The Design of Software (CLOSED)

A public forum for discussing the design of software, from the user interface to the code architecture. Now closed.

The "Design of Software" discussion group has been merged with the main Joel on Software discussion group.

The archives will remain online indefinitely.

Software factories and supply chains

Just checked out the great interview Jon Udell did with Ward Cunningham and Jack Greenfield ( and have some comments / connections here ( )

What do people here think of the Software Factory story? It sounds like people have wanted this component market for a long time but never quite achieved it.

Is this new strategy of Microsoft's likely to finally make it reality? And if so, what are the implications for software design and the programming profession?
phil jones Send private email
Monday, November 22, 2004
Sounds like the same thing we've been hearing for decades. When Ada came out, all code would soon be built from "software chips".

We have certainly come a long way from the days of writing everything from scratch in assembler/COBOL/FORTRAN/C. What they're predicting doesn't sound all that new to me.
Tom H
Monday, November 22, 2004
But the point is MS claim they have a "solution" to the reasons why there haven't been decent componenet markets before.

Is that solution new? Is it plausible?
phil jones Send private email
Tuesday, November 23, 2004
Is it a factory or a company town?
son of parnas
Wednesday, November 24, 2004
Strange, I'd say there's already a successful component market for things like COM components (and now .NET components).  Granted, these are usually "small" components, not major chunks of functionality.

The Java world had a tougher time w/ a component market, first with regular JavaBeans, then with EJBs.  The idea was to have large, reusable chunks of functionality that businesses could just grab off the shelf and use.

I think the obstacle isn't technology, per se, but the whole notion of "reuse".  In my 14 years in the software industry, I've seen all kinds of attempts at reusable components, both internal to companies, and with 3rd parties.  Its a tough market, because designing a component for "resuse" is really, really, hard. 

There are certain areas, like a PDF library, or a database access framework that can be reused, but only because their domain is very well understood, and doesn't require unique customizations for the most part.

Reuse of "business logic" is the tough one, since no 2 businesses are exactly alike.  I remember a few commercial attempts at a "e-commerce" library, but no-one could quite figure out how to use such a thing without a small army of consultants to customize it.

Not to suggest it can't be done, but I don't think the technical issues are the reason it hasn't happened on a larger scale.
Dave C Send private email
Sunday, November 28, 2004
Dave, that's what I think is interesting about this new MS approach.

Their interpretation is that the difficulty is in finding the right *combination* of components to build a particular type of application. And so basically they're gonna build Wizards to help people select and combine them.

Of course, that's an idea which might turn out to be very badly implemented - think clippy for enterprise developers  (shudder) - or it just might work.

But at least it looks like a more-or-less *new* idea for trying to get a component market going.
phil jones Send private email
Monday, November 29, 2004
Well, I think the idea to make software more like hardware or like construction.

Assemble some standard industry parts just in time, switch suppliers, use Web services, etc.

So Microsoft as the Home Depot of software.  With helpful attendants to show you which parts you need to build your deck or finish that bathroom remodel.

Not sure, but I'll wait and see what comes of it...
Dave C Send private email
Tuesday, November 30, 2004

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics
Powered by FogBugz