are very fortunate to discuss the chal-
lenges and opportunities, and often work with many of them on a rich variety of
research topics in this exciting field. We are deeply indebted to these collaborators
and colleagues from JD.COM, IBM Research, Tsinghua University, Simon Fraser
University, Emory University, and elsewhere, who encouraged us to create such a
book comprehensively covering various topics of Graph Neural Networks in order
to educate the interested beginners and foster the advancement of the field for both
academic researchers and industrial practitioners.
This book would not have been possible without the contributions of many peo-
ple. We would like to give many thanks to the people who offered feedback on
checking the consistency of the math notations of the entire book as well as ref-
erence editing of this book. They are people from Emory University: Ling Chen,
Xiaojie Guo, and Shiyu Wang, as well as people from Tsinghua University: Yue He,
Ziwei Zhang, and Haoxin Liu. We would like to give our special thanks to Dr. Xiao-
jie Guo, who generously offered her help in providing numerous valuable feedback
on many chapters.
We also want to thank those who allowed us to reproduce images, figures, or data
from their publications.
Finally, we would like to thank our families for their love, patience and support
during this very unusual time when we are writing and editing this book.

ix
Editor Biography

Dr. Lingfei Wu is a Principal Scientist at JD.COM

Silicon Valley Research Center, leading a team of
30+ machine learning/natural language processing
scientists and software engineers to build intelligent
e-commerce personalization systems. He earned his
Ph.D. degree in computer science from the College
of William and Mary in 2016. Previously, he was a
research staff member at IBM Thomas J. Watson Re-
search Center and led a 10+ research scientist team
for developing novel Graph Neural Networks meth-
ods and systems, which leads to the #1 AI Chal-
lenge Project in IBM Research and multiple IBM
Awards including three-time Outstanding Technical
Achievement Award. He has published more than 90 top-ranked conference and
journal papers, and is a co-inventor of more than 40 filed US patents. Because of
the high commercial value of his patents, he has received eight invention achieve-
ment awards and has been appointed as IBM Master Inventors, class of 2020. He
was the recipients of the Best Paper Award and Best Student Paper Award of sev-
eral conferences such as IEEE ICC’19, AAAI workshop on DLGMA’20 and KDD
workshop on DLG’19. His research has been featured in numerous media out-
lets, including NatureNews, YahooNews, Venturebeat, TechTalks, SyncedReview,
Leiphone, QbitAI, MIT News, IBM Research News, and SIAM News. He has
co-organized 10+ conferences (KDD, AAAI, IEEE BigData) and is the founding
co-chair for Workshops of Deep Learning on Graphs (with AAAI’21, AAAI’20,
KDD’21, KDD’20, KDD’19, and IEEE BigData’19). He has currently served as
Associate Editor for IEEE Transactions on Neural Networks and Learning Systems,
ACM Transactions on Knowledge Discovery from Data and International Journal
of Intelligent Systems, and regularly served as a SPC/PC member of the following
major AI/ML/NLP conferences including KDD, IJCAI, AAAI, NIPS, ICML, ICLR,
and ACL.

xi
xii Editor Biography

Dr. Peng Cui is an Associate Professor with

tenure at Department of Computer Science in Ts-
inghua University. He obtained his PhD degree from
Tsinghua University in 2010. His research interests
include data mining, machine learning and multime-
dia analysis, with expertise on network representa-
tion learning, causal inference and stable learning,
social dynamics modeling, and user behavior model-
ing, etc. He is keen to promote the convergence and
integration of causal inference and machine learn-
ing, addressing the fundamental issues of today’s
AI technology, including explainability, stability and
fairness issues. He is recognized as a Distinguished Scientist of ACM, Distinguished
Member of CCF and Senior Member of IEEE. He has published more than 100 pa-
pers in prestigious conferences and journals in machine learning and data mining.
He is one of the most cited authors in network embedding. A number of his pro-
posed algorithms on network embedding generate substantial impact in academia
and industry. His recent research won the IEEE Multimedia Best Department Paper
Award, IEEE ICDM 2015 Best Student Paper Award, IEEE ICME 2014 Best Pa-
per Award, ACM MM12 Grand Challenge Multimodal Award, MMM13 Best Paper
Award, and were selected into the Best of KDD special issues in 2014 and 2016,
respectively. He was PC co-chair of CIKM2019 and MMM2020, SPC or area chair
of ICML, KDD, WWW, IJCAI, AAAI, etc., and Associate Editors of IEEE TKDE
(2017-), IEEE TBD (2019-), ACM TIST(2018-), and ACM TOMM (2016-) etc. He
received ACM China Rising Star Award in 2015, and CCF-IEEE CS Young Scien-
tist Award in 2018.
Editor Biography xiii

