Google Book Search, um, Search

I've just blogged the "Google Book Search Search" bookmarklet.

I'm fairly exhausted right now. I know there will be questions. (I answered some in the blog comments.) It's a complex thing, for sure.

I'm going to sleep. I'm going to answer questions tomorrow, but not develop. I am itching to get back to the FWiki and collections. This bookmarklet took over my weekend!

Well done Tim, dead easy to use - plenty of AJAXy goodness in the bookmarklet itself and the data viewer/editor :)

No Ajax. All JS. JS that inserts JS into the HEAD, that inserts JS into the HEAD, etc. Altay taught me something about that. (It's crap, but it's crap in the "school of Altay.")

Bonus points for figuring out how I manage to persist it while the page changes... :)

It doesn't seem to be working in Firefox/Ubuntu. It loads a book in the LT box and then carries out a search but then it just anything. I've tried with 'Your books' and 'other people's books'

I'm seeing 98 members for whom it's working. (Holy moley, that's uptake!) Maybe a pattern will emerge. FF/Ubuntu I'm not sure about. The Ubuntu shouldn't matter, I don't thing.

3> Well, regardless, it's very clever - I'm not even going to try and figure out how it "sticks" while the GBS page changes! No bonus points for me :p

Bookmarklet doesn't recognize that I'm logged into LibraryThing. Only gives option to find 'other people's books'. (IE 7 and Windows Vista--can't load bookmarklet in FireFox from here)

Does collect IDs for other people's books.

I may be doing something silly or being too impatient, but like fluteflute in message 4 I don't seem to be able to get it working properly - for both mine and other books it looks up a first book within the search box that comes up when you use the bookmarklet (saying ID not found) and then sticks at that with a message under the book saying 'starting...'. I'm using Firefox v2.

One book is found, but after that it seems to be stuck... Using firefox on Windows 2000. Going to try with Internet Explorer.

Alright: Internet Explorer didn't recognised I was logged into LT and searching ID's for other's books got stuck also.

I'm getting the message "Google Book Search Search only works on Google Book Search: http://books.google.com" even when I am on the http://books.google.com/ page.

Looking at the FireBug console I can also see an error "GBSSstatus has no properties" in gbbs_response.php (line 61)

I seem to be having a similar problem to that of #8 & #9. Firefox 2.0.0.6 on Windows XP SP2. I click on Start in the search window, and it says:

"Book: The Sands of Time
Request: isbn:014131513X
ID: None

Starting..."

and the book is displayed in the Google Book Search window underneath. Nothing else happens.

I'm not sure I understand the use of this whole operation. What additional information will I ultimately get from this "linking with google" about my books?
Anyway, in any case I dislike the idea to have google having access in any way to my catalog - seeing how they deal with privacy issues. I wouldn't like to have to set my account to private because of that .. this is in my opinion a bad move.

I am having the same problem as andl in 11.

13: You aren't giving google access to your catalogue, you are setting up an automated process to search Google Book Search for every book in your catalogue. Anyhow, the additional information will be that from the work page you can search Google Book Search for that book - search the text of the book.

It says I'm not logged in; I am logged in. Is there a step I am missing?

IE6 on Winxp

This is all very unclear to me. Where is this bookmarklet? The only place I (think) I've found a link to it is from the blog entry, and all it did was make my browser (FF) busy, then nothing happened.

Second, despite reading the whole blog post, I still have no idea what the functionality is here. What does it do? And for what purposes are people expected to use it?

Sorry if I'm slow, but none of this is clear to me at all.

To use the bookmarklet, either right-click/ctrl-click the link and add it to the Bookmarks Toolbar Folder in your bookmarks, or just drag and drop the link itself onto the toolbar. This creates a button on the toolbar. Once that's done, go to books.google.com and click on the button you just created.

What it does is collect information about which works in the LT library can be searched on Google Book Search (presumably this could lead to links on book info pages and the like at some point, but right now all I see is that it's making a big list of the records you can link to).

Working fine for me. Really nice!

Worked for me, but it took a while to work out I had to right click (windows user) in order to add to my bookmarks...

It did fairly well in finding most of my books, but on reviewing the data I noticed some required editing... mainly with fiction it found and linked a omnibus/collection for some authors like Agatha Christie, and mine was the single book of one of those stories.
I noticed a higher degree of data errors like that with highly reprinted/collected authors than less prolific authors. *Shrug* no biggie (perhaps 20 additional errors out of 900) but worth noting.

I started it, and it seems to just hang ... I wonder if library size is a problem? Mine is 4K.

also, i just posted a bug relating to catalog displays that include the GBS results.

A note on how I got this to work.

I got this to work in Firefox. I am logged into both my Google account and my LT account. I have a tab open for each in my Firefox browser. In my LT tab I go to the LT blog and save the "Save this bookmarklet to your favorites" applet to my Browser bar. Just drag it up to the bar and when you get a green '+' sign drop it there. (I am on an Imac, btw.) Then go back to the Google tab and and find the Google Book Search page. Again, remember that you should be logged in to both LT and Google. Now, click on the "Save this bookmarklet to your favorites" link and the application titled 'Google Book Search Search' opens on (over?) the Google Book Search page. Click 'Start' in the "Google Book Search Search" app. and it should begin to search for all your LT books in the Google 'Library'.

The search seems to try more than once for books if it doesn't get an immediate hit, so I think it safe to trust the count (mine right now is '653 found 60 missed') only if it stops searching when the book is actually found. If there are duplicates then you will probably need to go and delete them when this search completes. I don't know how to delete duplicates nor how to put the found books in my google library from Tim's application because it is not finished searching yet. (I have 3190 books at LT!) I did have a few books (less than a hundred) that I entered last week when I first heard abut the new Google Library feature. The only benefit to Google Library, that I can see, is that some books (certainly nowhere near all) will be text searchable and readable. Usually, this means a 'Limited Preview' but that is better than nothing...

