George R. Fisher, III, M.D.

Obituary

George R. Fisher, III, M.D.

Age: 97 of Philadelphia, formerly of Haddonfield Dr George Fisher III

Dr. George Ross Fisher of Philadelphia died on March 9, 2023, surrounded by his loving family.

Born in 1925 in Erie, Pennsylvania, to two teachers, George and Margaret Fisher, he grew up in Pittsburgh, later attending The Lawrenceville School and Yale University (graduating early because of the war). He was very proud of the fact that he was the only person who ever graduated from Yale with a Bachelor of Science in English Literature. He attended Columbia University’s College of Physicians and Surgeons where he met the love of his life, fellow medical student, and future renowned Philadelphia radiologist Mary Stuart Blakely. While dating, they entertained themselves by dressing up in evening attire and crashing fancy Manhattan weddings. They married in 1950 and were each other’s true loves, mutual admirers, and life partners until Mary Stuart passed away in 2006. A Columbia faculty member wrote of him, “This young man’s personality is way off the beaten track, and cannot be evaluated by the customary methods.”

After training at the Pennsylvania Hospital in Philadelphia where he was Chief Resident in Medicine, and spending a year at the NIH, he opened a practice in Endocrinology on Spruce Street where he practiced for sixty years. He also consulted regularly for the employees of Strawbridge and Clothier as well as the Hospital for the Mentally Retarded at Stockley, Delaware. He was beloved by his patients, his guiding philosophy being the adage, “Listen to your patient – he’s telling you his diagnosis.” His patients also told him their stories which gave him an education in all things Philadelphia, the city he passionately loved and which he went on to chronicle in this online blog. Many of these blogs were adapted into a history-oriented tour book, Philadelphia Revelations: Twenty Tours of the Delaware Valley.

He was a true Renaissance Man, interested in everything and everyone, remembering everything he read or heard in complete detail, and endowed with a penetrating intellect which cut to the heart of whatever was being discussed, whether it be medicine, history, literature, economics, investments, politics, science or even lawn care for his home in Haddonfield, NJ where he and his wife raised their four children. He was an “early adopter.” Memories of his children from the 1960s include being taken to visit his colleagues working on the UNIVAC computer at Penn; the air-mail version of the London Economist on the dining room table; and his work on developing a proprietary medical office software using Fortran. His dedication to patients and to his profession extended to his many years representing Pennsylvania to the American Medical Association.

After retiring from his practice in 2003, he started his pioneering “just-in-time” Ross & Perry publishing company, which printed more than 300 new and reprint titles, ranging from Flight Manual for the SR-71 Blackbird Spy Plane (his best seller!) to Terse Verse, a collection of a hundred mostly humorous haikus. He authored four books. In 2013 at age 88, he ran as a Republican for New Jersey Assemblyman for the 6th district (he lost).

A gregarious extrovert, he loved meeting his fellow Philadelphians well into his nineties at the Shakespeare Society, the Global Interdependence Center, the College of Physicians, the Right Angle Club, the Union League, the Haddonfield 65 Club, and the Franklin Inn. He faithfully attended Quaker Meeting in Haddonfield NJ for over 60 years. Later in life he was fortunate to be joined in his life, travels, and adventures by his dear friend Dr. Janice Gordon.

He passed away peacefully, held in the Light and surrounded by his family as they sang to him and read aloud the love letters that he and his wife penned throughout their courtship. In addition to his children – George, Miriam, Margaret, and Stuart – he leaves his three children-in-law, eight grandchildren, three great-grandchildren, and his younger brother, John.

A memorial service, followed by a reception, will be held at the Friends Meeting in Haddonfield New Jersey on April 1 at one in the afternoon. Memorial contributions may be sent to Haddonfield Friends Meeting, 47 Friends Avenue, Haddonfield, NJ 08033.

Gunk, 27 Million Tons of It

Henry Hudson

When Henry Hudson reached the mouth of Delaware Bay in 1609, the river was so full of snags he simply went up to what is now the Hudson River in New York rather than try to wiggle his little sailboat up Delaware. By 1900, there had been enough dredging and removal of islands that the channel was 17 feet deep all of the ninety miles up to Philadelphia. One of the consequences was that the new river edge was down at Delaware (Columbus) Avenue, rather than up at Front Street. When you make it deeper, the width of a shallow river often narrows.

