Problem with the WTM database??

[Eric]

While doing some research in Hebrew I noticed something that looks odd. Before I report it as a bug, I was wondering if others could verify that it is a problem.

When using the WTM database the lemmas are separated by an "@" symbol and then the morphology codes follow.

In the book of Daniel from 2:4 to 7:28 the lemmas are separated by a "%" instead of an "@" sign. This seems to affect the searches as well.

Can someone verify that this is the way that their WTM looks and if they agree that this prevents proper hits when doing searches?

Thanks.

[Michael Hanel]

While doing some research in Hebrew I noticed something that looks odd. Before I report it as a bug, I was wondering if others could verify that it is a problem.

When using the WTM database the lemmas are separated by an "@" symbol and then the morphology codes follow.

In the book of Daniel from 2:4 to 7:28 the lemmas are separated by a "%" instead of an "@" sign. This seems to affect the searches as well.

Can someone verify that this is the way that their WTM looks and if they agree that this prevents proper hits when doing searches?

Thanks.

I should really look before speaking, but I'm going to take a stab anyway and let someone else clarify. But my guess is that the % is the division marker between word and morphological coding for text which is in ARAMAIC. So that's why it's different than the rest of the WTM which is written in Hebrew.

[terogs]

The % is the separator for Aramaic. I confirmed it in on-line help as well as looking at the database.

It also looks like a lemma-only search in the WTM will return the Hebrew since it assumes the @ separator. I guess you'd have to do an or search specifying both Hebrew and Aramaic morphology codes if you want to cross languages.

Tom

[Ben Spackman]

If you're specifying morphology, an "or" search won't work, because the coding is different. There's no Gt stem in Hebrew, for instance, but there is in Aramaic. Ditto for root searches, but for a different reason- several PNWS consonants went different ways in Hebrew and Aramaic, such as d-underline. In Hebrew, it became zayin (eg. zhb, "gold") but in Aramaic dalet (eg. dhb, "gold"). A Hebrew/Aramaic root search woul only be accurate if the root contains none of those consonantal differences.

[terogs]

You can search both Hebrew and Aramaic with an OR search if you use wildcards: for example /aaa@* aaa%* (where aaa represents a common root) will pick up occurrences of the root in Dan 2:4 - 7:28. I'm not sure how meaningful that is, but it works.