The Internet first appeared long after I had received my Ph.D. in physics, and I was slow to use it. I had been trained in physical library search techniques: look up the subject in Science Abstracts (a journal itself now made defunct by the Internet), then go to the archived full article in the physical journal shelved nearby. Now I simply search the topics in the Science Citation Index (SCI), and then go to the journal article available online. I no longer have to go to the library; I can access the SCI and the online journals via the Internet.
These Internet versions of journals and Abstracts have one disadvantage at present: my university can afford only a limited window for the search. I can use the SCI only back ten years, and most e-journals have not yet converted their older volumes to online format, or if they have, my university can often not afford to pay for access to these older print journals.
So the Internet causes scientific knowledge to become obsolete faster than was the case with the older print media. A scientist trained in the print media tradition is aware that there is knowledge stored in the print journals, but I wonder if the new generation of scientists, who grow up with the Internet, are aware of this. Also, print journals were forever. They may have merely gathered dust for decades, but they could still be read by any later generation. I can no longer read my own articles stored on the floppy discs of the 1980's. Computer technology has changed too much. Will information stored on the Internet become unreadable to later generations because of data storage changes, and the knowledge lost?
At the moment the data is accessible. More importantly, the raw experimental data is becoming available to theorists like myself via the Internet. It is well known from the history of science that experimentalists quite often do not appreciate the full significance of their own observations. "A new phenomenon is first seen by someone who did not discover it," is one way of expressing this fact. Now that the Internet allows the experimenter to post her data, we theorists can individually analyze it.
Let me give an example from my own work. Standard quantum mechanics asserts that an interference pattern of electrons passing through a double slit must have a certain distribution as the number of electrons approaches infinity. However, this same standard quantum mechanics does not give an exact description of the rate at which the final distribution will be approached. Many-Worlds quantum mechanics, in contrast, gives us a precise formula for this rate of approach, since according to Many-Worlds quantum mechanics, physical reality is not probabilistic at all, but more deterministic than the universe of classical mechanics. (According to Many-Worlds quantum mechanics, the wave function measures the density of Worlds in the Multiverse rather than a probability.)
Experimenters — indeed, undergraduate students in physics — have observed the approach to the final distribution, but they have never tried to compare their observations with any rate of approach formula, since according to standard quantum mechanics there is no rate of approach formula. Using the Internet, I was able to find raw data on electron interference that I used to test the Many-Worlds formula. Most theorists can tell a similar story.
But I sometimes wonder if later generations of theorists will be able to tell a similar story. Discoveries can be made by analyzing raw data posted online today, but will this always be true? The great physicist Richard Feynman often claimed: "there will be no more great physicists." Feynman believed that great physicists where those scientists who looked at reality from a different point of view than other scientists. Feynman argued in Surely You're Joking Mr. Feynman that all of his own achievements were due, not to his higher-than-other-physicists I.Q., but to his having a 'different bag of tricks." Feynman thought the future generations of physicists would all have the same "bag of tricks," and consequently be unable to move beyond the consensus view. Everyone would think the same way.
The Internet is currently the great leveler: it allows everyone to have access to exactly the same information. Will this ultimately destroy diversity of thought? Or will the tendency of people to form isolated groups on the Internet preserve that all important diversity of thought, so that although scientists all have equal access in principle, there are still those who look at the raw data in a different way from the consensus?