This forum is read-only - Please use our new Citavi Forum instead.
Dieses Forum dient nur zur Recherche - Bitte nutzen Sie das neue Citavi Forum für Ihre Fragen.

Citavi 5: expected but (apparently) missing features

Download and discuss pre-release versions of Citavi.

Citavi 5: expected but (apparently) missing features

Postby alainsr » 2014-12-05 15:53

Hello.

First, I have to say I was a bit afraid when Citavi 5 started opening my Citavi 4 project. I really hope it simply copied my project, not changing it in any way. Did it?

That being asked, I would like to thank you for the a great software and suggest a few features I really miss:

1. Missing fields: many styles (such as NP-405 for Portugal) require the user to use custom fields or adapt other ones (such as page count). Since we're moving towards a new project structure, I guess there's no reason for not improving fields. Fields for journal publisher and place are what I immediately think of, but if given chance all users will give you a hand plenty of them, and this should be considered.

2. This leads us to ISSN: if there's a(n internal) field for the journal's ISSN, it should be used to immediately pull precise info the journal, such as it's official name, publisher, place, etc.
2.1. There's always the issue of double ISSN (electronic and paper ones) for many journals but this doesn't seem to be a real problem.
2.2. What might be a problem is that older issues (or worse: older series!) of a journal might have been published by a different publisher, but I (amateurly) think there shouldn't be hard to cross this information with the date input from the user (something like if ISSN = 1234-5678 AND date < 2000 then publisher = Old Publishing House, else publisher = New Publisher).

3. And then there's exactly that thing, the date: Citavi 4 (and 5) doesn't seem be very clever recognizing dates. And why this is important? For three reasons: different types of references have different lenghts of date-data; different styles cite a same date differently; and one needs to cite a reference in different languages. Ex.: if a journal issue was published in October 2012, one might be required to cite it as "Oct. 2012", "10/2012", simply "2012", or "out. 2012" (in Portuguese, for example, where October = outubro); and a conference happened in August 30th 1990 may be cited with it's complete date ("30/08/1990", or "10/30/1990", or "Oct. 10, 1990", or "10 out. 1990", etc.) or just the year, and so on. So wouldn't it be great if Citavi understood what the (available) date actually is and cited according to each style? A slightly more (but not that much more) complicated is the case of intervals, when a journal issue correspond to a set of months or an event happened in a set of days, sometimes even starting in a year and ending in the following one (e.g., October-December 2000; Oct. 1-3 2000; Oct. 2000 - Mar. 2001; Dec. 30th 2000 - Jan. 3rd 2001). So in general there seem to be two types of date-data: instantaneous dates (10/10/2010, 10/2010, or simply 2010) and interval dates, and the last ones are simply a combination of two dates (start and end).

4. Batch editing: this would save so, but SO much time... I would love to perform a search and change any field for all results (e.g.: search for "Nature20**" and set all results as belonging to the Nature journal; or add author "Chaplin, Charlie" to all files containing "chaplin"), in the same we can already select multiple references and change their reference type.
4.1. Conditional search and replace would be taking this to heaven.

5. Short title: this should be customizable, because it has so many (unused) potentialities (which become a problem thereafter). Improving date recognition would allow having only the year in the short title, even if the date is 10/10/2010. And I might prefer having only the first author, or the first three, or the first and "et al.", or all of them. And it should be possible to reprogram the rule for the short title formation and apply it in batch, both to the whole project and for a selection of references only.

6. Batch renaming: the combination of short title customization and controllable batch operations would lead to a better control of files naming, but I have been experiencing a catch here, which is having several files for a single reference, each of them with individual properties that I need to recognize when choosing which file to view (which means I have to manually rename them everytime this happens). The two situations in which this happens the most are: (1) different versions of a single file, specially different languages (a same paper in English, French and Spanish); and (2) separation of complementary contents in different files (main contents and it's attachments, for example).

7. Folder watch: Citavi Picker is great, but having to use it as main import tool is not great. Sometimes one receives several files attached to an e-mail, or needs to manually split (we'll get to this below) a journal issue single file into several PDF files (one for each paper), or simply perform heavy in-browser research, downloading dozens of PDF files for later reading, indexing, etc. Mendeley and others have this feature, and it would be really nice to have it.

