An awareness of the competing benefits of upgrading components of an existing system versus creating an entirely new system is one that is particularly notable as you try to evaluate military technology.
Consider the case of Iran's Combattante III missile boats. The system design is three decades old, and as ships they aren't all that. At 249 tons, give or take, they are about one sixteenth the size of the smallest surface combatant in the American navy, the frigate, and about one thirty-sixth of the size of an American destroyer. But, they were designed to carry a key component, four Harpoon anti-ship missiles. Two of these missile boats have the same anti-ship missile capacity as many American destroyers and cruisers.
Some analysts argue that this doesn't mean very much because American warship have sophisticated countermeasures, in particular the Aegis fire control system, since the days when two similar missile boats sank an Israeli destroyer in 1967. And, the American made missiles the boats were built to carry have since been replaced with the Chinese version. But, missiles that travel up to 70 miles, at 600 miles an hour (i.e. the time from launch to arrival is typically under seven minutes), just 15 feet above the sea, which clearly have the ability to destroy a U.S. surface combatant if they do hit their target, are a threat to take seriously. You can also bet that the missile boats aren't nearly as expensive as an American warship. Also, there is little doubt that one of these missile boats could easily sink any unarmed and unescorted civilian ship in the Persian Gulf, a threat than in hostile times could place on huge burden of any Navy to escort the normal flow of traffic, even if that escort operations was 100% effective.
The U.S. has also taken naval resources are retooled them, turning submarines designed to launch nuclear missiles, into submarines that can launch conventional cruise missiles from just off the shore of the coast nearest the target, greatly reducing the time that a target has to detect and respond to the missiles launched.
Another situation where the component v. system issue comes up is in military aircraft. Historically, about 80% of all air to air combat incidents are single shot affairs. The loser is destroyed before there is time to launch return fire. As a result, many countries with modest defense budgets upgrade their avionics (so that they can see the opponent before the opponent sees them), and their air to air missiles, while retaining older aircraft that are doomed in a genuine dogfight with more modern jet fighters.
The component v. system issue comes up with American warplanes in Iraq and Afghanistan as well. In neither country is there even a single enemy aircraft and there are no fixed anti-aircraft artillery batteries. One could deliver the relatively small number of small bombs that are used at this stage of the respective conflicts with modified small civilian aircraft just as well. But, the sensors on these planes, designed for far more rigorous situations, have turned out to be very good in a reconnaisance capacity, located threats to ground troops or ground targets. The system as a whole isn't designed for the kind of conflicts the U.S. is currently facing, but certain components of that system have proven useful.
Perhaps one of the most notable cases of a new component dramatically changing an existing system is the AC-130, which is basically a C-130 cargo plane with a howitzer and a few smaller weapons mounted on it, making it a fire support platform for infantry soldiers needing reinforcements.
Of course, bad systems can be a problem. The B-1 bomber's practical usefulness was delayed for many years because its swing wing was difficult to make work. The M-1 Abrahms tank is some fine individual components, like good sensors, a powerful main weapon, and powerful engines. But, the system as a whole has been used far less than originally contemplated because at 70 tons, it is so heavy that it collapses weaker bridges, is too large to operate in more confined spaces, and can't be quickly deployed to distant battlefields. The powerful engines also end up spewing hot exhaust that makes it hard for infantry to support the tank in urban environments, and the designers who were focused on fighting World War III in the Fulda Gap overlooked such niceties as shielding for one of the machine gunners, an oversight that caused a significant share of the combat deaths in tank units in Iraq.
Looking beyond individual vehicles, one of the greatest problems the U.S. military has faced is poor communications between vehicles and infantry units and fire support controllers. While the military, in theory, has a system to handle this issue, in many cases civilian unencrypted walkie-talkies, civilian cell phones, and e-mail sent from within military vehicles using Microsoft Outlook, have had to step in to fill the gap.
Of course, the system v. component issue is hardly confined to military technology. The Blackberry and the similarly popular technology of Text Messaging with cell phones are pretty obvious combinations of a cell phone and e-mail, but until it was invented, the integration of the two had been limited to crude versions built at home by geeky hobbiests, rather than elegantly from the bottom up. The iPod is another example of truly elegant refining at the system level of something (MP3 players) which had been floundering until Apple imposed an improved user interface on the technology.
On the other hand, sometimes components can make a huge difference. For example, while it has long been possible to scan a photograph onto a computer and print it, and webcams like mine (which has about 0.1 megapixels of resolution) have also been around for sometime, the high resolution of current consumer digital cameras (the new affordable ones run 4 or 5 megapixels and are very light), and the advent of specialized photoprinters that use high quality, standard sized photopaper, often direclty interfacing with the camera rather than though an intervening computer, have been far more important that developments in computer systems themselves in making digital photography so popular that camera maker Nikon already does less than 5% of its business with film cameras, has ceased making film cameras all togther, and will stop selling them sometime this summer. High end camera lens, however, continue to be produced because unlike the portion of the camera that interfaces with the film, both digital and film cameras can use them.
The long predicted paperless offices and stores, on the other hand, is a case of components that have outrun the system. The individual technological elements necessary for a paperless office, like e-mail, scanners, wire transfers, ample data storage capabilities, digital voice mail, and word processors have all been around for quite some time. But, no one has come up with a really quality integration of the various components that lawyers and other office operators have been comfortable with, and the laws necessary to allow a paperless office to interface with the world outside the office have also not kept up. For example, the idea of authorizing a financial transaction with a signature written with a stylus on a computer screen, instead of with a pen on paper, is just starting to secure acceptance in the world of retail commerce. Processing checks without physically moving the little slips of paper that they are has alaso been a long time in coming. Automatic recurring electronic payments charged to (or credited to) credit cards or checking accounts are also just coming into their own. High speed internet access, which allows large documents to be transferred as e-mail attachments, is finally starting to overcome some of the interia that a multitude of different storage formats for data had previously held back.
It is almost always more expensive to buy an entirely new system, rather than a new component, so technology purchasers always need to ask themselves whether they really need a new system from scratch (which is typically far easier to do) or to use more judgment and expertise to upgrade an existing system.