I devote a fair amount of time to indulging my idle curiosities in gbooks, and it takes a lot of work and more than a little luck to find anything useful. It's a total mess. Great resource, but nowhere near what it should be.
I'm glad someone is drawing attention to the irresponsible sloppiness.
Is there a way to link each of your references to each of Google's Book references? A one-to-one mapping could solve almost all the issues the blog author raised.
Looks like a feature to formally link to a copy of a book/edition anywhere on the web is coming soon. (Informally linking from description hypertext or a list of related links is already possible.)
Whether Google backlinks or otherwise uses the info to improve their own metadata is of course up to them, but Google didn't get where they are by turning down free sources of crawlable quality information.
I think what Google is doing is fantastic, despite these problems. Some of the sloppiness is a byproduct of the attempts to automate the book scanning and metadata gathering. This has got to be a hugely expensive undertaking, so I can understand the shortcuts to reduce cost.
Until this gets cleaned up, what then, is the fallout from the metadata errors? Researchers will have to verify publishing dates and use other means of search for period pieces, but they will have access to historical texts over the internet. That's still pretty awesome.
I often need the very specific tiny bits of information that is trapped between pages of paper stored in some library somewhere. Today I need to fill out forms and afterwards wait days/weeks to get that information delivered to me.
The thought about all of the necessary resources for this (paper, transportation, storage, etc) makes you almost feel ashamed, because most of the time the amount of desired information is very small.
I would love to abolish this waste and have a service payed for by a flat rate that gives you access to all books, papers, phd-thesises, etc by keystroke. The service should not be completely free in order to maintain the incentive of writing well structured books as way of personal income.
Whether it is Google Book Search or something else, I think humanity definitely needs something like a global digital library.
I'd imagine in being allowed to scan a large amount of the works they are storing they presented the system in a way that would improve the way scholars did things. It's possible they gave no guarantees but if they did they have the responsibility to live up to them.
As mentioned they may have a monopoly on scanning the vast majority of older texts online due to the court settlement and deals with processors. If they are to be the gatekeeper and don't act responsibly then it is really a loss for everyone, regardless if they are bound to or not.
All this is not an argument against Google Book Search, but an argument for improving the metadata of said system. If the scholars are so distraught by this, they should offer to help review and correct Google's digital library. I'm sure they can come to some kind of arrangement.
They did not agree to this in the first place. Most of them were not happy about the deal google was given. I strongly believe it is google's job to ensure that their work is of good quality. This is similar to Wikipedia errors that find their way into newspapers and even research. The difference is that wikipedia isn't claiming to be a reliable source. Google is trying to build an exhaustive catalogue and they are the only one allowed to distribute a large amount of text due to a court settlement. Good metadata should be part of that deal.
And like the article mentioned, there already exist much better catalogues (or whatever they are called) such as the library of congress one. They might not be perfect but they sure beat scanning a book cover and picking the first date found. Why not just license it?
That's what I've often wondered. Why isn't this a government initiative to take the Library of Congress and provide a digital layer on top of it for public access. Then just turn the regular libraries into community centers without the overhead cost of book storage, purchasing, & cataloging. I don't know about anyone else, but the idea of this being in the hands of a single corporation is unnerving.
Why isn't this a government initiative to take the Library of Congress and provide a digital layer...
There's nothing in the Constitution that gives the Federal government the authority (let alone the responsibility to do this).
the idea of this being in the hands of a single corporation is unnerving.
And having it controlled by the government would make you breathe a sigh of relief? The opportunities for curating the collection as in 1984 are scary enough, but consider the accuracy of government data in other areas.
The NSA, Air Force, etc. are branches of the armed forces. While not explicitly mentioned by name, they're pretty clearly implied by Article I Section 8.
The FAA is a little grayer, and I wouldn't necessarily mind losing it. But I can see it sneaking in under the (much abused) Interstate Commerce Clause, also in Article I Section 8.
I can't put my finger on anything authorizing a giant library. Maybe under the "necessary and proper" clause, with a claim that they need a comprehensive library at hand in order to research for the government's other responsibilities (in real life, do they actually use it for that?).
A modern air force gives the President the ability to push some buttons and annihilate entire American cities. At his sole discretion. With no judicial review. And no legislation needed.
I cannot imagine how granting such unlimited power to the President is congruent with any rational reading of the Constitution. Do you really believe that the founders who insisted that the President couldn't be trusted with the ability to declare war would be happy giving the President complete authority to annihilate entire cities at his sole discretion?
