Refining My POSSE Setup

Dave Slusher responded yesterday to one of my posts tracking my experiments with implementing a POSSE strategy on my web site. He clarified a couple of things. POSSE actually stands for Post Own Site, Syndicate Everywhere. POSSE as a concept is being advanced by the Indie Web folks. I have been following the Indie Web constellation of ideas, as well. They claim that strategies like POSSE are in some ways an improvement over federation, a concept which as much as I like it has gained little traction. Federation simple means you post wherever you want and your info flows to federated services seamlessly. Email is probably the most popular example of a federated system.

With federation, there still may be barriers that interfere with simple sharing of information. The examples Indie Web give run along the sentiment none of the existing services directly support federation so you have to adopt something new and try to cajole all your existing online connections to do the same. As the short history of the idea has demonstrated, that’s just too much inertia to overcome for all but the early adopter set.

By contrast, implementing a POSSE approach puts your content in front of your friends and followers wherever they happen to be. It concentrates on the ownership of the origins of that content (the Post Own Site part) so that if there is some issue with a particular aggregator on the other end, it limits the impact on your ability to generate and propagate the information and messages you want. Perhaps a fine nuance but one I have been pondering since Dave’s correction.

I am now onto the second day of an ad hoc weekend project to make my existing site (and one new one) drive a POSSE strategy by simply adding to tools I had already been using, primarily WordPress. I think I have things dialed in a bit better and have found some simple solutions to the few workflows I laid out in my last post.

For the link sharing workflow, I realized I was overlooking that I could extend my RSS aggregator, Tiny Tiny RSS. Doing so makes more sense to me since my aggregator is the canonical store of my curation efforts, not my web site. Tiny Tiny RSS supports plugins, as so many tools do, and I found plugins to share directly from it to Twitter, Google+ and Facebook. You can follow my curated links directly out of my aggregator or continue to follow along on Twitter. Up until now, I had only really been sharing links there but I think I will experiment with pushing those links to Facebook and Google+. I will keep an eye on the reaction of my connections on those services and adjust as seems appropriate. Regardless, this workflow should work much better using more of the right tool for the job.

For the workflow of sharing blog posts, I returned pretty much to the typical configuration of SNAP. I restored the templates it suggests for each target service which is pretty much an excerpt and a link. I set these up to auto post anything within the categories I use for blogging, on both my sites.

For the lifestream workflow (thanks Chris Miller for reminding me of this term) I re-thought my approach. I like the idea of posts to a specific category being handled a little differently but tweaked that to try to make it a little more intuitive. I still withhold life stream posts from my main page and my feed, as I figure these introduce a bit more noise so should require a little more intent and action beyond just visiting my site. If you navigate into my archives or view the life stream category itself, you can now see these intermingled with the rest of my posts and all on their own, respectively.

That left cracking the length constrained vs. free form content problem. I didn’t want to be editing sharing templates post by post, that is a retrograde action, introducing more work rather than simplifying. I had the idea of adding a second config for each of the social services I use. Each new setup would have a template more appropriate for direct posting, that is just the contents of the post on my site without any link back or other extra information. SNAP’s free edition doesn’t allow adding the same social service more than once, however they are running a sale on the pro edition right now. For just under fifty bucks a year you get both the pro upgrade and support for Google+. I went ahead and bought the upgrade, figuring it was cheap enough over the course of a year and if it worked, would be well worth the monetary cost.

So in my SNAP settings as an example I have two entries for Twitter. The first uses the blog post template, the second a template that just shares the text of a post on my site. The first autoposts anything I publish to the usual categories for my blog. The second only autoposts when I publish to one of three life stream categories.

The last piece of the puzzle, at least for now, is that I have gone from one life stream category for everything to three. I kept the existing category so that I can write a short post and simply have it go everywhere. I added one for long form content and added it in SNAP only to those services that don’t have length constraints, like Facebook and Google+. The third category is for short form posts for, you guessed it, Twitter which has its character cap. If I have a longer message I want to share everywhere, I can write it once, adding it to the long form category. Then I can chunk it up and post it as a series of length appropriate entries to the short form category. Tying this all to autoposting to remote services and sites means that if I only have my phone, everything will still work simply by using the right category in the WordPress mobile client. This is critical since up until now, a lot of what I post on the various social services I post from my phone.

