I'll play devil's advocate. Of course John Langford thinks what he and his group were doing held a lot of promise, and the work was fun and challenging and super important. But in the end you also need to live in the present and deliver. In the words of Jerry Maguire, "Show me the money". He had years to show Microsoft the money, and in the same time other people at other firms showed their bosses the money, big time.

this is not true. even with ideas with provable/strong scalability, there was still a lack of resources (especially GPUs) to validate such ideas at scale. ultimately, "showing the money" still requires some initial faith from those higher up who control the resources.