US Air Force Chief Software Officer Quits (theregister.com) 81
The US Air Force's first ever chief software officer has quit the job after branding it "probably the most challenging and infuriating of my entire career" in a remarkably candid blog post. The Register reports: Nicolas Chaillan's impressively blunt leaving note, which he posted to his LinkedIn profile, castigated USAF senior hierarchy for failing to prioritise basic IT issues, saying: "A lack of response and alignment is certainly a contributor to my accelerated exit." Chaillan took on his chief software officer role in May 2019, having previously worked at the US Department of Defense rolling out DevSecOps practices to the American military. Before that he founded two companies.
In his missive, Chaillan also singled out a part of military culture that features in both the US and the UK: the practice of appointing mid-ranking generalist officers to run specialist projects. "Please," he implored, "stop putting a Major or Lt Col (despite their devotion, exceptional attitude, and culture) in charge of ICAM, Zero Trust or Cloud for 1 to 4 million users when they have no previous experience in that field -- we are setting up critical infrastructure to fail." The former chief software officer continued: "We would not put a pilot in the cockpit without extensive flight training; why would we expect someone with no IT experience to be close to successful? They do not know what to execute on or what to prioritize which leads to endless risk reduction efforts and diluted focus. IT is a highly skilled and trained job; staff it as such."
Chaillan went on to complain that while he had managed to roll out DevSecOps practices within his corner of US DoD, his ability to achieve larger scale projects was being hampered by institutional inertia. "I told my leadership that I could have fixed Enterprise IT in 6 months if empowered," he wrote. Among the USAF's sins-according-to-Chaillan? The service is still using "outdated water-agile-fall acquisition principles to procure services and talent", while he lamented the failure of the Joint All-Domain Command and Control (JADC2) to secure its required $20m funding in the USAF's FY22 budget. He was also quite scathing about the USAF's adoption -- or lack thereof -- of DevSecOps, the trendy name for efforts to make developers include security-related decisions at the same time as product-related decisions when writing new software. It appears the service wasn't quite as open-minded as its overseers in the wider DoD.
In his missive, Chaillan also singled out a part of military culture that features in both the US and the UK: the practice of appointing mid-ranking generalist officers to run specialist projects. "Please," he implored, "stop putting a Major or Lt Col (despite their devotion, exceptional attitude, and culture) in charge of ICAM, Zero Trust or Cloud for 1 to 4 million users when they have no previous experience in that field -- we are setting up critical infrastructure to fail." The former chief software officer continued: "We would not put a pilot in the cockpit without extensive flight training; why would we expect someone with no IT experience to be close to successful? They do not know what to execute on or what to prioritize which leads to endless risk reduction efforts and diluted focus. IT is a highly skilled and trained job; staff it as such."
Chaillan went on to complain that while he had managed to roll out DevSecOps practices within his corner of US DoD, his ability to achieve larger scale projects was being hampered by institutional inertia. "I told my leadership that I could have fixed Enterprise IT in 6 months if empowered," he wrote. Among the USAF's sins-according-to-Chaillan? The service is still using "outdated water-agile-fall acquisition principles to procure services and talent", while he lamented the failure of the Joint All-Domain Command and Control (JADC2) to secure its required $20m funding in the USAF's FY22 budget. He was also quite scathing about the USAF's adoption -- or lack thereof -- of DevSecOps, the trendy name for efforts to make developers include security-related decisions at the same time as product-related decisions when writing new software. It appears the service wasn't quite as open-minded as its overseers in the wider DoD.
armed forces need to kill up or out & IT (no b (Score:2)
armed forces need to kill up or out & IT staff should have an no boot camp / no combat / no rank way to get in.
And you don't want some JR officer who knows NO IT to out rank an 10+ year IT tech.
Re: (Score:2)
In English please?
Re: (Score:3)
up or out forcing people good at jobs into management and with armed forced you can also have rank things like there only X slots at that rank and the opening are out of your field. But also you can't have 100's 4 star general so some are just pushed out.
Re: (Score:3)
Re: (Score:2)
They are referring to the US Military practice of expelling members who fail to make promotions to higher ranks. The rules vary by branch of service I think, and also between officer and enlisted. But essentially you can't be a highly skilled worker bee for life. Pay is directly tied to rank, and you can't make rank past the most basic levels without getting turned into a supervisor/manager.