I also do not know how this integrates back with LT. Will anyone be able to now have a 'Limited Preview' (or in the rare case, a 'Full View') of books in my LT library that Google has found? Will I be able to do the same in the libraries of others on LT? That would be excellent!

Joe

>21 pomonomo2003: Done all that, result as #11

I have the same problem as 9 sophies_choice, it found one book and then seems to be stuck. I'm using firefox and winxp. Guess I'll try on my other browsers, too.

Same as >15 kurtabeard:, "You are not logged into LibraryThing."

Ok, oddly enough it worked fine in IE and Flock, which is a derivative of Firefox I think.

I just upgraded Firefox, is there something about the newest version that would make it not work?

Ditto. It won't load properly in the tab where I'm logged into LT, I get this tinly little window at the top with a URL in it for google book search. When in another tab, it does load however it only lets me search other people's books. am using Vista Home plus IE 7.

...Hmmm. I have a suggestion. (Mine is still working, btw, it is currently at 968 found/81 missed.) I wonder if the fact that I already had started accumulating books in 'My Library' at Google is important? Perhaps having a Google account and also Google's 'My Library' populated with some books is vital. I don't see that this was mentioned. Has anyone gotten this to work without having Google 'My Library' populated with a few books? Just a suggestion...

Joe

Just added my first two Google "My Library" books.

Still show me as not logged on to LT.

>27 pomonomo2003: I don't have anything in "My Library" on Google and it's working fine for me... using Firefox. Using IE it tells me that I'm not logged into LT. (For what's it's worth for anyone trying to debug, I'm using (shudder) Windows XP Professional on this computer (my Mac is tied up with my actual work...).

My failures are on Explorer with Windows XP.

... or just drag and drop the link itself onto the toolbar.

Yes. That's also a good way to do it. In most browsers you can just drag the link onto your toolbar. that's where I have mine.

Right now all I see is that it's making a big list of the records you can link to.

No, it puts a link on your social page, at the bottom. (I'm going to keep it off the book-info page for now. I'm in the process of redoing all the work-level pages and the social page is the most visited. Anyway, the data is collected socially.)

Also, you can add a column to your catalog. Edit your styles by clicking the "edit" link to the right of the five orange circles. Then add "Google Book Search" to the fields in your current display style.

It says I'm not logged in; I am logged in. Is there a step I am missing?

I'm thinking that the logged-in status doesn't work on IE6. I'll confirm this when Casey or Chris are around to help me with it. It works on IE7.

This is all very unclear to me. Where is this bookmarklet? The only place I (think) I've found a link to it is from the blog entry, and all it did was make my browser (FF) busy, then nothing happened.

Yes, it's on the blog post. You need to add it to your favorites. And you need to click it when you're on Google Book Search. If you click it when you're NOT on Google Book Search it should give you a message that it only works if you're over there.

Sorry if I'm slow, but none of this is clear to me at all.

No, this is hard stuff to explain. Bookmarklets are second-nature to about 2% of web users. The rest have never heard of them and the underlying logic—bookmarking a script, not a page—is odd.

... it found one book and then seems to be stuck. I'm using firefox and winxp.

It waits betwen five and ten seconds between requests, but I'm guessing you waited that long. If so, that sounds like a JavaScript error. If anyone has the JS error message, I'd love to see it.

Perhaps having a Google account and also Google's 'My Library' populated with some books is vital.

No. It doesn't use the "My Library" feature at all.

... Hmmm again. Okay, it doesn't look like having an already populated 'My Library' at Google is relevant. Two more points to consider. First, for people where it starts and then drops dead - Are you on dial-up? Maybe the connection is being dropped because of a lack of a timely response? Also, if you are running this from your workplace your IT Guru may have something set to stop non-work related Internet activity.

Also, for those who are being shown as not logged into LT - When you initially took the "Save this bookmarklet to your favorites" applet were you then logged in? It doesn't seem that this should matter though because I believe it would then offer the option to do other peoples books. - Not sure though.

Hopefully, the more data we gather from people where it is not working (and also from people that have ot working) will expose what the problems are for Tim and the team.

Joe

PS I didn't see Tim's reply before I sent this message. Sorry!

Edited to add PS.

I'm "not logged on to LT" using IE7.

for what it is worth, I'm having the same issue as others. On FF, on XP, all scripts etc enabled--the search finds the first book in my library and then just stops. no status update, nothing. the status bar reads "Status: 0 found / 0 missed / 3405 remaining" and never changes.

It started and stopped in FF - I am not on dial-up, and it worked fine in two other browsers, so it seems like it definitely had something to do with Firefox (for me).

Also, there's no visible JS error message, and I am not tech savvy enough to find one if it's hidden somewhere (though if someone told me where to look...).

FF/XP. It freezes after one loop. Could this be the error:

Fehler: theID has no properties
http://www.librarything.com/gbbs_response.php?t=1190042067720
Zeile: 116

I'm using FF/XP too. I've tried running FF's Error Console whilst running the bookmarklet. There are a *lot* of warning events in the console, but only 2 actual error messages.

Error: href.match(/vid=ISBN(\d{13}|\d{9}0-9Xx)/) has no properties
Source File: chrome://bookburro/content/logic.js
Line: 127

(which I think we can discount as being BookBurro related, not LT!)

and

"Error: theID has no properties
Source File: http://www.librarything.com/gbbs_response.php?t=1190042459140
Line: 116"

That error doesn't want to paste into the message correctly...

Error: theID has no properties
Source File: http://www.librarything.com/gbbs_response.php?t=1190042459140
Line: 116

Agh, I don't know why it keeps getting formatted differently to how I paste it. But if you mouse over the link it creates you'll see the full error - after the number in the URL, the number 116 is preceded by Line:.

>36 jcbrunner:

Interesting. I made a change that should solve this. Let me know if people are still getting this.

