Via the Baran lab's Twitter feed, here's a provocative article on whether total organic synthesis has a place in the modern world or not.
One may wonder why this situation has passed undisputed for such a long time. Currently however, wide parts of the chemical community look upon total synthesis as a waste of time, resources and talents. Behind the scene, it may even be argued that the obsession to synthesize almost any natural product irrespective of its complexity and practical importance has blocked the development of other more relevant fields. Therefore, it is high time to consider a reorientation of the entire discipline
That's a bit of a straw man in that paragraph, and I have to note it, even though I do feel odd sticking up for total synthesis (about which I've been pretty caustic myself, for many years now.). I don't think that there's been an "obsession to synthesize almost any natural product", although it's true that many new synthetic methods have used some natural product or another as demonstration pieces. But the author, Johann Mulzer, came out of the Corey group in the old days, and has spent his career doing total synthesis, so he's speaking from experience here.
He goes on to argue that the field does have a place, but that it had better shape up. Short syntheses have to take priority over "first syntheses", because (let's face it), just about anything can be made if you're willing to throw enough time, money, and postdocs at it. The paper is full of examples from Mulzer's own career (and others'), and if you read it carefully, you'll see some unfavorable contrasts drawn to some Nicolaou syntheses. He finishes up:
In conclusion, this article tries to show how various strategies may be used to streamline and to shorten otherwise long synthetic routes to complex target molecules. The reader may get the impression that it pays very well to think intensively about cascade reactions, intramolecular cycloadditions, suitable starting materials and so on, instead of plunging into a brute-force and therefore mostly inefficient sequence. After all, there is an iron maxim: if a target cannot be reached within, say, 25 steps, it is better to drop it. For what you will get is a heroic synthesis, at best, but never an efficient one.
A 25-step limit would chop an awful lot out of the synthetic literature, wouldn't it? But it's not fair to apply that retrospectively. What if we apply it from here on out, though? What would the total synthetic landscape look like then?