Navy Aims for “Compile to Combat in 24 Hours”

http://www.doncio.navy.mil/CHIPS/ArticleDetails.aspx?ID=10501

by frostmatthew • 8 years ago

53 38 8 years ago

doncio.navy.mil

lloydde3• 8 years ago

Why convolute the data format and software architecture? The 18 month to 24 hour is gross.

That is a lot of words for little real information. I can’t figure out the target audience.

My favorite line was “XML is a leading open standard for describing and tagging data in industry” which could be rewritten to “For more than 20 years...”

childintime• 8 years ago

If I were the enemy I too would have them choose XML.

lloydde3• 8 years ago

Maybe that is it! This is propaganda to have the enemy work on DTDs and XML transformations.

MaxGabriel• 8 years ago

The concept is based on web services, or micro-services, similar to those you would see on your smart phone

:thinking_face:

shp0ngle• 8 years ago

Agile, machine learning, cloud, microservices... I wonder if they will also add ICO and blockchain in there.

disqard• 8 years ago

...and don't forget Big Data. Or is it no longer in fashion?

NickNameNick• 8 years ago

If it fits on a single ship, it's not really big data... /s

raverbashing• 8 years ago

Your big data is on the blockchain now (or something like that)

dig1• 8 years ago

Nope. Everything should be on blockchain now.

modin• 8 years ago

Navy seeks to leverage XML for the benefits that commercial industry enjoys from using it, like the ability to efficiently tag and describe data for use in big data analytics, artificial intelligence and machine learning, [...]

It's there. In XML!

dogma1138• 8 years ago

Drone strike smart contracts might be a thing soon enough if we ever privatize armies at a large scale.

johnsmith2018• 8 years ago

I can't help but think of the beginning of Battlestar Galactica https://en.battlestarwikiclone.org/wiki/Cylon_computer_virus

trdtaylor1• 8 years ago

This is obscenely true and somewhat underrated.

Sad as it is, the prime reason for the strength of DoD computer systems, both afloat and in the air, is that it's difficult enough to use and interface with them-- 'Hacking' them, and getting the ability to pivot and deny/destroy is extremely limited. Sailor Joe Schmuckatelli's Fire control system is on a separate network because there was no common network at the time the contract was let for competition. It's both the worst and most extreme version of security through obscurity. Now you're introducing more connections, availability, and reliance on a core infrastructure-- Which still won't meet basic corporate cybersecurity / patch control doctrine.

forapurpose2• 8 years ago

The U.S. military has been very aware of this threat for a few years (though a legacy of disregard precedes that, as they saw no peer threats and were fighting low-tech insurgents). They're not idiots; it's obvious and they hack into their enemy's systems. In fact, they have been focusing on systems and training that enable units to operate while disconnected from the defense networks.

perl4ever• 8 years ago

I can't help but think of:

"The ultimate cause of our failure was a simple one: despite all statements to the contrary, it was not due to lack of bravery on the part of our men, or to any fault of the Fleet's. We were defeated by one thing only - by the inferior science of our enemies. I repeat - by the inferior science of our enemies."

AYBABTME• 8 years ago

Join the military to work on leading edge technology like XML and SAML? Why do people think the military has advanced technology? Aside for aerospace stuff, it seems like they're hugely behind the commercial world.

ethbro3• 8 years ago

The innovation is mostly in hardware, albeit they're trying to move as much of that as possible to CotS.

At a tactical level, better hardware with dated but tested software > inferior hardware with the newest software.

Edit: To elaborate, there are almost no startup spaces where someone is actively trying to degrade your systems as they're being used. That changes a lot of calculus.

forapurpose2• 8 years ago

That's not my understanding, based only on reading foreign policy publications. For many years, the U.S. military has seen software as the key and hardware as just a platform to carry it (to oversimplify a vast domain of technology). With the F-35, for example, many critics focus on the hardware while the military focuses more on the software. For example, if your signal processing is better then you detect the enemy before they see you, shoot first, and win. If your networking is better, then you share information faster, adjust operations and tactics too fast for the enemy to keep up with, and win.

nightski2• 8 years ago

Seriously? You are judging the project on the serialization format they use? Who cares what it does or how it serves it's users right?

