Last updates: Sat Dec 11 10:55:07 2021 … Thu Aug 22 06:26:42 2024

Fisheries research bibliography archive

This portion of the TeX User Group bibliography archive collects links to complete bibliographies of journals in fisheries research, as part of a collaboration with scientists at the Northwest Fisheries Science Center at the National Marine Fisheries Service, in Seattle, Washington, USA, a division of the U.S. National Oceanic and Atmospheric Administration (NOAA).

This collection includes almost 170,000 BibTeX entries for articles in 40 major journals. To keep the bibliography files of manageable size, and within the limits of common BibTeX implementations, coverage of journals with large article counts is split into decade-specific files.

For searching convenience, the complete collection is also available in a single file as an SQLite3 platform-independent database (about 770MB): fishery-journals.db. There is more about bibliographic markup in the BibTeX Information and Tutorial and about its SQL database interface in BibTeX meets relational databases. The sqlite3 program is public domain, portable across all common operating systems and CPU platforms, and requires no special computer administration privileges to install, use, or run.

A tabular status report is updated nightly: each line has a filename, the BibTeX label of the most recent item in that file, and that entry's timestamp. The report is sorted by timestamp from oldest to newest. Where bibliographies are split into decades, only the files from the most recent one or two decades are included in the report, because further updates for earlier decades are expected to be rare. For some publishers, we receive e-mail reports when new issues are published, in which case, updates are usually done within minutes, or at most a day or two. For other publishers, a human researcher must visit the journal Web sites occasionally to check for new issues or volumes. The oldest entries in the status report are what we check first.

Mirroring the collection

From Fall 2022, there is a shared repository with software and the bibliography files available here.

With a fast Internet connection, the single command

  % git clone https://github.com/nmfs-fish-tools/journals

takes a few seconds to make a local copy in the current directory of that data in the subdirectory journals, and the BibTeX files are stored as journals/tex/bib/*.bib.

The repository can be quickly updated at any future time with

  % cd /path/to/journals
  % git pull

The pull operation is fast because only changes from existing file versions are sent over the network.

Updates to the collection are frequent, so a recommended procedure is to set up a job that runs at suitable intervals, such as daily or weekly. For example, the Unix crontab -e command to edit your scheduled job list would let you create a single entry like this:

  # crontab format:
  #       mm    hh    dd    mon   weekday       command
  #       00-59 00-23 01-31 01-12 0-6(0=Sunday)
  12 3 * * * (cd /path/to/journals ; git pull)

The three comment lines document the format. The final line runs the update every night at 3:12. Change the third asterisk to 1 to do so only every Monday, or to 1,3,5 to run only on Monday, Wednesday, and Friday. Any command output is reported in e-mail to you.

Notes on title markup

Biologists classify life forms into a descending hierarchy of eight levels: domain, kingdom, phylum, class, order, family, genus, and species. Names within each level are based on Latin, once the common language of scientific communication. In these bibliographies, the most important are the last two, and in modern publishing practice, the genus is capitalized, the species is lowercase, and both are italicized. The genus and species are predominantly single words, but the species is sometimes two or three words, representing more precise classifications. Other members of the hierarchy are normally capitalized, and displayed in roman type.

Following the practice used for other journals in the TeX User Group bibliography archive, the genus and species are wrapped in a \bioname{…} macro with a default definition that typesets its argument in italics. However, the default is used only if the macro has not already been defined. The user could, for example, supply a definition that uses a different font or color, and possibly also creates an index entry, perhaps decorated with additional text that allows easy extraction of data for use in a separate index of biological names. For a description of how this can be easily done, see this document section.

When related biological names are referenced more than once in nearby text, the genus is often abbreviated after the first use: …Brown (\bioname{Salmo trutta}) and Rainbow Trout (\bioname{S. gairdneri})….

Authors commonly qualify biological names with the name of the discoverers or namers, and sometimes, the year of that event. Thus, we find a title with the text …\bioname{Priacanthus tayenus} {Richardson}, and \bioname{P. macracanthus} {Cuvier} and {Valenciennes}…, and another with …electric eel, \bioname{Electrophorus electricus}, {Linnaeus}, 1766 ({Gymnotiformes: Gymnotidae})…. That second example shows another common practice: adding a parenthesized phrase with the colon-separated order and family, both capitalized. Often, only the order is given there.

Some historical scientists are so well known that their names are further abbreviated: the father of modern taxonomy, the Swedish botanist Carl Linaeus (1707–1778), is often reduced to Linn., or further to just L., and the French naturalist Jean Léopold Nicolas Frédéric Cuvier (1769–1832) is shortened to Cuv.. Other common short abbreviations are F. for the Danish zoologist Johan Christian Fabricius (1745–1808), Lmk. for the French naturalist Jean-Baptiste Pierre Antoine de Monet, chevalier de Lamarck (1744–1829), and W. for the German taxonomist Johann Julius Walbaum (1724–1799).

