About Time (More on Reader Migration)
So I've been leaving it to the last minute because I was trying to figure out where Newsrob was heading... looks like NewsBlur for now, but maybe Feedly later.
I finally got around to looking at my star situation. Looks like it wasn't as bad as I feared; I was mostly using it to tag stuff when I was offline and couldn't stuff it into a bookmark service, so it tended to be relevant but also fortunately there wasn't a lot of it, as I periodically cleared my starred items and just had the latest batch to archive. My bookmark services aren't incredibly organized, but they still all function. So I just made a 'Starred' folder on Instapaper and loaded anything that seemed still relevant into that. I never used tags on Reader so now I don't have to mourn their loss.
I've been poking The Old Reader and Feedly -- the Old Reader has a completely irritating tic where if you can't remember where you've categorized a feed, it doesn't tell you and you have to hunt through all of your folders to find it if you want to move it. Feedly has gotten a lot less awful since the last time I saw it, so it's pretty clear they've been listening to their users. Neither has any way to export OPML that I have yet seen, so they are both roach motels. The Old Reader digs back deeper into feeds than Feedly/Reader do, which is both good and bad: Reader had cut me off when I went on a long sick reading hiatus, so I just lost a lot of stuff, which played merry havoc with things that need to be sequential like webcomics. (I've been rereading them manually, a bit at a time.) But now I have even more unread posts. Feedly is still syncing with Google so doesn't have that issue, and will presumably drop all of the older stuff anyhow.
Newsblur already contains an OPML export. I haven't tried the beta version of Newsrob with it yet because that will require fiddling. Greader does talk to Feedly, and I used to use it for things that Newsrob didn't handle well. But Newsrob is still my favorite because of its ability to do per-feed configuration of full web page articles, which fixes a lot of feeds that are malformed in various ways -- if it only gives an excerpt or is buggy, it follows the link and shoops that down, which means I can read the original web version offline. And read the skimpier and cleaner version on the less buggy apps. (Some I like to read both; the article in the stripped down feed version, and then click through to read comments, since the comments aren't always up to date on the cached article.) Newsblur's android clients are a joke, so now I'm thinking it was an advantage for the Newsrob team to pick Newsblur as the back end -- it adds a good client to a back end that sorely needs something nice on the Android front, whereas Feedly already has its own magazine-y thing as well as gReader talking to its API. Newsrob also promises that it will talk to other APIs. They're doing a three month OSS experiment and if the author likes the way it goes, then it will open permanently.
Anyhow, despite leaving it until the last moment I'm not losing all of my feeds, which is a big relief. And now I've recovered my starred items. But I am not perfectly sanguine because Feedly and The Old Reader are currently roach motels for feeds, and I am not yet certain that I will get Newsblur running in a way that satisfies me, though if it does that will probably be the one I wind up on. (I'm not committing until I feel like I am well-covered, but I'm happy to pay for both client and service once I think it will suit my needs.)
I'd love to take a closer look at tinytinyrss but I just do not have the functionality to handle all of the infrastructure for it -- I don't have a LAMP server lying around or the energy to get one running or maintain it. Fever is likewise out of my league and also the smart sorting would probably drive me to distraction; I want my feed queues to act like queues and not have any smarts. It does sound cool, especially if it can be toned down, but I do not have anything like the brain power to adapt to something so different, but I don't think it offers offline support anyhow, and I love being able to read offline and have my changes sync back up when I'm back on the network. I want a more or less drop-in replacement for Reader with offline support; anything else is gravy.
I finally got around to looking at my star situation. Looks like it wasn't as bad as I feared; I was mostly using it to tag stuff when I was offline and couldn't stuff it into a bookmark service, so it tended to be relevant but also fortunately there wasn't a lot of it, as I periodically cleared my starred items and just had the latest batch to archive. My bookmark services aren't incredibly organized, but they still all function. So I just made a 'Starred' folder on Instapaper and loaded anything that seemed still relevant into that. I never used tags on Reader so now I don't have to mourn their loss.
I've been poking The Old Reader and Feedly -- the Old Reader has a completely irritating tic where if you can't remember where you've categorized a feed, it doesn't tell you and you have to hunt through all of your folders to find it if you want to move it. Feedly has gotten a lot less awful since the last time I saw it, so it's pretty clear they've been listening to their users. Neither has any way to export OPML that I have yet seen, so they are both roach motels. The Old Reader digs back deeper into feeds than Feedly/Reader do, which is both good and bad: Reader had cut me off when I went on a long sick reading hiatus, so I just lost a lot of stuff, which played merry havoc with things that need to be sequential like webcomics. (I've been rereading them manually, a bit at a time.) But now I have even more unread posts. Feedly is still syncing with Google so doesn't have that issue, and will presumably drop all of the older stuff anyhow.
Newsblur already contains an OPML export. I haven't tried the beta version of Newsrob with it yet because that will require fiddling. Greader does talk to Feedly, and I used to use it for things that Newsrob didn't handle well. But Newsrob is still my favorite because of its ability to do per-feed configuration of full web page articles, which fixes a lot of feeds that are malformed in various ways -- if it only gives an excerpt or is buggy, it follows the link and shoops that down, which means I can read the original web version offline. And read the skimpier and cleaner version on the less buggy apps. (Some I like to read both; the article in the stripped down feed version, and then click through to read comments, since the comments aren't always up to date on the cached article.) Newsblur's android clients are a joke, so now I'm thinking it was an advantage for the Newsrob team to pick Newsblur as the back end -- it adds a good client to a back end that sorely needs something nice on the Android front, whereas Feedly already has its own magazine-y thing as well as gReader talking to its API. Newsrob also promises that it will talk to other APIs. They're doing a three month OSS experiment and if the author likes the way it goes, then it will open permanently.
Anyhow, despite leaving it until the last moment I'm not losing all of my feeds, which is a big relief. And now I've recovered my starred items. But I am not perfectly sanguine because Feedly and The Old Reader are currently roach motels for feeds, and I am not yet certain that I will get Newsblur running in a way that satisfies me, though if it does that will probably be the one I wind up on. (I'm not committing until I feel like I am well-covered, but I'm happy to pay for both client and service once I think it will suit my needs.)
I'd love to take a closer look at tinytinyrss but I just do not have the functionality to handle all of the infrastructure for it -- I don't have a LAMP server lying around or the energy to get one running or maintain it. Fever is likewise out of my league and also the smart sorting would probably drive me to distraction; I want my feed queues to act like queues and not have any smarts. It does sound cool, especially if it can be toned down, but I do not have anything like the brain power to adapt to something so different, but I don't think it offers offline support anyhow, and I love being able to read offline and have my changes sync back up when I'm back on the network. I want a more or less drop-in replacement for Reader with offline support; anything else is gravy.