Ideas for using the date fields
DiscussionsMore Power to the Date Fields!
Rejoignez LibraryThing pour poster.
Ce sujet est actuellement indiqué comme "en sommeil"—le dernier message date de plus de 90 jours. Vous pouvez le réveiller en postant une réponse.
1_Zoe_
I know there was a list of ideas somewhere, but I can't remember where.
Some thoughts:
1. Finish developing the Timeline
2. Show books that were Read, Started, or Acquired in the Most Recent Activity box on the profile
3. Add at least Date Acquired to the quick edit on the Add Books page
4. Create a "smart" Recently Read collection based on completion dates
5. Make the Currently Reading collection at least partially smart, so that adding a Start date automatically adds the book to CR, and adding an end date automatically removes the book from CR
Some thoughts:
1. Finish developing the Timeline
2. Show books that were Read, Started, or Acquired in the Most Recent Activity box on the profile
3. Add at least Date Acquired to the quick edit on the Add Books page
4. Create a "smart" Recently Read collection based on completion dates
5. Make the Currently Reading collection at least partially smart, so that adding a Start date automatically adds the book to CR, and adding an end date automatically removes the book from CR
2girlunderglass
I'm mostly interested in numbers 1,2, and 5. The timeline would be great - it's great already but I never remember where to go to find it in its "unofficial" state so I end up spending much more time looking for it than looking at it.
A smart currently reading is a big big big thing for me and mostly the reason I joined this group. Unlike you, however, I am not interested in books getting added automatically to CR when we enter dates, but the opposite. I would say it's not much hassle to just change the collection for a book to CR, but it IS more of a of hassle to have to go to "Edit Book" and add the start an end dates every time. So it would be great if we had the opposite: when we add a book to CR it automatically adds a start date and when we remove it and end date. (Some people have pointed out that removing a book from CR doesn't necessarily mean they've finished it - and to them I say: this could be easily solved with a pop-up. Say, when you remove a book from CR immediately a pop-up appears asking you if you have finished reading the book. YES would auto-enter the end-date, and NO would not enter an end-date , possibly erasing the start date as well?)
A smart currently reading is a big big big thing for me and mostly the reason I joined this group. Unlike you, however, I am not interested in books getting added automatically to CR when we enter dates, but the opposite. I would say it's not much hassle to just change the collection for a book to CR, but it IS more of a of hassle to have to go to "Edit Book" and add the start an end dates every time. So it would be great if we had the opposite: when we add a book to CR it automatically adds a start date and when we remove it and end date. (Some people have pointed out that removing a book from CR doesn't necessarily mean they've finished it - and to them I say: this could be easily solved with a pop-up. Say, when you remove a book from CR immediately a pop-up appears asking you if you have finished reading the book. YES would auto-enter the end-date, and NO would not enter an end-date , possibly erasing the start date as well?)
3_Zoe_
I think CR was actually supposed to work like that initially--it did in the beta--so hopefully Chris will add it back eventually.
Maybe the ideal situation would be an option that users could set: I'd like the date fields to influence CR, and you'd like CR to influence the date fields, so why not let us both choose to have it work the way we want? My main problem with using CR alone is that if I read a book quickly, it never gets into CR at all, whereas I can come back and fill in the dates later.
I also thought of another thing I'd like for the date fields:
6. The ability to search by date. This would be especially useful for group catalogues, where it's not possible to see the whole catalogue and just sort it to show recently-finished books.
Maybe the ideal situation would be an option that users could set: I'd like the date fields to influence CR, and you'd like CR to influence the date fields, so why not let us both choose to have it work the way we want? My main problem with using CR alone is that if I read a book quickly, it never gets into CR at all, whereas I can come back and fill in the dates later.
I also thought of another thing I'd like for the date fields:
6. The ability to search by date. This would be especially useful for group catalogues, where it's not possible to see the whole catalogue and just sort it to show recently-finished books.
4reading_fox
Currently Reading Connections.
Who has finished reading the same books as you within 1 month or so.
Who has finished reading the same books as you within 1 month or so.
