We've gone live with Common Knowledge. Go ahead and read the blog post.

We're looking for first reactions. This is very new. And it's not something a lot of others have done. We're not sure where it's going to go. We're only sure that we're hopeful. We see problems, but also potential.
RulesIf possible, let's go for first reactions, not detailed discussions of this or that topic. Feel free to start topics on these.

Also, I've starting a topic for bug reports. I don't want to mix bug reports with reactions. Fluteflute, can you remove your bug report? It'll be fixed in a sec.

nb: I renamed this group, formerly about WikiThing, and promoted it to a standing group. The topic is going to spider out in all directions. It needs a home.

Message edited by its author, Oct 10, 2007, 5:11am.

I was waiting for an appropriate thread to appear where I could post:

It's AWESOME!! And so much fun. =)

(Except the author common knowledge, which isn't working(?))

Message edited by its author, Oct 10, 2007, 5:01am.

There are various links pointing to /fwiki rather than /commonknowledge . (Change log link on author pages is wrong. And 'female' link and 'London, England' link is wrong.)

Yeah and the author pages one doesn't want to let me edit!

Message edited by its author, Oct 10, 2007, 5:05am.

Ah...I'll check on those. Be back in a minute.

Seems to be fixed. Let me know if you find any references to fwiki. We tried to catch them all, but the code is deep.

I discovered this new feature by accident on the author page of Adelbert von Chamisso. I entered the few facts I know about him.
I wonder what we should do when several authors share an author page. Whose data should we enter?

Short answer: The most popular one, which is the one who'll keep the name when we offer perfect disambiguation. If it's dicey, leave it out.

I love it! I didn't know what to expect, but this is really great: entering all those tidbits and info that booklovers love to know. And I love clicking on an award gives a list of winners. I'm going to play with this more in the weekend.

Little question though: I have filled in some information for a favourite author of mine: Stephen King .Is the information for the Bram Stoker award correct? I didn't know an other way to include all those days. Is there a consensus on that, on how to do this?

I'd like a way to show the information entered in the brackets in the awards field. Ordering by that information would be cool, too (chronological order, etc).

Excellent!

Dates: I've assumed that these should be entered in American format. I fear there will be confusion, though.

Yes, US format dates. There was much internal discussion before the launch and there is even more discussion in this talk thread:
http://www.librarything.com/talktopic.ph...

It seems like there are only the last 100 entries shown on the award page. This makes sense bandwidth-wise, I guess, but I'd appreciate a show-all option/next page option anyway because a "last 100 added Nobel Prize winners" only is of very limited use for me... :)

12> Fixed.

That was a left over from some development limits that I put on it. It's bad news when you automate the entry of 6,000 items and then want to check that some of them made it. :)

I do like this, and appreciate its potential. I guess the creation of new fields is going to be key to its success. How does this happen? At the moment, none of the fields seem particularly relevant to my non-fiction collection.

Fields will be added by LT staff as interest in them is presented. Adding of fields is, as Tim said, trivial. It takes just a minute or so.

If you have some ideas then we'd love to hear them. Remember that they must correspond to the item you want to add them to. For instance, "Cover Designer" is not WORK related, but EDITION related, which is why it isn't on the current list of data.

Let's let this percolate for the day, and then let's start a "more fields" discussion.

For authors, it would be great to have a field to add aliases used, or real name, if the author is commonly known by their penname.

Thank you so much for this - I have been trying to get my head around this concept but now I see it 'live' the potential is beyond my ability for words. Brilliant!

The column 'when' on recent changes isn't very useful just just showing the date. Could we at least have the hour as in 'one hour ago' or 'ten hours ago' if not an exact date?

My initial reaction: oh, damn! Now I REALLY don't want to go to work today. I want to stay home and play.

>16

Sorry didn't see this. I have already started a thread "Field suggestions".

The column 'when' on recent changes isn't very useful just just showing the date. Could we at least have the hour as in 'one hour ago' or 'ten hours ago' if not an exact date?

On Chris' list. It's an hour now, btw.

I just started looking at this, it seems really fun and useful, thanks!

For important locations, I'm wondering if there is any way to differentiate between real world locations and fictional locations.

Compasses work differently, I think.

Why do you guys always have to introduce features like this when I have to go to work...

It's as great as I suspected it would be, and there seems like so much potential too. Will it be possible at some point to search across multiple fields, say, if a real-life author is also a character in another book?

Right. That's the idea. Or who are the Female Chicago mystery writers who died in Italy? (Zip, I'm guessing.)