8. PDF splitting and/or assigning: a single PDF file for a whole edited book or journal issue is a very common find. This is very frustrating, because it requires one to manually split the PDF file, add each one to a new reference, etc. If Citavi could automate this, it would be great. But what would be greater, and would solve the suggestion number 9 (below), is recognizing the pages of a single PDF as belonging to different references. This might not be easy, I know, but wonder this: you scroll a 180 pages PDF file down, passing by all it's papers/chapters, and Citavi follows you showing the correspondent reference on the side.

9. Page number gap recognition: when adding quotations, Citavi thinks page 1 of the PDF file corresponds to page 1 of the reference, but that's (often) not the case. Sometimes it's just a matter of interval (adding +23 or -5 do the PDF page is enough to get the correct page number), sometimes it envolves more (two reference pages per PDF page, skipped or missing pages in PDF, etc., giving us simple equations or conditional equations, like rule X+Y for pages until N and rule X+Z after that), but I also think this should be a feature of a new Citavi.

I am sorry if this is too much, but I really think these features would fill 95% of a perfect software.

(The other 5% relate to legal references, which I didn't include here because they are really a huge mess, changing per-country etc.)

Best regards,

A. Rémy
alainsr
 

Re: Citavi 5: expected but (apparently) missing features

Postby Jennifer Schultz » 2014-12-05 17:59

Hello alainsr,

Thanks so much for your suggestions! Citavi does create copies for the conversion, so there's no need to fear that anything has happened to your Citavi 4 projects.

I'll address each of your points below.

1. Missing fields: many styles (such as NP-405 for Portugal) require the user to use custom fields or adapt other ones (such as page count). Since we're moving towards a new project structure, I guess there's no reason for not improving fields. Fields for journal publisher and place are what I immediately think of, but if given chance all users will give you a hand plenty of them, and this should be considered.


We're aware that many users would like additional fields. However, adding fields is a more involved process than one would think and we won't be able to add new ones until a future version, unfortunately.

2. This leads us to ISSN: if there's a(n internal) field for the journal's ISSN, it should be used to immediately pull precise info the journal, such as it's official name, publisher, place, etc.
2.1. There's always the issue of double ISSN (electronic and paper ones) for many journals but this doesn't seem to be a real problem.
2.2. What might be a problem is that older issues (or worse: older series!) of a journal might have been published by a different publisher, but I (amateurly) think there shouldn't be hard to cross this information with the date input from the user (something like if ISSN = 1234-5678 AND date < 2000 then publisher = Old Publishing House, else publisher = New Publisher).


Ideally, we'd also like to add journals as a reference type in a future version. Having this reference type would allow you to add a publisher and place for the journal and then we could also see if it might be possible to automatically add information using the ISSN.

3. And then there's exactly that thing, the date: Citavi 4 (and 5) doesn't seem be very clever recognizing dates. And why this is important? For three reasons: different types of references have different lenghts of date-data; different styles cite a same date differently; and one needs to cite a reference in different languages. Ex.: if a journal issue was published in October 2012, one might be required to cite it as "Oct. 2012", "10/2012", simply "2012", or "out. 2012" (in Portuguese, for example, where October = outubro); and a conference happened in August 30th 1990 may be cited with it's complete date ("30/08/1990", or "10/30/1990", or "Oct. 10, 1990", or "10 out. 1990", etc.) or just the year, and so on. So wouldn't it be great if Citavi understood what the (available) date actually is and cited according to each style? A slightly more (but not that much more) complicated is the case of intervals, when a journal issue correspond to a set of months or an event happened in a set of days, sometimes even starting in a year and ending in the following one (e.g., October-December 2000; Oct. 1-3 2000; Oct. 2000 - Mar. 2001; Dec. 30th 2000 - Jan. 3rd 2001). So in general there seem to be two types of date-data: instantaneous dates (10/10/2010, 10/2010, or simply 2010) and interval dates, and the last ones are simply a combination of two dates (start and end).


