The ASF infra team is encouraging folks to move away from the old roller blogging system.
We have moved our blog content into our website. You can find the new home with hopefully all existing content here: https://groovy.apache.org/blog/ Source files here: https://github.com/apache/groovy-website/tree/asf-site/site/src/site/blog I have a few more details in the users mailing list: https://lists.apache.org/thread/kv13scwzftljdqf42wfol3cokm2d906c In particular, it is hopefully even easier for anyone to create posts now! New posts are most welcome!! If folks can do a quick check that things look okay, that would be great. There are some additional things we could do (other suggestions are also welcome): * We could move blogs to its own repo and splice it into the existing website. This reflects that blogs and the website have different lifecycles but for now that seems a non-essential additional complexity. Anyway, that might be something for later. I am imagining JReleaser sending out tweets/toots whenever a new blog post goes out. * I have placed a tag cloud on the index page but not fully implemented the ability to click on cloud topics. If anyone wants to play, go ahead! * Each topic has keywords (tags). Currently, the "Related posts" list in each blog is generated by matching keywords and also looking for a Groovy version number in the blog titles. We could do more complex article content analysis to find additional related posts. * We could make more use of asciidoc file inclusion so that blog content comes from unit tested source files (like most of our documentation) but since blogs are often timely articles like announcements, they often reflect code at the time of the post, rather than something we would maintain over time. So, I haven't tried to do that when converting over the existing content. * There is a preliminary atom feed (https://groovy.apache.org/blog/feed.atom) but I must admit to not knowing the latest best practices for such feeds. The original feed in the roller system broke. It was using XML 1.1 features but declaring it as XML 1.0 which seemed to break readers. Does anyone use image/media/thumbnail entries in atom feeds? The old feed embedded the entire contents into the feed but now I just have a short description. Does anyone use RSS instead these days? * The existing site made it easy to deepen engagement: "You're viewing a weblog entry titled Groovy Dates And Times Cheat Sheet. If you like this entry you might want to: del.icio.us Bookmark it submit to dig digg.com Digg it slashdot Slashdot it technorati See who links to it" While somewhat outdated, I don't have an equivalent replacement yet. Does anyone know best practice or have any suggestions for this? * I have very little SEO/analytics experience. Do any folks have advice to offer in that area? Cheers, Paul.