I knew a guy when I was in who finally made his promotion to E5 and ended up with a line number of seven, out of more
Re: (Score:2)
Ah, now it makes sense, thanks! I know some corporates have addressed this problem by creating roles at management-level pay grades but without management responsibilities in order to retain skilled technical people in the position where they do the most good, generally with things like "Scientist" or "Fellow" or "Researcher" in the job title (whether they are or not), but I can see that would be an issue in the military. It would probably take at least one world war and several revolutions and insurgenci
The real mission of the military (Score:4, Insightful)
We recently learned that the military learned nothing from Vietnam, making exactly the same mistakes in Afghanistan.
The military exists to funnel money into the coffers of defense contractors, not to fight wars, which it appears to be bad at.
Re:The real mission of the military (Score:5, Insightful)
What we learned is that the politicians had no desire to let the military do what they do best, and that is to lay waste and win by destruction.
Re: (Score:1)
AKA, genocide
Re:The real mission of the military (Score:5, Insightful)
History does not repeat itself ... (Score:5, Interesting)
But it often rhymes.
In both Vietnam and Afghanistan we went in with our view of the world that ignored their reality on the ground. For Vietnam it was anti-communism, for Afghanistan it was anti-terrorism. In both cases we supported corrupt governments and had no real understanding of the people. In both cases very few of us even spoke their language.
And in both cases we predicted a massacre when the bad guys won, which never happened.
That is some very poor "rhyming" there (Score:2)
But it often rhymes. In both Vietnam and Afghanistan we went in with our view of the world that ignored their reality on the ground. For Vietnam it was anti-communism, for Afghanistan it was anti-terrorism.
That is some very poor "rhyming" there. In Vietnam we fought people who fought along side us against the Japanese. In Afghanistan we fought the people who attacked us in NYC and the people who decided to defend these attackers.
In both cases we supported corrupt governments and had no real understanding of the people.
In the Afghanistan case they had one corrupt government before our invasion and a different corrupt government after. The latter corrupt government would not attack us, unlike the former corrupt government. Again, quite different from any Vietnamese government.
And in both cases we predicted a massacre when the bad guys won, which never happened.
In the Vietnam case, "up
Re: (Score:2)
But it often rhymes. In both Vietnam and Afghanistan we went in with our view of the world that ignored their reality on the ground. For Vietnam it was anti-communism, for Afghanistan it was anti-terrorism.
That is some very poor "rhyming" there. In Vietnam we fought people who fought along side us against the Japanese. In Afghanistan we fought the people who attacked us in NYC and the people who decided to defend these attackers.
Actually the rhyme is even better than I had thought. In Vietnam we fought people that helped us fight the Japanese. in Afghanistan we fought people that helped us fight the Soviets. (We never really fought Al Quada, they are not the Taliban.)
And in both cases we predicted a massacre when the bad guys won, which never happened.
In the Vietnam case, "up to 300,000 people, especially those associated with the former government and military of South Vietnam, were sent to re-education camps, where many endured torture, starvation, and disease while being forced to perform hard labor."
https://en.wikipedia.org/wiki/... [wikipedia.org]
Sure tough, but there was no massacre, i.e. lots of people killed. I think it is clear that there will be no massacre in Afghanistan either.
And in both cases, I fear, history will be rewritten.
"Never will we again fight a war that we do not have the courage to wi
Re: (Score:2)
But it often rhymes. In both Vietnam and Afghanistan we went in with our view of the world that ignored their reality on the ground. For Vietnam it was anti-communism, for Afghanistan it was anti-terrorism.
That is some very poor "rhyming" there. In Vietnam we fought people who fought along side us against the Japanese. In Afghanistan we fought the people who attacked us in NYC and the people who decided to defend these attackers.
Actually the rhyme is even better than I had thought. In Vietnam we fought people that helped us fight the Japanese. in Afghanistan we fought people that helped us fight the Soviets.
