Slashdot Mirror
Deferred IT Maintenance Is a Ticking Time Bomb
snydeq writes "The underfunding of routine hardware replacement purchases and the degradation of aging enterprise apps pose systemic risk for many IT organizations, thanks to a ballooning 'deferred IT maintenance debt' in the decade since Y2K fears pushed enterprises to invest heavily in essential system upgrades, InfoWorld's Bill Snyder reports. And with sysadmins 'scrambling to keep systems up and running with budgets that barely cover the basics,' this 'IT debt' promises only to increase in the coming years, especially as IT continues to defer routine maintenance in favor of new 'cost-saving' initiatives, particularly around the cloud."
Deferring any maintenance can have calamitous effects.
I fail to see why this is newsworthy? Is it just because IT people whine louder?
If you are in the US---just look around. Infrastructure systems are crumbling away because of "deferred maintenance". It's not just IT. It's roads, bridges, state governments, municipalities, houses, businesses---it'severything!
And with sysadmins 'scrambling to keep systems up and running with budgets that barely cover the basics,' this 'IT debt' promises only to increase in the coming years, especially as IT continues to defer routine maintenance in favor of new 'cost-saving' initiatives, particularly around the cloud.
The point of using "the cloud" (a hollow buzzword, I admit) is that you can offload the servers, software, and maintenance to a firm that specializes in such things. Theoretically, taking advantage of the cloud where it fits your organization will offset the "maintenance debt" problem. YMMV, of course.
Too many CIOs of too many western corporations report to the CFO, not the CEO. There are WAY too many CIOs who come into organizations with an eye, or a reputation, for cost cutting instead of tech innovation. Pick up any copy of CIO magazine and look at the toadies who make the top CIOs in the nation, and ask yourself - what innovation did they bring to make that list? What business process did they improve with tech? Only a handful make the cut. Most are there because they are good at pinching out costs, kicking out the older IT workers and either outsourcing or bringing in college grads.
I routinely see job ads for experienced Java developers, people with hard core experience in integration, esp. with telephony or security technologies, need 5-10 good years, offering $70k tops. Good luck with that, but again it is the CIOs who get the jobs telling people they can staff cheaper, run leaner, cut the corners - that get the job because it is the CFO who is doing the hiring and the performance reviews.
The big corporation IT C-level execs are a fear driven lot, there are no Gates or Zuckerburgs in their midsts. The action is being with the cloud providers, or the web service providers themselves. Enterprise IT is really a shit place to be outside China. It's a world full of EDS consultants and chickenshit CIOs who won't think how a business could use IT to expand. And the social media space is going to tear a bunch of them new assholes, because none of them know how to leverage it. The startups do.
I suspect two reasons: 1)(and most important): This is being published by Infoworld, ergo it focuses on IT stuff. 2) Much of the worst rot in IT is largely invisible to the layman.
Slow computers with styles that were pretty neato back in 2000 are obvious to the poor office drones who have to endure them; but anything that new can, largely, be forklift upgraded for the cost of the new systems and some grunt labor. Turning a 3 year desktop refresh cycle into a 5 year(or 7 year, *cough* *cough*) desktop refresh cycle doesn't make anybody happy(particularly once warranties run out, the scavenging and improvising begins); but is architecturally a small problem. You don't really accrue much "debt" over time. The cost will be "1 forklift upgrade to present day PCs" whether that upgrade takes you one generation ahead or three.
It's the complex software, the highly specialized proprietary industrial controller cards, and suchlike widgetry where there is real hell to pay, and most of that is invisible...
Just one minor gripe with the parent - a lot of times, what should be weeded out isn't the "one-offs" (which are often times built way under budget with way more capacity and way less maintenance cost), but the actual official enterprise standard that got put in because some CIO was buddies with some sales rep. "One-offs" are a signal that the current standards (either of technology, or product development), are having problems. While not all "one-offs" may be worthy of keeping, when going through the weeds, don't assume the enterprise standard is perfect, and don't assume the one-offs don't have something to teach you.
Examples of enterprise standards that should be weeded out where I work -> Lotus Notes, StarTeam, Windows XP.
How, exactly, do enterprise apps degrade?
Do they suffer from bit-rot, and have some kind of half-life?
I understand that eventually apps will fail to be supported by the developers, won't potentially work on more modern operating systems, and in some cases require updating in order to be able to work correctly with the rest of the world.
But it's a bit disingenuous to call this "degradation". The app continues to do what it always did. You're just wanting more out of it than you did before. The app didn't change, you did.
Please read my Canon EOS tech blog at http://www.everyothershot.com
Depends on the level of "bespoke" in your house.
Scavenging for desktop parts is the "little devil". Scavenging for people who know how the bloody things work more than 3 years after for IT systems is the real nightmare.
If a system has been in the field for 3+ year nobody knows what are its real dependencies and what does it really take to augment, add capacity or do any changes. The people who knew have left, gone to pastures new or have forgotten what the problems used to be and no documentation can help you here (even if there is any suriviving docs on the design of the system in question). This is valid for almost all classes of IT and telecoms systems and is the real cost factor in IT "maintenance debt". If we use a real-life analogy IT maintenance debt is like a discounted mortgage. You pay virtually nothing for 2-3 years and after that the lender skins your hide.
Baker's Law: Misery no longer loves company. Nowadays it insists on it
http://www.sigsegv.cx/
Cost savings is the biggest expense to any large organization that does it.
