# Where Knowledge Lived: Amritsar, 1995

*Learning before the internet, and what a scarcity of answers taught*

Kanwar Partap Singh Gill, MD · November 5, 2026 · Roots

On 15 August 1995, India's Independence Day, Videsh Sanchar Nigam Limited, the state-owned carrier for the country's international telephone traffic, opened India's first public internet service. It began in five cities: Delhi, Bombay, Calcutta, Madras and Pune. B. K. Syngal, who led the launch, later recalled that Bangalore was left off the list although it mattered more than Pune, and that the service reached the wider internet through lines of 128 kilobits per second to a handful of countries. [His account](https://theprint.in/pageturner/excerpt/how-vsnl-launched-internet-services-in-india-in-1995-when-even-china-hadnt/361722/) has the pleasant matter-of-factness of an engineer describing a bridge he built. Amritsar was not on the list.

A launch list is a map of priorities. It shows where a technology is expected to pay first, which is to say where money, institutions and demand are already concentrated, and it predicts with some accuracy the order in which everyone else will be reached. The lag between a metropolis and a provincial city is rarely announced; it is simply lived, as the period in which one place has something and another does not yet. In any case, the arrival of a technology in a country is not the same as its arrival in a student's day. That year I was midway through my pre-medical studies at Khalsa College in Amritsar, which ran from 1994 to 1996. I make no claim here about whether I touched a computer in 1995, in one direction or the other; the point of this essay does not depend on it. For a student of the sciences in a Punjabi city that year, the knowledge one needed lived where it had lived for a very long time: in books, in other people, and in whatever one could persuade one's own memory to keep.

I want to describe that arrangement without dressing it up. It was not purer than what came later, and inconvenience does not become a virtue merely by having happened long ago. But it had a particular shape, and the shape left marks I can still find in how I think. My argument is that the important change since 1995 is not that answers became faster. It is that the work of learning was redistributed. Scarcity made sources visible and questions costly. Abundance has made comparison possible and questions cheap, and in doing so it has handed back to the learner a set of tasks that scarcity used to perform for him without his noticing: checking where an answer came from, whether it is still current, and whether it has actually been understood.

## The Book That Was the Subject

A textbook in those years was less a reference than a place where one lived for a while. The same chapter was read many times, until it was known the way a house is known: the diagram in the upper corner of a right-hand page, the paragraph that always needed a second reading, the table that would surely turn up in some form on the examination.

Because a student usually worked from one principal book per subject, its author's voice became, for practical purposes, the voice of the subject. Its emphasis became your emphasis and its examples became the examples. Where it was clear, you were clear. Where it was muddled, you inherited the muddle, often without suspecting that there was another way to put it. Its errors, where it had them, were your errors, carried forward with the same confidence as everything else on the page. Secondhand copies came with a previous reader's underlining, which was a kind of company and a kind of hazard: a stranger's judgement about what mattered, inked in permanently, and not always right.

A book also came with edges. It stopped at its last page. Nothing at the foot of a paragraph led anywhere else, so curiosity was either satisfied by what was printed there or carried, unsatisfied, to someone who might know more.

Plato saw the difficulty with such a book before there were any. In the *Phaedrus*, Socrates tells a story about the Egyptian god Theuth, who invents writing and presents it to the king, Thamus, as a remedy for memory and wisdom. The king is unimpressed. In Benjamin Jowett's translation, he tells the inventor that his discovery "will create forgetfulness in the learners' souls, because they will not use their memories," that it is "an aid not to memory, but to reminiscence," and that it offers learners "not truth, but only the semblance of truth." Socrates then adds the comparison that matters most to a student with one textbook. Written words are like paintings. The figures in a painting "have the attitude of life, and yet if you ask them a question they preserve a solemn silence." A written text seems to speak intelligently, but if you question it, it "always gives one unvarying answer," and when it is misunderstood or attacked it cannot defend itself, because it has "no parent to protect" it.

That is an exact description of a textbook in a provincial city in 1995. Its author was somewhere else, often in another city or another country and sometimes no longer living, and could not be asked what a sentence meant. The book gave the same answer on every reading. It could not notice that a particular reader had misunderstood it, and it could not try again in different words. Socrates thought this made writing dangerous; the student had no choice but to make it useful, and the only way to do that was to supply, by repeated reading, the questioning the book could not answer. One read a chapter five times not because the words changed but because the reader did, and each reading put a slightly better question to the same silent page.

