Seems like Fed IT innovators are tied to the rails as the big-iron locomotive careens down the track. A new MeriTalk study, “Innovation Inspiration: Can Software Save IT?,” tells us what we already know - we spend too much money updating and maintaining legacy systems. But, this isn’t all a doom-and-gloom story - there’s a glimpse of a super-hero option.

We’ve reached the halfway point of 2013, which also signals the beginning of the federal fiscal year end (FFYE). This is a good time to take a second and survey what we’ve seen so far and what we can expect over the next six months.

Constantly changing requirements are the bane of production at scale. Finding ways to balance the needed stability of a production environment while handling bugs and enhancements is the key challenge in operating and developing enterprise software.

tl;dr: DB and RPM storage are the big system design concerns, not memory or OS install. Average channel size is up to 15K packages at 250KB per, meaning 3.5GB of DB table space per channel. Actual RPM storage is about 30GB per channel. Make sure you can grow each partition easily.

I'll be posting a series of articles on tips, tricks and considerations on Red Hat's RHN Satellite system life-cycle management tool. Each new post will get a link here, so they'll be easy to find. I'll be putting things I've learned from my own mis-steps, inside track from friends inside Red Hat, and perspectives gained from talking with others who work with RHN Satellite on a daily basis.

Government IT managers lack the in-house resources to expertly implement IT software, hardware, and systems. And with more budget cuts on the horizon, that gap will only widen. Earlier in the year, we expanded our Professional Services offering to help our customers.

Adaptive brake lights are a great feature hampered by the fact that the real audience doesn't know it exists and doesn't know how to interpret it - expertly specified, elegantly executed, completely useless. How did this get from need to requirement to execution, yet turn out so badly? For obvious reasons, my mind went immediately to my 10+ years of system operations experience, supporting web applications.

Good post from Dan Risacher on acquiring systems versus providing services in the DoD. Applies widely to all government acquisition and dovetails with my views. Maybe that's why I like Dan.

In October, I used “We the People” as an example of how to get citizens engaged with government in an open manner. In November, those engaged citizens petitioned the government to consider building a Death Star. By January, enough signatures had been gathered to garner the administration's consideration and, in my mind, a well authored response. The exercise may have been a geeky back and forth which you may see as a joke, but I feel any citizen engagement is good engagement. Here's why.

In the final part of this series, DLT Solutions Engineering Team Lead, Matt Micene assures us that there is a happier future for Enterprise IT on the horizon.