Considering the Library of Congress was established in 1800, and overseen by many of the same people who wrote and ratified the Constitution (Thomas Jefferson personally donated his book collection to it), you must have an understanding of the Constitution quite different from the founders if you think the LOC isn't authorized by it.
It appeared from your original statement that things not listed in the Constitution were opposed by you. This includes the Department of the Air Force. The Constitution allows for the Department of the Navy and the Department of the Army but nothing about the Department of the Air Force.
I would say promoting the general welfare includes making all non-copyrighted materials publicly available. Clearly we differ on this but the assertion that it's unconstitutional is not so clear cut as you make it to be.
There's nothing which makes it off-limits either. In fact, Article 1 s.8 specifically grants the power to Congress '[t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries,' which is why the Library of congress exists to begin with, and if you want to sue someone for infringing your copyright it helps considerably to have your work already registered in said Library of Congress.
Making that information digitally accessible would seem to be a natural extension of that, in much the same way that being able to look up regulations, legislation, court decisions etc. online has vastly simplified and lowered the cost of research. True, there's nothing explicit about the internet in the Constitution. I sometimes wonder how you can bring yourself to use the internet, considering how it's an out-of-control government boondoggle and all.
which is why the Library of congress exists to begin with
Interesting, I was not aware of this. Can you cite evidence?
I sometimes wonder how you can bring yourself to use the internet
Exaggerate much?
In any case, the Internet's underlying technology was developed by the DoD. Today, the infrastructure is private. To be honest, I'm not clear on where bodies like ICANN fall, in the public/private space. Can someone enlighten me?
which is why the Library of congress exists to begin with//
Interesting, I was not aware of this. Can you cite evidence?
Surely. I'll put Google books links in the same order as the legal citations.
Beginning with the copyright act of 1790, 1 Stat 124, those seeking copyright protection were required to deposit a copy of the work in question with the office of the secretary of State. http://books.google.com/books?id=u90WAAAAIAAJ&dq=%22unit...
At this time, the Library of congress didn't even exist; it was established in 1800 when the US moved the government to it's current location in DC. The post of Librarian was made a paid position in 1802, and in 1815 the purchase of Thomas Jefferson's (much larger) personal library for ~$25,000 was authorized following a fire which destroyed the most of the books in the Congressional library. At this time it was not a public library, but only for use of members of the government - first Congress, then the executive, and later members of the judiciary. Although it kept expanding, it was basically a private reference library.
By 1844 many thought the US should have a national library of some sort - something Jefferson himself had urged, arguing that there was no subject on which congress might not at some point wish to research as part of its lawmaking activity. The generous bequest of James Smithson led congress to pass on 8/10/1846 an act to establish the "Smithsonian Institution, for the Increase and Diffusion of Knowledge among Men." Article 10 of this amended the Copyright act of 1831, directing that copies of all copyrighted works should henceforth be deposited with both the Smithsonian and the LoC, 9 Stat 106. http://books.google.com/books?id=TGcZAAAAYAAJ&dq=%22unit...
By 1859 neither the Smithsonian nor the LoC had established themselves as clear 'national' libraries, with varying complaints about the completeness of their collections, accuracy of the catalogs and so on. Development of LoC had been set back several years by another fire in 1852, as well as the need to service new, smaller, libraries for the sole use of the House and Senate. The situation had become confusing, so in the meantime Congress passed "An act for keeping and distributing all public documents," 11 Stat 138, which directed that copyright and patent materials still held by the Secretary of State be transferred to the Department of the Interior, apparently on the theory that the DoI was best placed to service requests for copies and catalogs from libraries in the individual states. http://books.google.com/books?id=vio3AAAAIAAJ&dq=1859%20...
The Civil war interrupted matters, and Lincoln's first appointee as Librarian spent most of the war out in the field After it had ended, a new librarian took over named Ainsworth Spofford, who was dedicated to the original idea of a national library. His efforts led to the passage of legislation bringing copyright deposits back to the LoC (13 Stat 540 (1865)); the Smithsonian deposit, which moved the whole Smithsonian collection into the LoC and for the first time provided in law that it should be fully accessible to the public except for one month a year during recess, to make repairs (14 Stat 13 (1867)), and amendment to the Copyright making registration with the LoC mandatory on pain of a fine (14 Stat 395 (1867)); and the Copyright Act of 1870, which left the Department of the Interior in sole charge of patents and trademarks, and made the L...
