DARPA, Microsoft, Lockheed team up to reinvent TCP/IP Arms globocorp Lockheed Martin announced today that it has won a $31m contract from the famous Pentagon crazy-ideas bureau, DARPA, to reinvent the internet and make it more suitable for military use. Microsoft will also be involved in the effort. The main thrust of the effort will be to develop a new Military Network Protocol …
I doubt I'll be the first to mention this, but does the military really think getting Microsoft involved is a good idea given their track record?
If I were militarily minded, then I would be asking for an internet system that
1. Is secure
2. Is stable
3. Can withstand attacks
Microsoft are a shining example of total failure on all three counts (heck, even if you pull the network cable out, mostly fixing 1 and 3, you still don't get 2, and an off-network computer is probably not much good for internet use...)
I was always amused at how the MS idea of TCP/IP differed from the rest of the world (such as how many addresses would be available for hosts on a given subnet) I will await the outcome of the research.
This was based on answers which had to be given for MS qualifications that workmates where studying for (ir going through various braindumps)
"reinvent the internet and make it more suitable for military use. Microsoft will also be involved"
= An oxymoron along the "military Intelligence" lines
"The main thrust of the effort will be to develop a new Military Network Protocol, which will differ from old hat such as TCP/IP in that it will offer "improved security, dynamic bandwidth allocation, and policy-based prioritization levels at the individual and unit level"."
= Just like TCP/IP but built securely from the ground up by Microsoft (or EXACTLY like TCP/IP!)
Oh no! There goes the planet...
MS (re)writing a communication protocol? The same muppets that think SMB is so good and well designed they re-created it in a new and just as bad way in SMB2 (differing application protocols still being munged together into one uncontrollable protocol) rather than using real standards. That's before NETBIOS / NETBEUI, WINS and the other assorted bits like it get taken into account.
What a stupid way to do this. Have 3 groups go into a secret huddle and try to invent something that meets very difficult to achieve objectives. The muppets probably also believe that by keeping the details of their new protocols secret that no one will be able to reverse engineer it. What is needed is a public competition along the lines of the ones that resulted in the DES and AES encryption standards.
This is assuming that their aims can't be met with IPv6, possibly with some additions.
See this to learn about AES:
http://www.moserware.com/2009/09/stick-figure-guide-to-advanced.html
So .. you seriously believe that Microsoft have a good record in this area? I'm not an MS fan, but I'd be very wary of letting Microsoft go anywhere near this purely on their past record. What the headline doesn't make clear though is that there are some "sensible" companies other than MS in there. However, I still think that a competition on the lines of AES would be better. It's not like the other companies involved actually need the money!
"Hello, Software. You're speaking to John. Calls may be monitored to help with future Body Counts. How can I help?"
"Hello? This is Brigadier Colins, on the left flank. We're coming under small arms and mortar attack and are able to return fire. Our rifles are saying 'Permission Denied'."
"Oh, give it a few minutes, Sir. You're Active Directory privileges are probably just waiting to propagate."
"Through this project, as well as our cyber Mission Maker initiatives, we are working to enhance cyber security and ensure that warfighters* can fight on despite cyber attacks."
Then why involve a company that can't even secure it's own systems. And what was wrong with IPV6. I thought that was supposed to address these kind of problems.
First, take all the DARPA folk and lock them in a room under heavy sedation to avoid function-creep and sporadic sci-fi scoping
Second, lock the MS folk in a seperate room and beat them sensless for their assumption that they're the best people for this kind of job
The Lockheed folk will escape isolation only if they scope the project in clear terms and do not allow any creep
Then change the purpose of the project to: "develop a unified standard network protocol, interoperable, but not dependant upon, nor influenced by any of the myriad of ad-hoc sort-of-all-works-kinda-maybe protocols developed since Arpanet.
That's what must happen before they can even begin to consider what they want the protocol to do. The absence of such a protocol is what makes what they, as most of us, want it to do, so diffcult to begin with.
Well, look at the time!
The task of developing secure packet switching protocols was mostly done by the CCITT in the 1970's.
Somebody should just give Lockheed a copy of the X.25 and associated standards, and they would be pretty much there.
IP pushed out X.25 because it was free to cheap; and more suited to porn-site surfing, not because it was superior in relaibility, security or speed.
Oh that's easy ... No interop with IPv4, no documentation, only one implementation (tied to an OS you didn't want, natch) and cracked within a few weeks once the military start using it over public networks.
Seriously, this *must* have been mis-reported. For all its faults, IP has stood the test of time and delivers on all of the requirements mentioned in the article, so why would an organisation as smart as DARPA go looking for a replacement?
Microsoft ??? WTF.
Hugh amounts of misbegotten funds does not indicate the actual capabilities to provide the military with what Microsoft could not provide to themselves or the world, ever in the past.
May I remind the newbies at DARPA, it is the transparency of many eyes that provides what security they actually have now. Not obfuscation of code that for what should be obvious reasons, proven by Microsoft over the years, does not work.
