When you're logged in to your account at unicyclic.com, you'll now get notified if someone comments on a post. It will subscribe you automatically when you create a new post, but you can edit your subscriptions too.
Comments also have their own feed now (which reminds me I should make the feed discoverable...)
I've spent so much time working on lettuceshare.org this year that my feed reader has started to feel a little neglected... Thankfully as the year draws to a close, co-ops tend to slow down and take a break, which has given me the chance to make some improvements here!
One problem I've had is that on mobile browsers the page gets cached minus the dynamically loaded content. Not very helpful, but easily fixed thanks to localStorage. On top of this the session markers for the reader module were reset when the page was loaded, which wasn't happening due to the caching. This now happens from javascript on page ready, which means it can also be smarter about resetting by looking at the last time it was done.
I like the grouping by feed that the Reader module provides, but it relied on item permalinks being from the same site. This isn't always the case (as in linkblogs) so the grouping is now done by the feed level link. This has the added benefit of being able to show the title of the feed at the top of each group, and the image associated with the feed. Looks much nicer in my opinion!
I was looking for some more information on this nail biting finish to the last senate seat in WA, and instead found Scott's blog. His last entry is a fascinating look at the limits of our industrialisation.
I'm really keen to integrate RSS further into my still very young blogging software. In the spirit of his post, I won't even offer any suggestions, just confirmation that I think some of his ideas are worth implementing. My favourite enabling idea of his is storing an endpoint to share url's to. He uses it on his link blog to share links on your own blog, but I think the same idea could be used for one click RSS subscriptions.
I also think embracing title-less posts is a good idea and hopefully catches on in a post-Google Reader world. (I will leave a title off this post to show my support...) In a world of status updates, a title could often be overkill.
The other posts linked to in the above post mention sequence id's on RSS items, I've had a go at implementing this by serving feeds dynamically. That lets me pass in the start and end positions in the feed as part of the query string.
The shutdown page that Google Reader now displays offers a link to a site displaying alternative readers. With Adblock installed the page says: "The free web is dependent on ads to continue to be free".
Some interesting reading on the eve of the Google Reader shutdown: The Retailification of Online Publishing. If the developer community hadn't allowed RSS, an inherently distributed technology, to centralise I think some of the issues here could've been avoided.
Another thoughtful piece is the war for the free and open internet. The privatization of the internet goes hand in hand with shutting down distributed technologies.
Maybe the shake-up in the feed reader ecosystem is enough to get things moving in the right direction. I get that discovery is the problem to which centralisation provides an answer, but I hope we can find other solutions. If the end of reader has meant anything to me it's the chance to reflect on what communication tools I use. I want more control over those tools, and plan on working to make that happen. There's great stuff happening, like the ideas coming from people at indie web, I look forward to discovering more.
unicyclic.com is finally up! This wasn't on the top of my to-do list, but with only a week to go before the #readerpocalypse I bumped it up so I had somewhere to take my feeds... I'm using dobrado reader, which depends on SimplePie to do all the hard stuff. There's still lots to do to make it a decent alternative, but at least I'm up and running on my own server with free software. :-)
The only thing I changed in the SimplePie library was to add a date to items if the parser didn't find one. The problem I have with the parser skipping dates is that I'm using SimplePie in multifeed mode, which will only sort items by date if they all have dates. So basically one bad item would make the feed unreadable. Good news is it's working great now!
Notifications
Comments also have their own feed now (which reminds me I should make the feed discoverable...)
Reader update
One problem I've had is that on mobile browsers the page gets cached minus the dynamically loaded content. Not very helpful, but easily fixed thanks to localStorage. On top of this the session markers for the reader module were reset when the page was loaded, which wasn't happening due to the caching. This now happens from javascript on page ready, which means it can also be smarter about resetting by looking at the last time it was done.
I like the grouping by feed that the Reader module provides, but it relied on item permalinks being from the same site. This isn't always the case (as in linkblogs) so the grouping is now done by the feed level link. This has the added benefit of being able to show the title of the feed at the top of each group, and the image associated with the feed. Looks much nicer in my opinion!
I'm really keen to integrate RSS further into my still very young blogging software. In the spirit of his post, I won't even offer any suggestions, just confirmation that I think some of his ideas are worth implementing. My favourite enabling idea of his is storing an endpoint to share url's to. He uses it on his link blog to share links on your own blog, but I think the same idea could be used for one click RSS subscriptions.
I also think embracing title-less posts is a good idea and hopefully catches on in a post-Google Reader world. (I will leave a title off this post to show my support...) In a world of status updates, a title could often be overkill.
The other posts linked to in the above post mention sequence id's on RSS items, I've had a go at implementing this by serving feeds dynamically. That lets me pass in the start and end positions in the feed as part of the query string.
Google Reader is gone
Sums up the whole saga nicely!
Following the Reader Deathwatch
Another thoughtful piece is the war for the free and open internet. The privatization of the internet goes hand in hand with shutting down distributed technologies.
Maybe the shake-up in the feed reader ecosystem is enough to get things moving in the right direction. I get that discovery is the problem to which centralisation provides an answer, but I hope we can find other solutions. If the end of reader has meant anything to me it's the chance to reflect on what communication tools I use. I want more control over those tools, and plan on working to make that happen. There's great stuff happening, like the ideas coming from people at indie web, I look forward to discovering more.
First Post
The only thing I changed in the SimplePie library was to add a date to items if the parser didn't find one. The problem I have with the parser skipping dates is that I'm using SimplePie in multifeed mode, which will only sort items by date if they all have dates. So basically one bad item would make the feed unreadable. Good news is it's working great now!