Series?
there's been many requests for this, mostly in catalog data - but on work author pages would be good too.

Fantastic! Thanks for all the work on this -- it feels to me like a very LibraryThing thing.

This is very cool -- I won't have time to play with it for a while, but I do look forward to doing so.

One thing that bothers me (although maybe I'll get used to it) is that disambiguation notices are now in the Common Knowledge section. Until we have real author disambiguation, I think it is useful to have the disambiguation notice right at the top of the page, so everyone can see it before they read the list of books/see the author picture.

Oh wow, excellent work Chris et al. If I don't watch out, this will keep me busy for hours (once I get to work, that is).

Feature Request: It would be really useful if we could sort data on the Browse pages by clicking on column headers (Field, What, Type), as well as a link to sort by order added (new stuff at the top, I suppose).

This is probably already on your list, Chris, but there you go.

Is there an way into the common knowledge : Bowse pages without going through the works/authors. It would seem that if I want to get the list of authors born in 1942, I have to find one first. (C. J. Cherryh)

This is fab, I asked a long time ago when we could get to "tag" authors in this way, and it's here! Woot! Hurray !!!

I think we're going to need to think about what people want from an Awards list, though. I was looking at the Hugos, for example, and you can't tell who wrote the book or what year it got the award, whether an author has won it more than once and if so what for, or sort it alphabetically or chronologically. For a lot of the big awards (like the Hugos) there are websites that will tell you all that, but it would be cool not to have to dip out and back in. And if the completists amongst us start adding Hugo nominees or shortlisted contenders (which they will) it's going to get even more confusing.

I can't post a first reaction yet because I haven't even looked at it - but I'm taking the afternoon off and was already planning a couple hours with LT, so I'm THRILLED this is out today! It sounds fantastic and I can't wait to try it!!!

Fran, awards are discussed in topics http://www.librarything.com/talktopic.ph... and http://www.librarything.com/talktopic.ph...
I thought an award down approach in the wikiThing would complement the author work up approach of common knowledge, but I'm told this will handle it eventually.

Oh, dear. Just when I had decided to get this LT addiction under control. Oh, well. *sigh*
Seriously, my first reaction is "Wow." And I've barely touched the surface. Thanks.

Wow, great stuff!

Question: In the past, when we've have requested entries like these for our libraries, we're generally advised to use tags. Although these options are not very structured, they are a heck of a lot more structured than tags. Does Tim et al. see this as some sort of evolution from the tag idea?

> 36

They're really different, though, aren't they? Tags are subjective and personal, whereas the information in "Common Knowledge" is intended to be objective and for the use of the community.

Message edited by its author, Oct 10, 2007, 12:12pm.

This is fantastic, and is going to be very addictive. (I agree that a "series" field on the work pages would be very useful, considering how frequently it's been requested as a catalog field).

I asked this question in a different thread, but maybe this is the place for it.

I'm having a hard time figuring out how to navigate through everything in the new Common Knowledge pages. For example, I clicked on the link for Hugo Award Winners in the blog. I get a page that is simply a list. There are no links on this page, except to the individual award winners. No link to move forward or backward in the list - i.e. "next" or "back" (presumably these aren't the only winners?), No link to anything explaining what the Hugo Award is, and most strange to me, when I try to move up a level and visit http://www.librarything.com/commonknowle... , thinking it would take me to a list of the different awards we've got information on, I got to a "page not found" template.

I guess I want to know if easier navigation is something we can look forward to, or are things going to stay this clunky?

And before I get accused of complaining, I really love this new Common Knowledge thing, I'm just trying to figure out how best to utilize it & navigate around. It seems intuitively as if things should be more interconnected than they are now, but perhaps I'm misunderstanding and that's still to come?

40> The list of Hugo Awards winners contains all that have been added by LT users (that might be an incomplete listing, of course), but it really shows all the information that is available to LT.
It used to be only the last 100 added, but it seems that changed.

I agree, it could be made much more useful... (and addictive?) :)

It's cool. I want to see the code. This is a mediawiki back-end?

... Is the "links" going to go into CK, or back into structured fields?

42, per the blog post (and a few other mentions) the back end is infogami, not mediawiki. And thus insanely cool.

#43, doh, thanks! i'm a little spacey today.

A couple of suggestions, and sorry if they're repeats - there's a lot of posting on this today:

(1) When you click on a CK "tag" it takes you to a "browse" page. It would be nice if you could sort by column heads. For example,
nebula CK page could permit sorting by "field", "what", "type". (Yep, repeat suggestion: see #30 and #47.)