5fyrefly98
I found one of the "idea" threads:
http://www.librarything.com/topic/30084
I would like more graphs and more stats on the work pages:
- Average length of time this book takes to read across members, and maybe zeitgeist listings of fastest/slowest books
- Graphs of book acquisitions/readings by month (and year?) (both personal, and by work)
- Either searching by date or smart collections by date read and/or acquired (i.e. limit search for "fantasy" for books acquired in 2008)
- Some others that I'm not thinking of at the moment, I"m sure.
ETA: Oh, and definitely a list of members who've recently read a book.
http://www.librarything.com/topic/30084
I would like more graphs and more stats on the work pages:
- Average length of time this book takes to read across members, and maybe zeitgeist listings of fastest/slowest books
- Graphs of book acquisitions/readings by month (and year?) (both personal, and by work)
- Either searching by date or smart collections by date read and/or acquired (i.e. limit search for "fantasy" for books acquired in 2008)
- Some others that I'm not thinking of at the moment, I"m sure.
ETA: Oh, and definitely a list of members who've recently read a book.
6girlunderglass
I'd like the date fields to influence CR, and you'd like CR to influence the date fields, so why not let us both choose to have it work the way we want?
Okay, if you insist :) *singing "My Pony Is Better Than Yours", folk ballad passed on by LT ancestors long ago*
Average length of time this book takes to read across members
Yes!
Okay, if you insist :) *singing "My Pony Is Better Than Yours", folk ballad passed on by LT ancestors long ago*
Average length of time this book takes to read across members
Yes!
7_Zoe_
Thanks for finding the link to that thread!
I'd definitely like to see data about average reading time on the works pages, and the zeitgeist listing would be interesting too.
pony ballad
Ha!
I'd definitely like to see data about average reading time on the works pages, and the zeitgeist listing would be interesting too.
pony ballad
Ha!
8VictoriaPL
I think these are all interesting. I don't utilize date started because I have a nasty habit of abandoning books halfway through. But since the start of 2009 I've filled in date completed.
I've seen the current timeline module and I think it would be an excellent feature if it was polished up a bit.
I've seen the current timeline module and I think it would be an excellent feature if it was polished up a bit.
10aethercowboy
I'm not sure if this is the right thread, but I just thought of a potential gotcha with date fields...
As everybody knows, Cervantes and Shakespeare died on the same day (April 23, 1616). However, Shakespeare died 10 days later.
This is due to the Julian Calendar's (Shake) difference from the Gregorian Calendar (Cerv).
So, how will more powerful date fields handle dates that are given under these specific calendars? Will, for example, all dates be considered Gregorian by default, and then, maybe a datafield on the date that allows one to select the calendar? And then, you could set a user preference to display all days in "format x."
This philosophy could then be extended to any other calendaring system. Provided there are sufficient background hooks to take into consideration conversions, then (e.g.) devout Discordians could enter their dates as "The Aftermath 50, 3173 YOLD" and the standard user gets December 8, 2007.
As everybody knows, Cervantes and Shakespeare died on the same day (April 23, 1616). However, Shakespeare died 10 days later.
This is due to the Julian Calendar's (Shake) difference from the Gregorian Calendar (Cerv).
So, how will more powerful date fields handle dates that are given under these specific calendars? Will, for example, all dates be considered Gregorian by default, and then, maybe a datafield on the date that allows one to select the calendar? And then, you could set a user preference to display all days in "format x."
This philosophy could then be extended to any other calendaring system. Provided there are sufficient background hooks to take into consideration conversions, then (e.g.) devout Discordians could enter their dates as "The Aftermath 50, 3173 YOLD" and the standard user gets December 8, 2007.
11fyrefly98
>10 aethercowboy:
Man, and I thought I was meticulous recording my dates finished back to 2003...
So tell me, what were you currently reading on April 23rd, 1616? (Either one.) ;)
Man, and I thought I was meticulous recording my dates finished back to 2003...
So tell me, what were you currently reading on April 23rd, 1616? (Either one.) ;)
13aethercowboy
>11 fyrefly98:.
Well, it's not just the Reading-related date fields, or so I would assume.
There are Birth Date and Death Date fields on the author pages (like on the Shakespeare and Cervantes author pages). Also, other cultures still use non-Gregorian calendars, including the Julian calendar, which still has some agricultural use.
Though, yeah, I wasn't reading much back in 1616. To be honest. I took that year off to visit the New World.