FF/XP works for me, fwiw. I'm logged into a google account, but no google library stuff. It's been running all morning, except once when I stopped it.

Minor quibble: It doesn't like titles with ampersands. It finds the book and the correct google books id, but it only grabs the google books title up to the & (so King's Nightmares & Dreamscapes just said 'Nightmares') and it has no data for confidence/availability.

Newp. but wow, the reply to message box is way up the the top of the screen now. what havoc!

Newp. but wow, the reply to message box is way up the the top of the screen now. what havoc!

and I got a double message as a bonus. heh. code. freaky stuff. I'll just ignore this new feature until it actually works.

I'm surprised that:

var theID = linktoanalyze.match(/id=(^&?+?)&/);

results in an undefined—rather than empty—theID, but I changed it to

window.theID = linktoanalyze.match(/id=(^&?+?)&/);

Let me know if you're still seeing that error.

>40 antqueen: Good. Thanks. I'll fix that.

Firefox/Ubuntu Error

window.theID has no properties
http://www.librarything.com/gbbs_response.php?t=1190043211810
Line 116

Blech. I wish I had this error, so I could look for it.

--> 7

That's what's happening with me, too. I'll take some time to work on other people's book then.

Question...

If I'm working on other people's books, how can I be sure I'm not duplicating what someone else is doing?

Still getting the same problem. I did notice this: before I get

window.theID has no properties
(now line 117 of gbbs_response.php)

I get:

Warning: assignment to undeclared variable linktoanalyze
on line 115.

Okay, Sandpiper, I'm going to sign in as you and see if I can reproduce it.

Damn. It's working for me.

Still doesn't work for me. It get's stuck after it has found Koninklijk Huis der Nederlanden :(. The bookmarklet doesn't even says it has found a book

I'm getting some pretty amusing results back from Google. Most are perfect matches; a few (with reasonable confidence ratings) are totally off. Looking through my results on the Google Book Search Search Data (what a mouthful!) page I've deleted a few of them by switching their status to 'Not Found', but I've left my favourite so far:

For some reason, Tolkein's 'The Fellowship of the Ring' (Touchstone is bringing up 'The Hobbit', so has been omitted) links to a book called 'The 1960s' by Edward J. Rielly...

Similarly, Harold Bloom's The Best Poems of the English Language brought up something quite non-poetic (link deleted).

(Other errors make more sense to me: for example, my copy of Byatt's Angels and Insects linked to the Google page of Billy Graham's book Angels (link deleted) -- clearly not the same book but I can see how a search-engine might get it wrong)

Are these ISBN problems (i.e. two books with one ISBN), LT data problems (I try to be careful with my entries but you never know) or Google search problems?

Ah well, Tim, at least you got a view of my books done for me! I just had another go with only one tab open in Firefox, just in case that was a problem at all. No joy. I'm going to give up for the moment and have another go when I get home from work. I'm pretty sure I have exactly the same version of Firefox at home, also on XP SP2, but I'll give it a whirl.

Hmm, think I'll wait until I get home tonight to experiment with this. Clearly, trying this at work (Windows XP Professional, IE6 as opposed to my iMac at home) is not a good idea, judging from previous posts! Maybe things will settle down over the rest of the day.

Not quite the same problem here: so this mightn't be the best place to post it.
I added the Google Book Search column to a view, and sorted by it (to and fro) to see if it had brought back anything. It hadn't and when I switched to another view (without Google Book Search as a column) I got the following error (showing no items)

SELECT a1.books_userid, a1.books_id, a1.books_yourrev, w_reviews, w_workcode, w_copies, a1.books_stamp, booksall_summary, GROUP_CONCAT(DISTINCT s1.s_text ORDER BY s1.s_text ASC SEPARATOR '@@@') as subjects, GROUP_CONCAT(DISTINCT s2.s_text ORDER BY s2.s_text ASC SEPARATOR '@@@') as subjects_work, a1.books_tags, booksall_comments, a1.books_lccallnumber, wd_lccn, a1.books_ISBN FROM bookstack as a1 LEFT JOIN workalias AS wa1 ON books_sameness_resolved = wa_fullcode LEFT JOIN work ON wa_workcode = w_workcode LEFT JOIN booksall ON booksall_id = a1.books_id LEFT JOIN subjectbooklink AS sbl1 ON a1.books_id = sbl1.sbl_bookid LEFT JOIN subjects AS s1 ON sbl1.sbl_subjectcode = s1.s_subjectcode LEFT JOIN subjectworklink AS swl1 ON wa1.wa_workcode = swl1.swl_workcode LEFT JOIN subjects AS s2 ON swl1.swl_subjectcode = s2.s_subjectcode LEFT JOIN workdata ON w_workcode = wd_workcode WHERE a1.books_userid = 'Donogh' GROUP BY a1.books_id ORDER BY googlebooksIDs.gb_view DESC, a1.books_stamp LIMIT 0, 20 Unknown column 'googlebooksIDs.gb_view' in 'order clause'

I don't know if Tim worked on the script behind the scene, but I disabled the Customize Google add-on and now the Google Book Search works fine.

@45 (Fx/Ubu)

Additional info:
parseLinks()
if (window.theID.length)

re: #20, #34, #35, #36 -- It's working for me now with Tim's latest changes.

... Since there are a number of works that it's not finding, I'm curious about future iterations. If I run through my current 4200, and then later on want to update to pick up (a) titles that I've added to my library and (b) titles that google books has added, will the script know to only scan through the works with no google data?

--> 31

It says I'm not logged in; I am logged in. Is there a step I am missing?

I'm thinking that the logged-in status doesn't work on IE6. I'll confirm this when Casey or Chris are around to help me with it. It works on IE7.

I got around this by closing IE6 and trying to run my books on Firefox. It's working just fine now!

