Is Chronodebt ever a good idea?
Chronodebt is the accumulated cost of remediating all IT assets that aren’t what engineering standards say they should be.
It’s what most of us have been calling “technical debt,” and I would too except that Ward Cunningham and his fellows at the Agile Aliance have already claimed it to mean “the implied cost of additional rework caused by choosing an easy solution now instead of using a better approach that would take longer.”
Anyway, the short answer is yes. Taking on Chronodebt is, in many circumstances, exactly the right choice. The conundrum preceded the advent of business computing by at least a century, when, during construction of the transcontinental railroad, the Southern Pacific Railroad, was faced with a shortage of hardwood railroad ties.
And so, instead of waiting for enough hardwood ties to continue construction, it took on massive Chronodebt in the form of ties made of cheaper and plentiful cottonwood (“Cottonwood now or hardwood too late?“, KJR, 4/28/2003).
The cottonwood ties would only last a few years, but with them the company could generate enough revenue to replace them. Without them it would never have completed enough track to sell a single ticket.
The significant difference between this and most modern companies’ Chronodebt is that the Southern Pacific Railroad paid its Chronodebt off.
Chronodebt, like most other forms of debt, is neither good nor bad as an absolute. As with most other decisions, context matters. In the case of Chronodebt it all depend on what stage of concept development you’re in.
Imagine your “concept” barely deserves the term — it’s really more of a notion. You think it has promise, but you don’t have much in the way of supporting evidence to support it.
It’s time to bet the farm!
And it is, but only if you’re betting someone else’s farm, “someone else” is someone whose friendship you don’t value very much, and you’ve checked with your lawyers to confirm you aren’t at risk when someone else finds themselves farmless.
If it’s your kids’ college fund it’s time to launch Excel, or maybe Access, or an ISP’s generic eCommerce development kit.
If it isn’t all about you … if we’re talking about a corporate setting and it’s a proposal to try something new and different for which there isn’t and can’t be much data to bring to bear on the decision … then it’s still time for Excel, or maybe Access. Or, because it’s a corporation, perhaps SharePoint, or some SaaS product whose licensing terms aren’t too expensive and onerous.
It’s time, that is, for Chronodebt, because doing things the so-called “right way” probably means missing the opportunity altogether. And in fact we might not be talking about Chronodebt at all. Chronodebt in this situation comes from the danger of success, because it only has to be paid off if the idea pans out. Success is, to push the metaphor to the breaking point, the usurious interest rate charged for underinvestment, which wasn’t underinvestment until success happened.
Chronodebt is a good idea during the exploratory phase of innovation management. It’s a bad idea when innovations start to prove out. That’s when it’s time to replace the kludges and prototypes you built the new concept on with more robust and scalable alternatives … time, that is, to pay down the debt, which means investing in sustainability.
That isn’t the whole story, though.
There are times when a company’s whole business model starts to approach its use-by date. Imagine, for example, you’re CEO of a metropolitan daily newspaper and your presses are a major source of corporate Chronodebt. Time to pay it off by replacing them with something more modern?
Probably not. Like it or not (I don’t), newspaper print circulation has been steadily declining for decades and the more important metric — advertising revenue — is in even sharper decline. The best and most advanced presses money can buy won’t sell a single additional newspaper, or, more importantly, attract more advertisers.
As CEO, you tell the god Chronos to take a hike.
If, on the other hand, your on-line news site or mobile app are Chronodebt-bound, that’s another story entirely.
None of this is particularly complicated. And yet, especially in IT circles, we do have a tendency to consider engineering excellence to be an unalloyed and immutable good.
Sometimes, prototypes and kludges are exactly what the situation calls for.
And sometimes the right answer, although painful, is limping along on your ancient legacy systems until they crumble into dust.
As is always the case with your articles, thoroughly insightful.
I would just add that the choice of Excel or Access in prototyping isn’t mutually exclusive, because they serve two different purposes, or else Microsoft wouldn’t’ be bundling them together.
Excel is best for computations, due to its rich function set and Equation Solver – which also does Linear and Non-Linear Programming.
Access is optimized for storing data that can be indexed for fast access and manipulated by SQL (which, despite Access’s SQL dialect being frozen in time as of 1992, isn’t that difficult to learn – Access’s Query By Example makes it the best tool for beginners to learn SQL).
