I'm of the mindset that developing a home-grown/in-house solution when there are time-tested & proven solutions out there is generally a mistake unless you're trying to innovate and offer it to the public. I worked at a place where the devs thought apache2, django, rails, etc. were too slow so they made their own web-server re-implementing the whole HTTP 1.1 protocol from scratch....
A UK government agency tried to release wartime aerial photos for free on the web and it was a disaster because the site was always overwhelmed with traffic. Now you can only get those photos by paying. This story could easily get picked up by a large news organisation and it needs to be hosted on something with lots of capacity.
There is rarely an existing solution for your specific needs. And sometimes, at some point it becomes more work to tweak an existing thing into what you want it to be, than just do it from scratch. Yes, you can't always know that beforehand, but that doesn't mean it's always safer to err on the side of taking something off the shelve.
That said, seeing how the images are creative commons licensed, it would still make sense to put them on flickr, too :P The more, the merrier.