Well, it's not just the Reading-related date fields, or so I would assume.
There are Birth Date and Death Date fields on the author pages (like on the Shakespeare and Cervantes author pages). Also, other cultures still use non-Gregorian calendars, including the Julian calendar, which still has some agricultural use.
Though, yeah, I wasn't reading much back in 1616. To be honest. I took that year off to visit the New World.
14_Zoe_
I hadn't actually thought about the other date fields, but I guess it makes sense to do stuff with them too (though I think LT is a bit better about that--there's already the "On this day" homepage module).
What other features should be developed for the CK dates?
What other features should be developed for the CK dates?
15fyrefly98
I wanted a CK field that was "timespan"... and so Tim added "Important events". I can see where he's coming from, sort of, but there are a lot of dates in books that don't necessarily correspond to important events.
For instance, we know that the Harry Potter books take place in 1991-1998 (because of the date on Nearly Headless Nick's deathday cake), but there's no Important Event to tie that to, and so no good way to enter it. I think it would be neat to see what books take place over my birthday, or a good way to find what books take place in the 1920s, or whatever.
For instance, we know that the Harry Potter books take place in 1991-1998 (because of the date on Nearly Headless Nick's deathday cake), but there's no Important Event to tie that to, and so no good way to enter it. I think it would be neat to see what books take place over my birthday, or a good way to find what books take place in the 1920s, or whatever.
16aethercowboy
>15 fyrefly98:.
I agree.
It would also be interesting to see how many books take place in, say, 2012, and then see how that number increases as that date approaches.
There should be a CK field for "timespan" such that you can enter it as single span, like:
1942-06-06 to 1950 (or something like that)
and then get a plus-button so you could add more fields.
This would be ideal for books like Cryptonomicon, in which half the book is WWII era, and the other half is Present-day.
I agree.
It would also be interesting to see how many books take place in, say, 2012, and then see how that number increases as that date approaches.
There should be a CK field for "timespan" such that you can enter it as single span, like:
1942-06-06 to 1950 (or something like that)
and then get a plus-button so you could add more fields.
This would be ideal for books like Cryptonomicon, in which half the book is WWII era, and the other half is Present-day.
17Helcura
>15 fyrefly98: fyrefly98
That's a great idea! I can imagine all kinds of interesting reading suggestions based off of timespan.
That's a great idea! I can imagine all kinds of interesting reading suggestions based off of timespan.
18fyrefly98
>16 aethercowboy: Oh, I like the "future" timespans... and then we get to problems like 1984. :)
>17 Helcura: I think it would be particularly interesting for global reading... say, finding historical fiction that dealt with the world in 1300s. You could find stuff for Europe via Important Events - Black Plague, Crusades, whatever - but books set in Japan or South America wouldn't deal with those Important Events, even though they might have been set in the same time span.
There are places where it would break down, obviously - The Trojan War works much better as an Important Event than it does as an exact date span. (And, if I had to guess, I'd say that Tim's classicist streak is what pushed him towards the Event model rather than the Dates model for dealing with stuff in the past.) But there are a lot of places where it wouldn't break down, and a lot of interesting uses for the resultant data.
>17 Helcura: I think it would be particularly interesting for global reading... say, finding historical fiction that dealt with the world in 1300s. You could find stuff for Europe via Important Events - Black Plague, Crusades, whatever - but books set in Japan or South America wouldn't deal with those Important Events, even though they might have been set in the same time span.
There are places where it would break down, obviously - The Trojan War works much better as an Important Event than it does as an exact date span. (And, if I had to guess, I'd say that Tim's classicist streak is what pushed him towards the Event model rather than the Dates model for dealing with stuff in the past.) But there are a lot of places where it wouldn't break down, and a lot of interesting uses for the resultant data.
19VictoriaPL
I think an awesome widget for the home page would say something like:
6 years ago on this date you read Ender's Game.
2 years ago on this date you read Harry Potter and the Goblet of Fire
6 years ago on this date you read Ender's Game.
2 years ago on this date you read Harry Potter and the Goblet of Fire
20_Zoe_
>19 VictoriaPL: Oh, I like that!
21TheoClarke
>19 VictoriaPL: That is wonderful!
23timspalding
Ha. That's a good idea.
