Upgrade StorageOlder Formats
(Newbie post - from a fossil husband assisting a researcher mate. Please phrase replies in terminology appropriate to amateur comp sci, ca. 1980. I.e., an object is something on one's mantelpiece.)
I'm posting this in semi-ignorance, after a quick perusal of the FAQ, and I already have a couple of questions.
1. How would one go about importing data from *really* old formats, like dBASE, or a generic CSV output from a typical 1980s database, with several fixed-length fields (mostly USASCII text or numerical), but with one big "notes" field - what used to be called 'free-form'?
2. What is the recommended procedure for backup of a scholar's critical Zotero data? What else can read it? Is there a util to extract text from a crashed file?
If these are ignorant questions, so be it. They might at least prompt additions to the FAQ, as well as the (eagerly awaited raspberries).
caveat: I myself actually haven't even *run* Zotero 1.03. Further apologies for that.
I'm posting this in semi-ignorance, after a quick perusal of the FAQ, and I already have a couple of questions.
1. How would one go about importing data from *really* old formats, like dBASE, or a generic CSV output from a typical 1980s database, with several fixed-length fields (mostly USASCII text or numerical), but with one big "notes" field - what used to be called 'free-form'?
2. What is the recommended procedure for backup of a scholar's critical Zotero data? What else can read it? Is there a util to extract text from a crashed file?
If these are ignorant questions, so be it. They might at least prompt additions to the FAQ, as well as the (eagerly awaited raspberries).
caveat: I myself actually haven't even *run* Zotero 1.03. Further apologies for that.
This is an old discussion that has not been active in a long time. Instead of commenting here, you should start a new discussion. If you think the content of this discussion is still relevant, you can link to it from your new discussion.
See:
http://forums.zotero.org/discussion/1224/csv-or-other-simple-import-technique/#Item_2
and note that dbase is a closed binary format that Firefox doesn't understand, so it is unlikely that it will be be supported.
(2)
See:
http://www.zotero.org/documentation/frequently_asked_questions#how_do_i_back_up_my_zotero_library
Snapshots and attachments are stored in the file system. Metadata is stored in a SQLite database. You can read it with Zotero or with SQLite. You can use SQLite to recover corrupted databases (either manually or you may upload it to Zotero for repairs).
Third idiotic question: Suppose one's research is currently in 19thC journals and gazettes, as primary sources. Neither Google Books nor the databases from which it draws subindex journal articles. I'm just the helper here, not the scholar. Is it then generally the case that one has to find **someone else's cite** online to get something useful to clip into Zotero?
Fourth question: Given: that the researcher's computer is WinXP, US English, with the International Keyboard enabled. Why does Zotero ignore accents (French, Italian) and, indeed, actually *omit* accented letters from its clipped text. This is a screen-scraper of some kind, right? How does one enable it for a language, assuming that what one sees on the cite screen looks correct, and also assuming FF knows what the page's encoding is (usually UTF-8, from casual observation)?
(4) Site "translation" depends on the specific site. It has nothing to do with the keyboard. Where possible, Zotero usually downloads standard bibliographic data rather than screen scraping. This data is often not in UTF-8. You can sometimes get UTF-8 by visiting a different site that has the same reference (or can manually edit the citation). If you have a few specific sites that you'd like to have UTF-8 support for, report them (and DEFINITELY report them if you can get UTF-8 citations from them into any other reference manager).
(5) This is a FAQ. There is no current ability to merge databases. There will be better support for this in the near-future. You have lots of options for the present (though few are officially supported): Use portable firefox on a USB stick, store your zotero on the network, use a third-party website (citeulike, connota, refbase, etc.) to provide a central store, make changes to only one database & use third-party sync software (or manually sync it), manual export/import of references.
>This is a FAQ
It very nearl is, right in your message (g). I'll explore these as time permits. Would there be anything wrong in doing a copy-overwrite of the current db to a portable FF, using that in the library, then back-overwriting the desktop's db with the portable one? Are all the files in one dir or tree, really? No indices lurking elsewhere?
Any suggested site to catch up with the last 20 years in library science? I'm half-serious.
Oh, and..
6. Assume one does a lot of work on a non-standard site, such as NYPL/CATNYP. How on earth did Zotero extract a LOC-standard call number from a NYPL page, when neither the displayed page nor the page source contained any such thing?
With firefox closed, you can replace the directory. You may want to take measures to backup before you replace it (in case you sync in the wrong direction, etc.).
(6) I'd have to see the particular page, but Zotero will often try to download a machine-readable record (i.e. a MARC record form a library catalog or an ISI/RIS/bibtex/whatever reference from academic journals) rather than scraping the page. Some sites don't embed the same information in the human-readable webpage as they do in these records.