Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!seal.cis.ohio-state.edu!ogden
From: ogden@seal.cis.ohio-state.edu (William F Ogden)
Newsgroups: comp.sw.components
Subject: Re: What is a reusable software component?
Keywords: software component definition
Message-ID: <82559@tut.cis.ohio-state.edu>
Date: 30 Jul 90 22:43:53 GMT
References: <27705@athertn.Atherton.COM> <533@helios.prosys.se> <27914@athertn.Atherton.COM>
Sender: news@tut.cis.ohio-state.edu
Reply-To: William F Ogden <ogden@cis.ohio-state.edu>
Organization: Ohio State University Computer and Information Science
Lines: 50

In response to comments from Anders, Paul writes:
    ...
>>* browser
>>
>>  That is, providing a `bt_user_search(tree)' which presents the data
>>  in the tree, and lets the user select an item for further computation.
>>
>>This is something I think most components that implement data structures
>>should have. At least those implementing `complicated' structures.
>
>It is debatable whether or not a component should provide a browsing
>capability or not.  It has no way of knowing what the client has stored
>there.  It could, however, provide bt_getFirst, bt_getLast, bt_getNext,
>and bt_getPrev functions that work in conjunction with the bt_search
>function to provide the functionality needed for the client to locate
>information and provide a more effective browser.

Now here we are getting into a real reusablity issue. In designing a
facility for reuse, it is important to try to distinguish between
primary and secondary operations on a structure. The Get_First, 
Get_Next, etc. are primary operations on the reusable structure of
which the B-tree is one realization. Traverse or print-out operations
are secondary, since they can be implemented easily and efficiently using
the primary ones -- provided Get_Next, etc. are included as primaries.

There are generally a vast number of potentially useful secondary 
operations on a structure, but any particular application will
only require a small subset. Moreover the next application may
need other strange secondary operations which nobody ever considered.
Under these circumstances, achieving facility reuse without modification
seems paradoxical until you see that with proper design, secondary
operations can be added without changing the underlying implementation.

Determining which operations are primary and which secondary is
somewhat like finding a basis for a vector space of an independent
set of axioms for a mathematical theory. You list all the operations
on a structure that you can imagine anyone wanting (of course you'll
turn out to lack imagination) then you figure out which ones can
be implemented using which others. A small (and generally useful)
subset, in terms of which all the others can be implemented, form
the nucleus of the reusable facility.

>It is also debatable as to whether or not a data structure such as a
>B-tree should provide the user interface code needed to support such a
>browser. 
 
Clearly it shouldn't. It involves either a set of secondary operations or
more likely a separate facility.

/Bill