The theoretical models of boy friend selection merit some discussion and analysis.  It seems to me that this would be very profitable in the appropriate selection of such useful companions, and to reduce the likelihood of any understandings later on.  It is my feeling that women implicitly operate on the selection and maintenance process by having adopted one of several models:

One of these is the Good Crystal Selection Model.  This model views boyfriends as selected usually with care, and maintained over long units of time.  Doing this is labor-intensive in the selection process, and it has problems in that you cannot factor in depreciation or maintenance costs when this one is used.  Also, the use of this model does not seem to be effective for those under age 25 or so. 

Of course, you might have adopted the Unfortunate Wedding Gift Model.  This model also assumes long-term or permanent status, but without the value and appeal of the Good Crystal Selection Model.  In this case, you feel that are stuck with him and make the best of it.  In a word, strenuously resist any aunt who wants to set you up with a "fine son of her friend."

There is something to be said for the Faucet Water Filter Model.  When this one is adopted, you assume that he will be your boy friend for a limited, specified amount of time (say three months), and will then be replaced with a new one.  I am told that this is commonly done by Hollywood actresses and pop music singers.  As the song goes, "We'll sing in the sunshine."  However, maybe those selected for a year or longer should come with a warranty.

Recent news has suggested that a Individualized Module Selection Model that is in the process of being developed by Microsoft may become available in a few years.  This one allows you to have built in the specific features you want.  On the face of it, it promises some big advantages.  Let's see, you can get one who is sensitive, cultured, athletic, good-looking, loves animals, thinks you're wonderful, etc., etc.  It will be at least a decade before this will be available, and it will likely have problems in initial costs, particularly for Version 1.0.  Some additional things to be concerned with is that the program may tend to crash inopportunely and that the features likely to be offered are likely to be desired primarily by a Pacific Northwest clientele.

I propose, however, the Bra Model of boyfriend selection.  Consider this: we have quite a bit more than one bra, why not have more than one boyfriend?  Having three or four might be advantageous.  Now, there's lots of ways that this could be implemented.  One way is with each one having his own day, selected by lot.  If this is done, then you have a boyfriend for Monday, another for Tuesday, etc.  My needs are modest: I could be satisfied with four different ones, to be used in strict rotation (with a relief pitcher to come in from time to time when needed.)  After all, I need some time en famille and to keep up with my studies.

Alternatively, you could have appropriate boyfriends to fit your moods.  Now, again the lingerie metaphor is applicable here.  In the old days, there were only two basic colors: white and black.  These colors have their counterparts with the "good-boy" (white) -- "bad boy" (black) dichotomy.  This was good in the old days: you have one to misbehave with, and one to show up with on Sundays at Maw-maw's.  However, today, there are a variety of bra colors and types to fit your mood and clothing as well as your figure.  On some days, cranberry is called for, and sometimes bamboo!  Why not have boyfriends likewise?  You could coordinate them to go with your lingerie moods.  Or you could have different ones for different activities.  Let's see . . . . you could have one for hanging out with in coffee houses, one to go to the races with, one who likes to dance, one who likes to fish, one to go to "show up and look formidable" occasions, one who likes chick flicks . . .