Nope, again, poor "rhyming". In Vietnam we literally fought along side them, it was not simply supplying them as we did in Afghanistan in Soviet days. WW2 Vietnam was more like the campaign against the Taliban in 2001. US Special Operations troops and air power supporting local forces that did most of the fighting. In short it was not a proxy fight like with the soviets.
(We never really fought Al Quada, they are not the Taliban.)
And now you are flat out wrong. In the 2001 invasion Al Qaeda was our primary target. The Taliban only became a target when they refused to
Re: (Score:2)
Afghanistan was not a failure of the military but a failure of strategic policy where the US got into a war with no plan for how that war could end.
To be fair, not every war can be entered with a plan for how to end it. After September 11, going after Al Quida in Afghanistan - who were behind these attacks and others - was obvious and necessary. This was not a situation like the second Iraq war.
That aside, that's 20 years to find out and execute what you want...
Re:The real mission of the military (Score:5, Interesting)
Re: The real mission of the military (Score:2)
Re:The real mission of the military (Score:4, Informative)
If military rules a country, we call it a junta.
If politics run a war, we call it Vietnam.
That's the lesson that should have been learned from Vietnam and that's the lesson we didn't learn: That politicians have NO business on a battlefield. Define the goal of your war, hand it over to the generals, let them do the strategic planning and stay the FUCK out of it!
Re: (Score:2)
I agree, but in both wars there were also military leaders painting a completely unrealistic rosy picture of the situation in the far-off country. So I think both the military and the politicians are to blame.
Re: (Score:1)
Re: (Score:2)
We went to Vietnam to engage in nation-building and try to build a regional ally and stop Communism from spreading. Those are not goals attainable through war. Especially when our local "allies" were just corrupt despots who didn't actually embody our values.
We went to Afghanistan to kill certain people. Those people are all dead. This goal was attainable, and we attained it. Nation-building afterwards was a legal requirement according to the various treaties that discuss the responsibilities of an occupyin
Re: (Score:2)
The etymology of the name "taliban" has zilch do with its present meaning as the name of a political and military/ militant group. It's more like the relationship between socialism and the National Socialist German Workers Party.
Re: (Score:2)
Right, but who cares? We didn't set out to scour the word from the face of the Earth. My point was simply that the etymology would make that exceptionally foolish, it would be like trying to get people to stop calling themselves "Baptist" or something.
We killed all the leaders who did the things we set out to kill them for doing. Done.
Realistically... (Score:5, Insightful)
"Please," he implored, "stop putting a Major or Lt Col (despite their devotion, exceptional attitude, and culture) in charge of ICAM, Zero Trust or Cloud for 1 to 4 million users when they have no previous experience in that field -- we are setting up critical infrastructure to fail."
With that attitude, you'll likely soon be demanding the men and women who run for political office demonstrate proficiency, outside of, pandering to accumulate votes... you'd have a bit better luck asking the hurricane to go around your State, Cajun.
Re:Realistically... (Score:4, Insightful)
Re: (Score:1)
"Please," he implored, "stop putting a Major or Lt Col (despite their devotion, exceptional attitude, and culture) in charge of ICAM, Zero Trust or Cloud for 1 to 4 million users when they have no previous experience in that field -- we are setting up critical infrastructure to fail."
With that attitude, you'll likely soon be demanding the men and women who run for political office demonstrate proficiency, outside of, pandering to accumulate votes... you'd have a bit better luck asking the hurricane to go around your State, Cajun.
you either have oversight or you are a worker. get a job at the dept. of _fill in the blank_, or join the air force.
6 months really? (Score:4, Insightful)
"I told my leadership that I could have fixed Enterprise IT in 6 months if empowered"
I can see how someone who says that, in an org with millions of users, might suffer some credibility problems.
Re:6 months really? (Score:5, Informative)
As someone who was deeply involved at the base level for a pair of organizations going through the Enterprise IT transition, I can confirm that middle management was the problem. The big level ideals were good, several O-6s said the exact right thing, authorized the exact right thing, made the correct decisions and calls. The little people were cool with whatever and would follow the correct lead. The problem was complete dipshit O-5s, O-4s, and E8/E7 that couldn't quite wrap their heads around a new paradigm that they didn't grow up in themselves.