(2) It would be sweet if we could actually click on "Common Knowledge" and "Browse" in the breadcrumbs trail ... I'd like to browse. (Thanks #45 - I did find that. But I was thinking that it would be useful if the breadcrumbs trail that shows up on CK pages actually linked there, instead of just plain text.)

(3) It would be also good if we could edit the page created, to add basic information about the item.

(4) Merging & redirecting ?

Message edited by its author, Oct 10, 2007, 2:40pm.

Message 44, #1

Not sure if this is what you're looking for, but try: http://www.librarything.com/commonknowle...

Message edited by its author, Oct 10, 2007, 2:01pm.

Pretty cool, but it seems biased toward fiction and contemporary books/authors. I'd like to see more options for non-fiction and older books (e.g. fields like, I dunno, key concepts used/discussed, historical events discussed, etc.). I'd also like to see -- and maybe this would be a good candidate for Common Knowledge -- something searchable for Tables of Contents, especially for edited books with multiple contributors. Also useful would be a field for links to online versions of the work (e.g. at Project Gutenberg and other sites). Should I repost this on the fields thread?

And I agree with others about wanting more specifics on the award pages - authors, works, and dates won.

>45 I think that lquilter is looking for the same thing I mentioned in message 30.

On a browse page (such as this one for London, England) it would be nice if we could sort it on the Field column (so that all places of residence are listed together) or on the What column (so that the works, authors, etc. are listed alphabetically), etc.

Elsewhere (here), Chris has mentioned that they may eventually add the ability to "drill down" into grouped data (based on parenthetical subgroupings), but that doesn't rule out the usefulness of a simple sort.

Another quick thought on the Browse page:

Is there really a need for the Field and Type to be presented as two different columns? They could be combined into one (the contents could contain something like "work : important places" and "author : places of residence"). This would simplify sorting (one less sort to write). Also, the Type provides the context for the Field, so having them together makes sense.

This would also free up space for another column -- perhaps one containing subgroupings or maybe the parenthetical information that was stripped out -- e.g., "(shortlist 1971)"

dunno, just an idea.

Wow, thanks for all of the comments and feedback. It's great to see everyone so excited by Common Knowledge.

The backend code is not infogami, it's a custom solution designed by all of us and written by myself. It has some similarities to infogami, but it is all new code.

The browsability of the data will increase drastically. Basically, we wanted to see how people were going to use the tool before we over-designed it. Well, really we are never sure IF people are going to use a tool. :)

As for new fields in CK...yes, many will be added.

Sorry, Chris! I misremembered the infogami connection -- even bigger thanks/congrats since it's all you.

May I suggest that the disambiguation notice on the author pages be moved back up to the top where it can be immediately seen?

Also, I foresee much back-ing and forth-ing on field entries where authors share a name. Whose information will go in the fields? I'm hoping you guys have something up your sleeves. (You usually do!)

Tim has said that the most popular author should win out and if there are two close contenders to not fill it in for now. I'm not sure how well that will go over though.

Tim, I think this problem raises the priority of implementing a solution for multiple authors with the same name.

Sorry, this is a bit late. Much to do around here...

I'm having a hard time figuring out how to navigate through everything in the new Common Knowledge pages

It would be nice if you could sort by column heads

We know. We need to make it better. I just woke up, and have to take a flight to Columbus, OH. I'm sure when Chris wakes up he'll dive into it too.

It's cool. I want to see the code. This is a mediawiki back-end?

No, we thought about it, but we went with something totally new. We're not releasing the code, mostly because it would be a pain and a security risk more than anything. Maybe we'll blog about the principles involved.

42, per the blog post (and a few other mentions) the back end is infogami, not mediawiki. And thus insanely cool.

No, no. Open Library uses Infogami. We're using "Thingami."

Pretty cool, but it seems biased toward fiction and contemporary books/authors. ... (fields proposed)

Go to here to suggest fields.

1. Disambiguation notices to where they were, with a link to edit them on the CK page.

Tim, I think this problem raises the priority of implementing a solution for multiple authors with the same name.

Agreed.

Message edited by its author, Oct 10, 2007, 4:12pm.

I've upgraded the Browse page a bit so that it has more information. This is just a stop-gap measure until I can get more (read: real) navigation and cross-linking in the browse pages.