I'm now at home, still with Firefox/XPSP2, and it's working for me. :-) Don't know whether that's due to slightly different setup on the machine, or Tim's changes. I'm thinking it's more likely the latter, since I can't really think what would be very different between the two machines. I might have the odd plugin difference, but little else. Unless it's an overzealous bit of anti-virus software on the work machine.

Okay two three minor things:

(1) The "nnnn remaining" doesn't update. (Firefox 2.0.0.6, OS 10.4), although "n found" and "n missed" update.
* update at 3pm - stopping, starting again on a new machine; the "nnn remaining" count updates. yaay!

(2) Possibly related: I hit "stop" at "24 found", then started again, to see if it would preserve work done. The "4266 remaining" stayed the same. Now i'm worried: Am I tied to keeping my computer open here at this cafe for the, umm, 8 seconds * 4000 titles = 8 plus hours that it will take to finish this off?
** update 3pm: seems to be okay since the count updates.

(3) I have yet to a see a "search this work" or anything like that on the works/social info/book pages. Do I have to wait until it's finished my catalog? ... that seems unlikely. I've checked a number of books that went through my google book search screen and showed as available, preview available, etc.

It's still not working for me. And I found the error console. I don't know if this will be at all helpful but here is what it said.

still not working for me, either. still finding the same book over and over again.

I'm still getting a "not logged in" message (I also tried earlier this morning), using IE7.

First it looked that result is zero, but when I repeatedly pushed the start link the program started to find my books. Now it has found almost four hundred books.

This is really fun! I have a English-Finnish dictionary and the Google Books shows me the places mentioned in this book (there are lots of them of course):
http://books.google.com/books?id=VDgdAAAAIAAJ&pgis=1

--> 58

It's still working well. I'm about about 80% done. When I'm completely finished, I'll probably need some help in figuring out how to edit the items that are incorrect.

At some other time, would it be helpful for me to download the information for "other people's books"?

--> 65

I spoke too soon...

It looks as if it's stuck at 306/327. Does that mean I'll have to run the entire download again?! :-(

May I *now* go grocery shopping?! :D

--> 66

I answered my own question. I stopped my download and restarted it. It is only picking up those books I've not yet downloaded. Phew!

I finished mine a while ago. It found 244 of the 266 in my library. I moved on to other's libraries now and will leave it running on and off for as long as I can.

Also, I stopped other's libraries and went back to my library, and it is only trying to find the rest of the books that previously were not found. So, for me, it started by trying to find 22 books. I ran that and it didn't show any missed. But I went back to do it again, and now it seems to keep trying to find the same 8 books. Moving back on to other libraries.

31> "No, it puts a link on your social page, at the bottom. "

are you talking about this page that you get to via the Social Information link?

http://www.librarything.com/work/395645&book=252811

cause I don't see the link to Google Books.

While LT and Google chat in the background, I looked at some of the results:

1. Internationalization issues: Google Book Search is returning language specific results ("Eingeschränkte Vorschau" should be "limited preview", gets "undefined" in LT).

2. German books are surprisingly well covered. The largest problem pose United Kingdom editions, as US editions often have different ISBN and titles (usually tacking on "... changed the world").

3. Sometimes, results are plain weird. Can we have a simple clear data button? At the moment, I have to erase data on three lines.

The Lord of the Rings gave me Deconstructing Tolkien: A Fundamental Analysis of the Lord of the Rings. How do I clear it?

Follow this GBSS data link, choose your books and edit.

I have edited my Display style to show 'Google Book Search' under Display style 'E' in my catalogue. Even though my search is still running if I go to 'E' I can see those books that the Google search found. However, if I go to any other Display (any not 'E') I get the following (for 'A' style):

SELECT a1.books_userid, a1.books_id, a1.books_yourrev, w_reviews, w_workcode, w_copies, a1.books_title, a1.books_author, a1.books_sort_author, a1.books_date, a1.books_tags, a1.books_otherauthors, a1.books_source, a1.books_rating FROM bookstack as a1 LEFT JOIN workalias AS wa1 ON books_sameness_resolved = wa_fullcode LEFT JOIN work ON wa_workcode = w_workcode WHERE a1.books_userid = 'pomonomo2003' ORDER BY googlebooksIDs.gb_view, a1.books_stamp LIMIT 0, 100 Unknown column 'googlebooksIDs.gb_view' in 'order clause'

Another example, under Display style 'C' I get:

SELECT a1.books_userid, a1.books_id, a1.books_yourrev, w_reviews, w_workcode, w_copies, a1.books_ISBN, a1.books_customcover, ac_exists, booksall_summary, GROUP_CONCAT(DISTINCT s1.s_text ORDER BY s1.s_text ASC SEPARATOR '@@@') as subjects, GROUP_CONCAT(DISTINCT s2.s_text ORDER BY s2.s_text ASC SEPARATOR '@@@') as subjects_work, booksall_comments, a1.books_tags, a1.books_lccallnumber, wd_lccn, a1.books_otherauthors, a1.books_publication FROM bookstack as a1 LEFT JOIN workalias AS wa1 ON books_sameness_resolved = wa_fullcode LEFT JOIN work ON wa_workcode = w_workcode LEFT JOIN amazoncovers ON books_ISBN = ac_isbn LEFT JOIN booksall ON booksall_id = a1.books_id LEFT JOIN subjectbooklink AS sbl1 ON a1.books_id = sbl1.sbl_bookid LEFT JOIN subjects AS s1 ON sbl1.sbl_subjectcode = s1.s_subjectcode LEFT JOIN subjectworklink AS swl1 ON wa1.wa_workcode = swl1.swl_workcode LEFT JOIN subjects AS s2 ON swl1.swl_subjectcode = s2.s_subjectcode LEFT JOIN workdata ON w_workcode = wd_workcode WHERE a1.books_userid = 'pomonomo2003' GROUP BY a1.books_id ORDER BY googlebooksIDs.gb_view, a1.books_stamp LIMIT 0, 100 Unknown column 'googlebooksIDs.gb_view' in 'order clause'

Does anyone else see this? Help!