I think the answer is the same as for the question of why is copyright law so badly broken as to create a huge orphan works problem? I mean, there are easy tweaks to the copyright system that would alleviate the orphan works problem without interfering with the minuscule fraction of very old works that are still profitable.
The answer in both cases is regulatory capture. The parts of the government that would deal with this have lots of interaction with publishers and authors lobbying them. And those groups hate all the digitization plans.
From the perspective of a content producer, technology is the worst thing ever. Every year, technology advances and increases the number of ways that people can spend time not reading the works of any one author. A mass book digitization program would swamp the market with old books which means fewer sales for recently published books. It might allow people to get into hobbies that would sap up their time and prevent them from buying new books! Can you imagine if Wikipedia didn't exist? Think of the billions of person-hours that wouldn't have been spent reading and writing Wikipedia articles...that time would have gone somewhere, and part of it would certainly have been spent reading best seller books purchased at a book store.
I don't want to take anything away from Google: they bet big that scanning all books would have been worthwhile so all the best for them.
But it seems the usual Google/hackery way of doing things where things are automatised (or semi-automatised), but when it shows (some) shortcomings, they just don't care.
Another example (in my mind) was: the Nexus One customer service.
Yeah exactly, they just don't want to do anything if they can't automate it. I understand their viewpoint but like you said it when it fails they don't seem to want to fix it. Google does lots of good stuff but they don't seem to like to keep pushing to hard. When the road gets though they usually let the project go.
The author is arguing that Google is going to be the only one that gets to do this in the first place. They've changed the game in that they can have no competitors. If Google wasn't doing this, it would still be possible for someone else to step in and do it right. In that light, they've changed quite a bit.
* note: I'm not really familiar with the court case here, so I have no idea if the author is making a valid assertion or not.
This article argues seems to be suggesting that Google have a unique responsibility - almost as if they have somehow removed something that already existed.
I disagree with their fundamental premise that Google is "the world's last digital library". Google's demonstration that such a thing is actually possible removes one barrier to entry. They have also set a precedent which reduces the negotiation required for future projects.
The fact that Google has done a poor job (so far) on the metadata is something that's worth highlighting - but it's not a reason to curse the whole project. This is fixable, possibly even a transient issue. GBS is an entirely new resource, which scholars have not had before; if they find it unsuitable for scholarly research, perhaps they could continue to use their existing tools or work to improve it, rather than slam the effort as "A Disaster For Scholars". "A Useful Tool Which Still Needs Improvement" might not be as snappy but the The Chronicle Of Higher Education doesn't look like the Daily Mail either.
> "I disagree with their fundamental premise that Google is "the world's last digital library". Google's demonstration that such a thing is actually possible removes one barrier to entry."
That was my first reaction too, but then I realized the article was implying that the real barrier to entry is the work of scanning billions of books into digital form.
After Google has done that, there will be little incentive for anyone else to incur that same expense to create a service that Google already has a 5 year head start on.
Alot of startups these days are about finding better ways of doing things in already-crowded domains. Google itself showed the way there. But I'm not sure I disagree with article in its assertion that doing this will be much more difficult, if not virtually impossible, in this domain of creating a digital library from scanned books. There really is a significant moat here.
The idea that the barriers to entry mean that Google will never have competition are very high is true today but won't be true going forward.
Once all books are published (or reprinted) in eBook form, the conversion is potentially something which can be automated for all new (or still in "print" books). This will cover the majority of titles of interest.
Additional titles can be scanned where they're of significant interest or projects similar to Project Guttenberg may do some of the work.
Google has a head start but in actual fact has done a lot of their work when it's high cost / high labour where future competitors won't need to invest this level of effort.
You could certainly see publishers providing access to their own works and, having access to the electronic versions of texts even for non-eBooks, they could do so relatively cheaply. Add a standardised interface over the repositories provided by each publisher and you have a decentralised repository, owned by the publishers which you'd imagine could compete with Google.
> Once all books are published (or reprinted) in eBook form, the conversion is potentially something which can be automated for all new (or still in "print" books).
I think that's unlikely to ever happen, though. There are still thousands of even relatively prominent movies available on VHS that have never been released on DVD, due to a mixture of disputes over rights, expense of mastering a DVD from the source material, worry about whether they'd be sufficiently marketable this many years past their release, worry about too many back-catalogue releases cannibalizing new releases, etc.
34 comments
[ 4.0 ms ] story [ 69.2 ms ] threadI'm glad someone is drawing attention to the irresponsible sloppiness.
http://openlibrary.org/
[I work for the Internet Archive, but on our web archiving rather than our book projects.]