Some scientific name phrases contain abbreviated words set in roman type, such as Chironomus plumosus f. semireductus, Here, f. is an abbreviation for forma, a secondary taxon rank below variety. Other examples include subsp. or ssp. for subspecies, and var. for variety. For more on these intricacies, visit this site.

Preparation of correctly marked up article title data for this collection proved particularly challenging for two reasons: (1) about a quarter of the titles contain \bioname{…} wrappers, but publisher Web data does not consistently indicate biological names with HTML markup for italics; and (2) there are large numbers of geographic names in the titles. Both types of names must be protected against the downcasing operation that many BibTeX styles perform to conform to the requirements of particular journal reference-list styles. One or more levels of bracing around a sequence of words prevents the downcasing. None of the publishers of the journals covered so far supplies any markup for geographic names, so the bibliographer must manually identify them and brace them.

To assist in this tedious job, a software tool, find-proper-names.c, scans bibliography files that have been prettyprinted by bibclean to have one key/value pair per line, and outputs a filename and line number, the matching word, and the preceding and following words that commonly occur in such phrases.

The internal list of more than 650 geographic names includes all of the countries of the world, all of the provinces, states, and territories in Australia, Canada, and the USA, plus common words like lake, ocean, and river. The tool would then report the title Lake Trout Stocks in Lake Superior, but the bibliographer would have to apply human intelligence to correct it to read Lake Trout Stocks in {Lake Superior}.

To find biological names that are missing macro wrappers, another software tool, make-bioname-sed-script.sh, extracts substitution patterns from BibTeX files that the standard Unix stream editor tool, sed, can later apply to add the missing markup. A current list of almost 8000 such patterns is supplied in bioname.sed and can be automatically applied with a command sed -f bioname.sed old.bib > new.bib.

Experience shows that this is substantially helpful, but is not sufficient, because the Latin genus name can also be a common noun, perhaps with a different meaning. Examples include acacia, bacterium, cancer, daphnia, danio, octopus, raja, rhea, salmonella, tilapia, and venus. For that reason, patterns that contain only a single word in the argument of \bioname{…} are excluded from the substitutions: the text of the journal article must be consulted to determine which of the two uses is meant.

Some preliminary work has been done to mark gene names with \gene{…} wrappers, but more checking is needed, because the biology and biochemistry literature does not use distinctive marking of genes, enzymes, proteins, viruses, and so on, and their names are cryptic. For more on this problem, see the encyclopedia article Gene nomenclature.

In Fall 2022, some initial work has added \geoname{…} and \stormname{…} wrappers in parts of the collection to indicate geographic, or place, names, and to identify named weather events (cyclones, hurricanes, storms, and typhoons). The first of those represents a significant addition, with about 30,500 instances in almost half of the BibTeX files by August 2024. The default definitions of those two wrappers do not change the output appearance in typeset and HTML displays, but may make it easier to extract specific raw data from the BibTeX files.

Despite the substantial effort that has gone toward adding markup in titles, there are likely to be a few instances where the protective braces, or a \XXX{…} wrapper, are missing. The bibliography file author will be grateful for reports of any such omissions.

Fisheries research online resources

During the course of this work, we recorded Web links for several online resources related to fisheries research and fish biology, and to geographic names. For reader convenience, we provide this alphabetized list:

• ACM BibTeX Formatting, with samples of BibTeX entries in computer science for various document types;
• Common and Scientific Names of Fishes from the United States, Canada, and Mexico, a 400-page resource book (ISBN-13 978-1-934874-31-8, DOI 10.47886/9781934874318);
• FishBase, an online resource for information about fish species that is frequently updated, and has been available for three decades;
• Geographic Names and the World Wide Web;
• Geographic Names Server;
• Journals in Eschmeyer's Catalog of Fishes, a database of fish-related publications from about 2600 journals;
• Official AFS fish name spellchecker, with a plain-text list of about 7000 correct words that appear in fish nomenclature;
• RAM Legacy Stock Assessment Database, an online, and downloadable, database of worldwide assessments of fish stocks;
• Sea Around Us Database, an online database of worldwide fisheries;
• Species names, an 8-page extract from the resource book about the use of biological names for fish;
• U.S. Board on Geographic Names. This site includes links to resources for foreign, underwater, and Antarctic place names, and the search facility supplies zoomable satellite maps;
• USGS geonames.
• World Register of Marine Species (WoRMS).

Fisheries research bibliography data files

In the following large table, the name in the first column is a hypertext link target, and is also the basename of a BibTeX bibliography file. The corresponding .sok file is the spelling-okay file: it has a list of correct words that augment those in common spelling program dictionaries, so that after bibliography updates, words that are not in either of those lists are possible spelling errors, though commonly, they are new personal names, new technical jargon, and new biological names. Those that are believed to be correct are merged into the .sok file, and the errors that remain are repaired in the .bib file. A subsequent run of spelling programs should then produce no output at all.

Each bibliography file is completely typeset with the LaTeX wrapper file ending in .ltx, and that produces page descriptions in TeX device independent (.dvi), compressed PostScript (.ps.gz and .ps.xz), and Portable Document Format (.pdf) files that describe typeset page images.