The idea of browsing and cross linking is at the heart of this entire idea and it's the one thing that I'd like to really explore. The only reason it is sparse at the moment is because we knew the data format would be in flux at the very beginning. We'll adapt and make improvements as quickly as we can.

The idea of browsing and cross linking is at the heart of this entire idea and it's the one thing that I'd like to really explore. The only reason it is sparse at the moment is because we knew the data format would be in flux at the very beginning. We'll adapt and make improvements as quickly as we can.

That's so what I wanted to hear! (Because I wasn't addicted enough to LT...) Thanks for all the hard work, guys :D

You've just taken a cool on-line cataloger and made it an even greater time-sink. This is addicting and fun to play with. And will be so much more fun when more data have been added by all us busy little beavers.

This is the kind of thing I'd hoped I could do when I originally signed up for LT. Great job, guys!

>56, My thoughts exactly. Now all of us can show off our obscure knowledge in scary detail. I love it on the author pages, but I worry it makes the work pages cluttered. I may change my opinion on this though. Incidentally, I love the feature to add links to an author page. Is this new or did I just never see it?

I've only played with a few books and one author and thought this very cool.

But I have a lot of anthologies, many with different authors. The CK doesn't let me put the characters in with the related stories, etc.

That is a fix I am hoping for someday, a good way to list anthologies...

Oh...and since this is the general topic for CK:

!!!! Do NOT surround your entries with quotes. The samples are given that way because we had to have some way of separating them and many of them have commas in them.

!!!! DO use the plus button (+) to add multiple items instead of putting them on one line.

If you put a bunch of entries on one line separated by commas then your work is in vain because it will be removed by somebody else doing it correctly.

#60 Chris, is it ok to spell out state names instead of using the two letter abbreviation? Heck, even as a US resident I get confused with MA, MD, ME, MI, MN, MO, MS, and MT.

Message edited by its author, Oct 11, 2007, 12:41am.

I don't think that there is a care either way from our point of view. The community will reach a consensus and we'll go with that. The two things that I mentioned in #60 are particularly bad because they break the whole system by globbing information together or by not allowing it to be grouped with other identical information.

The state abbreviations have the potential to do that, but I think that a common ground will be reached. In fact, the format of Portland, ME, USA is already becoming fairly standard in CK data.

60> Chris: The examples with quotes are extremely misleading! But why don't you just strip off any quotes entered by mistake?

60> I'm guilty of using the examples as the way things are to be entered!

Thanks for adding the instruction line and fixing what I entered!

P.S. I still would like a way to enter anthologies and I still miss "See Yours".

edit: I see how to do the "see yours" by clicking on the bottom book number as someone said in another thread. Cool...

Message edited by its author, Oct 11, 2007, 7:29am.

63: I might just do that (at least beginning and ending quotes).

64: No problem. You weren't the only one. We'll work through these little problems in the first week or so and maybe we can remove the little warning box in CK.

I am OBSESSED with this new feature! It is such a fun time-suck! I could barely tear myself away last night to sleep!

Looks interesting. Could we add a pseudonym field?

The problem with a pseudonym field is that the name of the author will always be the name that is in the author field of the majority of his books, which will almost always be the pseudonym. So what you really want is an Actual Name field.

However, it could come out the other way and the author on the page is the real name in which case an Actual Name field won't make sense. I suspect that pseudonyms will have to be factored into the new Author system that is being contemplated.

Potentially awesome. I have three requests and a question, all pertaining to work CK pages:

Request 1. A field for first-edition publication information. I'd like to know that Catcher in the Rye was first published by Little, Brown in 1951.

Question 1. I think this is a bug, actually. On the character page for Holden Caulfield (http://www.librarything.com/commonknowle...), two works are listed, both Catcher in the Rye. The two works--numbers 951 and 4053418, oddly--seem to be one and the same in some respects (they both have the same number of owners) and different in others (I only own the latter; one is listed on the character page as "Catcher in the Rye, The" and the other as "The catcher in the rye").

Request 2. Touchstones for characters, awards, all the new fields. Maybe wiki-style, like--pretend the parentheses are square brackets--(aw:Hugo) would show up, linked to the Hugo Award page, as "Hugo"; (ch:Holden Caulfield), (lo:New York), etc.

Request 3. Discussion pages for each work, book, etc., so disputes over, say, whether Maurice should be listed under Characters in Catcher can be easily accessed from the work's page.

