I’ve been catching up on some reading and found the comments to a post from Dana very interesting.

Here’s a summary (although the comments are worth a read):

  1. Dana writes about Zenoss, an open source IT management vendor
  2. Zenoss CEO, Bill Karpovich, comments that: “(1) we provide strong coverage of what organizations are *really* looking for and are expanding rapidly (2) we will never do 100% of what the Big 4 does since much of it has resulted in bloated, shelf-ware, and (3) we will do many things that they can’t/won’t do.”
  3. Tarus Balog, a contributor from a competitive open source IT management provider (OpenNMS) questions whether Zenoss is really enterprise ready

Okay, is some software bloated? Sure. Are there features inside of <insert_your_fav_sw_here> that you don’t use? Sure. But I think the Bill’s comment on “bloated, shelf-ware” is a bit broad. Maybe he meant to apply his analysis against small or medium sized customer requirements?

Good product management practices don’t usually allow for adding a feature just to beef up the “New Feature list”. Most of the time the feature is something that a large number of customers or a small number of your best customers have asked for. Product managers don’t just dream up these new features (we’re not that smart or creative). Just because a feature is not used by the majority doesn’t make it useless to the majority. A friend used to format headings in MS Word document manually (i.e. highlight, underline, bold, change font size, find next heading and repeat) until I showed him the “Format Painter” icon. Until I showed him what that icon did, it was a “useless feature”. The ability to cluster a web application is “not necessary” or “critical” depending on the customer type and specific project. The ability to consume a QuickBooks data resource is “not necessary” for Citibank, but critical for a smaller sized company (a plug for another friend).

Incidentally, I haven’t heard anyone complain about “bloated” automobiles. I’m not talking about form factor; I mean “too many features” inside this thing that should just take me from A to B. Do we really need rain sensing windshield wipers, multiple temperature zones inside the car, or headlights that pivot during turns? Why don’t we vote with our wallets for manufactures to build something more like the Polski Fiat 126p?

Different customers require different features for different uses. Heck, they even need different levels of quality within a given feature, i.e. I’m happy with a car that can go from 0 to 60 period. My friends in the automotive industry care about going from 0 to 60 in less than 5 seconds.

Let’s be careful when claiming that open source software will cure the world of bloated software without understanding the customer types and use cases that the software is aimed at.

[The pic is from Flickr user sirronwong]