>But on the Mac, these files are already cr terminated and although they
>would require no conversion, would appear to be unchanged by any attempt
>at a conversion -- it is already Mac format.  Other ports would not
>include the Macintosh-only "????.TEXT" file information.   So, I'm not
>sure that changing the Mac materials would have any impact one way or the
>other on downloading Mac stuff.
>
>For the record, I've never had any problem with any of the Mac downloads
>whatsoever, and I've taken no special care, one way or the other, to
>download in "binary" format.

Well, a few weeks ago Hans-Martin Mosner writes:
 > Hi,
 > I just downloaded 2.4b from the UIUC file server, and the image and
 > changes file don't seem to correspond. I remember that there was some
 > discussion on the list and that 2.4b was supposed to have fixed that...
 > So, it is not fixed. Just open a fresh 2.4b and look at
 > StringMorph>>drawOn: to see what I mean.
 >
 > Hans-Martin
 >
I then check with my copy, downloaded in a Mac in binary .SEA format,
running in a Mac and the function was not pointing to the correct source.
There are two explanations:
1.- The .changes file have this problems in first place.
2.- Maybe a few LFs was entered in the file and because the file is of type
'TEXT' then in the process of compress/decompress the LFs were deleted
changing the offset of the source.

Changing to a binary type remove this last posibility.