69: Good stuff there.
r1- We are looking at this as a possibility. There is a whole thread on Talk about possible new fields:
http://www.librarything.com/talktopic.ph...

q1 - Tim will have to answer you about that. He's more in tune with works and how they get melded together than I am. I invariably put my foot in my mouth when talking of such things.

r2 - That's an interesting idea. Although it would probably just be a single pseudo link like (ck: Hugo) that would send you to a page that browses all items pertaining to Hugo (that's the beauty in the system, cross-linking)

r3 - Also another interesting idea. There are technical issues to be worked out there (we don't want to have the Talk thread for a book/author/whatever existing before it is needed, for example). But it is an interesting idea.

I'd like to take these suggestions a little further, by requesting that we can put touchstones IN the CK fields.

For example, I think someone in another thread referred to connections between this author and others. If that could point to the other authors record (if he/she existed) that would be much more powerful, and could optionally be semi-automatically reciprocated.

Someone has also mentioned the possibility of a "books mentioned in this book" field, for which something similar would be required.

However, could I strongly encourage a slightly different strategy, so we don't end up with a zillion fields. I think some of these peoples should be done like like the semantic web. So instead of having several different fields (place of birth, place of death, place of marriage, place particularly significant in authors life, etc.), we should just have one: Places (though this, could of course, be added several times). But for this field we could have a dropdown menu which described the significance of the place, with all the previous options. This is much tidier in my view, and just as powerful.

You could do the same for people, and have a dropdown for spouse, child, parent, other family member, student, teacher, editor, publisher, friend. You'd probably need to add a freetext field in addition, to allow extra description, but that wouldn't matter. It would give the ultimate in flexibility and freedom.

I'm going to disagree here. My hometown, Cambridge, MA, where I went to school, Georgetown, where Bilbo lived, Bag-end, and where the ring was, in his pocket, are all "places." But thinking about them at this level of abstraction is not natural, it's artificial.

While it might make sense for LibraryThing to store data in the format you suggest, asking users to work like the semantic web (which, every year, is going to happen next year) foolish. It relates quite directly to some of the basic questions LibraryThing raises about classification in general. Library Science has spent a long time coming up with perfect, squared-away list of subjects. Tagging blows that away not by being more perfect, but by being less. If controlling the mess gets you too far from normal, first-instinct thought, you lose.

In this and all things it should be remembered that the average LibraryThing user is not a computer person, or a librarian. We are very diverse in age and computer experience. They're not all young, hip, computer people.

For pete's sake, one of our most active—and helpful—users is using IE5.5.

Message edited by its author, Oct 12, 2007, 8:08am.

I wasn't trying to suggest that CK should become complex at all. I want to make it simpler by reducing the number of fields.

I don't see what's difficult about entering a place in a text field, then selecting "Birthplace" from a dropdown menu. That's simple.

In my view, whilst I see the possibilities with the current system, it's just too limited. Why, as a user, should I want to see all the relationships with my home town - whether they be authors who lived or died there, books who mentioned the town, or books who were published there, or even books that were written there. That's not artificial, that's entirely natural.

But as CK stands at the moment, it doesn't look as though that can ever happen. The data is too compartmentalised. I'm suggesting you free the data up by reducing the fields to generic things like people, places, dates.

Message edited by its author, Oct 12, 2007, 9:30pm.

I'm not sure I understand why you think that the example you gave could not happen here on LT?

For instance I can look for a town where I've lived before and it will return books that are about that town, authors who have lived in that town, etc. Complete with the symantic relationship between the data and the subject (a place the author has lived) Example:
http://www.librarything.com/commonknowle...

Is that not what you were looking for? Or am I misunderstanding you?

I don't want to sound too negative. I'm sure you've bashed this through with your team many times. Our philosophies of data are obviously not the same, and LT would not have got where it did if your philosophy was like mine!

But, I just can't see how this data can be manipulated if the relationships are not clearly defined. Fields that are well defined (e.g. date of birth) it's straight-forward. I can see that when the data set is full enough, you could show me aggregated data of when my favourite authors lived, or which authors were born at particular times, or whose birthday is it today.

But what can you do with badly defined data, except display it? Are you really going to have ten fields saying : Spouse, Friend, Student, Lecturer, Child, Parent, Other relative, Secretary, Researcher, Lover/Mistress. You can't, so either the data is left out, or its just badly defined as People, or something similar. I'm suggesting allowing the user to enter any people who have any kind of relationship with an author. That's nice an messy - you should like it! But I'm also suggesting that they get the opportunity to describe that relationship. Then you have messy data that it's at least possible to keep tidy.