There is an irony here that I have always enjoyed: Socrates's warning against writing survives only because Plato wrote it down. A book cannot answer. But a reader who has lived with it long enough can begin to answer for it, and sometimes against it.

## The Price of a Question

When the book fell silent, the question went to a person: a teacher, a senior student, a classmate who had understood the thing first. A question meant for a person has a different texture from a question typed into a box. It has to wait for the person. It has to be formed well enough to be said aloud, which is harder than it sounds, because a confused student often cannot say exactly where the confusion begins. And it carries a small social price. To ask is to admit, in front of someone, that you do not know.

People paid that price unevenly. The confident asked freely; the shy rationed their questions or never asked at all. That is a real cost of any system in which knowledge lives mainly in people: access to it runs through personality, and through standing, and through the accidents of who happens to be sitting near whom. A question that went unasked because the asker was embarrassed was not an exercise in self-reliance. It was a loss, and in any classroom some of the brightest losses belonged to the quietest students.

What came back, when a question was asked, was more than an answer. A person's explanation arrives carrying the person: the impatience or the encouragement, the diagram drawn in the air, the moment of stopping to ask what exactly you do not follow. A good teacher could tell whether a student lacked a fact or lacked a picture of how the facts fit together, and those are different problems with different remedies. Here the *Phaedrus* has its other half. Against the silent written word, Socrates sets "the living word of knowledge which has a soul," the speech that can answer back, adjust itself to its listener, and defend what it means. A teacher at his best is that living word, and also the parent the orphaned book was missing.

He was not always at his best, and honesty requires saying so. Teachers could be hurried, mistaken, or simply unavailable. A generous explanation was a gift, not a guarantee, and a student who asked and was dismissed learned something too, often the wrong thing. The living word is wonderful when it is present. In 1995 its presence depended on where one studied, whom one knew and how much nerve one had, and a system that depends on those things distributes knowledge unfairly however devoted its teachers may be.

The price of asking had one benefit that is easy to miss, because it has since disappeared. A question that has to be spoken aloud to another person has to be formed first. The student must decide what exactly he does not understand, put it into a sentence, and judge whether it is worth someone else's time. That preliminary work is often where half the learning happens; a confusion precisely stated is already partly resolved. A search box asks for none of it. It accepts a fragment, a guess, a half-remembered phrase, and returns an answer to the words typed rather than to the question meant. The social cost of asking has fallen to nothing, which is a real gain for the shy. But some of that cost has been converted into an epistemic one: it is now very easy to receive a confident answer to a question one never quite asked.

## Copies of Copies

Knowledge also moved between students by hand. Notes were lent and copied, line by line, into one's own notebook, and each copy carried the copier's small errors and occasional improvements on to the next reader. In miniature, and with far lower stakes, it was the oldest problem of any manuscript tradition: how to keep a text faithful as it passes from one hand to another.

Consider a hypothetical chain, invented to show the mechanism rather than to report anything that happened. A teacher explains a process and adds, in passing, that it holds "in most cases." The first student to take notes writes the process down and drops the qualification, because it looked like a pause in speech rather than part of the content. A second student copies those notes and redraws the diagram, reversing one arrow in the haste of the night before an examination. A third, more careful, checks the diagram against the textbook and corrects the arrow, but has no way of knowing that a qualification ever existed, since the book does not mention it and the teacher's words were not recorded. By the fourth copy the note is cleaner, better organised and more confidently phrased than the teacher's original explanation. It is also less true, and nobody's name is on it.

The mechanism is worth stating plainly. Qualifications are the first casualties of copying, because they look like padding. Errors of drawing get caught when someone checks against a source; errors of omission do not, because there is nothing to check against. And attribution falls away fastest of all, so that a note which has passed through four hands arrives looking like nobody's opinion, which is to say like plain fact. Textual scholars have long used the shared mistakes in manuscripts to trace them back to a common ancestor; a copied error is evidence of lineage. A copied omission leaves no such trace.

This is not only a schoolroom problem. In 2016, Ron Sender, Shai Fuchs and Ron Milo published a short paper in *PLoS Biology* on a figure repeated for decades in textbooks and lectures: that bacteria in the human body outnumber human cells by about ten to one. When they traced the citations, they found that the papers they checked all led back to a single back-of-the-envelope estimate from 1972, based on assumptions about the bacterial density and volume of the gut. Recalculating with better measurements, they estimated roughly 38 trillion bacteria and 30 trillion human cells in a reference adult, a ratio closer to one to one. The old figure had not been fraudulent. It had simply been repeated so often, by so many careful people, that repetition had come to look like corroboration.

