As a project manager for Raytheon back in the 90’s, I can attest to this. We were responsible for ATC facilities across the Southeast. A requirement would originate at the FAA. They would develop a scope of work and send it to us. We would in turn hire a contractor to perform the work. Each individual location could have a different company perform the work. And I didn’t actually work for Raytheon. My check came from an engineering company in California. Each of these layers, of course, included a management fee and other duplicate line items. What a stupid way of doing things!
In addition to being financially inefficient, this practice pushes institutional knowledge farther out away from the government and into entities that the government may not even know exist. They contract a big prime like Raytheon expecting Raytheon to know how things work, but Raytheon doesn’t have that knowledge. Their subcontractors do. And then when something goes wrong at the government level, getting it remedied requires multiple bureaucratic hops.
In the D.C. area, a federal GS-14 job pays about $144k to $187k. A GS-15 pays about $169k to $197k. A decent senior software engineer in the private sector can easily make over $200k plus equity. So you’re asking a lot of engineers to take a large pay cut.
Now throw in the government bureaucracy and all the laws and regulations you’d have to deal with. I’ve been a software engineer for 15 years and that doesn’t sound too appealing.
Quite simply the government can’t hire good technical talent because its pay and culture are so inferior to what the public sector can offer.
One way to route around this problem is contracting. If you started hiring in house you’d end up with different and potentially far worse problems if you don’t change the pay and culture issues.
I understand the instinct to bundle the outage/reliability issues for the FAA's network with FAFSA website operations, but these situations represent very different resourcing challenges.
Contractor owned and operated network services are much more cost effective and more likely to reflect current standards than a government owned and operated service. It's why the government relies on telecom carriers to support all manner of telecommunications for the federal government.
It's when the government delays updating technical standards imposed on those contractor owned and operated services that lead to outages, underperformance, or costs that exceed those for commercially available services. For example, at the same time OMB issued memos for "cloud first" in Obama's first term, it also issued memos that mandated the use of Trusted Internet Connections - costly standards never designed for the high volume of traffic generated by putting government services in the cloud.
After the dust settles on the FAA outages, it'll be interesting to understand how the nature of the upgrades and government oversight impacted router upgrades. I would be surprised if OIG/GAO/congressional subcommittee would conclude that a government owned and operated telecommunications network would be the best way to provide service and receive timely upgrades.
In terms of having government owned and operated web services, that's a scenario much more likely to meet expectations for service and timely upgrades - especially with AI supported SW development. Still - obstacles remain in recruiting and managing staff under current civil service requirements, but it's far more likely for government to own and operate discrete web services than a nationwide telecommunications network.
Pahlka hits it correctly again. However, it also strikes me that there is a much more worrying deeper cultural and values disease that gives rise to this structure and poor decisionmaking. The outsourcing of thinking and consumerist mentality of the US has completely debilitated its governing institutions, and no amount of smart people making smart recommendations can change anything until that changes. Seems like it will get worse and we will be forced to rebuild from near or total catastrophe for that to change. Each society seems hellbent on learning or not learning this lesson the hard way
As a widower, trying to feel my way through the the haunted forest of Canada Revenue Agency, most of those I spoke to in website frustration, were anxious to assist, but unable as "that's another department's information I can't access", volunteering they knew the system was screwed up, but had no power to fix.
When government jobs are seen as premium grade and permanently secure, promotions are based on seniority, not merit, when hiring is by demographic quota, the motivation to deliver better results and become more efficient with input from the coalface workers is lacking. What administrator wants to shrink their budget by accomplishing more with less, surely not a career builder?
No worries in any case, the government is too big to fail.
As a project manager for Raytheon back in the 90’s, I can attest to this. We were responsible for ATC facilities across the Southeast. A requirement would originate at the FAA. They would develop a scope of work and send it to us. We would in turn hire a contractor to perform the work. Each individual location could have a different company perform the work. And I didn’t actually work for Raytheon. My check came from an engineering company in California. Each of these layers, of course, included a management fee and other duplicate line items. What a stupid way of doing things!
In addition to being financially inefficient, this practice pushes institutional knowledge farther out away from the government and into entities that the government may not even know exist. They contract a big prime like Raytheon expecting Raytheon to know how things work, but Raytheon doesn’t have that knowledge. Their subcontractors do. And then when something goes wrong at the government level, getting it remedied requires multiple bureaucratic hops.
In the D.C. area, a federal GS-14 job pays about $144k to $187k. A GS-15 pays about $169k to $197k. A decent senior software engineer in the private sector can easily make over $200k plus equity. So you’re asking a lot of engineers to take a large pay cut.
Now throw in the government bureaucracy and all the laws and regulations you’d have to deal with. I’ve been a software engineer for 15 years and that doesn’t sound too appealing.
Quite simply the government can’t hire good technical talent because its pay and culture are so inferior to what the public sector can offer.
One way to route around this problem is contracting. If you started hiring in house you’d end up with different and potentially far worse problems if you don’t change the pay and culture issues.
I understand the instinct to bundle the outage/reliability issues for the FAA's network with FAFSA website operations, but these situations represent very different resourcing challenges.
Contractor owned and operated network services are much more cost effective and more likely to reflect current standards than a government owned and operated service. It's why the government relies on telecom carriers to support all manner of telecommunications for the federal government.
It's when the government delays updating technical standards imposed on those contractor owned and operated services that lead to outages, underperformance, or costs that exceed those for commercially available services. For example, at the same time OMB issued memos for "cloud first" in Obama's first term, it also issued memos that mandated the use of Trusted Internet Connections - costly standards never designed for the high volume of traffic generated by putting government services in the cloud.
After the dust settles on the FAA outages, it'll be interesting to understand how the nature of the upgrades and government oversight impacted router upgrades. I would be surprised if OIG/GAO/congressional subcommittee would conclude that a government owned and operated telecommunications network would be the best way to provide service and receive timely upgrades.
In terms of having government owned and operated web services, that's a scenario much more likely to meet expectations for service and timely upgrades - especially with AI supported SW development. Still - obstacles remain in recruiting and managing staff under current civil service requirements, but it's far more likely for government to own and operate discrete web services than a nationwide telecommunications network.
Pahlka hits it correctly again. However, it also strikes me that there is a much more worrying deeper cultural and values disease that gives rise to this structure and poor decisionmaking. The outsourcing of thinking and consumerist mentality of the US has completely debilitated its governing institutions, and no amount of smart people making smart recommendations can change anything until that changes. Seems like it will get worse and we will be forced to rebuild from near or total catastrophe for that to change. Each society seems hellbent on learning or not learning this lesson the hard way
As a widower, trying to feel my way through the the haunted forest of Canada Revenue Agency, most of those I spoke to in website frustration, were anxious to assist, but unable as "that's another department's information I can't access", volunteering they knew the system was screwed up, but had no power to fix.
When government jobs are seen as premium grade and permanently secure, promotions are based on seniority, not merit, when hiring is by demographic quota, the motivation to deliver better results and become more efficient with input from the coalface workers is lacking. What administrator wants to shrink their budget by accomplishing more with less, surely not a career builder?
No worries in any case, the government is too big to fail.