Enough of 20, already
It would be fair to say we have had a difficult year. No individual has kept their routine, no family has avoided some upheaval or distress, no neighbourhood is without a sad tale, no business has survived unscathed – all of them affected by the most unpredictable phenomenon of modern times. Yet at the heart of all of this, banks still lend money, insurers still insure, retailers sell (albeit online), logistics businesses still ship packages, software still runs, mainframes still run the economy, and COBOL is crunching the numbers and managing the data. The only constant – apart from change – is how core business systems work.
Where did you say?
We heard that – interestingly – as everyone rushes towards a hybrid future in which trusted technology such as the mainframe rubs shoulders, quite happily, with newer platforms such as Cloud, that COBOL actually does not mind or care where it runs. So often that’s of course the mainframe - Z, but in a hybrid connected world that workload is also found on LinuxOne, Linux, AIX, Windows and all shapes and sizes of Cloud (for example, here’s an AWS hosted COBOL testing blog).
Where to use COBOL should really be a business decision; however, whether it works there is a technical certainty.
Why COBOL rocks (still)
Earlier in the year, initial reports of COBOL’s part to play in failing to assist with the pandemic were swiftly debunked by pieces such as Wired’s Can't File for Unemployment? Don't Blame Cobol, and Hackaday’s forthright COBOL isn’t the issue – a misinterpreted crisis. At Micro Focus – we agreed wholeheartedly – here’s what I said at the time.
In broader terms, the global COBOL community stood up, too. The mainframe community SHARE sought input from a variety of sources in the excellent piece - COBOL, a cornerstone of the mainframe. Further articles on the topic on the SHARE web site cemented that view.
Elsewhere, the Guide SHARE Europe (GSE) organization took an equally strident view – their take on the COBOL situation was to commission a session on COBOL as part of the GSE Virtual Conference in 2020 (entitled “Virtually Unstoppable”). I was delighted to give a keynote presentation on the language (“COBOL in its 7th decade”).
The Open Mainframe Project, the open source community for the mainframe, entered the fray earlier this year. A new Working Group for COBOL, which aims to promote the viability and vitality of the language with factually accurate information about usage and value, was established. Spearheaded by experienced COBOL practitioners and commentators. As founding members, Micro Focus led the keynote discussion from the recent OMP Summit, which discussed the new COBOL Working group. We also joined a video interview the Working Group held with popular open source and Linux media frontman, Swapnil Bhartiya.
Making the conversation truly mainstream, however, needs broader press interest. Forbes’ article “COBOL language: call it a comeback?” was as insightful as it was upbeat about its credentials, as was Technology Historian Mar Hicks’ absorbing and entertaining webcast, “Scapegoating COBOL”.
The upshot? Quite a busy year for the now 61-year-old language. COBOL made into the top 10 computer language searches on Twitter, according to the IEEE, while it retains its top 30 place in the TIOBE index. Another important indicator is the popularity of the Facebook COBOL Programmers Group, which in the last couple of years has nearly doubled to a very active 17,000+ members.
Ignoring Fake News
In a 2020 survey, 92% confirmed COBOL systems remain strategic. That is a convincing vote of confidence. As the excellently researched article entitled itself, this year we heard many voices explain, “Why COBOL will never die”. Another article laid bare a hitherto taboo topic, “That older code can not only be good, but in crucial ways superior to newer code, is at odds with a lot of Silicon Valley myth making”.
Such sentiment is a timely reminder to those who might predict or encourage the demise of this most trusted of technologies. Standish Group’s Chaos Report, drawing up the multi-decade research of over 50,000 participants, reported this year that those “replacing a software application and starting from scratch had a 26 percent success”. If you want to know how tremendously unreliable throwing away perfectly good systems can be, listen to this webinar.
In 2019, in its 60th anniversary year, I referred to COBOL the bedrock of digital transformation. A year later, and a tricky year at that, I see no reason to change that perspective.
Here’s our view as to why COBOL remains the foundation of core IT application modernization.
[A version of this blog first appeared on mainframedebate.com in November 2020].