Tim, this is still not working for me, I am getting the same message "Google Book Search Search only works on Google Book Search: http://books.google.com" which of course is the page I am at.

This worked fine for me when I was beta testing it, I had no problems at all. XP/FF.

73>See http://www.librarything.com/talktopic.php?topic=20344 .

I have the same problem as 74, did google put the kibosh on this already?

I just checked mine and it is still running for other libraries. (277 found, 22 missing, 1000 remaining)

#76

I have had the same problem right from the start of the day as has nperrin. People have reported the bookmarklet working subsequent to our initial failures.

Mine's still running for other libraries as well...

188 found - 40 missed - 1000 remiain

I'll check back later to see how it's doing.

I should say also that I tried the bookmarklet in IE 6 also and get the same exact error.

#75 - jjwilson61 thanks for the heads up!

There is another issue I've been seeing. Google stops the process saying:

"We're sorry...

... but your query looks similar to automated requests from a computer virus or spyware application. To protect our users, we can't process your request right now.

We'll restore your access as quickly as possible, so try again soon. In the meantime, if you suspect that your computer or network has been infected, you might want to run a virus checker or spyware remover to make sure that your systems are free of viruses and other spurious software.

We apologize for the inconvenience, and hope we'll see you again on Google.

To continue searching, please type the characters you see below:"

I type the characters, press enter, and then have to start the LT application (Google Book Search Search) again. I have 'bout 3000 books total with only about 1000 that still need to be 'googlized'. I've seen this a handful of times but Google (thus far) always allows me to restart. Obviously, this is not an LT problem. I only mention it because if anyone planned to run the 'Google Book Search Search' app. overnight on a large library I doubt it would finish for them...

Joe

Regarding editing the ones that are already done - 99% of my books are unabridged audio. I try to use the specific ISBN for the audio. The GBSS seems to just find the book. If it is linking to the correct book (just not an audio version) should it be marked "Exact Edition", "Other Edition", etc.? Right now almost all of mine say "Probably right". Most of them do go to the right book, just not an audio version (which it seems Google is sorely lacking in.)

82: Google wouldn't really be able to scan and upload audiobooks so you could search them full-text, so I doubt you would really find any audio editions correctly. It seems to me that would make it "Other Edition" but I don't know.

I finally got it to work and start searching in Firefox (after trying a few times in Firefox and having it stall with the first book); it's plugging away now. (I can't get it to work in IE7 at all; it doesn't recognize I'm logged in.) But I'm confused about what the results will be. Will I be able to edit the data it finds to remove links to incorrect editions and books? And, like others, I'm not seeing anything on the "Social Information" page for the books it has found.

#84 you can edit them by going to http://www.librarything.com/gbss_data.php.

But that's not what Tim said. Is the implementation just not finished yet?

I have the Google search in my book display presently. I edited my display so that it would show up as one of the options.

book display

#73: I didn't have this problem initially, but it has now appeared. I can't look at any display style except the one containing "Google Book Search".

To get rid of the style display problem, try this:

(a) edit your style options & get rid of the google book search column in any styles
(b) save
(c) log out of LT
(d) close your browser
(e) restart, login, and now view

that worked for me

wow - when i just went to gbss_data.php (my personal file) i got a large xml file with a lot of things like this:

Removed per Amazon TOS
Removed per Amazon TOS
202555
uBu0pWK4rzIC
Abusing Science: The Case Against Creationism 
026261037X
Exact edition
Limited preview

key element being "Removed for Amazon TOS"
... and then back to the default gbss_data.php I got this message:
Fatal error: Allowed memory size of 41943040 bytes exhausted (tried to allocate 82 bytes) in /var/www/html/inc_magicDB.php on line 521

and, having broken it, i will now take a break from the internet.

Much to respond to here. And much going on and getting done here. I wish there were ten of me.

Regarding >90 lquilter: Wow. The data is now so voluminous that it broke the PHP XML parser's memory limits. Wow!

Home and on Firefox.
Bingo.

--> 82

Adendum:

I cannot edit the data that's been downloaded. When I try to edit it, it says "saving" in green leters. However, when I go back to it later, the information has reverted back to the original incorrect information.

By the way, how many books have to be downloaded at this time with all of our members so far?

I'm getting the same error message on the FF console as others earlier today:

Error: theID has no properties
Source File: http://www.librarything.com/gbbs_response.php?t=1190078896152
Line: 116

>93 SqueakyChu:
SqueakyChu,
The same thing happened when I was using IE7 with Windows Vista. (I just finished reading comments on the blog and read your comment.)

On a different computer, XP with IE7 or Firefox, most of the feature has worked as described--I could search for my own LT books and I could edit. (I'm wondering about teh extra privacy features of Vista as a possible cause for the block. Or that I may have more pop-up blocks on the Vista computer.)

Lots of editing to do.

I never saw the Google Book Data on the social pages, just in my library listings. Probably a forest and the trees thing.

A few book appear not to have been searched by Google. How, or can you, redo those? When I try to restart the bookmarklet after running all the way through ny catalog, nothing happens even though there are some books with no data from Google - not even a 'no book found'.

1. Data is showig on the work pages now.
2. Large XML feed is working—56 Megs.

--> 97

My work pages are not showing the data.

On my social page, it says...

"We're not sure if this book is available from Google Book Search (help us find out)...."

and

"Warning: include(fwiki/inc_fwiki.php): failed to open stream: No such file or directory in /var/www/html/work.php on line 995 Warning: include(): Failed opening 'fwiki/inc_fwiki.php' for inclusion (include_path='/var/www/html:.:/php/includes') in /var/www/html/work.php on line 995 Fatal error: Class 'WikiItem' not found in /var/www/html/work.php on line 996"

This happens for books whose Google information is being displayed with no problem on the library display page.

I'm not sure why this is happening.