Amritsar has a much older and more serious example of what fidelity in transmission demands. In 1604 Guru Arjan, the fifth Guru, completed the compilation of the Adi Granth, with Bhai Gurdas writing it out as the Guru dictated, and installed it in the Harmandir Sahib at the centre of the city. I mention this with care, because scripture is not a textbook and its authority is of a different order from anything in a biology syllabus. But it means that the city where I studied had at its heart a book whose making is remembered as an act of careful compilation, with a named scribe and a named custodian. A student's borrowed notes were a long way from that. The contrast is instructive anyway: fidelity is never the default condition of a copied text. It has to be built.

## Memory Without a Date Field

The one reference always at hand was the one inside the skull. Examinations were closed-book. Whatever had not been carried into the hall did not exist there, and so a great deal of study was the slow work of moving knowledge from paper into memory and persuading it to stay.

The classical art of memory, which Frances Yates reconstructed in *The Art of Memory* in 1966, begins with a catastrophe. The poet Simonides, the story goes, stepped out of a banquet hall moments before its roof fell in. The bodies of the guests were crushed beyond recognition, and Simonides identified them by remembering where each had been sitting. From that story the ancient rhetoricians drew a method: to remember, assign each thing to a place in a familiar building, and attach a vivid image to it; to recall, walk through the building in the mind. Yates traced the art from the Roman handbook known as the *Ad Herennium*, through Cicero and Quintilian, into the elaborate memory theatres of the Renaissance. She also suggested one of the reasons it declined. Once printed books made copies plentiful, it was no longer necessary to carry whole libraries in the head, and the memory treatises dwindled into puzzles and curiosities.

That observation seems to me the key to the whole story of 1995 and after. Every new abundance of copies redistributes the work of memory. Print relieved scholars of memorising entire books and handed them a new task, the management of many books. The internet has relieved students of a great deal of memorising and handed them a new task in turn, the management of an almost unlimited supply of claims. Each shift is a gain. Each also leaves a residue of the old skill in places where it is still needed, and an unexamined assumption that the new arrangement has made the old discipline unnecessary.

Memory does more than store, and this is the part of memorisation that is easiest to undersell. Facts held in the mind can meet one another in a way that facts on a shelf cannot. A fact memorised for one examination can surface months later, unbidden, and make sense of something else entirely. Knowledge that has shared a room for long enough begins to talk to itself.

Memory has a flaw, though, that took me much longer to appreciate, and the art of Simonides makes it vivid. A memory palace is organised by place. It has rooms, corridors and images, and no calendar. Nothing in it records when a fact was put there or whether the world has since revised it. An old fact does not announce itself as old. It returns with the ease that repetition gave it, and that ease can feel like a certificate of reliability. The difficulty is not merely forgetting. It is remembering accurately something that should no longer be used in quite the same way.

There are several different ways in which a remembered fact can go out of date, and they deserve to be kept apart, because treating them all as proof that the old book was wrong misses how knowledge develops.

A statement may be replaced because new evidence contradicted it. When the Nobel Assembly honoured Barry Marshall and Robin Warren in 2005, its announcement noted that before their discovery of *Helicobacter pylori* in 1982, stress and lifestyle were considered the major causes of peptic ulcer disease. That was not a careless belief; it was the reasonable reading of the evidence then available. New evidence changed the answer.

A number may change because measurement improved. In 2004, when the international consortium published the finished human genome sequence, Francis Collins, then director of the National Human Genome Research Institute, [remarked](https://news.mit.edu/2004/humangenome): "Only a decade ago, most scientists thought humans had about 100,000 genes." The finished sequence brought the estimate of protein-coding genes down to between 20,000 and 25,000, and the counting has continued since, as methods have sharpened and the definition of a gene has itself grown more complicated.

And a category may be redefined while the underlying observations remain. Anyone at school in those years learned that there were nine planets. In August 2006 the International Astronomical Union, meeting in Prague, adopted a definition under which a planet must orbit the Sun, be massive enough to have pulled itself into a nearly round shape, and have cleared the neighbourhood around its orbit. Pluto fails the third test. It became a dwarf planet and, in the Union's words, the prototype of a new category of trans-Neptunian objects. Nothing about Pluto changed. The word did.

