Ctrl+K
🚧 This is a working draft and will change often. Do not cite!
Use the latest published version instead. 🚧

Overview of periodicals

22.1. Overview of periodicals#

On this page

22.1.1. What is a periodical?#

Periodicals are publications that are issued at regular intervals, like newspapers, magazines, or academic journals. Newspapers have their own category in Trove and are managed and delivered through a separate system, but what about everything else? This section helps you find and use all the other types of digitised periodicals available through Trove.

Here’s some examples of Trove’s digitised periodicals:

The Bulletin, 1880–1984

nla.obj-188284455.jpeg

Browse 5,418 issues >

The Home, 1920–1942

nla.obj-380634657.jpeg Browse 229 issues>

Dogs, 1962–1970

nla.obj-873894973.jpeg Browse 106 issues >

Sometimes it’s not clear whether a publication is a periodical or not. What about annual reports produced by government departments? Or almanacs that are updated each year? As with most things Trove, the boundaries are blurry. What counts as a periodical might depend on your own research interests.

22.1.2. Digitised and born digital periodicals#

This section is looking at periodicals that have been digitised by the NLA and its partners and delivered through Trove. In other words, they’re periodicals that have been converted from print publications to digital objects by scanning, imaging, and OCR. But there are a growing number of periodicals available through Trove that were born digital. These include digital publications (usually PDFs) that are uploaded by publishers under the National edeposit (NED) scheme. Recent editions of periodicals are likely to arrive in Trove via NED.

A search for nuc:"ANL:NED" with the format Periodical in the Books & Libraries or Research & Reports categories will reveal many of these titles. Using the API, the same search with category set to all returns 10,736 results.

Try it!

Since the 1990s, the Australian Web Archive has sought to preserve online journals through Pandora, its selective archiving program. The contents of the preserved journals are now rolled into the whole of domain web harvest that you can explore through Trove’s Websites category. But there are also records for individual journal titles in other categories, particularly Books & Libraries. Most web archive links in Books & Libraries have the form nla.arc-[some number], so searching for "nla.arc" and the format Periodical will reveal many archived titles. Using the API, the same search with category set to all returns 22,333 results.

Try it!

22.1.3. Finding digitised periodicals#

Strange as it seems, there’s no simple way to find which periodicals have been digitised in Trove, but there are a few useful strategies:

  • search for digitised periodical titles

  • search for digitised articles from periodicals and explore the title facet

  • use the Trove API’s /magazine/titles endpoint

These are described more fully in Finding digitised periodicals.

22.1.4. Pre-harvested dataset of digitised periodicals#

To make it easier to analyse and explore digitised periodicals in Trove, I’ve created and shared a dataset containing details of titles and issues. This dataset was created using the magazine/titles endpoint, but I’ve removed duplicates, excluded Commonwealth Parliamentary Papers, fixed some of the issue data, and added extra metadata fields.

Parliamentary Papers make up more than half the periodical titles returned by the API. They’re also treated inconsistently in Trove – some are ‘books’, while others are ‘periodicals’. It seems to make more sense to bring all the Parliamentary Papers together separately. This also makes it easier to explore the fascinating variety of journals, almanacs, magazines, and newsletters that have been digitised, without wading through innumerable annual reports from government agencies.

You can download the dataset from the GLAM Workbench or explore it using Datasette-Lite.

Expand

22.1.5. Titles of digitised periodicals#

The calculations and visualisations below are all based on the pre-harvested dataset which excludes Parliamentary Papers.

How many digitised periodicals are there in the dataset?

Hide code cell source 
import pandas as pd

df_titles = pd.read_csv(
    "https://github.com/GLAM-Workbench/trove-periodicals-data/raw/main/periodical-titles.csv",
    dtype={"start_year": "Int64", "end_year": "Int64"},
)

print(f"There are {df_titles.shape[0]:,} digitised periodicals in Trove.")

There are 908 digitised periodicals in Trove.

What words appear most frequently in the titles of periodicals?

Hide code cell source 
from wordcloud import WordCloud

wc = WordCloud()
wc = (
    WordCloud(width=800, height=300)
    .generate("\n".join(df_titles["title"].to_list()))
    .to_image()
)
wc
../../_images/c043a6015ad009996b7e66c393daeb69e2aa3282ccf4ebd99f8ecb79b2c7209c.png

Fig. 22.1 Words in the titles of periodicals#

Which titles have the most issues?

