Biblical Hebrew Forum

["JAMES CHRISTIAN READ" <JCR128 AT student.anglia.ac.uk>]

Hi Karl,

KWR: Where should the separator be? At the ends of the lines? If so, then
the return is already a unique character.

JCR: The separator should be between the form and the 
entry. The form and the entry would be the natural 
columns to use in a database driven version of your 
dictionary.

KWR: Making a quick electronic count of irregular forms and forms which
could come from two or more sources, I got a result of 777 entries in
my dictionary. The problem is that I know I am missing many such
forms.

JCR: How did you make such an automated count?

KWR: Because I don't know data base programming, I don't understand how
this would work.

JCR: I actually made a mistake. A database table is 
basically a set of rows (each row is unique) with 
columns. The columns hold different pieces of 
information. To achieve the functionality you suggest 
we would need to have one massive table with two 
columns. The first column would be the word we are 
looking up. The second column would be the root it is 
derived from. When looking for the definition of a 
word found in the text the application would ask the 
database to return the row which has the word in its 
first field. Taking the root from the second column 
it would be able to query a second table dedicated to 
your dictionary entries to return the desired 
definition.

However, as I said, I think it would be far more 
functional and user friendly if a definition was 
provided for every single word. This way we make do 
with one table instead of two. And each request would 
result in one query to the database rather than two 
and so gains in performance would also be achieved.

I don't think that hand feeding the student in this 
way will impede their progress in recognising forms. 
On the contrary, I think that with time, they will 
start to notice the similarities between forms and 
work out the identifying marks of, for example the 
3rd person masculine singular, for themselves. Such 
learning is more representative of the natural way of 
learning grammar and makes reading more accessible to 
non academic learners. There is a large category of 
learners who are linguistic geniouses in the sense that 
if you put them in any culture in the world they will 
soak up the language like sponges but are academically 
challenged in the sense that the very second you start 
talking about grammar their brains switch off and they 
start looking at you as if you were from another 
planet. My feelings are that reading the Hebrew text 
should be made accessible to all with an interest and 
such 'hand-holding' would open the text up to a much 
wider range of learners.

KWR: OK, it's in your concordance, but not your frequency table.


JCR: If it's in the concordance then it's in the 
frequency tables. They are both generated from the 
same data. Note that there are hundreds of pages in 
the frequency tables. You may not have found the 
bi-gram in question because you were expecting it to 
have a specific count. Please note that frequencies 
were not generated using a critical version of the 
Tanakh but using the xml Aleppo Codex and therefore 
all counts are specific to this text. I do plan to get 
around to doing something similar with Chris's online 
Leningrad Codex but haven't got around to it yet.

One way of finding the bi-gram quicker in the frequency 
charts is to first find the first word in the 1-gram 
column, click on it and search through the much 
reduced results. Hope this helps.

I suppose my whole architecture is lacking in 
documentation and I really should get round to 
publishing a page or two about how the concordance, 
reader and frequency charts can be used.

James Christian Read - BSc Computer Science 
http://www.lamie.org/hebrew - thesis1: concept driven machine translation 
using the Aleppo codex 
http://www.lamie.org/lad-sim.doc - thesis2: language acquisition 
simulationPlease note that frequencies 
were not generated using a critical version of the 
Tanakh but using the xml Aleppo Codex and therefore
all counts are specific to this text. I do plan to get
around to doing something similar with Chris's online
Leningrad Codex but haven't got around to it yet.

One way of finding the bi-gram quicker in the frequency
charts is to first find the first word in the 1-gram
column, click on it and search through the much
reduced results. Hope this helps.

I suppose my whole architecture is lacking in
documentation and I really should get round to
publishi