I see the search this book on the social page. Here is what I get when I click it:

Warning: include(fwiki/inc_fwiki.php): failed to open stream: No such file or directory in /var/www/html/work.php on line 995 Warning: include(): Failed opening 'fwiki/inc_fwiki.php' for inclusion (include_path='/var/www/html:.:/php/includes') in /var/www/html/work.php on line 995 Fatal error: Class 'WikiItem' not found in /var/www/html/work.php on line 996

Pushed some Fwiki code by mistake. Christopher is fixing.

--> 93

The edits look as if they're taking now.

I think this is a nice-to-have but certainly not crucial... If the member is signed into LT and that is being recorded beside each GBS ID retrieved - it'd be a nice little statistic to see somewhere how many IDs you've contributed (I'd like to know, on my sluggish ADSL connection) and perhaps a leaderboard (like the top combiners etc)? :)

Well, I got to work this morning and tried it again in FF/XP, but it still didn't work. So I tried uninstalling a couple of addons. First I uninstalled GreaseMonkey, restarted the browser. Bookmarklet still didn't work. Then I uninstalled the Customize Google addon, and restarted the browser again. Bingo! It's working now. I don't think I was actively using the Customize Google addon - it was one of those I installed, had a play with, then realised I didn't really need, but didn't get around to uninstalling.

I have got it to work this morning after doing nothing over night. I guess that there was a few tweaks in the javascript generated by the php pages. So it might work for everyone who had the same problem as me in #11.

I was one of those with the #11 problem; it's now working for me too.

Help for the tech-challenged, please. I'm not in the 2%, and need explanations of the explanations.

To try to get some idea what it is doing, I edited my catalog view and then clicked on Google Book Search. I found "no preview available" or "limited preview" next to 8 titles. (2 turned out to be for the wrong books. How does one change them?) I looked on the social page for the first of them --- Quartered Safe Out Here (touchstone not loading --- and found "We're not sure if this book is available from Google Book Search (help us find out)."

How did that get in my catalog? Please explain about the "other people's books" option. Does it search at random? For books in private accounts? I'd figure that someone(s) with whom I share those 8 books had run this thingie, and I got the results. But my catalog shows "no data" for the 40 titles I share with Tim, and I do see the data in his catalog.

Adding the column in my catalog did mess up the views. I had to sign out, disconnect from the net (I'm on dial-up) and go back in. I don't think I'll be looking at that view much for now!

#106, go to http://www.librarything.com/gbss_data.php. Here you will find the books gathered by the GBSS (Google BOOK Search Search), click on the 'Just you' link and you will see your books only. Click on the pencil to edit. There doesn't seem to be a way to delete. In the catalog view of my Library, in the Google Book Search column, I have seen the following:

no data
no book found
Full View (this is extremely rare, out of nearly 3200 books this occurred, from what I can see, about 30 times! Disappointing...)
Limited preview
No preview available (by far, the greatest number of my books are here.)
Snippet view

The above is the order they appear in my catalog when I sort the Google Book Search column. I would prefer this to be the default order:

Full View
Limited preview
Snippet view
No preview available
no book found
no data

Most people, I believe, would be interested in books with the 'Full view' and 'Limited preview' available. Conversely, those interested in troubleshooting their own library would be most interested in 'no data' and 'no book found'. A reverse sort of my proposed default order would give them these two categories first. Does anyone know the difference between 'no data' and 'no book found'? And, as myshelves mentioned,

"my catalog shows "no data" for the 40 titles I share with Tim, and I do see the data in his catalog"!

How is that possible?

Also, on the book social information page I am seeing at the bottom:

"We're not sure if this book is available from Google Book Search (help us find out)."

This is good; but when I click on the link "(help us find out)" I am brought to the top of the page that I am on (that is, the book social information page). I've seen this in Safari, Firefox and Opera. Where is that link supposed to go?

Joe

Fixed for me too as of this morning.

This is like reading about a color I've never seen. Got the window to open, but the search stops after one ISBN and that's it. Am at work now on IE7 on XPP. I'm giving up. Have at it folks.

Still not working for me in Firefox, either.

Running well for me. I'll do some more when I get home from work tonight. Using Firefox.

It's working for me now, the only thing I changed was added the Google search column to one of my views, though that could just be coincidence.

#107,

click on the 'Just you' link and you will see your books only.

Been there. :-) It didn't show me any books. Because I haven't run the thingie? Or because my catalog is private?

Tim has disabled it - theres a blog post about it.

then I better not stop mine anytime soon, it still seems to be running fine...

Disabling the Customize Google Options in Firefox unstuck everything for me. Now its working. Very neat!

Just as someone comes up with a fix that will probably help me (disabling Customize Google firefox extension) then its disabled! :-D

Here's the blog post.

Feel free to leave comments there or here. I'm not going to start another thread since I don't want to expand much on what I said there.

#73 & #88: Taking "Google Book Search" out of all display views appeared to correct the problem. But appearances are deceiving. This morning, my catalogue is inaccessible in all views. The system suffering this misfortune runs IE6. On one running Opera, everything looks fine.

119 > actually all you have to do is sort on a different column...

Tom,

Did you clear out the cache?

The sort thing should be fixed. I've got to go out for a lunch meeting—quite a rare event in Portland, Maine!—but will check here when I come back.

