gypsy,
I don't want to get involved, but I feel compelled to make sure you are on the right track, when Antonio talks about specific training.
In this sense, "specific" training is training around race pace (e.g. 95% to 105% of race pace).
Arguing for "specific" training isn't necessarily arguing against "general" training.
It's arguing that the "general" phase already include some workouts defined as a function of "specific" race pace.
The only real examples I can point you to is one of Renato Canova's detailed explanations of his training approach.
Lydiard takes an approach where you work on different qualities one at a time, i.e. aerobic first, then hills for strength, then anaerobic, then coordination then racing. He does aerobic first, because its effects last longer, because anaerobic training develops ability rather quickly, and because doing anaerobic training too soon can cause you to peak early. But in the early phases, over the course of a few months, everyone from 800m to marathons does the same marathon conditioning training, and hill training. It's only in the later phases (still up to about 3 months), that training is individualized for the athlete, and some workouts like time trials are included, that simulate race conditions.
But Canova talks about training that develops qualities related to supporting more directly the athlete's event. Aerobic support required for the 1500m runner is not the same as for the 10K or marathon runner. So maybe after a short preliminary introductory phase, the athlete moves to a general phase that supports aerobic conditioning, but also includes "circuits" (comparable to Lydiard's hill training) and interval training, then progress to a special phase (with circuits and intervals), then to a specific phase (with circuits and intervals). The concept is to progress from general fitness, first by "extending" the athlete's ability to run at race pace for longer distances, and then, work on increasing intensity.
Maybe some details are wrong somewhere, but for me the main contrast is that Lydiard defines training in terms of aerobic, then anaerobic development, in that order, while Canova defines training in terms of the qualities developed at different pace ranges around race pace, and how they support racing, and defines a training progression that is first extensive, than intensive.
Now to touch on something Victoria BC said -- it all looks the same. While that may be painting with too broad a brush, it's not completely wrong. There are certainly a lot of overlaps. Lydiard does focus on developing "specific" race qualities, in the later phases of training, with the use of time trials, and by including development races. It looks like the main problem with this approach, for many, is it takes too long to get there. You spend many months doing training that is not necessarily closely related to your racing. To me, this matters a lot less in high school, then it does for pros.
And Canova's plans certainly develops the aerobic and anaerobic capabilities in a way that produces results in important races. You might even find a trend that could be called highly aerobic in the general phases, progressing to an increasing anaerobic load. But there is no confusing that mid-distance and long-distance do very different general training.
You might also argue that the difference between the two approaches is marginal. But then again, a marginal improvement can be the difference between Olympic gold, and watching it at home.
I don't want to complicate the picture more, but Lydiard's approach has also been modified to handle competing for several seasons in the year. This is sometimes called a "complex" approach (because aerobic and anaerobic qualities are developed together, simultaneously week to week), but amusingly, it's much easier to describe, using "typical" weeks. Lydiards alternating "race-week" and "non-race week" is one example. "Deek" also used this "complex" approach.
I also want to touch on one last thing -- the time scale. A lot of the objections to "aerobic first, then anaerobic" seem to be applied on the scale of a single season, or maybe within a single year. But even if you are a fan of "post-Lydiard" modern methods that combine aerobic/anaerobic training year round, I still think that only works if you've already developed your aerobic system. So over the course of the career, I would certainly argue that it's still "aerobic development first", that a strong aerobic fitness is the prerequisite that allows you to consider a "specific" approach. Canova has also said something like "once you've built your aerobic house, then it's time to add furniture". Antonio thinks I don't understand the comment, but I interpret this exactly to mean that. In the early years, your aerobic systems take a long time to develop, over the course of years. But once you've done this, you don't nearly need to pay as much attention to aerobic development. What you've observed, is that some kids don't need to do it (as much), because they've already done other things which naturally developed their aerobic qualities.
Oh no. Did I just get involved? Don't expect me to post much more than this.