Hide code cell source 
from IPython.display import HTML

# Sort by issue count then take the top 20
df_titles[["title", "trove_url", "issue_count"]].sort_values(
    "issue_count", ascending=False
)[:20].style.format(thousands=",").hide()
title trove_url issue_count
The bulletin. https://nla.gov.au/nla.obj-68375465 5,418
Dun's gazette for New South Wales. https://nla.gov.au/nla.obj-724008889 2,408
The Rugby League news. https://nla.gov.au/nla.obj-598579045 1,568
The Australian woman's mirror. https://nla.gov.au/nla.obj-389050376 1,561
Education : journal of the N.S.W. Public School Teachers Federation. https://nla.gov.au/nla.obj-525471579 1,534
The school magazine of literature for our boys and girls https://nla.gov.au/nla.obj-748113553 1,170
The wireless weekly : the hundred per cent Australian radio journal https://nla.gov.au/nla.obj-589648812 1,069
ABC weekly https://nla.gov.au/nla.obj-890736639 1,033
Everyones. https://nla.gov.au/nla.obj-522690001 916
Overseas trading. https://nla.gov.au/nla.obj-855657180 884
Pacific islands monthly : PIM. https://nla.gov.au/nla.obj-310385031 840
Pix. https://nla.gov.au/nla.obj-389050162 826
Australasian photo-review. https://nla.gov.au/nla.obj-392290161 669
Current notes on international affairs. https://nla.gov.au/nla.obj-850514773 525
Uniken. https://nla.gov.au/nla.obj-211194649 522
Walkabout. https://nla.gov.au/nla.obj-566923190 494
Building : the magazine for the architect, builder, property owner and merchant. https://nla.gov.au/nla.obj-79483170 429
The Newcastle and Maitland Catholic Sentinel : the official organ of the diocese of Maitland. https://nla.gov.au/nla.obj-302666124 416
Architecture : an Australasian review of architecture and the allied arts and sciences. https://nla.gov.au/nla.obj-2967139558 361
K.C.C. kennel gazette : the official organ of the publishers, the Kennel Control Council. https://nla.gov.au/nla.obj-760030452 360

Not all of the titles in the dataset have digitised issues available on Trove. I’m not sure if they’re still in the process of being digitised, or if it’s an error of some sort. Here’s how mnay titles currently have zero issues

df_titles.loc[df_titles["issue_count"] == 0].shape[0]

109

22.1.6. Issues#

How many digitised periodical issues are in the dataset?

Hide code cell source 
df_issues = pd.read_csv(
    "https://github.com/GLAM-Workbench/trove-periodicals-data/raw/main/periodical-issues.csv",
    keep_default_na=False,
)

print(f"There are {df_issues.shape[0]:,} digitised periodical issues in Trove.")

There are 37,015 digitised periodical issues in Trove.

How are the issues distributed over time?

Hide code cell source 
import altair as alt

df_issues["year"] = df_issues["date"].str.slice(0, 4)

df_issues_years = df_issues["year"].value_counts().to_frame().reset_index()

# Add a link to the db of issues in Datasette
df_issues_years["db_link"] = df_issues_years["year"].apply(
    lambda x: f"https://glam-workbench.net/datasette-lite/?url=https://github.com/GLAM-Workbench/trove-periodicals-data/blob/main/periodicals.db&install=datasette-json-html&install=datasette-template-sql&metadata=https://github.com/GLAM-Workbench/trove-periodicals-data/blob/main/metadata.json#/periodicals/issues?date__lte={x}-12-31&date__gte={x}-01-01&_sort=date"
)

chart_issues_years = (
    alt.Chart(df_issues_years.loc[df_issues_years["year"] > "1800"])
    .mark_bar(size=2)
    .encode(
        x=alt.X("year:T", title="year of publication"),
        y=alt.Y("count:Q", title="number of issues"),
        tooltip=[
            alt.Tooltip("year:T", format="%Y"),
            alt.Tooltip("count:Q", title="issues"),
        ],
        href="db_link",
    )
    .properties(width=600, height=300, padding=20)
)

chart_issues_years

Fig. 22.2 Number of periodical issues by year#

Contents