Your example is interesting. The URL says you're searching for education places, but the page shows up all places. Does that mean the wiki is searching through every single field, or is the some mapping going on in the background so that it knows that certain fields are place fields?

You can already do this, to a point. Use the parenthetical syntax that we've provided.

If we add an important people field to authors:
Ashley Holland (spouse)
Dale Holland (father)

We're not throwing that parenthetical information away. We're just not using it yet. And if the community decides to use hierarchical paths in the parenthetical then all the better for them. And remember that you can have duplicates.

Ashley Holland (spouse:wife)
Ashley Holland (student)
or
Ashley Holland (spouse:wife|student)
or...or...or...

I think that there are ways of using CK that get what you want. The idea is that we wanted to provide a TOOL but not provide too many rules and not over-design it because we knew it would change. It is absolutely impossible for us to imagine all of the ways that people will use the tools that we give them (or if they will use them at all...we can't forget that aspect of the development of CK. We weren't sure it would be as big of a hit as it has. We hoped, but we weren't sure) The community has much power in this process.

But what can you do with badly defined data, except display it?

In a sense, that's true of Wikipedia. You can search it, but it's not semantically formatted for much else. By virtue of being fielded—not big gob, like wikipedia—Common Knowledge can do more. It can't do everything, but it can do something. And the lack of structure was key to Wikipedia's success. I think it's why CK is doing so well too. We're around 1,000 contributors now. That's reaching deep down into the regular people, I think.

As I see it, data has two elements. Structure is one element. The other is where you're going to get it. As a computer professional, it's tempting to pay a lot of attention to the first, and end up with no data for your perfect schema.

I've seen this happen over and over again in user-content software. It's so tempting to want that data, and want it perfect. That's a trap.

There was an early competitor—ConnectViaBooks—that made you fill out more than a page of detailed personal questions before you could use the site. If everyone had done it, it would have been cool and they might have become LibraryThing, or more But nobody wanted to go through that trouble, so they became nothing.

Message edited by its author, Oct 13, 2007, 1:16am.

A few suggestions to CK and the new author page:

1. CK on the author page is very good. But on the author page, I wish there was a link to jump directly to Users with books by X. Maybe the text "318 LibraryThing users " could be made into a link?

2. I am a bit less enthusiastic about CK for works. Who cares if a book is about Sara, John and Paul...? And most books will never have any CK anyway. The book descriptions are much much more interesting, but in order to see them one has to scroll down past a number of empty fields everytime. Yes, I know I will be able to change this when the page can be customized. But still I think you/we over-emphasize CK, just because it's brand new...

3. Finally, I really wish the words Members, Reviews and Conversations in the top line would be links. Yes, I know there are links in the column to the left. But is is kind of strange to read about a thing in one place, and then hunt for the place to click in another place.

I agree that Common Knowledge for works is way over-emphasized. I hope that's just because it's new. I know we'll eventually be able to rearrange things, but the default should still be as good as possible.

Also, there's way too much ugly CK editing stuff visible all the time. There should be one unobtrusive edit button in the top bar that brings up the editing instructions and the pencils beside each entry.

On the author pages, CK moves the ratings too far down. I think you should make a narrower rating box and put it under the author picture, since there's a lot of wasted space there.

> 79, 80

I beg to differ: I find CK for works enormously exciting, and can think (just off the top of my head) of a half-dozen kinds of questions that a mature CK-for-works could answer. For example . . .

1) Tracking historical figures' appearances across biography, non-fiction, and fiction

2) Tracking the comings and goings of supporting characters across a long fictional series . . . or cameo appearances by one author's characters in other authors' works.

3) Finding non-fiction references to (or fiction set in) a particular place.

And that's just with the existing fields. Add some CK/Works fields geared more specifically to non-fiction (era? organization?) and things could get really fun.

Long live CK for works! :-)

I'm torn as to the positioning of it. It may fall, although I don't want it to fall more than one slot, below reviews. Zoe is right about the chart-junk.

> 81

I agree that it has the potential to be really interesting in certain cases. But I think that most people would rather see the book description before the list of characters. The lists just take up so much space for the amount of information they actually provide, and it's not information that you're likely to want every time you look at the book.

1 and 3 can already be accomplished via tagmash. 2 might be interesting sometimes, but again, it would only be applicable to a small minority of works a small minority of the time. I think things that are likely to be used much more frequently, like book description, should have higher priority.

>81
The list of characters can be a spoiler issue, too:
Message 64.

