• Database software

    From Steve Hayes@21:1/5 to G6JPG-255@255soft.uk on Wed Jun 6 05:26:09 2018
    XPost: alt.windows7.general, soc.genealogy.computing

    On Tue, 5 Jun 2018 19:13:30 +0100, "J. P. Gilliver (John)" <G6JPG-255@255soft.uk> wrote:

    In message <pf4o0q$o80$1@news.albasani.net>, Ken Springer ><wordworks@greeleynet.com> writes:
    []
    I'm no database expert either, but I do know not every database need >>requires Access or other relational databases.

    And finding info about non relational databases seems to be like
    searching for hen's teeth.
    []
    I'm trying to understand what "relational" actually _means_ here. Rather
    than just asking, I did try to look at the wikipedia entry, but it made
    my brain hurt ...

    I _think_ it means data that's crosslinked in various ways. (I do
    genealogy, and I think the software I use is a relational database.) But
    this makes me wonder what a _non_-relational database would _be_, other
    than some sort of list or table.

    A relational database is one that links several tables so that you can
    save space for repeating data.

    Lineage-linked genealofy programs are relational. RootsMagic, for
    example, uses SQLite, and Legacy uses the Jet engine used by MS
    Access, so if you wanted to you could do other stuff with your
    database using Access or SQLite.

    In these programs you enter every person once, and there are linking
    tables that link the persons to spouses, children etc.

    In a non-relational database (also sometimes called a "flat file"
    database), you would enter a person with their spouse and children,
    but then would have to re-enter all those names for each record of
    each person.

    Non-relational databases are much easier to set up and use, and are
    often much more flexible in the kind of data that they can contain and
    you can seach for.

    I use two "large-text" databases for genealogical and other research,
    Inmagic and askSam. askSam is particularly good for dumping large
    amounts of disparate informnation and finding it again, so it is
    useful for raw material for research. Inmagic is more structured, but
    still more flexible than a relational database. I've used it, for
    example, to index family letters, typing in extracts or summaries of intormation from hard-copy letters people sent me. These often refer
    to several people, and if you saved them in a relational genealogy
    prtogram you'd actually end up with more duplication than you would in
    a "flat-file" database.

    Here's an example of an Inmagic report:

    From: Pearson, Ralph
    To: Steve & Val Hayes
    Date: 8-Jul-1992
    Had a letter from Mrs Anne Roberts in Carlisle who had
    connections with the Vipond family, and also Jeanne Shapcott
    of Queensland, who was descended from the Viponds of
    Garrigill. He was also in touch with a West Cumberland
    historian who had information about the Peile family of
    Parton. May Pearson, his father's second wife, died on
    16-Sep-1991. She was born on 14 Oct-1904. Gerald Elston (son
    of Edith Pearson) died on 14-Dec-1991. Charles Arthur
    Pearson (son of Charles Pearson) died on 5 Jan 1992. Sends
    account of the life of Daniel William Pearson (his
    grandfather). Had included something on Daniel's wife, Sarah
    Jane Walker, and about "Wonderful Walker", whose connection
    with the Walkers of Whicham may have been tenuous. Was
    researching Cecil William Pearson, son of Charles. He had
    joined the Northumberland Fusiliers (4th Battalion), but
    later volunteered for the Royal Flying Corps, and was killed
    on 3 Jan 1918 while serving with 57 squadron as an observer.
    Was buried at Amerika, close to Ten Brielen in Belgium.
    Grandson Kevin had graduated from Cubs to Scouts.

    My search argument for that was "peile w3 parton" (Peile within three
    words of "Parton"). Putting that kind of stuff in a lineage-linked
    program is actually more difficult.

    Hope this helps.


    --
    Steve Hayes
    http://www.khanya.org.za/stevesig.htm
    http://khanya.wordpress.com

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From J. P. Gilliver (John)@21:1/5 to hayesstw@telkomsa.net on Wed Jun 6 07:13:13 2018
    XPost: alt.windows7.general, soc.genealogy.computing

    In message <cejehdts0kd68rco1683ekhia0cotdrv93@4ax.com>, Steve Hayes <hayesstw@telkomsa.net> writes:
    []
    A relational database is one that links several tables so that you can
    save space for repeating data.

    I'm sure that there is more to the linking than just the saving of
    space. (Especially nowadays, when wasting space seems almost
    obligatory.)
    []
    still more flexible than a relational database. I've used it, for
    example, to index family letters, typing in extracts or summaries of >intormation from hard-copy letters people sent me. These often refer
    to several people, and if you saved them in a relational genealogy
    prtogram you'd actually end up with more duplication than you would in
    a "flat-file" database.
    []
    The one I use (Brother's Keeper) lets me link to files; in the case of a
    letter like the one you quoted, until I'd parsed all the individual
    facts in it, I'd just put the letter in a file, and link all the people
    it refers to to that one file. When I had time, I'd extract facts from
    it and file them under the affected persons, with the source credited -
    I can link a single "source" to lots of facts. (I would also remove the
    facts from the original as I went, until I'd eventually got it down to
    nothing, then delete it, with the individual facts - with extracts if
    necessary - filed under the people they refer to, and citing it from the
    list of sources [which are a flat file]; that's just me, and I know some
    people would keep the full original text file even once they'd copied
    the facts it contained to more appropriate places.)
    --
    J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

    But remember, in a permissive society, it is also permissible to stay at home and have a nice cup of tea instead. Andrew Collins, RT 2015/2/14-20

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Steve Hayes@21:1/5 to G6JPG-255@255soft.uk on Thu Jun 7 06:58:15 2018
    XPost: alt.windows7.general, soc.genealogy.computing

    On Wed, 6 Jun 2018 07:13:13 +0100, "J. P. Gilliver (John)" <G6JPG-255@255soft.uk> wrote:

    In message <cejehdts0kd68rco1683ekhia0cotdrv93@4ax.com>, Steve Hayes ><hayesstw@telkomsa.net> writes:
    []
    A relational database is one that links several tables so that you can
    save space for repeating data.

    I'm sure that there is more to the linking than just the saving of
    space. (Especially nowadays, when wasting space seems almost
    obligatory.)

    Yes.

    It also saves one from having to reenter data multiple times, and
    makes it easier to link the data in various ways.

    still more flexible than a relational database. I've used it, for
    example, to index family letters, typing in extracts or summaries of >>intormation from hard-copy letters people sent me. These often refer
    to several people, and if you saved them in a relational genealogy
    prtogram you'd actually end up with more duplication than you would in
    a "flat-file" database.
    []
    The one I use (Brother's Keeper) lets me link to files; in the case of a >letter like the one you quoted, until I'd parsed all the individual
    facts in it, I'd just put the letter in a file, and link all the people
    it refers to to that one file. When I had time, I'd extract facts from
    it and file them under the affected persons, with the source credited -
    I can link a single "source" to lots of facts. (I would also remove the
    facts from the original as I went, until I'd eventually got it down to >nothing, then delete it, with the individual facts - with extracts if >necessary - filed under the people they refer to, and citing it from the
    list of sources [which are a flat file]; that's just me, and I know some >people would keep the full original text file even once they'd copied
    the facts it contained to more appropriate places.)

    But by doing that you can miss the opportunity to find important
    research hints.

    That can be done in a relational database, depending on the design --
    and the main DISadvantage or relational databases is how much time has
    to be spent on design, and how changes in design are also awkward and time-consuming.

    But keeping source data in a "flat-file database can make it possible
    to find new connections between old data as new data are added.

    --
    Steve Hayes
    http://www.khanya.org.za/stevesig.htm
    http://khanya.wordpress.com

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)