3 ms·

I don't think there's necessarily any great irony here. It's certainly possible that the same metric could be both (1) foolish to optimize and (2) a reliable in

by jal278 12y ago

I don't think there's necessarily any great irony here. It's certainly possible that the same metric could be both (1) foolish to optimize and (2) a reliable indicator of success. Think of yourself trying to solve a maze -- because there are cul-de-sacs, if you took the naive strategy of directly minimizing distance to the goal, you'll just become stuck in a dead end. Yet if you do reach the goal (by another means) the distance will still be minimal (zero). This is basically the concept of local optima in optimization. Of course, there are other situations, where through careless optimization a measure can become entirely disconnected from the holistic concept it is designed to encourage. Think for example of using "lines of code" as a heuristic for programmer productivity; I could unroll loops to maximize that measure and achieve a much higher value than the most truly productive programmer. So the question is whether long-term shareholder value is of the first or second type. I think it may be difficult for it to be entirely disconnected from success, although I agree that it is foolish to optimize directly.

pash 12y ago
Yes, I take your point, which is more precise and better argued than the author's. That is, I think your distillation of his argument misrepresents it by stripping out the logical inconsistencies (and the irony). Montier's letter is rather more muddled than your comment. He suggests not only that optimizing returns in the short run is a poor strategy for creating long-term value for shareholders, but that the benchmark of share value (even over the long run) is a deficient measure of corporate performance. Everything Montier writes suggests that he believes companies did better (in some sense that he never quite articulates) in the era when managers ran their firms as they saw fit, before the idea took hold that they should try to maximize share value. So the irony of his argument is the incongruity of demeaning the very benchmark he uses to suggest that J&J has somehow been a better managed company than IBM; the major arc of his argument vitiates the evidence he uses to support it. It is ironic, in the comedic sense, that Montier seems to be so imbedded in the culture of maximizing shareholder value that he invokes that criterion, seemingly reflexively, in the course of his argument against it. But it's a logical and rhetorical mess, too. Montier's argument is analogous in your maze-solving formulation to claiming that the final distance from the exit is a poor measure of the performance of a maze-solving algorithm while at the same time maintaining that greedy algorithms are worse than random algorithms because it turns out that the greedy ones end up farther from the exit. If you accept the first claim, then the second is a non-sequitur, and advancing the second argument undermines the first.
ScottBurson 12y ago
OT, but I am reminded of this great quote: [I]f we wish to count lines of code, we should not regard them as "lines produced" but as "lines spent": the current conventional wisdom is so foolish as to book that count on the wrong side of the ledger. -- E. W. Dijkstra

by jal278 12y ago

pash 12y ago

Yes, I take your point, which is more precise and better argued than the author's. That is, I think your distillation of his argument misrepresents it by stripping out the logical inconsistencies (and the irony). Montier's letter is rather more muddled than your comment. He suggests not only that optimizing returns in the short run is a poor strategy for creating long-term value for shareholders, but that the benchmark of share value (even over the long run) is a deficient measure of corporate performance. Everything Montier writes suggests that he believes companies did better (in some sense that he never quite articulates) in the era when managers ran their firms as they saw fit, before the idea took hold that they should try to maximize share value. So the irony of his argument is the incongruity of demeaning the very benchmark he uses to suggest that J&J has somehow been a better managed company than IBM; the major arc of his argument vitiates the evidence he uses to support it. It is ironic, in the comedic sense, that Montier seems to be so imbedded in the culture of maximizing shareholder value that he invokes that criterion, seemingly reflexively, in the course of his argument against it. But it's a logical and rhetorical mess, too. Montier's argument is analogous in your maze-solving formulation to claiming that the final distance from the exit is a poor measure of the performance of a maze-solving algorithm while at the same time maintaining that greedy algorithms are worse than random algorithms because it turns out that the greedy ones end up farther from the exit. If you accept the first claim, then the second is a non-sequitur, and advancing the second argument undermines the first.

ScottBurson 12y ago

OT, but I am reminded of this great quote: [I]f we wish to count lines of code, we should not regard them as "lines produced" but as "lines spent": the current conventional wisdom is so foolish as to book that count on the wrong side of the ledger. -- E. W. Dijkstra