These are three different histories, and they call for three different kinds of checking. The useful question is therefore more specific than whether a remembered fact is outdated. What kind of claim was it? Which evidence supported it? What later work could have changed the answer, the confidence, or the meaning of the terms? A date on a textbook helps, but it is not enough. Different chapters age at different rates. Some explanations remain serviceable for decades; others become inadequate long before the binding wears out. The same unevenness exists in a person's memory. Continuing education cannot consist only of adding new material to an expanding collection. It must sometimes reopen a familiar drawer, inspect what has been kept there, and attach a new date or a narrower claim. Medicine makes this a discipline. A physician has to keep learning, and also keep unlearning, which is the harder of the two, because nothing in memory announces which parts of it have gone stale.

## What Abundance Repaired

I have written on the [Amritsar page](../../roots/amritsar/index.html) about the plainer costs of the old arrangement, in which geography and money decided too much. Those costs were real, and the gains since have been real too. It is worth naming them precisely, because a memoir of scarcity can easily drift into a lament for something that deserved to end.

The first repair is comparison. The deepest weakness of the 1995 arrangement was not slowness. It was the authority of the single source. The one book was the authority because it was the only book, and a student had almost no way to discover where it was idiosyncratic, outdated or simply wrong. A second book is the beginning of judgement, because disagreement between sources is the first evidence that a question is harder than either of them admits. Abundance makes the second book, and the tenth, available to anyone with a connection.

The second repair is access to the literature itself. On 26 June 1997, at a press conference in Washington at which Vice President Al Gore demonstrated the system, the National Library of Medicine announced free web access to MEDLINE through a new interface called PubMed. Before then, according to the Library's own bulletin, searching its online service required a registered account and searches were billed. The change sounds administrative. It meant that the index of the biomedical literature was, in principle, open to a student in any city with a telephone line and a modem, including cities that had not been on anyone's launch list.

The third repair is subtler, and it changes the relations between people as well as their access to texts. When a student can check a claim independently, authority in a classroom is redistributed. A teacher's statement can be compared with a dozen other sources before the next lesson, and a textbook's error can be found by a reader who has never met its author. This unsettles some teachers, and it should not. The best of them always wanted students who checked. What it removes is the kind of authority that rested only on being the sole available source, and that kind deserved to go.

The fourth repair belongs to the shy student, and I think it is the most humane of all. A question that once had to be asked in front of someone can now be asked privately, at any hour, as many times as necessary, without anyone knowing how basic it was. The unevenness I described earlier, in which access to knowledge ran through personality, has not vanished, but it has been greatly reduced. Whatever else abundance has done, it has made it far less likely that a student fails to learn something merely because he was too embarrassed to say that he did not know it.

"In principle" is doing real work in that sentence about the literature, and it should. Access still depends on money, on language, on infrastructure, on the price of a device and a data plan, and on the paywalls that stand between an abstract and the article behind it. A student today in a city like the one I studied in has incomparably more within reach than a student did in 1995. That is not the same as saying every student has the same reach. Abundance has not abolished the old inequalities. It has moved some of them and softened others, and anyone who describes it as a levelling should look carefully at who is still standing on the lower ground.

Jorge Luis Borges imagined the far end of abundance in "The Library of Babel," published in 1941. His library contains every possible book of 410 pages that can be written with twenty-five characters: twenty-two letters, the period, the comma and the space. Somewhere on its shelves, therefore, is every true account of everything, and every false one, and every sentence of nonsense, in endless hexagonal galleries. Its librarians are not enlightened by this. They are desperate. Some search for the one book that would be a perfect catalogue of all the others, and some believe in a legendary librarian, the Man of the Book, who has read it. A library that contains everything, without an index anyone can trust, is useless; total abundance turns out to be a new form of scarcity.

The internet is not Babel. It has indexes, and very good ones. But the Babel problem has returned in a subtler form. The scarce thing is no longer the sentence. It is the warrant behind the sentence: who said it, on what evidence, when, and with what qualifications. A fluent answer with no visible source is a Babel book that happens to be grammatical. It may be true. The difficulty is that its fluency is no evidence either way.

## The Oldest Complaint

There is a strong objection to everything I have said so far, and it deserves to be heard in its own terms. It runs like this. Calling the change a redistribution of work is too generous. Abundance has not merely handed the learner new tasks; it has weakened the capacities those tasks require. A mind that never has to hold anything, because everything can be retrieved, loses the furnished rooms in which facts meet and talk to one another. A student who never waits for an answer loses the patience in which a question ripens. On this view, the scarcity of 1995 was not just an obstacle that happened to teach some habits. It was the condition under which certain kinds of understanding could form at all, and its disappearance is a loss for which convenience is no compensation.