Dr. Jian Pei is a Professor in the School of

Computing Science at Simon Fraser University. He
is a well-known leading researcher in the general
areas of data science, big data, data mining, and
database systems. His expertise is on developing
effective and efficient data analysis techniques for
novel data intensive applications, and transferring
his research results to products and business practice.
He is recognized as a Fellow of the Royal Society of
Canada (Canada’s national academy), the Canadian
Academy of Engineering, the Association of Com-
puting Machinery (ACM) and the Institute of Elec-
trical and Electronics Engineers (IEEE). He is one of the most cited authors in data
mining, database systems, and information retrieval. Since 2000, he has published
one textbook, two monographs and over 300 research papers in refereed journals
and conferences, which have been cited extensively by others. His research has
generated remarkable impact substantially beyond academia. For example, his al-
gorithms have been adopted by industry in production and popular open-source
software suites. Jian Pei also demonstrated outstanding professional leadership in
many academic organizations and activities. He was the editor-in-chief of the IEEE
Transactions of Knowledge and Data Engineering (TKDE) in 2013-16, the chair of
the Special Interest Group on Knowledge Discovery in Data (SIGKDD) of the As-
sociation for Computing Machinery (ACM) in 2017-2021, and a general co-chair
or program committee co-chair of many premier conferences. He maintains a wide
spectrum of industry relations with both global and local industry partners. He is
an active consultant and coach for industry on enterprise data strategies, healthcare
informatics, network security intelligence, computational finance, and smart retail.
He received many prestigious awards, including the 2017 ACM SIGKDD Innova-
tion Award, the 2015 ACM SIGKDD Service Award, the 2014 IEEE ICDM Re-
search Contributions Award, the British Columbia Innovation Council 2005 Young
Innovator Award, an NSERC 2008 Discovery Accelerator Supplements Award (100
awards cross the whole country), an IBM Faculty Award (2006), a KDD Best Ap-
plication Paper Award (2008), an ICDE Influential Paper Award (2018), a PAKDD
Best Paper Award (2014), a PAKDD Most Influential Paper Award (2009), and an
IEEE Outstanding Paper Award (2007).
xiv Editor Biography

Dr. Liang Zhao is an assistant professor at the

Department of Compute Science at Emory Univer-
sity. Before that, he was an assistant professor in
the Department of Information Science and Tech-
nology and the Department of Computer Science at
George Mason University. He obtained his PhD de-
gree in 2016 from Computer Science Department
at Virginia Tech in the United States. His research
interests include data mining, artificial intelligence,
and machine learning, with special interests in spa-
tiotemporal and network data mining, deep learn-
ing on graphs, nonconvex optimization, model paral-
lelism, event prediction, and interpretable machine learning. He received AWS Ma-
chine Learning Research Award in 2020 from Amazon Company for his research on
distributed graph neural networks. He won NSF Career Award in 2020 awarded by
National Science Foundation for his research on deep learning for spatial networks,
and Jeffress Trust Award in 2019 for his research on deep generative models for bio-
molecules, awarded by Jeffress Memorial Trust Foundation and Bank of America.
He won the Best Paper Award in the 19th IEEE International Conference on Data
Mining (ICDM 2019) for the paper of his lab on deep graph transformation. He has
also won Best Paper Award Shortlist in the 27th Web Conference (WWW 2021) for
deep generative models. He was selected as “Top 20 Rising Star in Data Mining”
by Microsoft Search in 2016 for his research on spatiotemporal data mining. He has
also won Outstanding Doctoral Student in the Department of Computer Science at
Virginia Tech in 2017. He is awarded as CIFellow Mentor 2021 by the Computing
Community Consortium for his research on deep learning for spatial data. He has
published numerous research papers in top-tier conferences and journals such as
KDD, TKDE, ICDM, ICLR, Proceedings of the IEEE, ACM Computing Surveys,
TKDD, IJCAI, AAAI, and WWW. He has been serving as organizers such as pub-
lication chair, poster chair, and session chair for many top-tier conferences such as
SIGSPATIAL, KDD, ICDM, and CIKM.
List of Contributors

