Interesting concept, however if we accept all your premises as true, we still must contend with one of the most important concepts in data analytics: recency. With each passing year this dataset becomes less useful as a model for predicting future behavior, which is really the whole point of collecting and analyzing data in the first place right?
Absolutely. Current data will always be highly valuable for behavioral insights - perhaps I should have clarified that I don't think we'll just stop caring about new datasets. Rather, I'm saying that going forward into history, this period of time will always be considered high value for its relative completeness, authenticity, and bot-purity. So, I'm not saying "future data analysts will continue recycling this chunk of time" but more that future data archeologists will keep coming back to this period for its richness and validity in trying to model "real" human behavior patterns and interests.
Interesting concept, however if we accept all your premises as true, we still must contend with one of the most important concepts in data analytics: recency. With each passing year this dataset becomes less useful as a model for predicting future behavior, which is really the whole point of collecting and analyzing data in the first place right?
Absolutely. Current data will always be highly valuable for behavioral insights - perhaps I should have clarified that I don't think we'll just stop caring about new datasets. Rather, I'm saying that going forward into history, this period of time will always be considered high value for its relative completeness, authenticity, and bot-purity. So, I'm not saying "future data analysts will continue recycling this chunk of time" but more that future data archeologists will keep coming back to this period for its richness and validity in trying to model "real" human behavior patterns and interests.