The .html file produces a Web page display that looks almost identical to the BibTeX file, but has live hyperlinks, most notably with DOI (Digital Object Identifier) and URL (Uniform Resource Locator) links that a Web browser can follow to the publisher page for that article. That in turn may contain article abstracts, links to PDF files with article contents, and sometimes, pointers to Web pages for preceding and following articles in the journal issue. Often, there are also hypertext links to author pages, allowing the location of related articles by the researchers in journals from the same publisher. The BibTeX entries also contain bibsource values that provide Web links to the source bibliography archive, and journal-URL values that link to the publisher Web site for the journal.

Finally, the .twx (title word index) file contains an automatically generated index of all of the title words in the articles in the bibliography file. It is typeset at the start of the bibliography DVI, PostScript, and PDF files.

To give a flavor of the bibliographic data available, here is a fragment for an article from one of the journals covered in this collection:

  @String{j-J-FISH-RES-BOARD-CAN  = "Journal of the Fisheries Research Board of
                                    Canada = Journal de l'Office des recherches
                                    sur les p{\^e}cheries du Canada"}
  @Article{Moreau:1979:DPA,
    author =       "Guy Moreau and Claude Barbeau",
    title =        "Diff{\'e}renciation de populations anadromes et
                   dulcicoles de {Grands Cor{\'e}gones}
                   (\bioname{Coregonus clupeaformis}) par la composition
                   min{\'e}rale de leurs {\'e}cailles. ({French})
                   [Differentiation of anadromous and freshwater
                   populations of large coregonids (\bioname{Coregonus
                   clupeaformis}) by the mineral composition of their
                   scales]",
    journal =      j-J-FISH-RES-BOARD-CAN,
    volume =       "36",
    number =       "12",
    pages =        "1439--1444",
    month =        dec,
    year =         "1979",
    CODEN =        "JFRBAK",
    DOI =          "https://doi.org/10.1139/f79-210",
    ISSN =         "0015-296X",
    ISSN-L =       "0015-296X",
    bibdate =      "Thu Nov 25 11:02:39 MST 2021",
    bibsource =    "https://www.math.utah.edu/pub/tex/bib/jfishresboardcan1970.bib",
    URL =          "https://cdnsciencepub.com/doi/full/10.1139/f79-210",
    acknowledgement = ack-nhfb,
    ajournal =     "J. Fish. Res. Board Can.",
    fjournal =     "Journal of the Fisheries Research Board of Canada =
                   Journal de l'Office des recherches sur les
                   p{\^e}cheries du Canada",
    journal-URL =  "https://cdnsciencepub.com/journal/cjfas",
    language =     "French",
  }

All text in the BibTeX files is coded in ASCII, which forms the first 128-characters of the Unicode character set that, in September 2022, contains almost 150,000 characters. Accented letters are coded in TeX markup, although from about 2020, recent TeX engines can accept Unicode UTF-8 encoding directly. When titles are in a language other than English, a bracketed translation is supplied. This collection currently contains entries for about 370 articles in Dutch, French, German, Spanish, and Turkish.

The CODEN value is the six-character code assigned by the American Chemical Society's Chemical Abstracts Service that provides a compact handle for the journal name. The CASSI Web site search facility accepts journal names, CODEN values, and ISSN values, and returns a detailed publication record of the journal, often including Web links to the publisher journal site. Library online catalogs usually allow searches on those three values too, and return information about local holdings of the journal.

The ISSN value is the 8-character code assigned to the journal as its International Standard Serial Number. There may be more than one such number, if the journal has multiple publication forms, such as print and electronic.

The ISSN-L value is the 8-character code assigned as the Linking ISSN: it is usually the ISSN of the first publication form that the journal appeared in, and is the recommended form to use when only a single ISSN value is needed or permitted.

The ajournal and fjournal values supply the abbreviated and full names of the journal. Where available, the US Library of Congress catalog record supplies the abbreviated form, and that in turn, stripped of punctuation, with spaces converted to hyphens, letters uppercased, and prefixed with j-, defines the BibTeX string abbreviation for the journal name. Use of full journal names in reference lists is encouraged to avoid reader confusion, but if the author prefers shorter names, the @String{…} value in the example can be followed with something like

  @String{j-J-FISH-RES-BOARD-CAN  = "J. Fish. Res. Board Can."}

in an author's personal BibTeX file for the current document. The last such definition applies when BibTeX formats the entry according to the current bibliography style, so there is no need to destructively remove alternate definitions.

The bibdate field records the date and time when the BibTeX entry was first created. It might be updated if the entry is later corrected, but often is not. It is also recorded in SQL databases as the value of a generated bibtimestamp key with a value of the form YYYY.MM.DD.HH.MM.SS (for year, month number, day of month, hour, minute, and second, with timezone not recorded) that can be used in searches to restrict consideration to entries created before or after a particular time, or within a certain time interval. Thus, an SQL SELECT statement clause of the form (bibtimestamp > '2022') could be used to extract data for just the entries created since the start of 2022.