Posted on 04/15/2005 2:27:55 AM PDT by Zero Sum
"There is no weapon system in the world that comes even close to the visible symbol of enormous power represented by the battleship." -- Retired Gen. P.X. Kelly, USMC
WASHINGTON, D.C. -- Those words of the former Marine commandant resonate with me. In 1969, gunfire from the battleship USS New Jersey (BB-62) saved my rifle platoon in Vietnam. During her six months in-theater, the USS New Jersey's 16-inch guns were credited with saving more than 1,000 Marines' lives. The North Vietnamese so feared the ship that they cited her as a roadblock to the Paris peace talks. Our leaders, as they did so often in that war, made the wrong choice and sent her home. Now, 36 years later, Washington is poised to make another battleship blunder.
(Excerpt) Read more at townhall.com ...
The The Alaska &, I think, the Guam were completed.
You are talking theory and I'm talking practical experience. If the goal is to make a truly autonomus, fully aware system, that is able to recognize and correct mistakes not solely limited to it's own programming, it will not happen anytime soon, if ever.
Let's take Big Blue and Deep Blue, as examples. Big Blue was designed to play chess and Deep Blue to predict oceanographic events.
Big Blue could play chess simply by recording and then extrapolating the possiblilites that stemmed from every possible move at any given time. In effect, it's options were limited even if there are millions of possibilities. It then reacted to those moves in the way that seemed most logical based upon probabilities. It could be "faked out" by unorthodox play, or, if you kept at it long enough, you would eventually find a bug in it's programming that it could not respond to. It "learned" from it's mistakes because it recorded every move and then a person went back and told it where it went wrong, changing the programming to avoid the same errors. It could 'think' tactically but not 'strategically'. It could not discern strategy, only react to to it's opponent's moves. In terms of initiating a game, it generated a random first move.
Deep Blue was an even greater disappointment. There simply were too many variables to consider when predicting the movement of a tide, for example, and it's effects. It's shortcoming was that while it could mathematically generate billions of predictions, it's input parameters were limited by what it's programmers could tell it. It could not, for example, apply anything like an empirical method to test assumptions or sift facts.
It's not the mechanics that are lacking, it's the software, and that software comes from an unreliable source to begin with --- human beings. Even highly intelligent, mathematically-competent human beings make mistakes and assumptions that may not be factual or even sensical. As an example, I give you the great Y2K scam -- the computers didn't care what century it was, the SOFTWARE did. The limitation is always software. I don't know about you, but when I write code, I apply logic and not theory. If there's a fault in my logic, there's a fault in my software. I can program based on theory all day, but it doesn't mean I have a viable result when I'm done.
P.S. intersting debate though...
Whatever is going on in our heads, it’s not magic. It is knowable and machine intelligence will be achieved.
What machine intelligence turns out to be, it won’t be base on current architecture though.
Well, No.
As originally built, the BB61 class had 212,000 shaft horsepower. Four LM2500s would provide 120,000 shp.
Its common to have two engines per shaft, but it isn't common to rely on running both of them on the same shaft together. I've been on many gas turbine ships, and its been rare to have both up and running at the same time.
That said, I prefaced my statement by saying you would need two coupled per shaft to get the shp required. So for proper maintenance, you would require four per shaft.
With modern electric motors, it might be better to pull the shafts altogether, and go to pods. Gas turbines could then power generators.
The concept of a Dreadnought type ship dominating the seas are over. But there is still a place for a well armored weapons platform, loaded with todays mordern weapons along with a large gun or 2 for bombbardments, shells are cheaper then cruise missles.
Same type of work here too, it is how I got into all this.
What we found was that we could handle the volume, but could not react fast enough to changing market conditions. For example, a simple price shift on a stock might require different trading stategies, each of which required several additional layers of programming. If you had 13 contingencies for one issue, the computer did not 'instinctively' know which one to use.It still had to be told what to do, and that was with $3 billion worth of hardware and software.
This is a fundamental failure of design and theory. We designed such a system in a similar fashion with the same issues (though for a mere $1.3M in 1995 -- a bargain considering it was 500kLoC), and it still needed a dozen people to hold its hand. I took to calling such systems "computer-aided" rather than automated, since they couldn't actually do anything on their own, only help the operators make better decisions faster.