The top had it right. The bottom would follow orders. The middle....that is where the problem was...and if somebody would simply take a giant shit all over every idiot in the middle...they would figure it out.
I've literally walked out of a meeting with a Colonel agreeing with everything....and the next day the dipshit Major below him talking the other direction. Idiocy and incompetency everywhere.
Re: 6 months really? (Score:2)
You just described every large MNC which has existed for more than 20 years.
Re: (Score:2)
I'd be curious to know how this compares with the present-day Boeing Corporation. My impression--from a distance--is that the problem is at the top, not the middle (and definitely not at the bottom). Also, how does it compare with Intel (about which I know even less than I know about Boeing)?
Re: 6 months really? (Score:2)
I think the VW diesel fiasco gives us a clue that this behaviour is pretty much the same across industries.
Re: (Score:2)
Naive? Conway's Law is hard to undo (Score:5, Insightful)
Also, DevOps and Agile often just doesn't work in certain kinds of orgs. As a contractor I brought in a private-sector view of how things should be done into some gov't contracts, and I realize now it would require a complete overhaul of how things are done there. You have to find a way to make incremental improvements in project management because you are NOT going to be able to flush out everyone and start over with the management fad of the month (even if it works in some places).
Civil Engineers manage Civil Engineering projects (Score:4, Insightful)
Senior lawyers (partners) manage law projects (big cases). Doctors manage medical practices. Professors manage research projects. Senior accountants manage accounting firms.
But MBAs manage large software projects.
Might have something to do with it.
Re: (Score:2)
That can work if, and only if, the MBA does what he's good at: Cutting the red tape and handing the technical stuff to his senior devs.
Yes, I've seen that work. But it requires an MBA whose ego allows him to accept that he's not the master of everything anyway and everyone else is a moron that needs his divine guidance.
And no, I still need that MBA, you can't have him.
Re: (Score:1)
That's the thing, there are indeed those good at coordinating people regardless of original specialty. However, it's hard to identify such managers without actually putting them in place. Putting your ego in your pocket is Step 1.
You need to have domain knowledge (Score:2)
A senior Engineer/Lawyer/Accountant etc. may not know the details or be the domain expert. But they have a deep understanding of the general subject. This helps them sort out the crap from the bullshit. To know when one of their reports is talking sense, to know what questions to ask. Otherwise all an MBA has to go on is their opinions about other people's opinions, and that is where all sorts of strange decisions get made.
Another issue is that an Senior Engineer etc. may or may not make a good manager
Re: (Score:1)
Correction: "don't work in"
Re: 6 months really? (Score:2)
100k developers? (Score:5, Interesting)
Maybe if you count all the contractors too...
In large part, this strikes me as someone who was not familiar with the constraints of government procurement. A lot of practices that are common in industry are illegal when done by DoD, due to the procurement rules written by Congress. Often those rules exist to prevent someone from repeating a mistake (deliberate or accidental) done before.
I was lucky to work for 2 highly qualified, software literate, LTCs. They understood when the prime contractor was blowing smoke, and both were able to get into details when necessary. Perhaps more importantly, they both were able to assemble a good team on the government side to balance out the contractor, so we often had meaningful give-and-take at the technical level while the bosses hammered out costs, schedules and all the other things that makes government procurement so difficult.
Cultural problem (Score:5, Insightful)
Re:Cultural problem (Score:4, Insightful)
Chaillan went on to complain that while he had managed to roll out DevSecOps practices within his corner of US DoD, his ability to achieve larger scale projects was being hampered by institutional inertia.
This is pretty much true of any organization, not just the military. What to do? One can't steamroller over one's users. Saying I'm right, your wrong. Maybe make them part of the solution. Vest part of the success in them.
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
Nice fantasy but no thats not how the real world works. Outside contractors want to deal with people who can make the decisions and have the decisions stick. Not nerds running after the framework fad of the day. I say that with love as I have been a nerd most of my life but when I got to management I had to make a conscious effort to keep that side of my character in check so as to get things done.
You kept that character in check, or pretended it did not exist? Because managers that know nothing at all about IT managing IT projects are a thing. I don't care how exceptional a manager you are, you can't simply manage your way out of being way out of your depth.
