Is all that equipment in your data center going to be as obsolete as that Univac I, I first wrote software for in 1958, and if so, when?
Surely it will it be very obsolete in 50 years. How about 50 months? Fifty weeks? Fifty days?
You might be thinking "It is already obsolete. It's expensive and difficult to maintain. It lacks adaptability and I'm locked into vendors I don't like."
We assume all platforms are, or will become, obsolete. After all, in 40 years of business we've seen many come and go. Watch out if a vendor tells you that their plan is to have the best Mac or Windows based newspaper system. Being tied to a chip architecture, proprietary operating system, or any particular platform technology a third-party controls builds in obsolesce. It's eventually bad for you and vendors.
Customers on SCS support get regular software updates as part of the subscription service. Switching to a new supported platform is included at no additional charge for new software versions. By applying agile software development methodologies, we provide new releases early and often. There have been over 400 for Layout-8000 alone since 2006. (In contrast, one group executive told me that they were paying expensive support for an IBM mainframe-based mission-critical application for which they hadn't gotten a new software version in two years. Quite frankly, this sucks.)
You might wonder why we ported our software to ARM compatible devices, like the Raspberry Pi. No platform stays top dog for long. Pi's might just be the next great thing. Whatever that might be, we want our software to be portable enough to run on it. Thus we usually have a current reference platform (Intel/Linux) and a new one in the wings taking part in the nightly builds - just in case.
We've seen some platform solutions we don't like. For example, I'm not a fan of VMware. It solves a problem. Unfortunately, it carries along with it a number of significant others.
Say you have a room full of old, hard-to-maintain servers, perhaps each running a separate application. Get a big blade server. Put VMware on it. Move old applications and operating systems to VMs, and voila, the old equipment goes and everything left looks new. And isn't the load balancing great?
To bad it falls short on doing backups and fail-overs, keeps obsolete insecure and vulnerable operating systems in place and, for at least our purposes, does little better than free open source solutions, like KVM, which comes standard with Linux.
I don't get it. Computing power is cheap and is continually getting cheaper. Why pay lots to share computing power this complex way? The scary part is when you witness, as we have, how such complex platforms often are the source of more problems than they solve. Preservation of obsolete legacy systems, even (perhaps, especially) when virtualized, is dangerous.
The correct solution is always doing needed upgrades. We don't want to just provide frequent updates, we want to deploy continuously.
Newspapers are a declining industry with a legacy IT infrastructure. Much of this now limits essential productivity growth. Being agile enough to change, even with limited resources for implementing change, is key to adopting smart solutions for doing more with less.
Richard J. Cichelli
Send email to me to comment
What is implied by a "doing more with less" philosophy?
What constitutes less?
Picking appropriate technology enables business transformations
Paying lots to share computing power in a complex way?
A machine intelligence revolution that will rival the industrial revolution? (Warning. About 4 minutes.)
Blueprint for SCS enhancement of its search technology