Citavi does format dates according to the citation style already. If you're noticing any problems with a certain style, please let us know and we will fix it.

Within your project you can set the date input format to be whatever you want. Simply click Tools > Options > Formatting. On the Date format drop-down menu select Custom.

Interval dates unfortunately can't be interpreted by Citavi's citation styles as date ranges and need to be added to the project in the format they should later appear in in the bibliography.

4. Batch editing: this would save so, but SO much time... I would love to perform a search and change any field for all results (e.g.: search for "Nature20**" and set all results as belonging to the Nature journal; or add author "Chaplin, Charlie" to all files containing "chaplin"), in the same we can already select multiple references and change their reference type.
4.1. Conditional search and replace would be taking this to heaven.


We do have batch modify and search and replace features in Citavi, but you're right that they could be extended. Similar requests are already in our wish list for new features, but I'll add your suggestions as well.

5. Short title: this should be customizable, because it has so many (unused) potentialities (which become a problem thereafter). Improving date recognition would allow having only the year in the short title, even if the date is 10/10/2010. And I might prefer having only the first author, or the first three, or the first and "et al.", or all of them. And it should be possible to reprogram the rule for the short title formation and apply it in batch, both to the whole project and for a selection of references only.


The short title is customizable (please see this chapter in our manual), however you cannot batch modify them yet. I'll add this request to our wish list, too.

6. Batch renaming: the combination of short title customization and controllable batch operations would lead to a better control of files naming, but I have been experiencing a catch here, which is having several files for a single reference, each of them with individual properties that I need to recognize when choosing which file to view (which means I have to manually rename them everytime this happens). The two situations in which this happens the most are: (1) different versions of a single file, specially different languages (a same paper in English, French and Spanish); and (2) separation of complementary contents in different files (main contents and it's attachments, for example).


At the moment Citavi adds a number to distinguish the files. If you need a different name, you'd have to either change the name manually later or not use the automatic renaming feature for all files at once.

7. Folder watch: Citavi Picker is great, but having to use it as main import tool is not great. Sometimes one receives several files attached to an e-mail, or needs to manually split (we'll get to this below) a journal issue single file into several PDF files (one for each paper), or simply perform heavy in-browser research, downloading dozens of PDF files for later reading, indexing, etc. Mendeley and others have this feature, and it would be really nice to have it.


Thanks for your suggestion for a folder watch feature! I'll add this to our wish list as well.

By the way, you don't only need to use the Picker. In most online databases you can also export your references and then import them into Citavi. You can also import a folder of PDF files at once, and you can drag PDF files to the navigation list to import them.

8. PDF splitting and/or assigning: a single PDF file for a whole edited book or journal issue is a very common find. This is very frustrating, because it requires one to manually split the PDF file, add each one to a new reference, etc. If Citavi could automate this, it would be great. But what would be greater, and would solve the suggestion number 9 (below), is recognizing the pages of a single PDF as belonging to different references. This might not be easy, I know, but wonder this: you scroll a 180 pages PDF file down, passing by all it's papers/chapters, and Citavi follows you showing the correspondent reference on the side.


This would be a very nice feature, but may be technically impossible. Our developers know that better than I do, though, and I'll add it to the wish list as well.

9. Page number gap recognition: when adding quotations, Citavi thinks page 1 of the PDF file corresponds to page 1 of the reference, but that's (often) not the case. Sometimes it's just a matter of interval (adding +23 or -5 do the PDF page is enough to get the correct page number), sometimes it envolves more (two reference pages per PDF page, skipped or missing pages in PDF, etc., giving us simple equations or conditional equations, like rule X+Y for pages until N and rule X+Z after that), but I also think this should be a feature of a new Citavi.


Citavi pulls this information from the PDF file's metadata, as far as I know, so I'm not sure if there would be a way for Citavi to know the actual page numbers if they're not accurate or not listed at all there. I'll add this to our wish list, but again it might not be possible.

Thanks again for all of your helpful suggestions!

Best regards,
Jennifer

P.S. Yes, legal references can be very tricky!
Jennifer Schultz
Citavi Customer Service
 


Return to Preview and Beta Versions