Now, the proposal is to deepen the channel to 42 feet, a number mandated by the present size of cargo container ships. Another limiting factor is the construction of the bridges, so the Port of Philadelphia is moving South of the Walt Whitman bridge. That's potentially of great value to the longshoremen who live in that region, although whether it will really bring prosperity is up to them, depending on whether they restrain their aggressive wage and work-rule proposals. There are serious students of the Philadelphia economy who maintain that the economic decline of Philadelphia is more traceable to the intransigence of the longshore unions than to any other factor. Since that comment is specifically made in comparison with the railroad brotherhoods, it is a dramatic accusation indeed.

If you deepen the channel to 42 feet, 800 feet wide (1300 feet at bends in the river), you can be calculated to bring up 27 million tons of sludge. You have to dump that stuff somewhere else, and the current plan is for Philadelphia to build a retaining wall out into the river next to the Packer Avenue terminal area, and dump Philadelphia's share of the stuff behind it. In time, the water will drain out of the gunk, and quite a few acres of dry land would make its appearance. Some engineers question whether the force of the river would permit this. Environmentalists have objections to this project relating to stirring up pollutants lying dormant on the river floor, but without likely effect on the tin ears of those who are presently congratulating themselves on obtaining Federal money to accomplish this "big dig".

The really serious obstructions are coming from the State of New Jersey, which would acquire 9 million tons of gunk as their fair share. Right now, New Jersey is raising taxes and cutting state spending because of a budget deficit, so they are not anxious to take on another big project, particularly one whose benefits will have to be shared with Pennsylvania. Pennsylvania was momentarily sympathetic with this problem until it was learned that New Jersey is actively promoting a FIFTY-foot channel in the Hudson River. Immediately it becomes obvious that there is not enough money for two projects, and there are more New Jersey voters up near the port of New York than down around the Port of Philadelphia. Both New Jersey and Pennsylvania have Democrat governors, while New York has a Republican one. Ordinarily, this would be a decisive point, but the preponderant location of voters up in North Jersey seems to trump that. Keep watching the Saturday papers, on the editorial page down below the fold, the place newspapers ordinarily reserve for retractions, apologies, and local political truths.

What's going on here is attempted exploitation of geographical advantages. Philadelphia is at one of three navigable openings in the Atlantic coast barrier islands adjoining the New York-Washington megalopolis, or five openings if you call it a Boston-Richmond megopolis. Obviously, a seaway opening in the middle is superior to one at the ends, so it really comes down to a New York and Philadelphia competition, with Baltimore a poor third because European ships have to go down to Norfolk and then come come back up the Chesapeake, like Lord Admiral Howe in 1777. There's a huge amount of rail and truck traffic North and South, so crossing the T with ocean traffic arriving in the middle could make quite an economic center. Passenger rail traffic from Philadelphia to Pittsburgh and beyond is pretty anemic, but freight traffic is healthy and could be more so with cargo supplied by container ships. This is the dream, New York is the enemy, New Jersey is the villain, and the longshoremen are the main beneficiary. It is even possible to imagine eighty dollars per hundred in wages for Workman's compensation, but that would be cynical.

Because of the New Jersey problem, proposals have been made to fill up abandoned coal mines with dredging sludge and let the water seep out wherever it, please. Somehow, this isn't thought to be practical, and other suggestions seem to be very welcome, a rather unusual circumstance in itself.