The view of my life stream category on my own sites may at times appear a little repetitive. That original category contains the other two, but it is closer to what I think is ideal. Hopefully that makes sense. Or if it doesn’t, trust that it makes sense to me. Or just visit my web site and explore for a bit. At least I think this will all work for right now. It is good enough for a weekend of noodling around, anyway. I will continue to tinker, sharing updates as I come up with improvements or interesting possibilities.

Initial Impressions of a POSSE Setup

I’ve only been playing with SNAP, a WordPress plugin, a few hours, but have some initial thoughts.

NAND Cat Has a Posse, used under CC-BY thanks to Flickr user Paul Downer

NAND Cat Has a Posse, used under CC-BY thanks to Flickr user Paul Downey

I started down this path thanks to Dave Slusher who wrote about POSSE which stands for Post to Own Site, Share Everywhere. I like this concept a great deal. Investing the typical time more and more folks do in communications and information only to have that effort evaporate at the whim or circumstance of the platform, tool or channel of the moment seems very foolish to me. I had been experimenting with Bridgy but still manually sharing all of my posts and posting shorter thoughts directly to all these sharing outlets like Twitter and Facebook.

I stood up a second site for another set of interests of mine. Doubling my existing workflow was not appealing in the slightest. I decided to take the plunge with the WordPress plugin, SNAP, that Dave mentions. For the most part, I really like it. With a little fine tuning, content showing up on sharing networks looks native but cleanly and clearly originates from my site. I will warn that this is a power tool, which affects the effort to set up and how smoothly it supports more than one workflow.

In terms of the installation, just getting it into WordPress is easy enough. Connecting it to other services may require a good deal more effort. I have only connected Twitter and Facebook so far as Google+ requires some additional bits. For both Twitter and Facebook, I had to sign up as a developer and essentially create new applications with each of those for each of my sites. The SNAP documentation for this is superb but this may be outside a lot of people’s comfort zone.

For the primary workflow of simply sharing regular blog posts, SNAP is great. You can configure templates for messages to each with some pretty clear replacement parameters (although finding the list from the plugin is pretty much impossible, I bookmarked their documentation page.) Unlike other tools I had tried previously, messages can be tailored so they look native to the sites on which they appear. This is a huge plus as the current crop of popular sharing sites increasingly penalize anything coming in that doesn’t smoothly fit into their design, flow and expectations. Mismatched updates often get down ranked, defeating the point of sharing everywhere.

I have two other workflows I am still working on. The first is simply using my site, as Dave discusses, as the source for my usual social updates. I have a couple of plugins I use, as he suggests, to have a hidden category for purely social content on my site. Posting to those then only shows on the intended target and if someone follows a link back I provide. Unfortunately, the differences in content size limits makes this a bit clunky. WordPress supports one or two ways to break up content and SNAP can take advantage of those. But if I want a long update on services that support that chunked into three or more shorter updates, there is no good support for that. I am contemplating going back to what I was doing, but doing it via my own site–writing one long post then just simply copying that out into several shorter messages, massaged to work better on services with tighter limits.

If SNAP would add a character count to WordPress, for starters, then introduce a way to add markers that are invisible on my site but SNAP uses to break up posts into smaller pieces, as needed, that would be splendid. I haven’t looked to see if their Pro version does this or it is on their roadmap. I also haven’t looked to see if they have a robust way to suggest features. I have, after all, been using it less than a day. I will give them the benefit, continuing to investigate and push even if I have to fall back on some manual labor in the meantime.

The other workflow I use is Twitter specific, for sharing links out of my RSS aggregator. Usually I simply compose a tweet with the title of an article, its link, a via if warranted and then in any space left a comment. Predicting how SNAP will mangle such a post if composed in WordPress is proving difficult as it isn’t leveraging Twitter’s own URL shortener or offering its own. Again, the lack of character counting on posts is frustrating, I feel like using WordPress/SNAP for this is a bit like aiming blind. I am less concerned for this workflow since my aggregator is my canonical source for link curation and has its own way to share a feed of what I have shared, with my comments attached.

I have only just started using SNAP’s own comment import. Dave recommended Bridgy over this feature but that service doesn’t appear to support more than one site per sharing service, a use case I now inhabit. Also, it has always bugged me that it was a service rather than a tool I could host and run myself. I did like that Bridgy used an emerging, open standard, web mentions, so I may look for a third option that has the best of both. I’ll share more thoughts as I have more experience with the import feature.