Sonny Software

I often enter multiple editions of a book in my database, so as to keep track of different printings of the "same" text. This helps to distinguish, for example, first editions from later reprints. Here's an example:

Wylie, Elinor. The Venetian Glass Nephew. New York: George H. Doran, 1925.

Wylie, Elinor. The Venetian Glass Nephew. Chicago: Academy Chicago Publishers, 1984.

Unfortunately, Bookends's method for identifying duplicate entries does not consider date of publication, publisher, or ISBN, all of which may be relevant in this context, so it considers the above entries to be duplicates. (This feature seems to have been written with journal articles in mind.)

It would be nice to be able to flag those data in a "Remove Duplicates" operation so as to minimize the number of false duplicates I have to verify manually.

TH

Hi,

Adding Publisher as an option would be easy. The others not so much, because the fields differ between reference Types, and the Remove Dups function is not Type-discriminating. Perhaps you'd like this to be added as a feature, too, although that would make things considerably more complex (and slower). But something to think about...

Jon
Sonny Software

Jon wrote:Hi,

Adding Publisher as an option would be easy. The others not so much, because the fields differ between reference Types, and the Remove Dups function is not Type-discriminating. Perhaps you'd like this to be added as a feature, too, although that would make things considerably more complex (and slower). But something to think about...

Jon
Sonny Software

Yes, I think that type discrimination is probably the ideal here. Users would, I think (I would, certainly) be willing to put up with a slower Remove Dups cycle if there were, say, a popup menu to set the type considered by Remove Dups. My intuition is that this would require a more complex interface for the function, and that might pose problems, but I think it would still be very useful. As it stands, the function is not terribly useful for me, but YMMV.

TH

tharpold wrote:
Yes, I think that type discrimination is probably the ideal here. Users would, I think (I would, certainly) be willing to put up with a slower Remove Dups cycle if there were, say, a popup menu to set the type considered by Remove Dups. My intuition is that this would require a more complex interface for the function, and that might pose problems, but I think it would still be very useful. As it stands, the function is not terribly useful for me, but YMMV.

TH

EndNote has a preference panel for duplicates in which you can select, via checkboxes, the fields you want used to compare for duplicates. My suggestion would be to duplicate that, as it's most flexible, and, any fields a user checks in there, should be then set up as indexed fields in the biblio database which would help with the speed issue, I think.

The default would be Author, Year, Title and most of us would be happy with that but we could alter it however we like this way. (BTW: It also has an option to ignore spacing and punctuation which might be good to throw in). What do you think?

I think Bookends has the same thing, and we introduced it several years before EndNote did (this is one of several cases where they did a Microsoft to our Apple -- although I have to admit it goes both ways sometimes).

Check out Remove Duplicates for yourself and see...

Jon
Sonny Software

Jon wrote:I think Bookends has the same thing, and we introduced it several years before EndNote did (this is one of several cases where they did a Microsoft to our Apple -- although I have to admit it goes both ways sometimes).

Check out Remove Duplicates for yourself and see...

I was looking for it in the wrong place -- as a preference--so I didn't find the options right away although I had seen the function in the reference menu.

Thanks,

I have the same problem as tharpold. When I enter several editions of the same book (and that usually means: same Author, title, publisher) the references only differ in publication years. So it would be easiest to include publication year/date in the options dialogue for checking duplicates

Hareiko

I think that the 'Remove Duplicates' command is hampered by the fact that only one reference can be shown at any time. So it is necessary to jump between the duplicates and to look manually for all differences. This is a waste of time, the software should do this work and display in detail the differences between the references selected in the duplicates list (maybe limited to two references). The window which lists the duplicates could get a split view with the duplicates list at the top and the detailed differences of the two first selected references at the bottom. Another software that I used some years ago had a merge feature which created a single reference from two references and kept all data. This did indeed save some copy/paste actions.

Rudi