See this article: "The COBOL Brain Drain" in ComputerWorld. This article is very, very well written and repeats a number of common fallacies.
The fallacies lead to expensive, labor-intensive in-house software development and maintenance. Either there's a lot of effort poking at the COBOL code. Or there's a lot of extra code to "wrap" the COBOL code so that it's never touched.
"Migrating large-scale systems built in Cobol is costly and risky." A popular position. But the risks are actually quite small; possibly non-existent. The risks of not converting are usually higher than the risk of conversion.
The perception of the COBOL code is that it's filled with decades of tricky, nuanced legacy details that are hard to fully understand. This is only partially true.
A great deal of the tricky code is simply redundant. COBOL is often written with copy-and-paste programming and blocks of code are simply repeated. It's also important to note that some of the code is no longer exercised in the first place.
Mythical Risk
The "risk" comes from the perceived opacity of the tricky, nuanced legacy details. It doesn't appear to be clear what they mean. How can a project be started when the requirements aren't fully understood?
What appears to be the case in reality is is that this tricky code isn't very interesting. Most COBOL programs don't do much. They can often be summarized in a few sentences and bullet points.
Detailed analysis (641,000 lines of code, 933 programs) reveals that COBOL programs often contain several commingled feature sets.
- The actual business rules. These are often easy to find in the code and can also be articulated by key users. The real work is usually quite simple.
- A bunch of hackarounds. These are hacks to work around bugs that occur elsewhere in the processing. Sometimes a hackaround introduces additional problems which require yet more hackarounds. All of this can be ignored.
- Solutions to COBOL data representation issues. Most of these seem to be "subtype" issues: a flag or indicator is introduced to distinguish between subtypes. Often, these are extensions. A field that has a defined range of values ("A", "C" or "D") has a few instances of "*" to indicated another subclass that was introduced with a non-standard code for peculiar historical reasons.
Once we separate the real code from the hackarounds and the representation issues, we find that most COBOL programs are trivial. About 46% of the lines of code (74% of the distinct programs) involves programs that read one to four files to write a report. In effect, these programs do a simple "relational join" or query. These programs have single-sentence summaries.
The hackaround problem is profound. When looking at COBOL code, there may be endless copy-and-paste IF-statements to handle some issue. There may be whole suites of programs designed to work around some issue with a third-party package. There may be suites of programs to deal with odd product offerings or special customer relationships.
The remaining 26% of the non-trivial programs follow a normal distribution of 1/4 simple, 1/2 moderately complex, and 1/4 regrettably and stupidly complex. That final 5% of the programs will also be a whopping 20% of the lines of code. They are the few big programs that really matter.
Risk Mitigation
The risk mitigation strategy involves several parts.
- Data Profiling. The COBOL data may have considerable value. The processing is what we're trying to rewrite into a modern language. A profile of every field of every file will reveal that (a) most of the data is usable and (b) the unusable portion of the data isn't very valuable.
- Triage. We can summarize 80% of the code in simple sentences. 46% of the code is single-sentence summaries. 34% of the code has multiple sentence summaries. The remaining 20% requires in-depth analysis because the programs are large; they average of 2400 lines of code each.
- Test-Driven Reverse Engineering. Since a 5% of the programs do the real updates, it's important to collect the inputs and outputs of these few programs. This forms a core test suite.
- Agile Methods. Find the user stories which really matter. Find the COBOL programs that implement those user stories.
The most important risk reduction strategy is to take an Agile approach. We need to prioritize based on value creation. All COBOL programs are not created equal. Some fraction can simply be ignored. Another fraction produces relatively low value, and can be converted last.
The second most important risk mitigation is to build conversions from legacy COBOL files to the new, preferred databases. The data is preserved.
There's almost no risk in rewriting the 46% of low-complexity COBOL lines of code. This code is trivial. Note that some of this code actually has no business value. Simply ignoring the no-value code also reduces risk. Since we're using live files to drive testing, we can easily prove that our new programs work.
It's not risky to focus on the 20% of high-value COBOL lines of code. This contains most (or all) of the processing that the business needs to have preserved. They can articulate the user stories; it's easy to confirm that the COBOL does what the business needs. It's easy to use live data to drive the reverse engineering.
The remaining 34% of the code base may actually involve a small amount of overlooked complexity. There may be a nuance here that really matters.
This overlooked "nuance" is something that the users didn't articulate, but it showed up in our unit testing. We couldn't reproduce an expected output because we didn't correctly locate all the processing steps. It wasn't in our summary of the 80% of moderate-to-low complexity programs. It wasn't in our detailed analysis of the 20% subset of hyper-complex, large programs.
We've reduced the risk by reducing the volume of code that needs to be examined under the microscope. We've reduced the risk by using live files for parallel testing. We've reduced the risk by collecting user stories.
The remaining risks are ordinary project risks, unrelated to COBOL or legacy data.
The Other Great Lie
Another popular fallacy is this:
The value-add is to create a nimble business. Purging the COBOL has a lot of advantages.
- It reduces the total number of lines of code. Reducing costs. Improving time-to-market for an IT solution to a business problem.
- It reduces the number of technologies competing for mind-share. Less thinking about the legacy applications is less time wasted solving problems.
- It reduces the architectural complexity. If the architecture is a spaghetti-bowl of interconnections between Web and Legacy COBOL and Desktop, then following the spaghetti-like connections is simply a kind of intellectual friction.
The COBOL does not need to be purged all at once through a magical "big-bang" replacement. It needs to be phased out.
Agile techniques need to be applied. A simple backlog of high-value COBOL-based user stories is the place to start. The prioritization of these stories needs to then be clustered around the data files.
Ideally all of the programs which create or update a given file (or related group of files) can be rewritten in a tidy package. The old files can be used for Test-Driven Reverse Engineering. Once the programs have been rewritten, the remaining COBOL legacy applications can continue to operate, using a file created by a non-COBOL application.
Each file (and related cluster of programs) is replaced from high-value down to low-value. Each step creates a more nimble organization.