Miltiadis Allamanis
Microsoft Research, Cambridge, UK
Yu Chen
Facebook AI, Menlo Park, CA, USA
Yunfei Chu
Alibaba Group, Hangzhou, China
Peng Cui
Tsinghua University, Beijing, China
Tyler Derr
Vanderbilt University, Nashville, TN, USA
Keyu Duan
Texas A&M University, College Station, TX, USA
Qizhang Feng
Texas A&M University, College Station, TX, USA
Stephan Günnemann
Technical University of Munich, München, Germany
Xiaojie Guo
JD.COM Silicon Valley Research Center, Mountain View, CA, USA
Yu Hou
Weill Cornell Medicine, New York City, New York, USA
Xia Hu
Texas A&M University, College Station, TX, USA
Junzhou Huang
University of Texas at Arlington, Arlington, TA, United States
Shouling Ji

xv
xvi List of Contributors

Zhejiang University, Hangzhou, China

Wei Jin
Michigan State University, East Lansing, MI, USA
Anowarul Kabir
George Mason University, Fairfax, VA, USA
Seyed Mehran Kazemi
Borealis AI, Montreal, Canada.
Jure Leskovec
Stanford University, Stanford, CA, USA
Juncheng Li
Zhejiang University, Hangzhou, China
Jiacheng Li
Zhejiang University, Hangzhou, China
Pan Li
Purdue University, Lafayette, IN, USA
Yanhua Li
Worcester Polytechnic Institute, Worcester, MA, USA
Renjie Liao
University of Toronto, Toronto, Canada
Xiang Ling
Zhejiang University, Hangzhou, China
Bang Liu
University of Montreal, Montreal, Canada
Ninghao Liu
Texas A&M University, College Station, TX, USA
Zirui Liu
Texas A&M University, College Station, TX, USA
Hehuan Ma
University of Texas at Arlington, College Station, TX, USA
Collin McMillan
University of Notre Dame, Notre Dame, IN, USA
Christopher Morris
Polytechnique Montréal, Montréal, Canada
Zongshen Mu
Zhejiang University, Hangzhou, China
Menghai Pan
 Another Random Document on
Scribd Without Any Related Topics
The Project Gutenberg eBook of QRM-
Interplanetary
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: QRM-Interplanetary

Author: George O. Smith

Illustrator: William A. Kolliker

Release date: May 6, 2022 [eBook #68005]