A paid saboteur could not achieve as much damage as the implementation of Microsoft at any level of a secure system... why do you think they were not in on the development of the internet in the first place.
Get a brain.
If you use _ALL_ features in TCPv6 you get that without any problem. You need network support for a couple of them which is not there, but for that you probably need Cisco or someone else doing network kit, not MSFT. For f*** sake it took me half an hour to sit down and draft a design that can do that recently. Not like we are talking rocket science here. It is bleeding obvious.
It is sad when people cannot read the spec for something that is standardised, available and most of all _IMPLEMENTED_ by them already and instead of that do "An elefant is a mouse designed to government specifications". Not that they are alone. They are clearly in the same boat with a few others - 3GPP, ETSI and ITU come to mind.
By the time they get all the initial design documents approved there will be something in private industry that already does all this. Trust me, I've been there. The Govt is all in favor of innovation but not at the expense of full mind-numbing documentation providing full employment for all civil servants involved.
I'm a huge MS fan, but I'll also be at the front of the line to say they fuck up a lot, a whole lot in fact. Fuckup's are part of business though, you don't see MS running for the bankruptcy courts and they have scads of liquid cash so they are obviously doing something right on the business front.
Their business fuckup's aside, MS has some of the greatest minds in technology working for them, they far outpace Google in that respect, possibly only IBM has more super smarties than MS so maybe it's not such a bad thing if MS is involved.
Who knows and who cares, this'll never happen anyway, it's just a great way to scarf some tax dollars for a senseless project.
"improved security, dynamic bandwidth allocation, and policy-based prioritization levels at the individual and unit level".
Yep, sounds like what the whole industry has been doing, and specifying in standards track RFCs, for the last 15 years. It'll take some good marketing to dress up implementing those into something that can be sold for lotsa money to the Dept of Dexpense.
Paris, cos she can also spend lots of money to cover her a...
...why slag Microsoft off? They actually produce stable software, especially with the last couple of releases, most of the issues come about because they don't talk to third party's very well (if at all).
If you ran entirely Microsoft software on a PC, you'd not see a fraction of the crashes & issues you encounter day to day.
Microsoft® to harness the distributed computing power of all PCs so they can defend & attack.
With 70-80% of CPUs on the interweb by the short & curlies, SKYNET will have dominance and be near impossible to shutdown without doing that freaky green digital rush thing like cousin Neo.
Mines the one with the EMP in the pocket.
This post has been deleted by its author
"IPv6 does all this"
No it doesn't - go learn about IPv6.
<rant about microsoft but no cisco being involved even though it's partly routing-related tech>
Juniper is, says so in the article.
<microsoft will screw this up because they can't write secure code type comment>
Asside from you're wrong, least they have people who can actually attempt it unlike say.. apple.. who lift bits of BSD then STILL screw it up.
A paid saboteur could not achieve as much damage as the implementation of Microsoft at any level of a secure system... why do you think they were not in on the development of the internet in the first place.
If by the Internet you are talking about Arpanet, the main reason, for me anyway, why microsoft where not involved was bacause that was happening in the late 60's (68 i believe) and microsoft was founded in the mid 70's (75 I believe).
Good at business yes. Good at software... not so much.
Perhaps when Korea fires it's long range nuclear missiles at the USA the new protocols will automatically issue an injunction to prevent the use of missiles until a full review can be implemented or perhaps an IP infringement writ based on the use pointy shaped missiles. Maybe it will just lobby for a change in US law making the deployment of foreign made missiles in US airspace illegal.
Maybe they will just try to buy Korea.
"and policy-based prioritization levels at the individual and unit level" in other words central control over content and who can receive the data. That sounds something the music and movie industry will really get behind (if not China or Iran).
This is an extremely bad development. One can bet on it that once the hard and software is build, the rest of the industry is gonna push this network to consumers as well. This can be the end of a "free" internet.
Or is that my Orwellian paranoia playing up again...
We've got DARPA, Lockheed and MS working together on this one. These are three of the best organisations at putting the "dead" into "deadline", so the chances of anything occuring on this are pretty damn minimal. My best guess, they'll come up with something that works, kind of, by about the same time engineering progress has given us superconducting comms cables and quantum-dot processors which invalidate everything they've done. So it's not like it'll affect anyone. And that's $31m which isn't going into researching new nukes, which is also good news.
This post has been deleted by its author
Excellent swallowing of the corporate spiel, dude!
"The anti-Microsoft crowd were quick off the mark."
Who says that they are all "Anti-Microsoft"? I don't doubt that some of them are, but most folk are aware of the woeful efforts made by Microsoft to produce something even approaching secure, plus their dubious efforts to produce a credible network setup before chucking it all up in favour of TCP/IP. Some of us are realists, ya know!
"Let's not forget that Microsoft is a incredibly succesful business and like it or not, warfare is also big business so I'd say MS is well suited to the task."