Whether Google backlinks or otherwise uses the info to improve their own metadata is of course up to them, but Google didn't get where they are by turning down free sources of crawlable quality information.
Until this gets cleaned up, what then, is the fallout from the metadata errors? Researchers will have to verify publishing dates and use other means of search for period pieces, but they will have access to historical texts over the internet. That's still pretty awesome.
The thought about all of the necessary resources for this (paper, transportation, storage, etc) makes you almost feel ashamed, because most of the time the amount of desired information is very small.
I would love to abolish this waste and have a service payed for by a flat rate that gives you access to all books, papers, phd-thesises, etc by keystroke. The service should not be completely free in order to maintain the incentive of writing well structured books as way of personal income.
Whether it is Google Book Search or something else, I think humanity definitely needs something like a global digital library.
I wasn't aware Google was under any such obligation. If it is, from where did the obligation come, when, and by whose authority?
As mentioned they may have a monopoly on scanning the vast majority of older texts online due to the court settlement and deals with processors. If they are to be the gatekeeper and don't act responsibly then it is really a loss for everyone, regardless if they are bound to or not.
And like the article mentioned, there already exist much better catalogues (or whatever they are called) such as the library of congress one. They might not be perfect but they sure beat scanning a book cover and picking the first date found. Why not just license it?
There's nothing in the Constitution that gives the Federal government the authority (let alone the responsibility to do this).
the idea of this being in the hands of a single corporation is unnerving.
And having it controlled by the government would make you breathe a sigh of relief? The opportunities for curating the collection as in 1984 are scary enough, but consider the accuracy of government data in other areas.
The FAA is a little grayer, and I wouldn't necessarily mind losing it. But I can see it sneaking in under the (much abused) Interstate Commerce Clause, also in Article I Section 8.
I can't put my finger on anything authorizing a giant library. Maybe under the "necessary and proper" clause, with a claim that they need a comprehensive library at hand in order to research for the government's other responsibilities (in real life, do they actually use it for that?).
I cannot imagine how granting such unlimited power to the President is congruent with any rational reading of the Constitution. Do you really believe that the founders who insisted that the President couldn't be trusted with the ability to declare war would be happy giving the President complete authority to annihilate entire cities at his sole discretion?
Really?
I would say promoting the general welfare includes making all non-copyrighted materials publicly available. Clearly we differ on this but the assertion that it's unconstitutional is not so clear cut as you make it to be.
Making that information digitally accessible would seem to be a natural extension of that, in much the same way that being able to look up regulations, legislation, court decisions etc. online has vastly simplified and lowered the cost of research. True, there's nothing explicit about the internet in the Constitution. I sometimes wonder how you can bring yourself to use the internet, considering how it's an out-of-control government boondoggle and all.
Interesting, I was not aware of this. Can you cite evidence?
I sometimes wonder how you can bring yourself to use the internet
Exaggerate much?
In any case, the Internet's underlying technology was developed by the DoD. Today, the infrastructure is private. To be honest, I'm not clear on where bodies like ICANN fall, in the public/private space. Can someone enlighten me?
Surely. I'll put Google books links in the same order as the legal citations.
Beginning with the copyright act of 1790, 1 Stat 124, those seeking copyright protection were required to deposit a copy of the work in question with the office of the secretary of State. http://books.google.com/books?id=u90WAAAAIAAJ&dq=%22unit...
At this time, the Library of congress didn't even exist; it was established in 1800 when the US moved the government to it's current location in DC. The post of Librarian was made a paid position in 1802, and in 1815 the purchase of Thomas Jefferson's (much larger) personal library for ~$25,000 was authorized following a fire which destroyed the most of the books in the Congressional library. At this time it was not a public library, but only for use of members of the government - first Congress, then the executive, and later members of the judiciary. Although it kept expanding, it was basically a private reference library.
By 1844 many thought the US should have a national library of some sort - something Jefferson himself had urged, arguing that there was no subject on which congress might not at some point wish to research as part of its lawmaking activity. The generous bequest of James Smithson led congress to pass on 8/10/1846 an act to establish the "Smithsonian Institution, for the Increase and Diffusion of Knowledge among Men." Article 10 of this amended the Copyright act of 1831, directing that copies of all copyrighted works should henceforth be deposited with both the Smithsonian and the LoC, 9 Stat 106. http://books.google.com/books?id=TGcZAAAAYAAJ&dq=%22unit...