Most recently updated: June 10, 2022

Language: English

Original publication: United States: Street & Smith Publications,

Incorporated, 1942

Credits: Greg Weeks, Mary Meehan and the Online Distributed

Proofreading Team at http://www.pgdp.net

*** START OF THE PROJECT GUTENBERG EBOOK QRM-

INTERPLANETARY ***
QRM—INTERPLANETARY
 By George O. Smith

Illustrated by Kolliker

[Transcriber's Note: This etext was produced from

Astounding Science-Fiction October 1942.
Extensive research did not uncover any evidence that
the U.S. copyright on this publication was renewed.]
 QRM—International code signal meaning "Interference" of
controllable nature, such as man-made static, cross
modulation from another channel adjoining or willful
obliteration of signals by an interfering source.
Interference not of natural sources (designated by
International code as QRN), such as electrical storms,
common static, et cetera.
—Handbook, Interplanetary Amateur Radio League.
Korvus, the Magnificent, Nilamo of Yoralen, picked up the telephone
in his palace and said: "I want to talk to Wilneda. He is at the
International Hotel in Detroit, Michigan."
"I'm sorry, sir," came the voice of the operator. "Talking is not
possible, due to the fifteen-minute transmission lag between here
and Terra. Interplanetary Communications will not permit audio.
However, teletype messages are welcome."
Her voice originated fifteen hundred miles north of Yoralen, but it
sounded as though she might be in the next room. Korvus thought
for a moment and then said: "Take this message: 'Wilneda: Add to
order for mining machinery one type 56-XXD flier to replace washed-
out model. And remember, tobacco and sublevel energy will not mix!'
Sign that Korvus."
"Yes, Mr. Korvus."
"Not mister!" yelled the monarch. "I am Korvus, the Magnificent! I am
Nilamo of Yoralen!"
"Yes, your magnificence," said the operator humbly. It was more than
possible that she was stifling a laugh, which knowledge made the
little man of Venus squirm in wrath. But there was nothing that he
could do about it, television still being distant by the same five years
that it was behind in 1929.
To give Korvus credit, he was not a pompous little man. He was
large—for a Venusian—which made him small according to the
standards set up by Terrestrians. He, as Nilamo of Yoralen, had
extended the once-small kingdom outward to include most of the
Palanortis Country which extended from about 23.0 degrees North
Latitude to 61.7 degrees, and almost across the whole, single
continent that was the dry land of Venus. He was a wily monarch,
making his conquest of the wild and lawless country by treaty, and
by double-double-crossing those who might have tried to double-
cross him. Armed conquest was scorned, but armed defense was
desirable in the Palanortis Country—and Korvus had defended
himself up and down the inhabitable Northern portion of the planet.
His conquest had been a blessing to civilization, and though publicly
denounced, it was privately commended. Those who could have
stopped it did all they could to delay and intercept any proceedings
that would have caused the conquest of Korvus' intended country
any trouble.
Korvus' message to Terra zoomed across the fifteen hundred rocky
miles of Palanortis to Northern Landing. It passed high across the
thousand-foot-high trees and over the mountain ranges. It swept
over open patches of water, and across intervening cities and towns.
It went with the speed of light and in a tight beam from Yoralen to
Northern Landing, straight as a die and with person-to-person clarity.
The operator in the city that lay across the North Pole of Venus
clicked on a teletype, reading back the message as it was written.
Korvus told him: "That is correct."
"The message will be in the hands of your representative Wilneda
within the hour!"

The punched tape from Operator No. 7's machine slid along the line.
It entered a coupling machine and was stripped from the tape and
repunched upon a tape that was operating at better than a thousand
words per minute. Operator 7's tape then left the machine to be
rolled into a file roll and placed in the vaults below the city. It was of
no use save as a reference from now on.
The coupling machine worked furiously. It accepted the tapes from
seventy operators as fast as they could write them. It selected the
messages as they entered the machine, placing a mechanical
preference upon whichever message happened to be ahead of the
others on the moving tapes. The master tape moved continuously at
eleven hundred words per minute, taking teletype messages from
everywhere in the Northern Hemisphere of Venus to any of the other
planets in the Solar System. It was a busy machine, for even at
eleven hundred words per minute, it often got hours behind.
The synchronous-keyed signal from the coupling machine left the
operating room and went to the transmission room. It was amplified
and hurled out of the city to a small, squat building at the outskirts of
Northern Landing.
Here it was impressed upon a carrier wave and flung at the sky.
But not alone. Not unguarded. The upper half of the building carried
a monstrous parabolic reflector, mounted on gimbals. The signal was
focused into a beam. The beam was made of two components. The
center component was a circularly polarized, ultra-high frequency
wave of five centimeter waves, modulated with the keying signals of
the teletype coupling machine. The outer component was a radially
polarized wave of one centimeter waves. A radio frequency armor.
It was hurled at the sky, a concentric wave, out of a reflector, by a
thousand kilowatt transmitter. The wave seared against the Venusian
Heaviside Layer. The outer component bored at the ionization. It
chewed and it bit. It fought and it struggled. It destroyed ionization by
electronically shorting the ionization. And, as is the case with strife, it
lost heavily in the encounter. The beam was resisted fiercely.
Infiltrations of ionization tore at the central component, stripping and
trying to beat it down.
But man triumphed over nature! The megawatt of energy that came
in a tight beam from the building at Northern Landing emerged from
the Heaviside Layer as a weak, piffling signal. It wavered and it
crackled. It wanted desperately to lie down and sleep. Its directional
qualities were impaired, and it wabbled badly. It arrived at the relay
station tired and worn.
One million watts of ultra-high frequency energy at the start, it was
measurable in microvolts when it reached the relay station, only five
hundred miles above the city of Northern Landing.
The signal, as weak and as wabbly as it was, was taken in by eager
receptors. It was amplified. It was dehashed, de-staticked, and
deloused. And once again, one hundred decibels stronger and
infinitely cleaner, the signal was hurled out on a tight beam from a
gigantic parabolic reflector.
Across sixty-seven million miles of space went the signal. Across the
orbit of Venus it went in a vast chord. It arrived at the Venus
Equilateral Station with less trouble than the original transmission
through the Heaviside Layer. The signal was amplified and
demodulated. It went into a decoupler machine where the messages
were sorted mechanically and sent, each to the proper channel, into
other coupler machines. Beams from Venus Equilateral were
directed at Mars and at Terra.
The Terra beam ended at Luna. Here it again was placed in the two-
component beam and from Luna it punched down at Terra's Layer. It
emerged into the atmosphere of Terra, as weak and as tired as it had
been when it had come out of the Venus Layer. It entered a station in
the Bahamas, was stripped of the interference, and put upon the
land beams. It entered decoupling machines that sorted the
messages as to destination. These various beams spread out across
the face of Terra, the one carrying Korvus' message finally coming
into a station at Ten Mile Road and Woodward. From this station at
the outskirts of Detroit, it went upon land wires downtown to the
International Hotel.
The teletype machine in the office of the hotel began to click rapidly.
The message to Wilneda was arriving.
And fifty-five minutes after the operator told Korvus that less than an
hour would ensue, Wilneda was saying, humorously, "So, Korvus
was drunk again last night—"

