• Kyna Hamill

    Kyna Hamill is the director of the College of Arts & Sciences Core Curriculum and a senior lecturer in the College of Fine Arts School of Theatre. She specializes in Baroque theatricality, theater and visual culture, and theater and war. Profile

    She can be reached at kyna@bu.edu.

Comments & Discussion

Boston University moderates comments to facilitate an informed, substantive, civil conversation. Abusive, profane, self-promotional, misleading, incoherent or off-topic comments will be rejected. Moderators are staffed during regular business hours (EST) and can only accept comments written in English. Statistics or facts must include a citation or a link to the citation.

There are 6 comments on POV: We Need to Keep Talking about Books, Not Ban Them

  1. Most information are hidden in books, hence reading books exposes to the right knowledge and the writers thought. Books shouldn’t be kept out of reach.

  2. Thank you for this opinion piece and for the comment (so far.) I really appreciate the research that went into the article and the resulting quotes. However, I’m also worried this article doesn’t do enough to provoke the level of discussion that the topic merits., We have to admit that current book bans are destroying the careers of librarians and teachers, undermining DEI initiatives, rewriting racial history in this country, destroying safe learning environments, and undermining democratic principles. Not all is doom and gloom however, as there are plenty of ironies that have resurfaced recently. Perhaps my favorite is Kellen Keller’s letter to Nazi authorities, when her books being burned by Nazi’s in the 1930s.
    At BU, at US universities in general, and in large cities like Boston, we have to do better. For decades we have been putting library books into storage or dumpsters in the name of solving space problems, expertise in both bookstores and used bookstores has been replaces by influencer reviews on Amazon, and college bookstores no longer have the feel of cutting-edge faculty research, but more of the ghost written political memoirs that make best seller lists.(I haven’t even thought about the potential of generative Ai for writing bogus books!) Indeed, we need to do a lot more to restore book culture and I hope this article sparks a much wider debate.

  3. Dear All: while we discuss the issue of banned books, hundreds, perhaps thousands of physical books are getting thrown in the recycling dumpsters behind Mugar Library as part of the transition of the library to the electronic era. Maybe those books could find a home somewhere, or be given away to students?

  4. A fascinating perspective; thanks for your POV.

    I’m particularly interested in your reference to Simon Pegg, who states “I’m not worried about it […] all it will do is produce mediocrity […] because it can only aggregate things that it knows. […] I think using it would be lazy and it would just undersell the audience.” As a threshold question, you contradict your own source in suggesting that he’s stating AI will create an age of mediocrity, which is the opposite of what he states — is this your own argument instead, or what do you think in terms of “settling for mediocrity”?

    I think there’s a few important points to be made here:
    I agree with your mischaracterization of your reference to Simon Pegg, in that Simon Pegg overestimates “the audience”.
    Simon Pegg (and your POV) further underestimate the incredible progress we’re likely to make with AI in terms of all the unsolved problems we’re working on to improve the state of generative AI.
    I think there is something “in our books to draw on for AI as a tool for writing” in previous technological revolutions.

    As to the last point, consider the effect of the record label on popular music. It used to be that if you wanted to listen to music, you often had to create it yourself or with others, and an individual performance or particular implementation of a musical idea was far less commodifiable at scale. The modern hegemonic record label and the scalability of recorded music rendered music production an industrialized, factory-like process, producing increasingly rigid adherence to simple musical structures.

    The reduction in musical innovation, i.e. the homogeneity of modern popular music, has not led to a substantial movement away from it — mass media producers, aiming to produce what people will be the most engaged with, must know their audience doesn’t need more complexity for them to produce it. Using this as an example, I think Simon Pegg potentially grossly overestimates the interest of the everyday layman in terms of interest in high art. He’s almost certainly right to not be worried about his own job, but I think even mediocre AI-produced art has the potential to take a decent market share or niche in places where cheap, convenient entertainment and enjoyment are higher priorities than artistic innovation.

    Finally, while the state of generative AI may be to produce truly subpar results as supervised learning relies on essentially aggregating “things that it knows,” not only will the methods advance, but large media companies have the potential to invest in a disruptively different method of training gen AI: reinforcement learning with human feedback. RLHF has already seen lots of use in fine-tuning generative models, such as GPT-4, but its current limitation is that it is expensive and consumes human effort to use. But, large media producers have the scale to enable such large investments. Once successful enough, it could become a virtuous cycle of the big get better and hence bigger: by using at-scale user engagement or other metrics as the reward in RLHF, one can envision a media company that is able to produce ever-more engaging and fascinating media that is not limited by the “things that it knows,” but only by that which optimally engages us, whether that be to explore new creative heights or exploit our time for ad revenue.

Post a comment.

Your email address will not be published. Required fields are marked *