Oh Please! That is SUCH an easy one to fix! You either run XP Mode in Pro or just load up XP VMs. No you want to talk about "IT debt" try some of the places I walk into, where there is ALWAYS a "mission critical app" that is this horribly mangled piece of badly coded VB+Access mess of no comments anywhere junk, and then they expect YOU to deal with it! Hell one place I walked into in mid 09 had a NT 4 box running a VB3 "app" because each guy they brought in took one look at that beast and said "fuck that!".
Man I can hear the real programmers right now screaming out in pain just at the thought! You want to watch a "real" programmer wet his pants in fear you hand him a huge 14 page VB mess written by a half a dozen guys over the years, NONE of whom ever heard of a comment, with shit all over the place and nothing indented or even calling in a logical order, unless "insane band aid" is considered logic.
You want to know why there is an ever increasing IT debt I'd say that is a BIG part of it. All across the country you have this huge mess of apps written by some Joe Schmo that was bought ages ago and nobody knows how to live without and it DON'T run on anything but what it was written for and even then it is fussy as hell. And that of course don't even take into account the lovely crap like that ISA C&C controller written for DOS 3 that runs a $75,000 piece of machinery made by a company that has been DOA for a decade plus! I have stared into the abyss pal, and not only did it stare back it gave me the finger to boot!
ACs don't waste your time replying, your posts are never seen by me.
In the age of BS corporate leadership, who *doesn't* want to be the guy who cuts costs by 25%, gets promoted up into the suites, then lets one of his successors take the fall when the shit hits the fan? I'm more concerned with our public infrastructure BTW.
I swear to God...I swear to God! That is NOT how you treat your human!
Silly young grasshoppah. There is no "just do..." in IT. The mythical solution you're referring to is a cruel joke told by vendors.
huge 14 page
So, a little script, then?
Something on my table right now: 15k (in-file - and probably significantly more on the printed page) lines of PHP3 with nasty embedded SQL up the wazoo. It ties into half a dozen (literally, 6) other 'mission critical' applications and is customer facing as well as providing internal network management functionality. And this is small fry compared to some of the crap out there.
You want to know why there is an ever increasing IT debt I'd say that is a BIG part of it. All across the country you have this huge mess of apps written by some Joe Schmo that was bought ages ago and nobody knows how to live without and it DON'T run on anything but what it was written for and even then it is fussy as hell
I couldn't agree more. We've had entirely too many Boy Geniuses in decision making places in IT who think they've got something special and unique which will have Totally Awesome Results. They don't bother to think through their decisions.
The proper approach to something like this isn't to fix it. It's to replace it outright with something that does 90% of the task, better, with 50% fewer inter-dependencies by modularizing things as much as possible. Re-implementing, feature for feature, is quite often quicker. Just make sure you don't make the mistake of so many before you and re-implement it - poorly. If you can't do it, find someone who can.
IMO, the key to a successfully maintainable software infrastructure is to KISS and leave things as White Box as possible. When you can't keep things generic, you keep things isolated and modular. When you need something custom purpose that your users rely upon, you make damn sure it's standards based and that there are alternatives available.
(I don't even want to THINK about where we will be with things like Sharepoint in 3-5 years. Likely, another lengthy, drawn out, and costly migration project. This time, maybe back to something like, oh, NFSv5.)
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
Ohhhh...sounds like the abyss has given you the finger as well. And to be fair we are talking about 14 pages of badly written VB4 with about half of the pages nothing but fricking GOTOs bouncing all over the place.Now I'll admit when I've had to do something quick and dirty I've thrown in the occasional GOTO but the GOTOs all went to the same place whereas this massive pile of shite had more twists and turns than a bad detective novel!
But since you have seen the "true horror" of IT I'm sure you can see why I sometimes want to bash my head on the desk when dealing with FOSSies. They always seem to think "All you have to do is replace Windows and Office, and all will be hearts and flowers!" when it is NEVER Windows and Office that is the problem. Hell most companies can't even upgrade to the latest windows version for fear of that giant reeking mess of garbage code they've come to depend on will come falling down like a house of cards in a hurricane.
And while I agree 110% that the goal should ALWAYS be KISS, the problem is the PHBs at these places will never ever in a million years shell out what it costs to actually get all the data out and build a REAL solution, not until they have the crap they are depending on fall apart like some giant train wreck from hell. Like that VB4 app I mentioned early on, I ended up just jamming the thing in a Win2K VM and letting it rot for the next guy. Not because that is what I wanted to do, but because you couldn't pay a college kid what they wanted to pay to have it rewritten, much less get a REAL programmer of any skill.
And that to me is the problem in a nutshell and why the shit isn't gonna get any better for the foreseeable future. It is because businesses in the USA can't pull their heads out of the stock page to realize that working long term solutions cost money no different that decent roads or schools, and look how well THOSE are faring here in the US. IT has it even worse because as long as the thing works that day they don't care if it is a single power surge away from taking down the whole company. To them ANY expense related to IT is just a "waste" unless it is something like an iShiny for themselves or the CEO. That is why I ended up getting out of dealing with corporate, because I frankly got tired of people with impossible problems that wanted to pay pennies to fix years worth of neglect. I mean with the corporate attitude in the USA, is it any wonder nobody young is going into IT anymore? You'd have to be nuts!
ACs don't waste your time replying, your posts are never seen by me.