Concerning "forked" data, I would like to point out that the forking should not distinguish between the language of the user interface but the language of the book. I am most active on library.de, nevertheless I own a lot of (original) english (french...) books as well as translations. I would prefer to have the CK in the language of my individual copy. This would make contributing a lot easier too by using the names that are actually provided in the copy one owns.

>83 I don't like using tags because it mushes a bunch of information that becomes meaningless in one blob. Whereas with Common Knowledge, there is some structure to the data. I don't go looking in tags for "Darkover" for instance. Whereas, I would do it in Common Knowledge.

I think things that are likely to be used much more frequently, like book description, should have higher priority.

I agree with Morphidae in 86, but I also disagree with this assumption. You can get a book description anywhere - Amazon, wikipedia, etc etc. But where else can you find out exactly which Jane Austen characters appear in the Thursday Next books? I would think book description would be the least popular field in common knowledge since it's the least unique and exciting.

I also think the book description should be up higher. But then Tim showed me the length of some of those book descriptions! Uh, no. I don't need 10 pages worth.

A suggestion...

Being that some LT members can be quite verbose in thier descriptions, I'd like to see all entries as beginning snippets ended by "more...". I personally tend to avoid reading extensively long reviews or synposes.

I agree with SqueakyChu that the descriptions shouldn't be visible in their entirety, but I still think they should be more prominent.

I would think book description would be the least popular field in common knowledge since it's the least unique and exciting.

I guess we just see the purpose of LT differently, so this discussion probably isn't going to get anywhere. But I'd ultimately like to have the book information I use most often available on LT without having to go to another site. Yes, it would be fun to find out which Jane Austen characters appear in the Thursday Next books, but that doesn't apply equally to all books. In general, if I'm looking at a book I haven't read on LT, I want to find out what it's about and whether people liked it (description and ratings). If I have read a book, I probably want to see who else has it and what they thought of it. I'm not denying that Common Knowledge can be exciting, I just think it's not likely to be what people are looking for most frequently on the works page.

I might not mind the placement of CK so much if it made more efficient use of the space. I'd rather see the place lists and character lists going horizontally, for example. The current set-up looks like it should be edit mode, but once the information has been entered it could be displayed better. Even just reducing the spacing between the lines would be better, and empty fields shouldn't be displayed.

I don't like using tags because it mushes a bunch of information that becomes meaningless in one blob. Whereas with Common Knowledge, there is some structure to the data.

Maybe I'm just being dense, but I don't understand this at all. I think that tagmash is much more structured because the results are ranked and because you can refine what you're looking for. So when "tracking historical figures' appearances across biography, non-fiction, and fiction" the ones where the historical figure is a main character would probably show up above the ones where he's a secondary character, and you can specify whether you want biography, non-fiction, or fiction. It's CK where I feel like everything is just mushed together.

> 90 If you look at a tag listing for a book you are unfamiliar with, how do you know which items are places? characters? You don't. All the information for a book is in one "paragraph."

But I'm not trying to say one thing is better than another, only that I have a strong preference for having categorized lists over blocks of comma-delimited text. YMMV.

Message edited by its author, Oct 14, 2007, 4:29pm.

Oh, I see what you mean now, thanks for clarifying. I was thinking of instances where you knew what data you wanted (e.g., fiction set in London) and were trying to find books, rather than looking at a particular book and wanting data.

> 90-92

I agree with Morphidae's preference for a list over a block of text (and second YMMV), but there's also another issue for me . . .

Tag Mash will only reveal what people have tagged. Main characters and principal settings are likely to be tagged, but minor characters and settings that figure briefly or peripherally in the story are far less likely. Common Knowledge seems to me to actively invite the recording of such information by detail-mad people like me. I suspect (but can't yet prove) that CK will in time yield a much more "fine-grained" picture of what's in a given book.

Concrete example: Future president William Henry Harrison is a fairly significant character in Orson Scott Card's fantasy novel Red Prophet. 871 people have it in their libraries, but none of them have tagged it "William Henry Harrison" or any variation. I'm sure, however, that when some Orson Scott Card fan gets around to doing the CK works page for it, Harrison will be listed.

Concrete example #2: Henry Giroux's The Mouse That Roared (about the Walt Disney Corporation) contains extended analyses of the films "Pretty Woman" and "Good Morning, Vietnam." Neither work shows up in the tags for the book (nor would I expect them to), but both would likely show up in a hypothetical "Other Creative Works Discussed" field in CK.

