THE OBSOLESCENCE OF GOVERNMENT INFORMATION AND COMMUNICATIONS SYSTEMS
THE OBSOLESCENCE OF GOVERNMENT INFORMATION AND COMMUNICATIONS SYSTEMS
Elon Musk recently criticized the state of
U.S. government computer systems, describing them as outdated and inefficient.
His remarks have sparked discussions worldwide, including in the Philippines,
about the state of government technology infrastructure. What exactly did he
mean? Based on his comments, he seems to be referring to both outdated hardware
and software, including the continued use of COBOL by some U.S. government
agencies.
COBOL, a programming language developed in
the 1950s, is still widely used today, particularly in banking and government
institutions. While the language itself is considered obsolete in the modern
era of AI-driven computing, the applications built with COBOL continue to
function effectively. This raises the question: Is it really the software that
is obsolete, or is it the hardware supporting it?
Musk may have been pointing to the reliance
on legacy systems—Local Area Networks (LANs) that use old servers and
workstations instead of leveraging cloud computing. Cloud-based systems are
regularly updated and less prone to obsolescence, making them a more modern
solution. If that is the case, then the real issue is not just software but the
entire ICT infrastructure of government agencies.
However, obsolescence is not purely about
age. In the ICT sector, the key concern is the Service Level Agreement (SLA)
between the client and the service provider. Even if hardware and software are
not brand new, they are acceptable if the agreed-upon performance metrics are
met. The implication is that software and hardware should not be unnecessarily
old or inefficient to the point of failing to deliver required services.
A useful analogy is the car rental industry.
When you buy a car, it is brand new, but when you rent one, it is usually a
used model. What matters is not whether the car is new but whether it runs
efficiently. Similarly, the focus should not just be on whether government ICT
systems are old but whether they function efficiently and securely.
In the Philippine government, the
depreciation period for computer hardware is typically set at five years, based
on guidelines from the Commission on Audit (COA). This means that hardware
older than five years is considered obsolete. A similar depreciation period
applies to vehicles, yet many government agencies continue using vehicles well
beyond this timeframe if they remain serviceable. The same principle applies to
computers—many agencies use them far beyond their supposed obsolescence.
In contrast, private corporations often
retire computers within three years, as declining performance directly affects
productivity and competitiveness. In the Philippine government, where
productivity standards are often unclear and competition is nonexistent, there
is less pressure to upgrade systems.
From a modern perspective, any software that
does not incorporate artificial intelligence is already falling behind.
AI-driven applications improve efficiency, automate processes, and enhance
decision-making. Additionally, energy efficiency is another critical factor—any
hardware that consumes excessive power is inefficient and arguably obsolete.
The way forward for the Philippine government
is to shift towards cloud computing, which eliminates the need for maintaining
old servers. Government agencies should also explore "Software as a
Service" (SaaS) and "Equipment Lease Agreements" (ELAs) instead of
purchasing software and hardware outright. This approach ensures that systems
remain up to date without requiring frequent large capital expenditures.
Elon Musk’s criticisms serve as a wake-up
call. If the U.S. government is struggling with outdated systems, what more for
countries like the Philippines? It is time to rethink how we manage government
ICT infrastructure to ensure efficiency, security, and long-term
sustainability.
Ramon Ike V. Seneres,
www.facebook.com/ike.seneres
iseneres@yahoo.com, 09088877282, senseneres.blogspot.com
04-16-2025
Comments
Post a Comment