A few times now I have worked for organizations that are basically IT shops — that is, their core business now depends on being very good at doing Computer Things. However, many of these places were pretty terrible at computers. Let’s keep picking on Telstra, because its an easier example.
Telstra was originally a part of the Post Master General’s Department of the Commonwealth of Australia’s government. That is, it was a branch of the Post Office. This means it inherited a lot of Government Culture, mostly around wearing suits and being terrible at making sensible business decisions.
Over time telephony became a big deal, and the phone branch was split out into what we called at the time Telecom Australia. Somewhere along the line it was rebranded to Telstra. I would say that was associated with the government selling it off into private hands, but I don’t think that is true — there was a long phase where the overseas operations of Telecom Australia operated under the name Telstra, and that all predated privatization.
Throughout this entire history the hard bits of telecommunications were switching (getting your call to the right place), and long distance (what AT&T called “long lines”). These things where done in chonky mechanical ways often with quite large amounts of electricity so the electrons wouldn’t fall out half way, if they weren’t done by a nice young lady sitting in a cupboard and manually moving cables on a plugboard. That is, telecommunications was a serious Electrical Engineering endeavor in a slightly weird relationship with really big construction projects like digging a hole between Sydney and Perth. Big buildings full of lots of electromechanical things making exciting clicking and clunking noises, except for the young ladies (who I assume did not run directly on electricity and did not click and clunk).
…and then one day Telstra woke up and all of that was no longer true. All the high voltage mechanical things had become packet switched networks (the Internet and friends), and either all the really long holes had been dug or the digging had been outsourced to three layers of subcontractors by some guy in accounting who probably thought Jack Welsh was a good person.
We should note that the telecommunications industry fought long and hard against TCP/IP and the Internet with things like Frame Relay and X.500. They just lost because their ideas were dumb. Mostly we suffer through the echoes of these battles with things like ITU standards that are both not freely available, and cover protocols people really care about like RDP, X.509 public key cryptography, and H.264 / H.265 video compression.
Now all of this is a slight ranty way of saying that this is why I think Telstra is so bad at computers now. They’re a government department that is not part of the government, who is good at things they don’t do any more, and is too arrogant to realize that perhaps they could steal some ideas from people who are int fact good at computers. That’s how you end up with a national phone network outage that might have killed people because of an ancient time server that no one has applied the vendor patches on.
The national Telstra outage last week was triggered by maintenance work that caused the system’s clocks to go back two decades, the telco says… Ms Brady told the hearing that the outage potentially could have been avoided had a 15-year-old server — which could have been replaced for $30,000 — been updated earlier.
…
Telstra executives said the SSU 2000 server that caused the outage was still supported by Scientific Devices, which supplies the technology from Microchip.https://www.abc.net.au/news/2026-07-17/telstra-national-outage-senate-inquiry/106927200
So this is Michael’s unifying theory of making good management decisions — it is important that an organization understand what it is now, not just what it might once have been.