The goal was to eventually replace about 20 people's brains with a computer, but we couldn't get there from where we were. The system had to be able to adapt to major shifts in the market it operated on in real-time and detect them early enough that it could maximize return. We designed a new system from scratch that used a highly adaptive topology optimization algorithm that we researched and developed specifically for that project. Using the new generalized algorithm greatly reduced the code size, to about 150kLoC, and tests on historical data showed a ~10% improvement in net profits versus the human operators for the same periods of history due to superior decisions on the part of the machine. The better solutions are simpler and more general, not more complicated, but to cover the same cases correctly and adaptively you need to take a more inductive approach. The nature of the algorithm was such that it tended to always make a good decision even though there was no way to guarantee that it always made a good decision.
The irony is that the Global 100 company for which this system was designed for decided that they were uncomfortable removing humans from the decision loop, particularly since we could not guarantee that it would always make good decisions (as though humans do), and shelved it despite its demonstrable superiority and adaptivity compared to the human operators. Note that I would consider that design obsolete today as the math behind such things has come a long way since (it barely even existed then), but it was an interesting project that really got me interested in theoretical mathematics and the number of unsolved problems of application that exist there.
We still work off of pre-set instructions -- when the price of "X" reaches "Y" then do "Z". Only human beings are capable of making the leap that when "X" is not "X" but in fact "A", that "Y" and "Z" get thrown out the window.
I know of a couple good hedge funds that do not use axiomatic decision theory in their systems. I'm working on a new state-of-the-art system design for this business right now that uses context-sensitive non-axiomatic reasoning to make decisions (the esoteric mechanics of such elided for the sanity of the reader). There is a trade-off here: such systems lack the brittleness of the type of system you are working on, but they also cannot be easily verified and audited to give a particular result under all possible conditions which adds some uncertainty and unpredictability to the mix. Fortunately for you, generally inductive systems are not used at the moment in the markets (very, very tough theoretical engineering problem), as even a modest one would provably eat axiomatic rule-based systems for lunch on the open market.
By 1943, those problems were resolved. And by 1945, we were producing some interesting high-tech torpedoes. To correct an earlier post of mine, the cruisers with 15 6" guns were the Brooklyn-class.
In my opinion, the American ships built around 1935 or thereafter (or significantly rebuilt) were the only ones that had substantial AAA. And even then, the ships had to be numerous and tightly-grouped to provide effective fire, unless you're talking about the Iowa-class and the Massachusetts-class battleships or the Essex-class carriers.
Otherwise, a large number of aircraft could still be overwhelming. The best defense against incoming bombers has always been fighter aircraft.
I've also heard the story about the Argentine Navy's somewhat casual attitude about condition zebra.
And you're right about what Belgrano would have been able to do to any British force on the island.
Look out/Look out/Look out for Jimmy Valentine ...
The problem is the complete absence of strict theory in software design. Virtually all programmers design software without having any idea whether 1) the design is theoretically valid for the intended result, and 2) whether the design is a theoretically optimal implementation per the specs. Given this, it is no wonder that software is so crappy. I had to put the smack down on a programmer today who was trying to do something using a multi-version concurrency control design that is disallowed by transaction theory if you want perfectly correct behavior all the time in MVCC. Yes, the problem was obscure, esoteric, and would not even show up under many conditions and probably most testing. Yet everyone intuited that the design was correct even though I could prove that it was fundamentally broken by going to the theory, which no one bothers to do never mind actually understanding it. A lot of what is wrong with software is that it is treated like literature rather than math by programmers. All software design is an implementation of strict mathematics, and if you can't explain why a software design is correct in terms of mathematics then you don't understand the software design and it is likely broken.
AI has suffered from this badly. For the entire 20th century, not a single AI theorist could demonstrate why their particular theory du jour should work other than their demonstrably useless intuition about the nature of the solution. There was never an exhaustive mathematical basis for their design theories.