Re: (Score:2)
What? A good manager can't manage anything? (Score:4, Funny)
Re: What? A good manager can't manage anything? (Score:3)
Re: What? A good manager can't manage anything? (Score:3)
The conceit among phbs is that phbs hold the world on their shoulders. All their phb friends and colleagues agree.
Re:What? A good manager can't manage anything? (Score:4)
A good manager knows what management is all about: Finding the right people for the right job and giving them the right amount of freedom to do their job.
That also means that a good manager doesn't need to know how to do X. He only needs to have someone who can do X and hand him everything that person needs to do it, remove the bureaucratic obstacles that need his management muscles to move them and otherwise keeps out of his hair.
In other words, a good manger doesn't tell you what to do, he asks you what you need done.
Re: (Score:2)
Re: (Score:1)
I thought everyone knew a good manager could mange anything. (assuming you want that thing to crash an burn that is).
but that does not mean they will be allowed to.
Translation (Score:3)
EDMICS (Score:4, Insightful)
Years ago, I had a great experience working on an early project to begin the US Navy's transition for paper manual (bulky, heavy, and almost irretrievable in a ship at sea) to digital storage. Think about how many manuals there are on a carrier: one for not only everything on the carrier, from the galley, to flight support, to the engines, electronics, and crew support, but for every bird (fixed wing and rotary wing) she carries.
We worked for a captain (equivalent for a full colonel in the Air Force) in weapon systems engineering who was accustomed to dealing with "tech types". We had a working, fully demonstrable instance on time, and within budget. Of course, when the bureaucrats in Naval Publishing and Printing found out they had conniptions. It took them, and their pet contractor, years to get back to where we had already been.
It is possible to find competent mid-level officers, but they need the backing of senior officers and the latitude to get the job done. Senior Air Force officers appear to be so aircraft-focused that they have no idea what to do with something like IT.
Re: (Score:2)
Re: (Score:2)
'twould be interesting to hear more about that. I was in the Navy maybe a decade or two before that, when everything was paper, with the exception of one poorly thought out system: we were supposed to schedule major maintenance jobs six months in advance, then submit paper descriptions (each character had to fit in a separate box); a week later we got back the printouts, which were worthless. We had the same information on paper notebooks, the fact that it was in a mainframe computer somewhere did nothing
more EDMICS (Score:2)
NSWSES in Port Hueneme was a guest of the Seabees. There was a huge repository of IBM aperture cards. An aperture card was a standard-size card with punch codes for the item description (identifier, classification level, ...) plus a frame of film with the actual drawing. When someone needed a drawing, for component or system verification, or to perform an update of some sort, a print of the drawing was requested, and after the security checks, ..., it was delivered to the requesting party.
We had a scanne
What'd I just say? (Score:3)
Okay, this isn't related on paper to the US Digital Corps, but this is exactly what I said happens with this stuff: too much red tape and a culture of not staffing based on proficiency in necessary skill.
The way these generalist officers usually operate is they oversee a permanent or quasi-permantent contractor that makes the actual technical decisions.
Leadership tends to go from the middle out, rather than top down. The middle of course gets influenced by the top, but the top changes every 4 to 8 years while some in the current middle have been in their jobs since the 80s.
Re: (Score:2)
You'll find that the same is true for politics and bureaucracy.
Re: What'd I just say? (Score:2)
Yes, I have found exactly that throughout my career. I have drawn the conclusion from these experiences that the overall "best" system is one where there is a large number of smallish institutions rather than a small number of largish institutions.
Call it the Unix Philosophy for Government and Economics.
Re: (Score:2)
It should also be the policy of governments.
Because the difference between a large number of small businesses in your country and a small number of large businesses in your country is who controls whom. If you have to pass legislation they don't like, small businesses have to grin and bear it, because they and their family are located here. Large businesses just flip you off and move on to another country.
Re: What'd I just say? (Score:2)
No, if it's truly egregious they'd band together and elect politicians more favorably disposed to them. Difference would be that it would be a real grass roots campaign instead of astroturf controlled by a small number of powerful actors.
Boss or Technician (Score:4, Insightful)
I've also bemoaned technologically incompetent bosses. And, when I did, I missed the point.