Let's ask ourselves whether we want to return Philadelphia to its old industrial mightiness, or whether we want to encourage the development of a service economy, computers and all that. One way to measure success in the container cargo race is to count the unloading cranes. Philadelphia has about five of them, and most of the time they are sticking straight up, unused. There are many times that many in Seattle, and in Yokohama, there are over a hundred. The port of Kobe has far more, too many to count as you go past on the bullet train. However, there's a secret truth about container ships. When they get to Seattle, there is no cargo to fill them with for a return voyage, and in fact, the empty container pile-up is a rather serious problem. Bill Gates is shipping lots of software to Japan, but it doesn't fill cargo containers, and the economy of the region is going to have high transportation costs until someone figures out a bulk cargo product to ship from Seattle. This is exactly the situation of a century ago when the New York Central, the Baltimore and Ohio, and the Pennsylvania RR were battling it out for transcontinental supremacy. The Pennsy won that battle because oil was discovered in Bradford Pennsylvania, and refineries were built on the Schuylkill to process the oil. The Pennsy had thus found a round-trip cargo for filling its empty East-bound freight cars, and it beat out the other two railroads which didn't. Cheaper freight rates became possible, and therefore other industries prospered in the region.

Right here is a topic which somebody at the Wharton School had better start talking about. No matter how much software and other service industry prosperity a city region may support, it has to find a way to supply bulk cargo to all those container ships that are bringing in the BMW's for the service industry hotshots to drive.

Meanwhile, what does New Jersey do with 9 million tons of gunk?

Please Touch

Please Touch Museum

There had been rumors for some time that the Please Touch Museum was planning to move from 21st Street to larger quarters, but recently its Executive Director Laura Foster appeared at a luncheon at the Franklin Inn Club to announce definite plans. The Museum moved into Memorial Hall in West Fairmount Park in the fall of 2008.

Memorial Hall Fairmount Park

There are over 400 children's museums in the world, and the first one was started in Brooklyn in 1899. Just why Philadelphia waited until the Bicentennial Celebration in 1976 to start one, is not clear. It's particularly unclear when you hear of its explosive success. Growing rapidly during an era when museums of all sorts are seeing declining attendance, the Please Touch Museum will be making its fourth move in thirty years, each time to larger quarters because they needed more room. Sooner or later, expansionism will get its comeuppance of course, and Memorial Hall is one awfully large building to fill. And to heat, and to paint, and to air condition. The price is right, however. The City Administration, which approached the museum with a proposal, has offered an 80-year lease for a dollar. When you hear that they have occasionally had 1500 visitors in a single day, however, and annual attendances approaching 200,000, almost anything seems possible.

Liberty Torch

There are certain limits. Nothing frightens a 4-year-old like a herd of 10-year-old boys racketing about, so there are segregations necessary. In other situations, these little kids not only can't touch, but they also can't see and they want to see badly. And the sociology is interesting. The kids may well clamor to come when they hear other kids talking, but in general, it is the parents who get the idea that a museum trip would be fun. And the parents seem motivated by theories of upward mobility, of giving their child a "leg up" on the competition. The museum is certainly filling a need, but you have to wonder where our society is headed if a picnic in the park is mainly a good idea if it gives junior a leg up.

Meanwhile, keep tuned. To fill up that monstrous Memorial Hall will take publicity, and these gals sound as though they mean to have a lot of it.

Google Earth Tour of Franklin Locations

On the front page of Philadelphia Reflections is found a button which will download Google Earth, and if you follow instructions on the left column, will give you a satellite tour of every blog let on the site. At least, it will when we get it finished; it's only about half complete at present. If you are unfamiliar with this approach, we suggest you download the Earth program from the Google site and get acquainted by locating your own house, or Independence Hall, or the Vatican.

In addition, every Topic (listed in the left-hand column of the front page of Philadelphia Reflections) will contain a button which generates a tour of the geoTags of that particular Topic, providing there are three or more such tags. You will generally get the best results from tours developed by unknown authors if you turn off ALL of the layers provided in the lower section of the left-hand panel of Google Earth, although you might turn them on, one at a time if you want to enhance the effects. Generally speaking, the route of Interstate 95 seems a little out of place among the local wanderings of Ben Franklin.

Take a satellite tour of nearly
every place Ben Franklin ever visited.

You should also become familiar with KMZ files and KML files. Keyhole markup language gives instructions to Google Earth, allowing authors like Bob Florig to organize tours of a particular subject. KML files are quite large, so they get compressed to make them easier to send over the Internet. Compressed files of KML are designated KMZ, referring to Winzip the decompressor. Other decompressors will often work, too, especially Stuffit for Apple users. The extra step of decompression is a nuisance, and it is possible to have the file do things itself, to become known as a self-extracting file. Self-extracting files are often, but not always, designated as EXE files.