By 1859 neither the Smithsonian nor the LoC had established themselves as clear 'national' libraries, with varying complaints about the completeness of their collections, accuracy of the catalogs and so on. Development of LoC had been set back several years by another fire in 1852, as well as the need to service new, smaller, libraries for the sole use of the House and Senate. The situation had become confusing, so in the meantime Congress passed "An act for keeping and distributing all public documents," 11 Stat 138, which directed that copyright and patent materials still held by the Secretary of State be transferred to the Department of the Interior, apparently on the theory that the DoI was best placed to service requests for copies and catalogs from libraries in the individual states. http://books.google.com/books?id=vio3AAAAIAAJ&dq=1859%20...
The Civil war interrupted matters, and Lincoln's first appointee as Librarian spent most of the war out in the field After it had ended, a new librarian took over named Ainsworth Spofford, who was dedicated to the original idea of a national library. His efforts led to the passage of legislation bringing copyright deposits back to the LoC (13 Stat 540 (1865)); the Smithsonian deposit, which moved the whole Smithsonian collection into the LoC and for the first time provided in law that it should be fully accessible to the public except for one month a year during recess, to make repairs (14 Stat 13 (1867)), and amendment to the Copyright making registration with the LoC mandatory on pain of a fine (14 Stat 395 (1867)); and the Copyright Act of 1870, which left the Department of the Interior in sole charge of patents and trademarks, and made the L...
The answer in both cases is regulatory capture. The parts of the government that would deal with this have lots of interaction with publishers and authors lobbying them. And those groups hate all the digitization plans.
From the perspective of a content producer, technology is the worst thing ever. Every year, technology advances and increases the number of ways that people can spend time not reading the works of any one author. A mass book digitization program would swamp the market with old books which means fewer sales for recently published books. It might allow people to get into hobbies that would sap up their time and prevent them from buying new books! Can you imagine if Wikipedia didn't exist? Think of the billions of person-hours that wouldn't have been spent reading and writing Wikipedia articles...that time would have gone somewhere, and part of it would certainly have been spent reading best seller books purchased at a book store.
But it seems the usual Google/hackery way of doing things where things are automatised (or semi-automatised), but when it shows (some) shortcomings, they just don't care.
Another example (in my mind) was: the Nexus One customer service.
* note: I'm not really familiar with the court case here, so I have no idea if the author is making a valid assertion or not.
And if instead you were thinking of having Google paying the scholars for that, then could not a random intern at Google do it just as well?
I disagree with their fundamental premise that Google is "the world's last digital library". Google's demonstration that such a thing is actually possible removes one barrier to entry. They have also set a precedent which reduces the negotiation required for future projects.
The fact that Google has done a poor job (so far) on the metadata is something that's worth highlighting - but it's not a reason to curse the whole project. This is fixable, possibly even a transient issue. GBS is an entirely new resource, which scholars have not had before; if they find it unsuitable for scholarly research, perhaps they could continue to use their existing tools or work to improve it, rather than slam the effort as "A Disaster For Scholars". "A Useful Tool Which Still Needs Improvement" might not be as snappy but the The Chronicle Of Higher Education doesn't look like the Daily Mail either.
That was my first reaction too, but then I realized the article was implying that the real barrier to entry is the work of scanning billions of books into digital form.
After Google has done that, there will be little incentive for anyone else to incur that same expense to create a service that Google already has a 5 year head start on.
Alot of startups these days are about finding better ways of doing things in already-crowded domains. Google itself showed the way there. But I'm not sure I disagree with article in its assertion that doing this will be much more difficult, if not virtually impossible, in this domain of creating a digital library from scanned books. There really is a significant moat here.
Once all books are published (or reprinted) in eBook form, the conversion is potentially something which can be automated for all new (or still in "print" books). This will cover the majority of titles of interest.
Additional titles can be scanned where they're of significant interest or projects similar to Project Guttenberg may do some of the work.
Google has a head start but in actual fact has done a lot of their work when it's high cost / high labour where future competitors won't need to invest this level of effort.
You could certainly see publishers providing access to their own works and, having access to the electronic versions of texts even for non-eBooks, they could do so relatively cheaply. Add a standardised interface over the repositories provided by each publisher and you have a decentralised repository, owned by the publishers which you'd imagine could compete with Google.
I think that's unlikely to ever happen, though. There are still thousands of even relatively prominent movies available on VHS that have never been released on DVD, due to a mixture of disputes over rights, expense of mastering a DVD from the source material, worry about whether they'd be sufficiently marketable this many years past their release, worry about too many back-catalogue releases cannibalizing new releases, etc.