Completion of Korvus' message to Wilneda completes also one

phase of the tale at hand. It is not important. There were a hundred
and fifty other messages that might have been accompanied in the
same manner, each as interesting to the person who likes the
explanation of the interplanetary communication service. But this is
not a technical journal. A more complete explanation of the various
phases that a message goes through in leaving a city on Venus to go
to Terra may be found in the Communications Technical Review,
Volume XXVII, Number 8, pages 411 to 716. Readers more
interested in the technical aspects are referred to the article.
But it so happens that Korvus' message was picked out of a
hundred-odd messages because of one thing only. At the time that
Korvus' message was in transit through the decoupler machines at
Venus Equilateral Relay Station, something of a material nature was
entering the air lock of the station.
It was an unexpected visit.
Don Channing looked up at the indicator panel in his office and
frowned in puzzlement. He punched a buzzer and spoke into the
communicator on his desk.
"Find out who that is, will you, Arden?"
"He isn't expected," came back the voice of Arden Westland.
"I know that. But I've been expecting someone ever since John
Walters retired last week. You know why."
"You hope to get his job," said the girl in an amused voice. "I hope
you do. So that someone else will sit around all day trying to make
you retire so that he can have your job!"
"Now look, Arden, I've never tried to make Walters retire."
"No, but when the word came that he was thinking of it, you began to
think about taking over. Don't worry, I don't blame you." There was
quite a protracted silence, and then her voice returned. "The visitor is
a gentleman by the name of Francis Burbank. He came out in a flitter
with a chauffeur and all."
"Big shot, hey?"
"Take it easy. He's coming up the office now."
"I gather that he desires audience with me?" asked Don.
"I think that he is here to lay down the law! You'll have to get out of
Walters' office, if his appearance is any guide."