You are here invited to take a tour of every site Benjamin Franklin is known to have visited as if you were an interplanetary alien riding a flying saucer. Double-click the blue link to download a copy of Google Earth if you don't already have one, followed by a self-extracting KMZ file constructed by Bob Florig and used with his kind permission.

There's one other feature you should know about, called overlays. Bob took an 18th Century map of Philadelphia and substituted it for the satellite map of contemporary central Philadelphia. That lets you see Philadelphia as Franklin saw it, and by changing overlays, also allows you to see the little red-brick buildings which remain standing among the skyscrapers. Both he and I are uncertain about the copyright status of the old maps and may have to remove them if the author identifies himself and protests.

Robert Morris and the Lee Brothers

* * *

Morris was a monarchist, announcing he liked the king and did not want to change him. The Lees, one surmises, though they might make pretty good kings, themselves.

But we get far ahead of the story. The immediate mystery is Morris' behavior in July, 1776.

Robert Morris

The dominant feeling during 1775 had been that Americans needed to prepare for war while doing everything possible to avert it. If England prevented both the manufacture and importation of gunpowder in the colonies, there was no way to be prepared for war except to smuggle gunpowder. Not to smuggle it would leave the colonies weak, almost inviting abuse from London. In several elections, before then Morris had been the nominee of both the radicals and the conservatives. The radicals could see he was an energetic, efficient and close-mouthed international shipowner; when a Secret Committee was proposed, he obviously would be an effective member of it. But he was loudly opposed to war with England and was formidable in a debate. The conservatives might have felt he would keep the hotheads on the committee from wandering from or expanding its narrow charge. He explained his dual position as seeking "Constitutional Liberty" rather than independence, but that was shrugged off as just so much blather. It would be twelve years before people did understand what he wanted, and that he was entirely serious about it. That the Lee brothers didn't ever understand, didn't bother Morris at all. That the Lee brothers couldn't comprehend his unwillingness to charge headlong into battle, unarmed and unprepared, was equally mysterious to Morris. The Lee brothers would well have understood Napoleon's remark that victory was 10% based on surprise. But what Napoleon is thought to have said was that victory was 90% based on supplies.

Arthur Lee

The later appointment of Arthur Lee to the Secret Committee can be similarly explained as a counterbalance to Morris. There were no Southerners on the initial committee, so a hothead like Arthur Lee could be counted on to resist the influence of conservative merchants and perhaps Arthur, outnumbered, was even urged to overplay his hand. From the moment he was appointed, he was demanding that Benjamin Franklin be removed from the committee. Not only does this help us understand his disruptive behavior when he later joined Franklin at the Paris negotiations, but such misjudgment of Franklin's patriotism illustrates how extreme Lee's suspicions really were. Since the original motion for independence had been introduced in Congress as the Virginia Resolution by his brother Richard Henry Lee, we sense this family was perpetually anxious to slay dragons. And finally, we can sense the likelihood that Robert Morris and Arthur Lee probably provoked each other.

Richard Henery Lee

To tolerate for the moment the rather extreme Lee brothers, it must be said there was every reason to suppose Morris hoped to enrich himself from gun running. But that is an entirely different accusation with a different defense. The ambivalent behavior of Morris and Dickinson at the Congress is admittedly puzzling, but in the end, defensible and honorable. The accusation of graft and self-dealing was plausible for uncomprehending people but seems legitimate today only because American standards of leadership ethics have gravitated toward the Virginia viewpoint. That viewpoint can be summarized by George Washington refusing to accept a cent for his years of public service. Very few politicians will today tell you this is a practical position, but nevertheless, many Americans wish it could be. Lacking any insight into what Morris was talking about with "Constitutional Liberty", the Lees were perhaps understandably driven to invent plausible explanations. These seldom proved to be fair. Morris certainly had the sympathy of the commercial world at the time for what was to them quite normal behavior. However, the rural population, especially Cavalier Tidewater Virginians, will probably never yield the point.