Does a boss have more than passing competence in the fields that his subordinates work in? Well, if you think the answer is "yes", it follows that every president or prime minister or other leader of an entire country has to be at least partially incompetent. Because a president's job covers so many areas that no one could be great or even good on all topics. And, there have been effective presidents and prime ministers and kings. So, thinking that subject matter expertise is required in a boss would be a wrong conclusion, so ... there must be something wrong with my premise that the boss has to know the work.
I'm almost 60 and I've had a lot of bosses while working mostly as a programmer or systems administrator. Good ones. Bad ones. Technical ones. Not so technical ones. And, you know what - there's been very little correlation between who has been a good boss and who has had technical chops! When a boss is both good at management and at technology, that's great. But, one of the worst bosses I've ever seen was a promoted techie. I was quite thankful to be a peer not his subordinate. I've had good bosses who weren't very knowledgeable in IT, but were good bosses.
It's almost as if management was a skill and that some people had technical skills and some people had management skills and some people had both. Let's not talk about those that had neither...
The technical boss can be a micro-manager and focus too much on his own experiences and capabilities. The non-technical boss might make decisions based on what peers or salesmen say. Or, either might be a great leader.
A good boss has to know how much to delegate, how much to listen to his people, how to chose which person to listen to, when to put his/her foot down, etc. etc. If they know how to "manage" their advisors, they can be quite effective while only having a very basic technical understanding.
TL;DR Ability to boss IT workers is poorly correlated with IT expertise.
That said, I want to think that an MBA might negatively correlate with *anything* useful, but that just might be personal prejudice. :-/
Re: (Score:2)
A competent manager needs to be able to, at minimum, recognize and if possible increase the competency of their direct reports. Ideally they can step in for a direct report or at minimum find a competent replacement should the need arise.
Re:Boss or Technician (Score:4, Insightful)
A good boss doesn't need to be better than you in your job. He has a different one. His job is to make sure you have the resources you need at the time you need them. That's what I need him for. If he can't provide that, he may as well not exist at all.
Re: (Score:2)
He also needs to be able to recognize when you are incompetent, and find your replacement. FWIW, very few officers in the military can do that, unless the underling is so incompetent that they get court martialed. Otherwise they stay in place until they're rotated out. As my wife (who worked as a civilian in the Army) says, you fight with the army you have.
How fortuitous (Score:2)
To find this little musing [9cache.com]. Coincidence?
I's a difficult job (Score:2)
You need people who know how to handle 8 inch floppy disks.
Sounds like every IT job I've ever seen (Score:3)
Sounds like every IT job I've ever seen.
"stop putting a Major or Lt Col (despite their devotion, exceptional attitude, and culture) in charge of ICAM, Zero Trust or Cloud for 1 to 4 million users when they have no previous experience in that field -- we are setting up critical infrastructure to fail.
Sounds like every IT job I've ever seen. He's lucky they didn't give him an office in the basement, under the stairs next to the toilet.
Re: (Score:2)
He's lucky they didn't give him an office in the basement, under the stairs next to the toilet.
At least nobody will come into your office just to chat.
Don't bash it 'til you tried it.
Re: (Score:2)
He's lucky they didn't give him an office in the basement, under the stairs next to the toilet.
At least nobody will come into your office just to chat.
No, but their smell will come into your office once they've shat
Re: (Score:2)
Doors can be closed, ya know?
Which also increases the chance of nobody coming into your room.
Yeah, but (Score:1)
Having served for a decade and dealt with plenty of this sort of thing (in all fields, not just IT), I have no problem believing the criticisms are valid.
That said, had social media existed (in the form it does now), I don't think I would have taken to it to broadcast my complaints to the whole world.
US Air Force respects four special carreers (Score:4)
Re: (Score:2)
Don't forget that 1) pilots wear the Universal Management Badge (aka Wings). So unlike #2,3, and 4, where the AF never mixes them, they treat pilots like they can do literally anything. And the problem is there are so very few General Officer positions for anyone without wings and that means there's very little place for incredibly qualified non-pilots to grow their careers towards. And that means the heavy hitters around the table making decisions aren't the experts either.
So the experts leave before the "
Typical "government" (Score:1)
It's even more idiotic than you think (Score:2)