Note that pure mathematics does not always proscribe a good engineering solution, but that is a different issue. It is generally possible to show that a good engineering solution exists if the math is correct, though it may take a bit of work to derive an engineering solution. The difference between theory and practice is that proper practice is often a theoretically suboptimal or constrained implementation of theory. The ideal and real are not fundamentally different, the real should be the ideal re-derived with real constraints (ideally...).
Architecture is irrelevant. If it is possible at all, then it must also be possible on vanilla silicon. This was established in the 1930s(!), though this canard refuses to die.
Computer science is far from complete. We still discover new families of algorithms, which almost always seem "obvious" in retrospect despite almost a century of computer science not discovering them.
http://www.combinedfleet.com/baddest.htm
Interesting site on the whole best of the battleship issue with lots of underlying explanation for the analysis. Not sure it is definitive, but it gives plenty of information to argue over.
Alot of people think that arsenal ships will make carriers obselete.
I saw a pic of the Taiwanese navy trainging. They hit a cruiser with a missile and it folded like a jacknife.
If WW2 shells and bombs could get a battleship I'm sure some modern bombs could.
bump
Got any pointers? A Google search is not very enlightening. (Amusingly, the third hit is your FR posts page).
Roux! Makes gumbo well, gumbo! ;^)
Seriously, and not taking any side on this issue, but I do remember my old roomie's dad showing me a dictionary from 1900. In it, the definition of "airplane" was: a horseless motor carriage with wings that will never work....
To some degree...it wasn't fixed properly during WW-2,
especially in the various Destroyer classes which fired the 21" Mark 15's.
article excerpt:
Criticism of the destroyer launched Mk.15 is almost nonexistent. This is strange because the principal differences between the Mk.14 and the Mk.15 were in the size of the warhead, the fuel load, three speed vice two speed and slightly slower high speed, 45.0 k vice 46.3 k. One might speculate that it is even more difficult to distinguish misses from duds in a high speed destroyer attack than it is in a more measured submarine attack. The Mk.15 did, in fact suffer from the same defects and they were rectified in essentially the same way that those of the Mk.14 were. The Mk.13 was a slower speed torpedo so it did not have the contact exploder problem and it used the Mk.4 exploder which did not have the magnetic influence feature.
The Impact Exploder
Once the depth problem had been fixed and the magnetic influence feature of the Mk.6 exploder deactivated, it came the turn of the impact exploder to demonstrate its merit. Unfortunately the initial result was a plethora of duds, solid hits on targets without warhead detonations14. This problem was suspected earlier, but it was not until the other two problems had been eliminated that there was unequivocal evidence of a problem with the impact exploder. This difficulty was a further frustration for the operating forces, but fortunately it was quickly diagnosed. The key to the problem was again the increased speed of Mk.1415. The impact portion of the Mk.6 exploder was exactly the same as that which had been used in the Mk.4 and Mk.5 exploders. The Mk.4 worked entirely satisfactorily in the 33.5 knot Mk.13 torpedo. What was overlooked was that in going from 33.5 knots to 46.3 knots the inertial forces involved in striking the target at normal incidence were almost doubled. These greatly increased inertial forces were sufficient to bend the vertical pins that guided the firing pin block. The displacement was sometimes enough to cause the firing pins to miss the percussion caps, resulting in a dud.
***************************************************
The Westinghouse Mark 18 electrical torpedo went into service in submarines during 1943.
The drawback on the 18 was its speed...which was half of what a Mark14/15 could achieve.
The standard wet/heater torpedo had a 17%+ higher hit rate on Merchant ships than the Mark 18.
The Fletcher class DD's of WW-2 had 2 torpedo launchers... fore and aft of the second funnel stack.
Several Fletchers came apart violently from secondary internal explosions after kamikaze impacts.
Too much ordinance when coupled with 5 inch mounts and 40 mm tubs..plus ammo lockers....and the DD's boilers.
Fletchers were tough....but their were key frames ...which if failed from pressure wave or heat spelled doom for the plucky can.
Some destroyer crews who witnessed other cans erupting hundreds of ft into the air off Okinawa,were tenative to sleep inside their ships after.
Officers on these DD's had a difficult time getting them back below decks....and mitigating tensions.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.