Morris repeatedly advised his partners he was serving his nation, while of course engaging in private profit. Even using present standards about self-dealing, it must be admitted the need for secrecy at the time necessitated hiding the smuggling within legitimate businesses. Nowadays we throw people out of a public office on suspicion of enriching themselves and throw them into jail on clear proof of it. There is also a curious feature of extreme mindsets, which seems to justify making false accusations. The underlying theory is that behind every great fortune is a great crime, so you might as well hang him for it without further proof. Morris would have been astonished at the thought of taking such risks without hope of restitution: "Do you want me to smuggle this stuff or don't you?" might well have been his answer.

And then there were those two aristocratic Frenchmen, Penet and Pliage, appearing in Providence RI in a boat loaded with gunpowder, looking for George Washington to sell it to. Everyone at the time assumed they were just adventurers, out for the money. After the passage of two centuries, it may be within bounds to ask the French to look into their records to see if maybe P&P were sent here to stir up a little trouble? There proves to be quite a literature about them, linked to that glamorous Renaissance Man, Beaumarchais.

TF-IDF in Python and MySQL

TF-IDF: Term Frequency-Inverse Document Frequency

Using Python and Scikit-Learn to analyze blogs in a MySQL database

Downloaded from a jupyter notebook into *.py format


import mysql.connector
import re
import pandas as pd
import numpy as np


# dict to group blogs by keywords

from collections import defaultdict
d = defaultdict(list)


# everything

plot_title = "Non-Terse-Verse Philadelphia Reflections"

query = '''SELECT title, blog_contents, table_key 
                   FROM individual_reflections
                  WHERE LOWER(title) NOT LIKE '%terse%' AND 
                     LOWER(title) NOT LIKE '%znote%' AND
                     LOWER(title) NOT LIKE '%front stuff%' '''

def remove_html_tags(text):
    """Remove html tags from a string"""
    import re
    clean = re.compile('<.*?>')
    step1 = re.sub(clean, ' ', text)
    step2 = step1.replace('\n', ' ').replace('\r', '')
    step3 = ' '.join(step2.split())
    
    text=step3.lower()
    text=re.sub("","",text)
    text=re.sub("(\\d|\\W)+"," ",text)
    text=text.strip()
    
    return text

mydb = mysql.connector.connect(
  host="",
  user="",
  passwd="",
    database=''
)
cursor = mydb.cursor()

cursor.execute(query)

df_idf = pd.DataFrame([])

for (title, blog_contents, table_key) in cursor:
    blog_contents = remove_html_tags(blog_contents)
    title = remove_html_tags(title)
    df_idf = df_idf.append(pd.DataFrame({'table_key': table_key, 'title': title, 'body': blog_contents}, index=[0]), ignore_index=True)
    
df_idf['text'] = df_idf['title'] + " " + df_idf['body']

docs_test = df_idf['text'].tolist()

cursor.close()
mydb.close()

df_idf.head()


from sklearn.feature_extraction.text import CountVectorizer
import re
 
 
#get the text column 
docs=df_idf['text'].tolist()
 
#create a vocabulary of words, 
#ignore words that appear in 85% of documents, 
#eliminate stop words
cv=CountVectorizer(max_df=0.85,stop_words='english',max_features=10000)  # cv = cv
word_count_vector=cv.fit_transform(docs)                                 # cv_fit = word_count_vector
list(cv.vocabulary_.keys())[:10]


from sklearn.feature_extraction.text import TfidfTransformer
 
tfidf_transformer=TfidfTransformer(smooth_idf=True,use_idf=True)
tfidf_transformer.fit(word_count_vector)


def sort_coo(coo_matrix):
    tuples = zip(coo_matrix.col, coo_matrix.data)
    return sorted(tuples, key=lambda x: (x[1], x[0]), reverse=True)
 
def extract_topn_from_vector(feature_names, sorted_items, topn=10):
    """get the feature names and tf-idf score of top n items"""
    
    #use only topn items from vector
    sorted_items = sorted_items[:topn]
 
    score_vals = []
    feature_vals = []
    