I concur with _Zoe_ that there are page-layout issues involved here (intrigue the casual visitor looking for a description, but don't overwhelm them). I want to stress, though, that for me (and, I suspect, for others), CK for works is not just detail-for-detail's sake. It represents "value added" on a potentially breathtaking scale . . . something I've dreamed about ever since I read about a (fictional) homebrew version of something similar in Charles de Lint's Moonheart 25 years ago.

>64 and 81, I had the same realization about revealing spoilers by listing characters. I decided to avoid doing anything that could be construed as spoilerish even if doing so means there is an incomplete list of characters.

>94 As I was listing characters for a book I started to put deceased after a characters name but in the end didn't. Reason: There is quite a bit of discussion about this character before another one discovers his body. So, even though he is dead before the beginning of the novel I decided against mentioning that.

> 64, 81, 94

The ultimate answer to the spoiler question (if there is such a thing) may depend on how Common Knowledge--and our/Tim's conception of it--evolves. If it's to be "like an online conversation" there may continue to be a strong presumption against spoilers. If it's to be "like an online reference source," a presumption in favor of completeness may come to seem more important (in most of the print references on books & films that I use, "spoiler" is a null concept).

Just a thought.

96: Yes,avoiding spoilers defeats much of the purpose of using this as a reference. If I am obsessed with a person/character and want to read everything they appear in, books in which they are spoilers would be the ones I was least likely to hear about any other way than CK. Take mysteries, for example. Say I'm into a particular villain. But whenever the villain is the culprit in a particular book, you don't find out until the end when the mystery is solved. So it would always be a "spoiler" to have him as a listed character. But you might really want to know all the mysteries where he appears.

I suspect that it would be a losing battle to keep spoiler information out of CK since if you leave it out someone else is sure to put it in. If you don't want to be spoiled you probably shouldn't check out CK.

Which leads me to think that CK probably shouldn't be embedded into the work page the way it is, but you should have to follow a link from the work page to get to it (also solves the problem of it getting larger and larger as more fields are added and they get filled in more completely so that it overwhelms the work page).

Which leads me to think that CK probably shouldn't be embedded into the work page the way it is, but you should have to follow a link from the work page to get to it

Yep, this is exactly what I was thinking.

Incidentally, in the earlier comments about how it shouldn't be so prominent, I was totally confused for several days because I had no idea CK was on the main work page - I never scroll down and rarely view the "main" page. Having it available only via the link on the left seems fine to me.

"Which leads me to think that CK probably shouldn't be embedded into the work page the way it is, but you should have to follow a link from the work page to get to it"

Agree 100%. I find it distracting and annoying to have CK come up on the main page which is not even a work page really, not for editing at least, so that I'll be looking for a review or description and have to go PAST all these blank lines of workspace. It's a totally different kind of thing and doesn't belong there. So, that's my first response: I don't like it because it is obtrusive. It might be okay, it probably has its uses, but its current location is such a major drawback that I would rather have it gone altogether than keep it where it is.

Even down at the very bottom of the main page would be a major improvement over the present situation.

100: See message 3
http://www.librarything.com/talktopic.ph...
Things will change, but the placement was important at first.

Hey, we are having an anniversary!

Yay. Go us!

Just 1.6 million facts later. Congratulations to all of the members for making it a success. I'm proud to have worked on it.

.... as long as we're talking about CK, could the CK clouds be regenerated? I only know they're out of date because some -- like publisher's editor -- display publisher names very big, but when I click to go fix them, they are already fixed (and have been for a while).

They should be regenerating every night. I'll have to check on that and see what's up.

I just found this thread again. I hadn't read from it since Oct. 13, 2007!

So I wonder if the CK clouds got regenerated? Maybe the problem is that some of the clouds get missed in the regeneration. The ones I look at are "blurbers" and "publisher's editors". Both fields include -- and have for a long time -- non-people (e.g., "Booklist") that in fact are not listed for any books.

Well, since we finally pushed out the new Get This Book, Local Books, and Quick Links features I can get back to tracking down a few of these other problems. This is on the top of my list.

Do you think it would be possible to add some kind of disclaimer to the CK section on the work page, explaining that these fields are not personal data but common data? Since it's on the work page, it might seem to new members as if they're only entering data for their own purposes, rather than data that's actually a fielded wiki.

Another point - would it be possible to have a way to split long award pages over several webpages, and/or to filter them by year?

It does hide very large lists—only the first five, I think.

I mean the award page (not the CK section of the work page). For instance:
http://www.librarything.com/bookaward/Ne...
http://www.librarything.com/bookaward/Ne...