Yes, deleted cache; it made no difference. Here is the error message: Warning: array_flip(): The argument should be an array in /var/www/html/catalog_bottom.php on line 1424 Warning: Invalid argument supplied for foreach() in /var/www/html/catalog_bottom.php on line 1425 Warning: join(): Bad arguments. in /var/www/html/catalog_bottom.php on line 1537 SELECT a1.books_userid, a1.books_id, a1.books_yourrev, w_reviews, w_workcode, w_copies, a1.books_ISBN, a1.books_customcover, ac_exists, booksall_summary, a1.books_publication, booksall_comments, a1.books_tags, GROUP_CONCAT(DISTINCT s1.s_text ORDER BY s1.s_text ASC SEPARATOR '@@@') as subjects, GROUP_CONCAT(DISTINCT s2.s_text ORDER BY s2.s_text ASC SEPARATOR '@@@') as subjects_work FROM bookstack as a1 LEFT JOIN workalias AS wa1 ON books_sameness_resolved = wa_fullcode LEFT JOIN work ON wa_workcode = w_workcode WHERE a1.books_userid = 'TomVeal' GROUP BY a1.books_id ORDER BY a1.books_title_resolved, googlebooksIDs.gb_view DESC LIMIT 0, 50 Unknown column 'ac_exists' in 'field list'

I can have a view with the column in it and see the data, but NOW it won't let me sort with the Google column (error below). Switching to a different column for sorting will get rid of the error.

Warning: array_flip(): The argument should be an array in /var/www/html/catalog_bottom.php on line 1424 Warning: Invalid argument supplied for foreach() in /var/www/html/catalog_bottom.php on line 1425 Warning: join(): Bad arguments. in /var/www/html/catalog_bottom.php on line 1537 SELECT a1.books_userid, a1.books_id, a1.books_yourrev, w_reviews, w_workcode, w_copies, a1.books_title, a1.books_author, a1.books_sort_author, a1.books_datebought, a1.books_dewey, wd_dewey, a1.books_lccallnumber, wd_lccn, googlebooksIDs.*, a1.books_ISBN, a1.books_customcover, ac_exists FROM bookstack as a1 LEFT JOIN workalias AS wa1 ON books_sameness_resolved = wa_fullcode LEFT JOIN work ON wa_workcode = w_workcode WHERE a1.books_userid = 'readafew' ORDER BY googlebooksIDs.gb_view DESC, a1.books_datebought DESC LIMIT 0, 100 Unknown table 'googlebooksIDs'

Further to #73, #88, #119 and #123: I tried logging out and looking at my library as an "outsider". That produced the same error message.

Try it now, if you can.

125: I had that problem before and it went away when I deleted all my LT cookies.

#126: Now I can see my library again. Thanks.

Now that the GBSS has been disabled, is there a way to manually add the link? I have quite a few that I'm sure are there (my Shakespeare springs to mind) that do not as yet have the links on their social data page. Can I manually do a search and add the link?

129> In Tim's blog post he mentioned the possibility of an interface being built to allow this: "The links members discovered will be kept, and the data is available. We will be adding new tools for members to edit and add Google book ID information by hand, if necessary."

Are there any social data pages that actually link to the Books site or have a search box for a work? Though most of my titles are included (albeit mostly without text or preview) and the xml file contains info on them, the social data pages for the books read:

Search this book
We're not sure if this book is available from Google Book Search (

help us find out

).

I had already begun creating manual links in the comments field and ahve continued--now cleaning up the data for my books in the LT file at the same time. The interface for the LT file still works and is available at http://www.librarything.com/gbss_data.php

Are there any social data pages that actually link to the Books site or have a search box for a work?

Not any more. I did previously. I only have snippet views left in my library field. Perhaps this malfunction has something to do with Tim's disabling the GBSS in preparation for a different method of transferring this information to LT. Anyone know?

Check out the LT Blog for the latest on Google ans LT. Tim has put something new up.

About Tim's latest entry on the collaboration between LT and Google (I originally posted my comment on the blog, so it concerns what he said there and what members commented):

Interesting to notice the absence of any substantial comment by Tim about all the privacy worries overwhelmingly expressed in this blog. Doesn't he care? is Google's $$$$$$ check just too good to take members worries into account?

I definitely vote against this move. I can't believe LT is even considering this. I am a google user (search) but only because there are almost no alternatives with the same quality in those fields. Do you people really want to end up having to use one and only one company's tools? Some might say they don't care (microsoft has that kind of monopole in his fields) but since google's business is mainly information gathering and SELLING and considering its well-known user -unfriendly privacy policies, I'd rather like to avoid as much as possible having all my information in google's hands.

Furthermore, I do agree with people wondering why so much time and energy is being spent on those projects of rather small effective use (I think I can count on the fingers of my hands the amount of books in my catalog that would actually be "in-text" searchable thanks to google and I'm guessing that would apply to the catalogs of most LT users), when all this time and energy could be spent actually improving LT's features as well as adding new ones, ones that have been asked for over and over again. I remember having read some 4 weeks ago that in one week (that would have been 3 weeks ago) we would at at last see collections and wishlists & co being implemented... still waiting.

Anyway I think "nathan" sums it up rather well.

Interesting to notice the absence of any substantial comment by Tim about all the privacy worries overwhelmingly expressed in this blog. Doesn't he care? is Google's $$$$$$ check just too good to take members worries into account?

(1) I think Tim did that purposely so comments would be a list of peoples' preferences instead of a debate with him personally.

(2) I'm sure he cares.

(3) I'm sure he'd like to make some $$$ from LT as well. LT is a fantastic service but not a charity.

(3) well there are a lot of ways to make money without having to deal with google. I just hope the whole purpose of LT wasn't just to buid up a substantial customer database in order to eventually sell it to google-like companies as so many start-ups are doing it these days :-(

In the end LT's value is it's content and that has been provided by it's members. I think they should have a say in what is done with the information they are providing. But this might be a lost battle, I guess.

>136 pivox:. I too have reservations about the closer ties with google. I am one of those who would not like this change to happen. However I have not attributed the same negative thoughts to Tim or LT as implied in your previous post.

The above statement "I just hope the whole purpose of LT wasn't just to buid up a substantial customer database in order to eventually sell it to google-like companies as so many start-ups are doing it these days :-(" I see as a very cynical interpretation of why Tim started LibraryThing. It also appears to view all the work done by Tim & co over the past few years as an exercise in simply tricking members to fulfil a secret plan.
I hope it wasn't your intent in posting the previous statement to imply they would behave unethically like that?

