Fun_People Archive
24 Mar
Talmud and FORTRAN, or the separation of church and state machines 

Date: Wed, 24 Mar 93 16:03:48 PST
To: Fun_People
Subject: Talmud and FORTRAN, or the separation of church and state machines

 From: bostic@vangogh.CS.Berkeley.EDU (Keith Bostic)
 From: 9999sc01@uhdvx3.dt.uh.edu (Alexandre Khalil)
 From: grabiner@math.harvard.edu (David Grabiner)
-----------------------------------------------------------------------------
[	Preface to the discourses from the soc.culture.jewish newsgroup

    If you aren't Jewish, didn't grow up in New York City (or Philadelphia or
Boston or ...), and didn't learn much Yiddish some other way, you may want to
hear my attempts at the definitions of a few terms:

	ASSUR		The god of war and supreme national god of Assyria
			(also ASHUR, ASSHUR, or ASUR)
	BRACHA		a prayer
	CHALLAH		an eggy leavened bread (eaten during SHABBAT)
	COHANIM		*
	HILLEL		1st cent. B.C. to 1st cent. A.D. Jewish teacher
	KASHER		v.t. to make kosher
	KAVANAH		**
	SHABBAT		the Jewish Sabbath
	SHULE		the local synagogue or temple
	TALMUD		the authoritative body of Jewish tradition
	TORAH		the body of wisdom and law contained in Jewish
			Scripture and other sacred literature and oral
			tradition
	TREIF		not kosher, tainted (also TRAIF or TRAFE)

* Okay, so I'm not Jewish either, but at least I can tell that COHANIM is
probably a Hebrew (not Yiddish) term for the plural of something related to
Cohan (a sect? sages? Talmudic scholars?). 
** Another Hebrew term, most likely.  Piety?  Conviction?

Peter Langston		]
-----------------------------------------------------------------------------

 From: David Grabiner <grabiner@zariski.harvard.edu>
 Subject: Talmud Fortran continued (was Re: b'Ezrat Hashem)

It's good to see that there are other scholars contributing to Talmud
Fortran, which first appeared on s.c.j. a few months ago with a claim
that COBOL was assur.

In article <1993Mar8.195355.7928@ttinews.tti.com>, Reid Kneeland writes:

> Also, for programs beyond a certain size, you have to set aside some
> code that is never executed, and say the bracha for challah.
> (Fortunately, the former tends to happen anyway.)  Israeli versions of
> lint recognize /*CHALLAH*/ as a synonymn for /*NOTREACHED*/.

This, however, conflicts with the requirement that any program to be
used in the Temple must be without blemish, and thus free of bugs.  At
first, programmers refused to write long programs for use in the Temple.
But the School of Hillel says that only the program which is used must
be perfect; the source code need not belong to the Temple.  Thus the
Sages allow a compiler which separates the Challah from the program, and
allows the free bytes to be used by the Cohanim.
-----------------------------------------------------------------------------

 From: faigin@aero.org (Daniel P. Faigin)
 Subject: Re: Talmud Fortran continued (was Re: b'Ezrat Hashem)

On 9 Mar 93 02:27:57 GMT, grabiner@math.harvard.edu (David Grabiner) said:
> This, however, conflicts with the requirement that any program to be
> used in the Temple must be without blemish, and thus free of bugs.  

This raises a question. As I recall, you are not (on certain days) permitted
to separate the good from the bad. How does this apply to debugging programs
on those days, or on using formal verification methods?
-----------------------------------------------------------------------------

 From: warren@nysernet.org (Warren Burstein)
 Subject: Re: Talmud Fortran continued (was Re: b'Ezrat Hashem)

In <FAIGIN.93Mar9084011@soldan.aero.org> faigin@aero.org (Daniel P. Faigin) writes:

>On 9 Mar 93 02:27:57 GMT, grabiner@math.harvard.edu (David Grabiner) said:

    This raises a question. As I recall, you are not (on certain days)
    permitted to separate the good from the bad. How does this apply
    to debugging programs on those days, or on using formal
    verification methods?

Since removing some bugs always entails inserting other bugs in their
place, it is permitted.

	"If there really was a Jewish conspiracy to run the
	world, my rabbi would have let me in on it by now.
	I contribute enough to the shule."
			    - Saul Goodman
-----------------------------------------------------------------------------

 From: faigin@aero.org (Daniel P. Faigin)
 Subject: Re: Talmud Fortran continued (was Re: b'Ezrat Hashem)

On Wed, 10 Mar 1993 06:49:26 GMT, warren@nysernet.org (Warren Burstein) said:

> Since removing some [program] bugs always entails inserting other bugs in
> their place, it is permitted.

I should note that it was pointed out to me that this restriction (separating
bad from good) only applies on Shabbat and festivals. Being Reform, I do use
the computer on Shabbat, but not to do work. Thus, I don't program, but I do
read news. This has the implication that one cannot use KILL files on shabbat
(although, thinking about it, you may never be able to use them because of the
torah injunction Thou Shalt Not KILL), unless they are run via crontab.
-----------------------------------------------------------------------------

 From: ask@cbnews.cb.att.com (Arthur S. Kamlet)
 Subject: Re: Talmud Fortran continued (was Re: b'Ezrat Hashem)

In article <3200@israel.nysernet.org> warren@nysernet.org writes:
>In <FAIGIN.93Mar9084011@soldan.aero.org> faigin@aero.org (Daniel P. Faigin) writes:
>>On 9 Mar 93 02:27:57 GMT, grabiner@math.harvard.edu (David Grabiner) said:
>    This raises a question. As I recall, you are not (on certain days)
>    permitted to separate the good from the bad. How does this apply
>    to debugging programs on those days, or on using formal
>    verification methods?
>
>Since removing some bugs always entails inserting other bugs in their
>place, it is permitted.

Do you really believe insertion of new bugs is performed with
sufficient kavanah to be permitted?
-----------------------------------------------------------------------------

From: jones@pyrite.cs.uiowa.edu (Douglas W. Jones)
Subject: Re: Talmud Fortran continued (was Re: b'Ezrat Hashem)

warren@nysernet.   faigin@aero.org wrote:                     To all this,
org answered                                                  jones@cs.uiowa.
thus:              This raises a question. As I recall,       edu replied:
                   you are not (on certain days) permitted
Since removing     to separate the good from the bad. How     So long as
some bugs always   does this apply to debugging programs      there is less
entails inserting  on those days, or on using formal          than one part
other bugs in      verification methods?                      in 60 of bugs
their place, it                                               in the code,
is permitted.  a_s_kamlet@att.com replied to this with this   it is kosher,
question:  Do you really believe insertion of new bugs is     so there is
performed with sufficient kavanah to be permitted?            no need to
                                                              deliberately
look for bugs to be removed; this greatly diminishes the severity of the
question faigin raises, but it also raises other problems.  If software is
created using an environment that has previously been used with treif
software, does the created software become treif even if it contains no
bugs?  How does one kasher one's software tools after they've become
contaminated?  This is particularly important for debugging tools, since
these are deliberately brought into contact with bugs and bug parts, and
therefore, I assume that they become treif with every use!
[=] © 1993 Peter Langston []