    # word index and corresponding tf-idf score
    for idx, score in sorted_items:
        
        #keep track of feature name and its corresponding score
        score_vals.append(round(score, 3))
        feature_vals.append(feature_names[idx])
 
    #create a tuples of feature,score
    #results = zip(feature_vals,score_vals)
    results= {}
    for idx in range(len(feature_vals)):
        results[feature_vals[idx]]=score_vals[idx]
    
    return results


# mapping of index 
feature_names=cv.get_feature_names()

# human-readable list of keywords for each blog
myfile = open('.../tf-idf.txt', 'w')

# explode-able/split-able list of keywords for each blog
kwfile = open('.../kwds.txt', 'w')

header = '''
==========================================
10 most-significant keywords for each blog
=========================================='''
print(header)
myfile.write(header+"\n")

for i in range(len(docs)):
 
    # get the document that we want to extract keywords from
    doc=docs[i]

    #generate tf-idf for the given document
    tf_idf_vector=tfidf_transformer.transform(cv.transform([doc]))

    #sort the tf-idf vectors by descending order of scores
    sorted_items=sort_coo(tf_idf_vector.tocoo())

    #extract only the top n; n here is 10
    keywords=extract_topn_from_vector(feature_names,sorted_items,10)

    # now print the results
    print("\n=====Title=====")
    myfile.write("\n=====Title=====\n")
    
    print(df_idf.iloc[i]['table_key'], df_idf.iloc[i]['title'])
    myfile.write("{} {}\n".format(df_idf.iloc[i]['table_key'],df_idf.iloc[i]['title']))
    
    kwds = str(df_idf.iloc[i]['table_key'])+'|'
    
    print("\n===Keywords===")
    myfile.write("\n===Keywords===\n")
    for k in keywords:
        print(k,keywords[k])
        d[k].append((df_idf.iloc[i]['table_key'],
                     keywords[k],
                     df_idf.iloc[i]['title']))
        myfile.write("{} {}\n".format(k,keywords[k]))
        kwds += k+','
    kwfile.write(kwds[:-1]+"\n")
        
myfile.close()
kwfile.close()



import operator

word_list = cv.get_feature_names();    
count_list = word_count_vector.toarray().sum(axis=0) 

word_freq = sorted(dict(zip(word_list,count_list)).items(), key=operator.itemgetter(1), reverse=True)

x = []
y = []
for word,freq in word_freq[:25]:
    x.append(word)
    y.append(freq)
    
from matplotlib import pyplot as plt
plt.figure(figsize=(15,15))
plt.plot(x,y)
plt.title("Word Frequency in "+plot_title,fontsize=18)
plt.xticks(rotation=45,fontsize=18)
plt.yticks(fontsize=18)
plt.grid()
plt.show()
plt.savefig('.../Word_Frequency.png')


# # Group blogs by keywords


myfile = open('.../Keyword_Popularity.txt', 'w')

header = '''
==================================================
Most-Frequent Keywords with the blogs they specify
==================================================
   (blog key, keyword importance, blog title)
..................................................\n'''
print(header)
myfile.write(header+"\n")

x = []
y = []

for k in sorted(d, key=lambda k: len(d[k]), reverse=True):
    if len(d[k]) == 1: break
    x.append(k)
    y.append(len(d[k]))
    print(len(d[k]),"blogs contain the keyword","'"+k+"'")
    myfile.write("{} blogs contain the keyword '{}'\n".format(len(d[k]), k))
    print('---------------------------------------------')
    myfile.write("---------------------------------------------\n")
    for t in sorted(d[k], key=lambda tup:(-tup[1], tup[2])):
        print(t)
        myfile.write("{}\n".format(t))
    print ('*****\n')
    myfile.write('*****\n\n')
    
myfile.close()


num = 50

plt.figure(figsize=(15,15))
plt.plot(x[:num],y[:num])
plt.title(str(num)+" Most Popular Keywords in "+plot_title,fontsize=18)
plt.xticks(rotation=90,fontsize=12)
plt.yticks(fontsize=18)
plt.ylabel("Number of blogs with this keyword",fontsize=18)

plt.grid()

plt.savefig('.../Keyword_Popularity.png')
plt.show()