Selling software which is of some value

At, after working with a lot with software for over twenty years,  we have envisioned a clear idea about open source. This is the second article of a series tagged “opensource”: the previous one is Open Source for the infrastructure; in such article we started to do some tiny consideration of the destiny of big closed huge framework. Things are getting bigger and bigger, and maintaining big code library is a high cost for software house

In the 2008, when I look at software I ask only a thing: to be simple. Embarrassingly simple. The success of Mac stuff (like ipods, macosx and so on) is based vastly on simple design of ready to use, not annoying stuff.

For this reason, the software is not anymore the focus. Core software should be open source: things like word process are not the future of the business. Things like FaceBook are the future of business.

What you try to sell is your GUI design experience (see things like FogBugz or WordPress), your networking idea (like Facebook), your focus on search agents (like Google, because too much information is useless by definition).

When bad is good

We like WordPress, even if by a software point of view:

  • WordPress is rigid, it works only with MySQL
  • WordPress does a lot of external call, for instance using anti-spam or so on, so it is not secure 


  • WordPress GUI is done very very well: effective, and polite.
  • The Wordress crew ask you via polls what do you want
  • The WordPress site organize plugins, and themes
  • When installed you get the only needed plugin: a networked antispam, very very important if you plan to do blogs.
So Drupal is better, but for us WordPress is perfect.

When outsourcing counts

At we have decided to use FogBugz as bug tracking software. The reason are simple: it is cheap and easy to install.

Anyway FogBugz configuration feature are a less then Mantis.

But setting up Mantis required about 8 hours of work, a security assessment, while FogBugz was ready n less then 5 minutes. Fogbugz interface was carefully designed to be effective, while Mantis GUI is quite ugly.

The same considerations are valid also for something like Trac

Conclusion: do not be an engineer

When you design infrastructural service, think like an end user before designing your software.
When you design your GUI this is a must, but also when you design your core business this is important.

Engineers (like us) tends to solve problems in a cost-effective manner, which is good if you need to produce something, but it is bad if you want to sell it to a broaden audience.

Software is changing its skin: in the 2010, customers will enforce their expectations about what they call “User Experience”:

User experience design is a subset of the field of experience design which pertains to the creation of the architecture and interaction models which impact a user’s perception of a device or system. The scope of the field is directed at affecting “all aspects of the user’s interaction with the product: how it is perceived, learned, and used.” [1]

(from Wikipedia)