Thread: Scan support
View Single Post
  #31  
Old 09-21-2007, 02:52 PM
Mark Walsen (markwa)
Guest
 
Posts: n/a
Default Hello Barrie and Clifton, I

Hello Barrie and Clifton,

I share your disappointment that MusicXML import is not yet included in Composer.

The priority of new features is determined basically by this forumula: P = (I x N) / C
where
P = relative priority, higher values are of higher priority
I = the importance of the feature to those customers who would benefit from the feature
N = the number of customers who would benefit from the feature
C = cost to implement the feature

The I (importance) value of the MusicXML feature is clearly very high to folk who want to import scanned sheet music into Composer. The Importance will also be high to folk who want to work in Composer with MusicXML files handed off by other notation programs that export MusicXML.

The N (number of customers) who find this important is somewhat small. My guess is that only 1% of current Composer customers would find MusicXML importing useful; and adding that feature to Composer would attract only an additional 1% or 2% of customers. That gives a total N value of 2% to 3%, which is actually high for a feature that has such a high I (importance) value for the customers who care about the feature.

The C (cost) of this feature is pretty high, even though I've already implemented about 50% of it. That's the catch. It brings the total value of P = (I x N) / C down below some other important features I've been working on for the last half year and will soon announce.

Another problem here is that whereas I have been able to hire other developers to do other large chunks of work, MusicXML importing is a feature that I can do 10 times faster than a hired developer, because it is so intimately related to the core of Composer's programming code.

I suspect that this overall P = (I x N) / C will change, though. MusicXML has already achieved widespread acceptance in the music software industry. Although I do not anticipate that there were be a growing number of users who want to scan music and import it into Composer, I indeed anticipate that there will be a growing numbers of users who want to import MusicXML that has been exported from other notation programs. That will increase the N value in the formula. Also, I'm exploring finding another developer to help me with "intimate" programming in Composer. That could decrease the C (cost) value, if I don't have to do that work.

The above is a long explanation, but it is my first exposition on how I determine priorities. I might publish this priority method in a more prominent spot in the forum... unless I get feedback that my method of prioritizing is all wrong ;-)

Cheers
-- Mark
Reply With Quote