// notes from 11 / 13 / 2026

What is art?

Art takes many forms. At its core, art is how people make meaning, express emotion, and share experiences ordinary words fail to capture. A piece of art might be beautiful, uncomfortable, playful, or serious. Perfection matters less than evidence of a person thinking and feeling through the world.

Two forms of art matter to me: public works and private works. Public works are finished pieces shown in galleries and other shared spaces, where they meet an audience and start a conversation. People photograph, review, and remember these pieces. Many people treat them as the “real” art because they are polished and on display.

Private works are unfinished pieces the artist does not yet intend to show. They live in sketchbooks, folders, and studio corners. They are experiments, half-ideas, and notes to the self. Private works are often messy, contradictory, or unresolved, and those qualities make them honest. They show the process of learning, doubting, and trying again. They remind us art is an ongoing practice as much as a product.

John Berger (1926 to 2017) clarified this difference for me. He described a drawing as a private work, tied only to the artist’s own needs. In his view, drawing does not begin as an attempt to impress an audience. Drawing is a way to think on paper, understand a subject, or work through a feeling. A drawing holds meaning even if no one else ever sees the page, because its first purpose is to serve the artist’s inner life.

This view explains why the rise of AI does not have to diminish human creativity. An AI image or text is a kind of public output, generated in seconds and shared with ease. AI output does not replace the private, inner work a person does when facing a blank page or a half-finished idea. The value of art lies in the final result and also in the time, attention, and self-knowledge gained while making the work. AI imitates this interior process on the surface without experiencing the struggle.

I also made my own artwork depicting “hope.” The piece shows a dark room with objects scattered around, from a guitar to drawing supplies, like hobbies someone dropped. These abandoned tools suggest a pause, or a loss of confidence. Nothing in the room is destroyed or gone forever. The tools are waiting. The hope in the piece lies in the return: someday, the person who left these hobbies will pick them back up with new energy or understanding.

The quiet room full of paused projects gives me a way to think about both art and AI. AI generates a song or a drawing in seconds. AI does not know how a person feels after setting down a guitar for months and then, one day, reaching for the instrument again with shaky hands and a changed heart. Human art lives in this fragile moment of return, in the decision to keep creating despite doubt or distraction. AI helps with technique, offers suggestions, and handles repetitive tasks. AI does not replace the human experience of hope filling a room like the one in my artwork.

Art and AI are able to coexist. AI is another tool in the room, alongside the guitar and the pencils, sometimes useful and sometimes ignored. What matters most is the person still choosing, hesitating, hoping, and making something drawn from a life lived. As long as this person remains, human creativity is not diminished. Human creativity finds new ways of expression.

2020 US Census

Census information lives in several places. Knowing which one holds your answer saves you time. The Census Bureau’s website (census.gov) publishes official counts and reports. The data portal (data.census.gov) holds tables you search by place and topic. Title 13 of the U.S. Code holds the laws. Court rulings and Government Accountability Office reports cover the 2020 disputes.

Apportionment. The count divides the 435 House seats among the 50 states. The 2020 count found 331,449,281 residents. Texas gained two seats. Colorado, Florida, Montana, North Carolina, and Oregon each gained one. California, Illinois, Michigan, New York, Ohio, Pennsylvania, and West Virginia each lost one. New York fell 89 people short of keeping its seat.

Redistricting. States use block-level counts to redraw congressional and state legislative districts. Public Law 94-171 requires the Bureau to deliver this data to every state.

American Community Survey (ACS). The ACS replaced the old census “long form” in 2005. The Bureau sends the survey to about 3.5 million addresses every year. The ACS asks about income, education, housing, commuting, and language. Federal programs used census data to distribute more than $2.8 trillion in fiscal year 2021.

The main mandate

Article I, Section 2 of the Constitution requires a count of every resident every ten years. The first count happened in 1790. The 2020 form asked for name, age, date of birth, sex, race, Hispanic origin, relationship to the householder, phone number, and whether you own or rent your home.

Title 13 protects your answers. The Bureau must not share your personal responses with the IRS, the FBI, immigration authorities, or any other agency. An employee who leaks them faces up to five years in prison and a $250,000 fine. Individual records stay sealed for 72 years. In 2020, the Bureau also added statistical noise to published tables, a method called differential privacy, so no one is able to trace a table back to you.

Online self-response. 2020 was the first census with an online option. About 80 percent of households who answered on their own did so online.

Mailed forms. Households with weak internet access received paper forms. Others got a paper form after several reminders.

Phone. You had the option to call and answer in English or 12 other languages.

In-person enumerators. About 67 percent of households responded on their own. Enumerators visited the rest, an operation called Nonresponse Followup.

Whatever works. When no one answered the door after repeated visits, enumerators asked a proxy such as a neighbor or landlord. The Bureau also filled gaps with administrative records from the IRS and Social Security, and with statistical imputation as a last step.

COVID. The Bureau suspended field work in March 2020, weeks before Census Day on April 1. Door knocking started months late. Counting ended October 15 after a court fight over the deadline. The Bureau delivered apportionment numbers on April 26, 2021, almost four months past the legal deadline of December 31, 2020. Redistricting data arrived in August 2021 instead of March.

Political pressure. A July 2020 presidential memo ordered the exclusion of unauthorized immigrants from the apportionment count. The Supreme Court declined to rule on the memo in December 2020, and President Biden rescinded the order in January 2021. The administration also pushed to shorten the counting period.

Citizenship question. The Commerce Department tried to add a citizenship question. In June 2019, the Supreme Court blocked the question in Department of Commerce v. New York, calling the stated reason “contrived.” Critics argued the debate alone lowered response among immigrant and Hispanic households.

Closed campuses. The census counts college students where they live at school on April 1. Campuses closed in March 2020 and students went home. Many parents counted them at home, and many dorms sat empty. The Bureau asked colleges for housing rosters to fix the gap.

Homeless population. The Bureau counts people at shelters, soup kitchens, and outdoor locations. COVID pushed this operation from late March to September 22 through 24, 2020. Shelters had reduced capacity and people had moved, so counts were harder to trust.

Non-response. The Bureau’s own review found uneven results. The count missed about 5 percent of Hispanic residents, 3.3 percent of Black residents, and 5.6 percent of American Indians living on reservations. The count overstated non-Hispanic white residents by 1.6 percent and Asian residents by 2.6 percent. Six states had undercounts, including Texas and Florida. Eight states had overcounts, including New York and Minnesota.

When help isn’t helpful. I read this note as a point about help with poor results. Proxy answers from neighbors often lacked age, race, or household size. Parents counted students at the wrong address. Differential privacy protected people but distorted numbers for small towns and tribal areas. If your class meant something different by this note, tell me and I will revise this section.