I take the objection seriously, partly because I find some of it in myself. But it is worth noticing how old it is. It is, almost word for word, Thamus's complaint to Theuth: the new technology will produce forgetfulness, an appearance of wisdom without the reality, learners who seem to know much and know little. It is also, in Yates's account, part of the story of print, after which the great memory systems decayed because they were no longer needed. Each time, the complaint was partly right. Something real was lost when speech was written down and when manuscripts were printed: certain feats of memory, certain intimacies between a teacher and a text. Each time, too, the complaint mistook the loss of a particular practice for the loss of a capacity. People did not stop remembering after print. They remembered differently, and they elaborated disciplines, such as the index, the footnote and the critical edition, to manage the abundance that print created.

That seems to me the right shape for an answer now. The objection is correct that some things which scarcity forced will not happen unless someone chooses them. A mind will not be furnished by accident if retrieval is always available; a question will not be left to ripen if an answer is always a second away. The objection is wrong to conclude that the only remedy is scarcity. The remedy is to name what scarcity used to do and to do it on purpose: to memorise what needs to be immediately at hand, to leave some questions open deliberately, to read one source deeply before reading twenty quickly. These are disciplines, and disciplines can be chosen. Barriers cannot be chosen; they are simply imposed, and they fall hardest on the people least able to climb them.

## Locating, Inspecting, Testing

Scarcity, I said earlier, performed certain tasks for the student without his noticing. The chief of these was provenance. You knew where an answer had come from because you had gone and fetched it yourself: this book, that teacher, a classmate's notes. You could weigh it accordingly. An answer now tends to arrive smooth and complete, and the work of asking where it came from has to be done deliberately, against the grain of the screen. It has to become a habit, because it is no longer a by-product of the effort of finding.

It helps to separate three kinds of work that the old arrangement bundled together. Locating a statement means finding where it is said. Inspecting it means asking who says it, on what basis, and as of when. Testing it means asking whether it applies to the case in hand. In 1995 locating was so expensive that it could pass for the whole job. Today locating is nearly free, and the danger is the reverse: that a statement located is taken as a statement inspected and tested.

Research on how people judge sources online has found something that should humble anyone who thinks that a scarcity education produced good judgement. Sam Wineburg and Sarah McGrew of Stanford watched ten professional fact-checkers, ten historians and twenty-five Stanford undergraduates evaluate unfamiliar websites. The fact-checkers did best, and they did it by leaving: when unsure of a site, they quickly opened other tabs to see what other sources said about it, a practice the researchers called lateral reading. The historians and students tended to read vertically, staying on the page and judging it by its own appearance, and were often taken in by signs of legitimacy such as a professional name, a logo, or a long list of scholarly references. The fact-checkers also showed what the researchers called click restraint, looking over a page of search results before choosing what to open.

The historians are the part of that finding I keep returning to. Their discipline trains them to read a document closely, attending to what it says and how it says it. That is exactly the habit scarcity trained in 1995: read the one book, and read it well. It is a valuable habit, and on the open web it can be a liability, because the question is no longer what this source says but whether this source should be trusted to say it. Neither vertical nor lateral reading is enough alone. A student needs the depth that scarcity forced and the breadth that abundance allows.

Punjabi keeps a distinction that English tends to blur. ਜਾਣਕਾਰੀ, *jāṇkārī*, is information, the state of being informed about something. ਸਮਝ, *samajh*, is understanding, the capacity to grasp how a thing works and to judge it. One can have a great deal of the first and little of the second; the language does not let them be confused. Testing a statement, the third kind of work, belongs to *samajh*. In medicine it is the point at which a fact found in a guideline or a study meets a particular patient, with a particular history, for whom the population studied may or may not be a fair stand-in. A finding is always a finding about someone: the people who were studied, at the ages and in the circumstances in which they were studied, measured in the ways the investigators chose. Whether it holds for the person in front of you is a separate question, and the answer is not printed in the paper. It has to be reasoned out from what is known about the differences between that person and the people who were counted. No retrieval system performs that step. It is the step for which the rest of the work exists.

## Keep the Discipline, Lose the Barrier

Some lessons of the scarce years seem to me worth carrying forward, provided they are carried without the obstacles that taught them.

Scarcity taught a student to rank questions. When every answer cost something, whether a trip, a wait, a favour or a moment of exposure, one learned which questions deserved the price. The habit of asking whether this is the question that matters outlasts the price, which has since fallen to nothing.

Scarcity made provenance visible. That visibility is gone, and the habit of looking for provenance has to be rebuilt on purpose, laterally as well as vertically.

