Incorrect Language

[Yaakov Jannis]

If one opens a text, say BHS, and aligns a parallel Hebrew text, then opts for a comparison of "words," the comparison results are expected. If one then adds a Targum in parallel, the program attempts to compare and thus highlights "words" in the Targum against the Hebrew. This is an error in logic, for clearly Accordance has equated Aramaic and Hebrew as one language. This is a bug in logic. Please fix.

 

thanks.

[rokas]

Sorry, couldn't resist not to answer this (although this is just a guess):

 

Accordance doesn't compare languages (meaning behind text), it compares text (letters/characters) only, because meaning requires interpretation, which Accordance cannot do. So it's not a bug in logic.

 

But what you ask for actually makes sense - it would be good if we could choose which panes should be compared to the main one and which ones shouldn't.

 

For now you can add Targum as a separate tab, place them side by side and tie them - please see the attachment (I got two English Bibles compared in one tab and German one in another tab, for otherwise German would be all blue) - looks almost as good

Edited June 21, 2015 by rokas

[Yaakov Jannis]

I too can't resist to reply.

 

You are aware that any Computer Science introduction course explains the polysemous quality of "bug"? If not, look it up. Even the first line at the Wikipedia page captures some of the issues covered by "bug."

 

Here, this is clearly a design bug, for Accordance is attempting to compare letters characters of two different semitic dialects as if they were the same. That's a bug. Accordance should, if programmed correctly, bypass an Aramaic text in comparison to a Hebrew text—for it is clearly not the same language.

 

You say, "looks almost as good," which the key word is almost. Fix the bug and it will be good.

[Michael J. Bolesta]

A similar phenomenon occurs if one selects English and Spanish in parallel, and check the compare box. The program does not distinguish languages using the same alphabet.

[Yaakov Jannis]

Yes, exactly, Michael.

 

So, Roman = Roman is not the same as English = Spanish. Hence it is illogical to assume that shared script equals same language.

[rokas]

You are aware that any Computer Science introduction course explains the polysemous quality of "bug"? If not, look it up. Even the first line at the Wikipedia page captures some of the issues covered by "bug."

 

Here, this is clearly a design bug, for Accordance is attempting to compare letters characters of two different semitic dialects as if they were the same. That's a bug. Accordance should, if programmed correctly, bypass an Aramaic text in comparison to a Hebrew text—for it is clearly not the same language.

 

I looked it up. Bug = when software behaves in an unintended way, gives unexpected result, etc.

 

Now, Accordance compares only characters. Characters in Aramaic are the same as those in Hebrew (if not, forgive me my ignorance). It also doesn't have languages attached to texts - Accordance doesn't "know" that one text is in Hebrew and another is in Aramaic.

 

So, is it a bug? I don't think so. I think it's rather you or I who behave in an unintended way when we force Accordance to compare Hebrew with Aramaic or English with Spanish or German. In other words, we are not supposed to compare texts with different languages. I am not sure whether Computer Science introduction courses explain it, but if not, look up "garbage in, garbage out" (exactly because it's logical)

Edited June 21, 2015 by rokas