There was some more silence. The communicator was turned off at
the other end, which made Channing fume. He would have preferred
to hear the interchange of words between his secretary and the
newcomer. Then, instead of having the man announced, the door
opened abruptly and the stranger entered. He came to the point
immediately.
"You're Don Channing? Acting Director of Interplanetary
Communications?"
"I am."
"Then I have some news for you, Dr. Channing. I have been
appointed Director by the Interplanetary Communications
Commission. You are to resume your position as Electronics
Engineer."
"Oh?" said Channing. His face fell. "I sort of believed that I would be
offered that position."
"There was a discussion of that procedure. However, the
Commission decided that a man of more commercial training would
better fill the position. The Communications Division has been
operating at too small a profit. They felt that a man of commercial
experience could cut expenses and so on to good effect. You
understand their reasoning, of course," said Burbank.
"Not exactly."
"Well, it is like this. They know that a scientist is not usually a man to
consider the cost of experimentation. They build thousand-ton
cyclotrons to convert a penny's worth of lead into one and one-tenth
cents' worth of lead and gold. And they use three hundred dollars'
worth of power and a million-dollar machine to do it with.
"They feel that a man with training like that will not know the real
meaning of the phrase, 'cutting expenses.' A new broom sweeps
clean, Dr. Channing. There must be many places where a man of
commercial experience can cut expenses. I, as Director, shall do so."
"I wish you luck," said Channing.
"Then there is no hard feeling?"
"I can't say that. It is probably not your fault. I cannot feel against
you, but I do feel sort of let down at the decision of the Commission.
I have had experience in this job."
"The Commission may appoint you to follow me. If your work shows
a grasp of commercial operations, I shall so recommend."
"Thanks," said Channing dryly. "May I buy you a drink?"
"I never drink. And I do not believe in it. If it were mine to say, I'd
prohibit liquor from the premises. Venus Equilateral would be better
off without it."
Don Channing snapped the communicator. "Miss Westland, will you
come in?"
She entered, puzzlement on her face.
"This is Mr. Burbank. His position places him in control of this office.
You will, in the future, report to him directly. The report on the
operations, engineering projects, and so on that I was to send in to
the Commission this morning will, therefore, be placed in Mr.
Burbank's hands as soon as possible."
"Yes, Dr. Channing." Her eyes held a twinkle, but there was concern
and sympathy in them, too. "Shall I get them immediately?"
"They are ready?"
"I was about to put them on the tape when you called."
"Then give them to Mr. Burbank." Channing turned to Burbank. "Miss
Westland will hand you the reports I mentioned. They are complete
and precise. A perusal of them will put you in grasp of the situation
here at Venus Equilateral better than will an all-afternoon
conference. I'll have Miss Westland haul my junk out of here. You
may consider this as your office, it having been used by Walters.
And, in the meantime, I've got to check up on some experiments
down on the ninth level." Channing paused, "You'll excuse me?"
"Yes, if Miss Westland knows where to find you."
"She will. I'll inform her of my whereabouts."
"I may want to consult you after I read the reports."
"That will be all right. The autocall can find me anywhere on Venus
Equilateral, if I'm not at the place Miss Westland calls."

Don Channing stopped at Arden's desk. "I'm booted," he told her.