More important, Excel can link to Access data, so you have the best of both worlds. In fact, many business schools (including the one in the university where I’m based) teach the use of MS-Office in dept (which means knowledge of VBA and the object modes of the component packages), so that Access can also split out MS-Word form letters or Outlook E-mail content (e.g., for “past-due” notices) or monthly statements to customers).
There are national-level contests (sponsored by Microsoft, I presume) where teams from each school are given the detailed specification of a problem to which they have to design a prototype, but fully functional, solution within 8 hours.
As you accurately pointed out, Office-based solutions don’t scale to a large enterprise, but as a prototype (or even a complete solution for a small business) it lets you hit the ground running.
So true: “Sometimes, prototypes and kludges are exactly what the situation calls for.And sometimes the right answer, although painful, is limping along on your ancient legacy systems until they crumble into dust.”
It also helps if there is management acknowledgment that those are deliberate decisions and not just because someone dropped the ball.
Part of the problem is that it is some times hard to calculate the cost of this “debt”
In your printing press example you left something out. That is maintenance. As things like printing presses age their maintenance costs go up. And after a while the parts become less available. Then there is the issue of up time. And so the cost of keeping the presses operational gets higher over time. And this can become a significant value.
Take your cotton wood tie example, they could have replaced them with more cotton wood. And there were probably ways to make the broken one continue to operate. And those at times were probably being used. But the cost of doing that will go up over time.
At some time it may not even be possible to replace the unit at all. This happened with a client of mine. They had been informed that their existing electrical meter system was kind of failing. Not really failing, but was weak and they should plan on upgrading because it was going to be failing one day soon. Well that day came. And so they tried to order a new meter base. Except that type of meter base was no longer available. In fact nothing that would allow the use of the existing main feed wiring to their breaker boxes was available.
The incoming electrical wiring was so old that they no longer made meter bases that did what this one did and which was wired the way this one was. The expectation was that when it failed they would simply buy another one and have an electrician install it. Like a few hours, much less than a day. No to get power to the various parts of the plant they had to start at the utility poles and have the utility run them new runs in to a completely different location. Then a modern style meter base had to be installed. Then electricians had to install new wiring from the meter base to the breaker and fuse boxes. They were down and without power for almost a week. Had they known that this was the real cost of the old equipment they could have planned the new upgrade.
To do the upgrade on a less priority basis would have been much cheaper. And while they were doing it they could have planned to removed the fuse boxes that still existed and replace them with breaker boxes. Instead of almost a week of the facility being shut down, they would have had the few hours that they thought it would take. AND that could have been scheduled and probably done after hours.
So at times you don’t know the real “chronocost”. In the printing press example. One thing, do they could they have any other printing customers. Could they have more if they had better capacities. Most newspapers I know of run their presses on other jobs more than they do for their news paper. One of them I know of has the entire system duplicated and normally never uses one for their use, but only rents it out for production of materials with a long lead time. So if there is a failure in the primary set they can go to it and still meet all the needs.
But assuming that they don’t do outside work, what would the cost be of a several day shutdown if something major fails on the press and they can’t just order another one. One client has an old piece of equipment. And a couple times it has cost them some business because they were down for several days while they took the old piece to a machinist and had a new one machined from scratch.
Then do you know the cost of replacement. I know of one incident with a university. They WWII time frame had a piece of equipment designed and then a computer to manage and control it for the tests being run. They spent a million in that day’s money to build this. Everyone just assumed that is what it would take to replace it. So they had a staff of four people to acquire and replace vacuum tubes. A few times a tube was no longer even available, so they worked with the electrical engineering department to come up with a circuit to replace the tube. They had no plans for this thing so if something failed they really had trouble actually finding and replacing the failed part.
Then the fateful day came when the transfromer for the building was struck by lightning. Stuff in the building was like destroyed. Light fixtures, the refrigerator. All gone. And there was enough melted parts and missing parts that it was impossible to repair the million dollar, today multi million dollar computer system. Luckily the equipment it was attached to was beng worked on and for safety electrically shut down, so little damage there. It was researched into how to replace this thing. People started looking into what grants were available to fund replacing it. They ended up about three months later with a apple IIE. Well and the four guys were transferred or retired.