Not to mention that XML is actually not all that bad of a technology. Not my first choice, but it is fine.

dean177• 8 years ago

Why isn't it your first choice?

ramoz• 8 years ago

It matters though, doesn't it? Context is military data transmission, and if things like compression and speed matter, shouldn't they aim to use something like protobufs?

stillmind• 8 years ago

Especially as it pertains to the cyber world the military is slow to adopt, yes. IMO this is largely due to the amount of vetting, documentation, the number of people who must approve it, and the security measures taken in order to use a product or process. This is frustrating, because as soon as a product or process is approved there is almost always been a shift in the industry to a better way of doing the same task or better product. The workaround here always seems to be the pairing up the military with commercial industry, which isn't hindered in the same way. However, you then risk having the commercial industry not following proper protocol for securing information and leaking secrets to foreign countries, amongst other issues. Problems are abound, and the military is still trying to figure out the best way to approach innovation in the cyber world.

otakucode2• 8 years ago

I'm active in this space and the vetting and documentation is going to be what makes this never happen. They will say they want agile, they want continuous integration, they want speedy development........ but they will turn around and say they want exactly the same amount of contract deliverable documentation, want exactly the same process of requesting approval (which takes 9+ months to get) for any new technology, library, or module, etc. Several years ago the government said Agile development should be used for any software projects going forward - without dropping a single one of their documentation requirements. The fact that one of the pillars of Agile is "Working software over comprehensive documentation" is something they seem to think they can just ignore.

TickleSteve2• 8 years ago

Having worked in defence for over 15 years, I can say with certainty that in general the level of technology in your average military product is well behind the times.

This is by design since the products tend to have 20 year support requirements and you fix the level of technology at the design stage (say 5 years development).

Its therefore commmon to be ten years out of date.

yread• 8 years ago

What's wrong with SAML?

ethbro3• 8 years ago

"This transformational architecture is called “Compile to Combat in 24 Hours,” and was recently piloted this spring aboard several Navy units. The concept is based on web services, or micro-services, similar to those you would see on your smart phone, and use of a new cloud architecture Navy is developing and testing, and fielding this capability quickly and securely. It involves decomposing legacy/monolithic applications into agile micro services, developed in the cloud and synchronized with the tactical cloud shipboard. The cloud environments are designed for machine learning and artificial intelligence capabilities and will “package and prioritize” data exchanged between the ship and shore clouds to get commanders the information they need for improved decision making and lethality."

... that poor, confused PR author.

lloydde3• 8 years ago

It’s poor, confused PR all the way down. The linked origin article is also glorious, http://www.doncio.navy.mil/CHIPS/ArticleDetails.aspx?ID=1021... with classics like “ubiquity of micro web services in the commercial world and in each individual smartphone“ which may be trying to say something about RESTful APIs.

ethbro3• 8 years ago

That's the tragedy: you can almost see the conversation between an engineer and the author.

Lost in translation, indeed.

mark_l_watson2• 8 years ago

Good read! Also good architectural and delivery decisions, considering the constraints they work under.

TickleSteve2• 8 years ago

</sarcasm> surely?

mark_l_watson2• 8 years ago

No sarcasm. I was a defense contractor (programmer) from the 1970s into the 1980s and I understand the constraints they work under.

everyone• 8 years ago

TLDR -> The US navy will be making their warships much more conveniently hackable.

ker0jo• 8 years ago

I wouldn’t have thought moving infrastructure from on ship servers to ‘the cloud’ would be a tactically sound move.

otakucode2• 8 years ago

You can have quick development. You can have reliable, predictable systems. You can not, ever, have both.

jonnydubowsky• 8 years ago

I would love to hear feedback from the officers whose daily operational experience is effected by these improvements. I wonder what kind of beta testing they do, and if they continuously refine the architecture and design choices after receiving rounds of feedback from active combat experiences?

Navy Aims for “Compile to Combat in 24 Hours”

Related Stories