"Leaving Venus Equilateral?" she asked with concern.
"No, blond and beautiful, I'm just shunted back to my own office."
"Can't I go with you?" pleaded the girl.
"Nope. You are to stay here and be a nice, good-looking Mata Hari.
This bird seems to think that he can run Interplanetary
Communications like a bus or a factory. I know the type, and the first
thing he'll do is to run Interplanetary Communications into a snarl.
Keep me informed of anything complicated, will you?"
"Sure. And where are you going now?"
"I'm going down and get Walt Franks. We're going to inspect the
transparency of a new type of glass."
"I didn't know that optical investigations come under your
jurisdiction."
"This investigation will consist of a visit to the ninth level."
"Can't you take me along?"
"Not today," he grinned. "Your new boss does not believe in the evils
of looking through the bottom of a glass. We must behave with
decor. We must forget fun. We are now operating under a man who
will commercialize electronics to a fine art."
"Don't get stewed. He may want to know where the electrons are
kept."
"I'm not going to drink that much. Walt and I need a discussion," he
said. "And in the meantime, haul my spinach out of the office, will
you, and take it back to the electronics office. I'll be needing it back
there."
"O.K., Don," she said. "I'll see you later."
Channing left to go to the ninth level. He stopped long enough to
collect Walt Franks.
Over a tall glass of beer, Channing told Franks of Burbank's visit.
And why.
Only one thing stuck in Franks' mind. "Did you say that he might
close Joe's?" asked Franks.
"He said that if it were in his power to do so, he would."
"Heaven forbid. Where will we go to be alone?"
"Alone?" snorted Channing. The barroom was half filled with people,
being the only drinking establishment for sixty-odd million miles.
"Well, you know what I mean."
"I could smuggle in a few cases of beer," suggested Don.
"Couldn't we smuggle him out?"
"That would be desirable. But I think he is here to stay. Darn it all,
why do they have to appoint some confounded political pal to a job
like this? I'm telling you, Walt, he must weigh two hundred if he
weighs a pound. He holds his stomach on his lap when he sits
down."
Walt looked up and down Channing's slender figure. "Well, he won't
be holding Westland on his lap if it is filled with stomach."
"I never hold Westland on my lap—"
"No?"
"—during working hours!" finished Channing. He grinned at Franks
and ordered another beer. "And how is the Office of Beam Control
going to make out under the new regime?"
"I'll answer that after I see how the new regime treats the Office of
Beam Control," answered Franks. "I doubt that he can do much to
bugger things up in my office. There aren't many cheaper ways to
direct a beam, you know."
"Yeah. You're safe."
"But what I can't understand is why they didn't continue you in that
job. You've been handling the business ever since last December
when Walters got sick. You've been doing all right."
"Doing all right just means that I've been carrying over Walters'
methods and ideas. What the Commission wants, apparently, is
something new. Ergo the new broom."
"Personally, I like that one about the old shoes being more
comfortable," said Franks. "If you say the word, Don, I'll slip him a
dose of high voltage. That should fix him."
"I think that the better way would be to work for the bird. Then when
he goes, I'll have his recommendation."
"Phooey," snorted Franks. "They'll just appoint another political pal.
They've tried it before and they'll try it again. I wonder what precinct
he carries."
The telephone rang in the bar, and the bartender, after answering,
motioned to Walt Franks. "You're wanted in your office," said the
bartender. "And besides," he told Channing, "if I'm going to get lunch
for three thousand people, you'd better trot along, too. It's nearly
eleven o'clock, you know, and the first batch of five hundred will be
coming in."
He wasn't quite accurate as to the figures. The complement of Venus
Equilateral was just shy of twenty-seven hundred. They worked in
three eight-hour shifts, about nine hundred to a shift. They had their
lunch and dinner hours staggered so that at no time was there more
than about two hundred people in the big lunchroom. The bar, it may
be mentioned, was in a smaller room at one end of the much larger
cafeteria.
Venus Equilateral Relay Station was a modern miracle of
engineering if you liked to believe the books. Actually, Venus
Equilateral was an asteroid that had been shoved into its orbit about
the Sun, forming a practical demonstration of the equilateral triangle
solution of the Three Moving Bodies. It was a long cylinder, about
three miles in length by about a mile in diameter.
There was little of the original asteroid. At the present time, most of
the original rock had been discarded to make room for the ever-
growing personnel and material that were needed to operate the
relay station. What had been an asteroid with machinery was now a
huge pile of machinery with people. The insides, formerly of spongy
rock, were now neatly cubed off into offices, rooms, hallways, and so
on, divided by sheets of steel. The outer surface, once rugged and
forbidding, was now almost all shiny steel. The small asteroid, a tiny
thing, was far smaller than the present relay station, the station
having overflowed the asteroid soon after men found that
uninterrupted communication was possible between the worlds.
Now, the man-made asteroid carried twenty-seven hundred people.
There were stores, offices, places of recreation, churches,
marriages, deaths, and everything but taxes. Judging by its
population, it was a small town.
Venus Equilateral rotated about its axis. On the inner surface of the
shell were the homes of the people—not cottages, but apartmental
cubicles, one, two, three, six rooms. The rotation made a little more
than one Earth G of artificial gravity. Above this outer shell of
apartments, the offices began. Offices, recreation centers, and so
on. Up in the central portion where the gravity was nil or near-nil, the
automatic machinery was placed. The gyroscopes and the beam
finders, the storerooms, the air plants, the hydroponic farms, and all
other things that needed little or no gravity for well-being.
This was the Venus Equilateral Relay Station, sixty degrees ahead of
the planet Venus, on Venus' orbit. Often closer to Terra than Venus,
the relay station offered a perfect place to relay messages through
whenever Mars or Terra were on the other side of the Sun. It was
seldom idle, for it was seldom that both Mars and Terra were in such
position that direct communication between the three planets was
possible.
This was the center of Interplanetary Communications. This was the
main office. It was the heart of the system's communication line, and
as such, it was well manned. Orders for everything emanated from
Venus Equilateral. It was a delicate proposition, Venus Equilateral
was, and hence the present-on-all-occasions official capacities and
office staff.
This was the organization that Don Channing hoped to direct. A
closed corporation with one purpose in mind, interplanetary
communication!
Channing wondered if the summons for Walt Franks was an official
one. Returning to the electronics office, Don punched the
communicator and asked: "Is Walt in there?"
Arden's voice came back: "No, but Burbank is in Franks' office.
Wanna listen?"
"Eavesdropper! Using the communicator?"
"Sure."