I could say that too, though I'd probably put a smiley at the end to indicate the satirical content! Let us not forget how Microsoft got to that position of "success" and why they might be suited to warfare. At least, part of the company anyway!
"Plus they have some very smart software/hardware engineers and computer scientists working for them around the world."
Oh yes. Can't forget that. Mind you, they don't have quite as many as they used to have...
"There is no such thing as bug free software, just fit for purpose software."
That's an old one. Actually, the correct phrase is "there is no such thing as a final product, just one in a high state of debug". The term "fit for purpose", however, is legalese and is open to all sorts of interpretation. The sort of thing that makes lawyers (and their bank managers) very happy! The trouble is that you can't afford a bug when you are in the field, aiming your whatever at the latest "enemy of freedom" or whoever the US has ants in its collective pants about this week.
"MIcrosoft does great fit for purpose software; the uqbuity of Microsoft solutions in everyday life is proof of this. So while people will complain about flaws, that fact it they do software that for the best part get's the job done."
See above. Microsoft sells software but not always its own. Some of it is source it from elsewhere, either by commission or by aggressively acquiring. In some ways, it means that Microsoft doesn't have to take the blame for every single bug it introduces, but it does call into question whether the company is actually capable of producing an item which is "fit for purpose".
"Also, if something needs reinventing and proprietizing for specific military use, who better?"
Tell me, would *you* be happy with the idea of a third party taking control of your military? Such fictional bodies as SKYNET and OCP could happen. Suffice to say that, if it was me, I'd be extremely suspicious of any third party company, especially one the size and track record of Microsoft, getting their digits involved with the defense of a country. Especially one the size and track record of the US.
Don't get me wrong I'm not against spending on research in to crazy stuff: That is where the future is to be found. My only beef is with the idea that the questionable business practices that have kept Microsoft at the top of the tree for so long will transfer over to Military technology. Failure is not an option and there is no recourse to law when it doesn't do what it was supposed to.
The thinking behind ARPAnet was to create a command/control network that would survive a nuclear attack and further that if the enemy knows your command/control will survive a nuclear attack they will be less likely to launch one. The survivable network concept was openly published in scientific journals to prove to the enemy that it worked. Trust in Microsoft's products is not very high. If Microsoft are involved in the creation of the next generation command/control network the trust in it's effectiveness is diminished and so the safety of the world is diminished also.
Leaving the participants in this contract out of my comment (for now) I do not understand the concept behind this excercise.
That is if you a) If you intend to send data over existing WAN infrastructure using new transport protocols (thus replacing TCP/IP) than everything in between requires to understand that protocol. Which defeats the efficiency factor required for military purposes - that is, they may transport using TCP/IP suite that everyone else "speaks" -- or -- something "else" (a replacement of TCP/IP) that everyone else "speaks". Uhm..?
b.) If they intend to use a separate WAN infrastructure (specific for military purposes) than you do not per se require a new transport protocol suite. That is, you can secure the traffic without developing new transport protocols. All in all, TCP/IP remains a very efficient protocol suite.
The only reason I can see that this contract would make sense is: Because TCP/IP is reasonably safe transport protocol suite - that is - it is not trivial exercise to monitor, intercept end relay Internet communications using TCP/IP -- a "new" transport protocol that makes such activities easier (at least for one party -- in this case military) would indeed be handy.
If the reason behind this is increasing transport security (or even increasing of the efficiency) -- you do not require to replace TCP/IP -- of at least conceptually not.
Having said that, from history, at the end of the day it is military who has given us a lot of technological progress and thus knock yourselfs out guys.
Coming back to contract winners - the interesting question is whether they were the only ones bidding....? :)
Leaving the participants in this contract out of my comment (for now) I do not understand the concept behind this excercise.
That is if you a) If you intend to send data over existing WAN infrastructure using new transport protocols (thus replacing TCP/IP) than everything in between requires to understand that protocol. Which defeats the efficiency factor required for military purposes - that is, they may transport using TCP/IP suite that everyone else "speaks" -- or -- something "else" (a replacement of TCP/IP) that everyone else "speaks". Uhm..?
b.) If they intend to use a separate WAN infrastructure (specific for military purposes) than you do not per se require a new transport protocol suite. That is, you can secure the traffic without developing new transport protocols. All in all, TCP/IP remains a very efficient protocol suite.
The only reason I can see that this contract would make sense is: Because TCP/IP is reasonably safe transport protocol suite - that is - it is not trivial exercise to monitor, intercept end relay Internet communications using TCP/IP -- a "new" transport protocol that makes such activities easier (at least for one party -- in this case military) would indeed be handy.
If the reason behind this is increasing transport security (or even increasing of the efficiency) -- you do not require to replace TCP/IP -- of at least conceptually not.
Having said that, from history, at the end of the day it is military who has given us a lot of technological progress and thus knock yourselfs out guys.
Coming back to contract winners - the interesting question is whether they were the only ones bidding....? :)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