24fyrefly98
>19 VictoriaPL: How about:
5 years ago on this date you acquired Ender's Game
...and then if it's also in your "To Read" collection, it appears:
5 years ago on this date you acquired Ender's Game, so get reading, slacker!
:)
5 years ago on this date you acquired Ender's Game
...and then if it's also in your "To Read" collection, it appears:
5 years ago on this date you acquired Ender's Game, so get reading, slacker!
:)
25_Zoe_
>24 fyrefly98: I like that ;)
>23 timspalding: I was just thinking someone should suggest that idea in RSI. It's nice to see that you're actually reading this group :)
>23 timspalding: I was just thinking someone should suggest that idea in RSI. It's nice to see that you're actually reading this group :)
26reading_fox
I'd love timespan, particularly for near future SF - see how accurate those predictions weren't.
27amarie
I work with a digital library collection that handles date spans by allowing one to enter a range (say 1850-1856) which the software than expands to 1850; 1851; 1852; 1853; 1854; 1855; 1856; This so that a search for items specifically dated 1853 will retrieve it correctly. Not sure if this is a truly reasonable way to handle the programming part, but maybe it might be useful.
>24 fyrefly98: I love the idea of a "reminder" to get reading that new book! My unofficial summer goal is to whittle down new unread books so I can get back to using the library more (those darn bookstore bargain tables!).
>24 fyrefly98: I love the idea of a "reminder" to get reading that new book! My unofficial summer goal is to whittle down new unread books so I can get back to using the library more (those darn bookstore bargain tables!).
28aethercowboy
>27 amarie:.
The problem, though, with this method, is that we already enter dates as "YYYY-MM-DD". We'd need a different token for spans, like "--" or " to " or something else that won't readily be parsed weirdly.
The only case I see it actually being an issue is when people enter date spans like "2005-06" and mean 2005-2006 and not June, 2005.
If the parser trusted that the user was using the right length date strings, the only time we'd have problems with parsing correctly is from 12 B.C. to A.D. 12 (assuming we only have 12 months).
Depending on how that data is used, it would actually seem to make more sense recording just the start and just the finish, as it would take up less space. The test to find if a particular event happens in that span would be simpler, too, as you'd just have to make sure it happened AFTER the beginning but BEFORE the end, instead of what would appear to be a for-loop checking each year. And graphing it would be a snap too, as it would be BEGINNING - ORIGIN for the start point and END - BEGINNING for the length (both multiplied by arbitrary length values representing a certain span of time).
The problem, though, with this method, is that we already enter dates as "YYYY-MM-DD". We'd need a different token for spans, like "--" or " to " or something else that won't readily be parsed weirdly.
The only case I see it actually being an issue is when people enter date spans like "2005-06" and mean 2005-2006 and not June, 2005.
If the parser trusted that the user was using the right length date strings, the only time we'd have problems with parsing correctly is from 12 B.C. to A.D. 12 (assuming we only have 12 months).
Depending on how that data is used, it would actually seem to make more sense recording just the start and just the finish, as it would take up less space. The test to find if a particular event happens in that span would be simpler, too, as you'd just have to make sure it happened AFTER the beginning but BEFORE the end, instead of what would appear to be a for-loop checking each year. And graphing it would be a snap too, as it would be BEGINNING - ORIGIN for the start point and END - BEGINNING for the length (both multiplied by arbitrary length values representing a certain span of time).
29fyrefly98
>28 aethercowboy: That's how I envisioned it working - each "line" of the CK is actually 2 fields, a beginning date of the timespan and an end date. Hitting the + would get you a new line with two more fields.
Filling in the just the beginning on any one line would indicate a point, not a span (i.e. so you could indicate that a book took place in 1969 without having to specify 06 Jan 1969 - 15 Oct 1969, if that info's not in the book.)
Filling in the just the beginning on any one line would indicate a point, not a span (i.e. so you could indicate that a book took place in 1969 without having to specify 06 Jan 1969 - 15 Oct 1969, if that info's not in the book.)
30TimSharrock
>27 amarie:
this has limitations for books that cover long time periods, eg restaurant at the end of the universe or tau zero
this has limitations for books that cover long time periods, eg restaurant at the end of the universe or tau zero