I too agree that members should have the opportunity to provide their thoughts and opinions about changes such as this.
I interpreted the blog post by Tim raising these questions and asking for comments as a conduit for that to happen.
In line with all the work he and the team do every month in making members aware of upcoming changes and in actively asking for feedback.

--> 136 (edited to correct this line)

However I have not attributed the same negative thoughts to Tim or LT as implied in your previous post.

My thoughts of Tim and his work are in no way meant to be negative. I only have positive thoughts and admiration of Tim's creation and work here at LT. I'm simply trying to be realistic.

There are several ideas that run around in my head. One is that it seems almost impossible to fight the battle with internet cyber-behemoths when one has a such a beautiful and idealistic style as Tim and company.

Second is that Tim and company must ensure all of their work remain profitable considering the more staff he hires and resources that are put in place with every new feature added. I believe Tim's ultimate goal is to make public libraries open to all (if I'm reading him correctly). These libraries are slow in coming to him. He would also like to take advantage of resources of Google- and/or Amazon-like companies because they already have some good resources in place. He'd like to share resources with them, not sell out to them nor be "swallowed" up by them.

In the end LT's value is it's content and that has been provided by it's members. I think they should have a say in what is done with the information they are providing.

Just because LT content has been provided in great part by the membership does not in turn *guarantee* the members' right to control what happens to LT. We do all our work here as volunteers because we love what Tim and company are doing. In turn, Tim asks for our input into features we want and how much he should integrate with other internet commercial sites.

I have faith in Tim's work here and find the process of openness he uses with LT's development absolutely fascinating. I've never seen such a thing before with any previous internet group in which I've participated. That was one of the reasons I stopped by for a look and immediately bought a lifetime membership. I've learned so much here...with the latest opportunity having been the open invitation for all members to jump right in to learn and work a wiki. It was fun to do!

Sorry! Didn't mean to make such a long post...

believe Tim's ultimate goal is to make public libraries open to all (if I'm reading him correctly).

Public library collections are almost all open to all now, as you put it; go to any library website for small city and up (small towns not quite yet) and you can search their OPACs. Or do you mean something different?

--> 139

You're right. They are almost all open. Tim wants individual libraries to use the LT database within their own library systems. He wants LT to be the premiere place for libraries to seek additional book information to complement their own catalogues.

Check out the Thingology blog (See entry for Sept. 4, 2007, for example) for three new libraries he just added this month.

Interesting to notice the absence of any substantial comment by Tim about all the privacy worries overwhelmingly expressed in this blog. Doesn't he care?

If I didn't care, I wouldn't have raised the issue so clearly, or so starkly. I was also clear that I can't talk back about this in depth. You can imagine why this might be case. (Recent dental surgery?)

I'm not ignoring concerns. Not at all. But I do think SOMEof the expressed worries come from a misreading of the idea. If LibraryThing allowed members to seach across their library with Google, it would be ENTIRELY voluntary. You'd need a Google account. And you'd need to specifically both sites to create the synching.

is Google's $$$$$$ check just too good to take members worries into account?

I didn't say anything about money either. Do you really think Google is going to pay LibraryThing for this?

On libraries and openness, I've been a champion of open data for some time. (Google LT and OCLC, for example.) In some sense open data is good for LibraryThing, although it would probably easier to keep quiet--you can be damn sure OCLC won't be dealing with us any time soon; I morphed their logo into the Death Star! More generally, having a company whose interest is aligned with openness is not a bug, it's a feature.

The GBSS bookmarklet was designed to collect a piece of data that Google isn't making public, that LibraryThing as well as libraries would love to have their hands on. I'm worried that Google is building up a secret unique identifier--something totally new to the book world.

Read the Thingology blog for more thoughts on that.

>141 timspalding:

"Do you really think Google is going to pay LibraryThing for this?"

YES! Don't you think the LT mailing list is worth a pretty penny? You're doing their selling for them by widely discussing this brand new product of theirs with an avid reading and book-loving audience who are probably at the top of their target audience priority list. Squeeze them till the pips squeak!

--> 141

Thanks for the explanation about the unique identifiers. I had no idea *that* was the problem until I read your last Thingology blog earlier today. I can see why we cannot allow this to happen after reading the example (i.e. the Harvard students collecting ISBNs from the campus book store) about which you wrote.

My daughter sells her campus textbooks online and buys what she can on-line before school starts each semester. Perhaps most Harvard parents can afford to have their kids buy only from the campus bookstore, but since we (my daughter, husband, and I) are paying out-of-pocket for my daughter's college education and associated expenses at a fairly pricey state university, any little amount of money she can save us truly helps.

Don't you think the LT mailing list is worth a pretty penny?

Um, there is no LT mailing list. Signup only requires a username and password. Google Book Search is also not super new anymore, and has already gotten a hell of a lot more publicity than LibraryThing could give it.

Plus, Tim and other users spend a lot of time on Talk talking about various competitors. It's not usually very flattering and that didn't change when people started talking about Google Books's My Library.

Well, I don't think GBS ML has any uptake so far. Maybe it will someday.

>Um, there is no LT mailing list.

We should offer to sell people "all the demographic information we have on users." Also, I promise to sell whatever it is I know about cold fusion.

146: Oh, I'll buy! Cold fusion... interesting. lol

141: up to this point I hadn't read any comment on your part whether this linking with a google acount would be opt-in or not (even though users had commented largely on that subject). So happy to hear it will be just an opt-in.

Google will always want to get more user information, as far as they can and are allowed to. So if they are able to link LT users and thus their reading habbits ("tell me what you read and I'll tell you who you are") to a google account they will happily do so. And if not, well all the better for those who ARE concerned about their privacy.

Anyway, just for having you taking position on some of the worries expressed in Talk and Blog, my overdoing it a bit was way worth it, I think.