coombe . id . au ~ colonial newspapers ~ historical mystery ~ family history

 

(Uncommon) TROVE Tips

 

I was a big user of the National Library of Australia’s TROVE from its early days. And my information technology background equips me with a database searcher’s mind. (I find stuff that surprises even me!) This all helped me in 2014 – along with heaps of adrenalin - to the laurel wreath of inaugural (and currently reigning) Troveia champion in the guise of my Gasworks Hotel. Enough bragging. Anyway I hope someone else wins next time. So here I share some Trove tips and hope we all continue to surprise ourselves with our finds.

 

My hope too is that government and bureaucracy think of TROVE as being as valuable as its regular users already regard it. I also think that TROVE is now sufficiently widely used that it needs a stakeholder structure to see that it improves with every released version.

 

These are random tips. If you want structured, read TROVE’s help. Here are some important parts of TROVE Help: Finding Things in TROVE, Digitised Newspapers, Searching in Newspapers.

 

Making a List of Lists

You can read how to create lists in Trove help centre. To start your list, you need its first item, such as a newspaper article. But to create my first “list of lists” I needed to work around this. First I opened a random newspaper article with which to create my list. I then added my sublists by adding them as a web pages. I then deleted the original random article from the list. Here is my first list of lists about my research on S. T. Gill: http://trove.nla.gov.au/list?id=92925. Here is another very impressive list of lists http://trove.nla.gov.au/list?id=31895.

 

Text Correctors Make Better Searchers

Most of my time is in the Digitised Newspapers zone. I am often correcting articles – one of my regular favourites is Police Court, keeping up with who was swearing drunk last night (well, 150 years ago).

 

Article text in TROVE is generated by a process called optical character recognition (OCR). An image of a page is interpreted by computer software looking for letters (and words). The software is far from perfect and there is even a TROVE Forum thread for amusing (mis-) translated text! My favourite is “farted with his father affectionately” – should have read “parted ...” from an 1874 article for my Deaths at Josephine’s Gasworks Hotel.

 

Correcting text makes me a better searcher. As well as fixing text so that it can be subsequently found in searches, it also helps me think like OCR. So keep in mind that the “Williams” you are looking for may be hiding there as “VVilliams”. Also consider Mclntire (numeric one in place of upper case I), Ho1den, 13eaver, 0Brien (zero, not “O”).

 

And you would have often seen “tbe” in place of “the.” So perhaps it’s time to search for “8m1tb”. Now can you see “Smith”? If you put your mind to it you can probably come up with many ways of misrepresenting the text that eludes you!

 

How Much Did My Ancestor Earn?

Try this labour market search and you will get a lot of market intelligence on the going rates for certain occupations. For better relevance, apply filters for decade and maybe state. (See image at right. Poor compositors! Should a female cook disguise herself as a man for better wages?)

 

Cleaning Images

Sometimes you want to save an image of part of an article to save or share, but the article has unwanted highlights. A search will result in the search terms being highlighted in the selected article. One way to remove these is to select article details then use the “article identifier” link. Another is to view the entire page which will also get rid of non-article noise. Then capture the screen.

 

 

 

8 Jun 2016, 16 Aug 2016