An aptly named "TimesMachine" from The New York Times offers up online images of historic newspapers from the 1850s through the 1920s. You can view sample papers at http://timesmachine.nytimes.com/browser, but to browse through the 70-year spread of history you need to be a home delivery subscriber of the paper.
Besides the boon to students and history buffs alike, the new site gets my gears going about the online media business and the huge promise afforded by services like Amazon's S3.
That's because while I agree with Silicon Alley Insider that the offering won't be a 'huge money-maker' for the Times, it doesn't have to be. From what NYT engineer Derek Gottfrid describes, once the papers were all digitized he was able to use Amazon's web services to quickly get the rest of the work done.
In a video interview with Robert Scoble, Gottfrid says "Once we moved it up to Amazon for a couple hundred dollars... we were able to do it in about 24-36 hours... It was a great use of Amazon for us to be able to churn through the data."
Gottfrid doesn't specifically describe how he used Amazon's services, and I don't want to give Amazon too much free advertising here. But this seems like more justification of my high hopes for the new platforms afforded to enterprising developers, whether they're at resource-starved companies or in their own living room, for getting great stuff up online.
Along with Amazon, Google is also jumping into this space. I previously wrote about its Engine Platform, which likewise aims to allow anyone with a great idea to bring it to life online.