Scarcity allowed a question to stay open. An unanswered question in 1995 might sit for days, waiting on a library, a teacher, a later chapter. Some such questions were abandoned for purely practical reasons, which was a loss and not a lesson. Others, carried around unresolved, turned themselves over in the waiting and came back sharper. An abundance that answers everything at once can deprive a question of that time.

And scarcity, by accident, taught the need to date one's knowledge, because it was so obvious that the one book had a year printed on it. That habit, more than any other, needs to be kept, since memory will never keep it for us.

What I will not defend is the barrier itself: the list of five cities, the price of a book, the luck of a good teacher, the nerve required to ask, the distance between a provincial college and the literature of the world. None of those taught anything that could not have been learned better without them. The task is to keep the discipline the barriers happened to impose and to be glad that the barriers are lower.

Somewhere a biology textbook printed in 1995 is still sitting on a shelf, ready to tell whoever opens it, in good faith and with perfect composure, that the human genome holds something like a hundred thousand genes. It was a good book for its year. Like every written thing in the *Phaedrus*, it gives one unvarying answer and cannot hear what happened next. Neither, unless I make the effort, can the part of my mind that books like it helped to furnish.

## Evidence and Editorial Note

**Documented findings.** The essay relies on the author's public biography: pre-medical studies at Khalsa College, Amritsar, 1994–1996. VSNL's public internet launch on 15 August 1995 in Delhi, Bombay, Calcutta, Madras and Pune, the omission of Bangalore and the 128 kbps international links: B. K. Syngal's memoir as excerpted by ThePrint (linked) and *Voice&Data*. Plato, *Phaedrus*, Benjamin Jowett's translation, for the quoted passages: Ellopos online text. Sender, Fuchs and Milo, "Revised Estimates for the Number of Human and Bacteria Cells in the Body," *PLoS Biology* 14(8): e1002533 (2016). The compilation of the Adi Granth by Guru Arjan in 1604, with Bhai Gurdas as scribe, its installation in the Harmandir Sahib, and Bhai Buddha as its first custodian: RE:ONLINE festival reference. Frances Yates, *The Art of Memory* (1966), including Simonides, the *Ad Herennium* and her account of the art's decline after print: the book, as summarised in a scholar's published reading notes. The 2005 Nobel Prize to Marshall and Warren, the 1982 discovery of *H. pylori* and the prior belief in stress and lifestyle: Nobel Assembly press release. Francis Collins's 2004 remark and the revised estimate of 20,000–25,000 protein-coding genes: MIT News, 29 October 2004 (linked). IAU Resolutions B5 and B6 (26th General Assembly, Prague, August 2006) on the definition of a planet and Pluto's classification: text of the resolutions. Free MEDLINE access through PubMed announced 26 June 1997, Vice President Gore's demonstration, and prior billing: *NLM Technical Bulletin*, May–June 1997. Borges, "The Library of Babel" (1941), its structure and the search for a catalogue: standard reference accounts of the story; it is described, not quoted. Wineburg and McGrew, "Lateral Reading and the Nature of Expertise" (*Teachers College Record*), participants, lateral reading and click restraint: Stanford News report on the study. Closed-book examinations, single principal textbooks, secondhand copies and hand-copied notes are described as general features of study in India in that period. The Punjabi words ਜਾਣਕਾਰੀ and ਸਮਝ are glossed from ordinary usage. **Vendor representations.** None.

**Recollection and tradition.** That the author was a pre-medical student at Khalsa College in 1995 is part of his published biography. No specific textbook, edition, teacher, classmate, library, question or study scene is described; the account of books, people and memory is a general description of learning in that era, written in the author's reflective voice. The author's personal access to computers or the internet in 1995 is not asserted either way. The chain of copied notes is explicitly hypothetical. The Adi Granth is mentioned as Sikh history and tradition; no claim is made about its text.

**Analysis and forecast.** The thesis that abundance redistributed rather than simply accelerated the work of learning, the readings of Plato, Yates and Borges, the three kinds of revision, the distinction among locating, inspecting and testing, and the lessons drawn from scarcity are the author's interpretation as of November 6, 2026. **Independence.** This essay is not sponsored by any institution, company, health system or technology developer.



## Related reading

- [The Library Desk](https://personal.kpsgill.com/essays/the-library-desk/)
- [Where Does This Fact Live?](https://personal.kpsgill.com/essays/where-does-this-fact-live/)
- [Wrong Beautifully](https://personal.kpsgill.com/essays/wrong-beautifully/)