@brentsimmons I am looking forward hopefully to syncing capability (Inoreader). I love the interface already, but without a sync system, I can't use NNW full-time.
@brentsimmons I had similar ideas for the feed in my app -- you are my RSS reader hero after all ;) Headline image is the first sensible image from the article, so only a PNG or a JPEG, for example. I even set the accepted content header to include only those, but CDNs usually ignore it. Also, constant cell height was both a perf and scannability optimization. And if the content snippet is longer than the cell, I apply a slight shadow, which makes for a nice indication that there's more inside.
@brentsimmons ignoring tiny gifs and huge images takes a few unnecessary downloads sometimes, but it's worth it - the feed looks rich in both content and pretty pixels to look at. And the headline image is only fetched when on Wi-Fi, since it would be too much of a burden on the cell connection.
@brentsimmons How about if there is a summary in the post, would that work as the second line instead of the first line of the content? Also, if you want richer understanding of posts, the indieweb way is to use microformats to add them - you can look for a u-featured or u-photo image, and there are ways to cue other post types too - see indieweb.org/post-type...
@kevinmarks It may already be using summary, when present in the feed, instead of the first line of the content. It should be, anyway. (But summaries are rarely present.)
As for looking for u-photo and similar — I’ll make that a feature request. (JSON Feed also has a way of specifying a featured image.)
@macbirdie When I did this with NetNewsWire Lite 4, it was pretty cool — except when it wasn’t, and it turned into a thing I had to keep tweaking. And there was the problem with noses-but-no-eyes when cropping. I may revisit this later, of course!
@brentsimmons I've been meaning to use face detection in Core Image to at least center to some faces, I'll see how that works out
@brentsimmons the summary/content distinction (vs ambiguous description) was an Atom thing, but has been made a bit moot by the rise of microblogs. Its there in activity streams, and h-feed with the name/summary/content split. Useful for feeds generated from other sources too, or when the content is more complex.
@brentsimmons Small copy edit: I think you have a missing word at the end of the last paragraph of the "Layout" section. I think "but I like it if seems that way." probably wants to be something like "but I like it if it seems that way."