On 30.12.2010 14:17, Makarius wrote:
Last summer I've made another round in clearing out the confusion,
working towards a stateless model based solely on the master directory,
which is the location of the enclosing theory file when traversing the
import graph. Thus the implicit use of current dir
On 31/12/2010, at 12:17 AM, Makarius wrote:
> On Wed, 22 Dec 2010, Gerwin Klein wrote:
>
>> I would think that only one search path is necessary, but I don't understand
>> what is meant by master_dir, I missed that part of the discussion.
>
> This thread is getting almost as entangled as the h
