If every book has a reader, then why are bookstores and libraries not carrying every single book published, even if it's only sold or loaned out once every five years? It's the tyranny of physical space - the limitations of space have forced physical bookstores and libraries to restrict their offerings to mostly hits. With ebooks - and when (not if) we perfect the ebook reader - there's no reason to only stock the hits. For users, it means gaining access to an infinite bookstore or library.
By "every book has a reader", Aaron was quoting one of Ranganathan's Five Laws of Library Science*. Meaning, for every book that exists, there must be some one out there who would find it useful or interesting.
What got me thinking was what Aaron so eloquently put as "the tyranny of physical space"; how bookstores (and libraries) are unable to carry all that they might want to stock because of physical space constraints.
He suggests when there's a prefect eBook reader, then all digital books should be made available, i.e. "digitally stocked", by the bookstore. Or libraries.
I asked myself what would justify not stocking all titles. I came up with these inter-related reasons:
- Sheer number of available titles
- Cost of digital storage
For simplicity, let's say there are one billion (1,000,000,000,000) eBook titles.
Assuming the 80/20 rule, 80% of the eBooks are not likely to be accessed most times. This means eight-hundred-thousand-million (800,000,000,000) titles.
(Wait, did i get that right? My brain simply cannot process beyond one-million... anyway, I'm trying to say it's a freaking heck lot of unaccessed titles!)
Because of sheer numbers, some titles might never be accessed even if a reader wanted them. Because the user might fail to use the right keyword. Or did not go through all the search results page.
Of course my logic could be flawed.
- I'm not sure if I've applied the 80/20 rule correctly. Maybe it could be 50% or 10% or 0.5% that do not "have its reader".
- Some of those eBooks might
notbe hosted on somebody else's server (e.g. eBooks in the public domain). The bookstore merely has to link to them and does not incur digital storage costs for all eBook titles. - Even if it's a freaking-heck-of-alota-unaccessed-titles, the data storage cost for them might be easily offset by the sales of those titles that do get downloaded. The bookstore might want to incur that cost as bragging rights.
All this is academic and all so rambly. I don't think there's any reliable source of what constitutes "all the eBooks in the world" (at least I've not checked if there have been esimates on this).
So maybe Aaron is right.
There really is no reason for the infinite eBook library not to exist.
[* ASIDE: I remember being blown away by the profundity of those 5 laws. Ranganathan is a genius. If I were to run a library school programme, I'd make the 'Five Laws of Library Science' a compulsory module. An indoctrination course, if you will. To me, if you don't accept and appreciate the 5 Laws, then you should reconsider being a librarian, imho.
Also remembered this earlier post: Ranganathan on Melvil Dewey, 1964 tape recording transcribed]
Interesting point about the 80/20 rule, but as you said, the cost of digital storage is low and I don't see why we should only stock the 20% of ebooks that are accessed most often. You brought up a good point about people not being able to look for the sheer amount of stuff if we were to stock everything. I think it's just a matter of time before we eventually develop better filters (search engines) to get to the things that people want. Oftentimes, we talk about the information overload with the deluge of digital content without realizing that the amount of information has in fact been increasing exponentially over time since the printing press was invented. The problem is more of the fact that we haven't developed good enough filters to keep up. With better filters, we might even be able to move the 80% to the top!
ReplyDeleteHi Aaron, I think in the end, it may be down to practicality. A bookstore, or a library, could content with just providing "as many as they can" rather than gunning for "all in the known world".
ReplyDeleteI don't see storage as being the issue. If you assume we're only storing the text, it's really negligible - probably <1MB for the vast majority of texts. Scanning them is probably much more expensive than storing them for 100 years.
ReplyDeleteI don't really see "as many as they can" vs "all in the known world" being incompatible. One is a goal to shoot for, the other is what you do on a day-to-day basis.
Totally agree that better search and filtering will be needed to find what you want out of an ocean. Already services like the Online Books Page are trying to organise these ebooks, but it's no Amazon. Brewster Kahle's OpenLibrary and of course Google Books are supplying search.
But probably the biggest barriers for now are non-technological: getting access to the books, getting permission to scan them - copyright issues, etc., getting them scanned for low cost. (There was a recent Economist article about Brewster Kahle that delved into this.)
And someday down the road, probably multilingual search will be the issue - I want to find out about some issue in Japanese history but don't speak much Japanese. How can I gain access to historical archives and find out what I need to know?
Great article. It would be more convenient if books have virtual copies.
ReplyDeleteEvery person have their own preference. I for example prefers to read on a hard copy.