In which I mention Aaron Pareki’s Now page primarily as a test of WebMentions, and also a test of post titles with a word after the date (rather than before, the way I’d been doing it).
In related news, I’m planning to start using the “image” and “link” post types for sharing photos and links.
All appear on my website as well as in my microblog (which is really the point).
In part of my continuing preparation for using Manton Reece’s micro.blog stuff, I went ahead and installed the Webmentions plugin for WordPress on this site.
So, if you use Webmention (whether on a WordPress blog or some other kind of site), you should get notified when I link to your stuff, and I should get notified when you link to my stuff.
This will be so much better than having these discussions on some lame blend of blog comments, Facebook, and Twitter.
I’m moderately active on Twitter and Facebook. I post photos on both Flickr and Instagram. I try to keep my profile reasonably up to date on LinkedIn. I even use Google Plus. I hate all these things.
(Actually, I don’t hate Flickr—it’s pretty good.)
There are a lot of things I hate about them, beginning with big corporations deciding which ads appear next to my words (not to mention keeping the money from those ads). I also hate the way they keep making their services worse (the needs of the venture capitalists outweigh the needs of the users). But the worst thing is: I can’t find stuff I remember writing!
“Oh, yeah—I put that one on Facebook.”
“Oh, yeah—that one was a tweet.”
“Oh, maybe I wrote that back when I had my LiveJournal going?”
My partial solution, for a while now, has been to put almost everything—all but the shortest bits—on my own blog. Then I link back to them on Twitter and Facebook and Google Plus. That’s still unsatisfactory in several ways, but especially for those short bits—tweets and Facebook posts—that don’t get their own blog entries.
There’s no good solution. My blogging software supports “asides” or “status” posts which are supposed to be for things like Twitter or Facebook posts. I used those briefly, and didn’t like it. Those little posts cluttered up the main flow of my blog. Worse, different blog themes displayed them differently. (Maybe I’ll try posting an “aside” or a “status” again after this post, and see if they’re better now.)
I even considered setting up a Diaspora node for a while, and then arranging to have things I posted there flow out to Twitter, Facebook, and Google Plus. Then at least everything I posted would be just one of two places (my website and my Diaspora node). That turned out to be too much work.
Just yesterday I ran across something that may go some way toward solving this problem: Indie Microblogging: owning your short-form writing, which I have backed on Kickstarter (video below).
I don’t yet know if it will solve my problems, but I’ll try it out and see if it works.
One key feature of the Indie Microblogging thing that makes me think it might be satisfactory is that it’s built on RSS. (The fact that it provides RSS feeds is the reason I don’t hate Flickr, and the fact that they don’t is a big piece of what’s wrong with Twitter, Facebook, and the rest.)
Anyway, check out Indie Microblogging, and see if it’s something of interest to you, too:
Just visit philipbrewer.net using https instead of http and you can browse the site secure in the knowledge that the pages will be encrypted in transit. As if that mattered for a public website. But still—might be useful, and costs nothing except a whole bunch of cycles on your computer and my hosting service’s computers.
I had actually turned on encryption some time ago for the admin pages, so that I could securely administer the site even when my access to the internet wasn’t secure (over public WiFi, for example). But I hadn’t pulled the trigger to route general traffic over https because the certificate I used was self-signed, which meant that I could trust it, because I knew which certificate I’d installed, but the general public couldn’t tell the difference between my site and a fake site set up by a some perfidious fraudster. The new Let’s Encrypt certificate is signed by a well-known issuer, so any modern browser will show the handsome green padlock.
I started having odd troubles with the theme I’d been using for this website, so I thought I’d look into alternatives.
I tried quite a few, most of which had one problem or another. Among the things that made me reject themes today:
- Header image was completely integrated into the design, so it couldn’t be changed.
- There was no way to show the full text of each post on the front page.
- Post titles were forced to all-uppercase.
- A full third of the page was dedicated to showing a featured image for each post, except my posts often don’t have a featured image—in which case the theme showed a dumb icon instead.
This theme isn’t perfect either. In particular, when I use an image in a post and mark it as the featured image, it appears twice, which is usually not what I mean to have happen, but I’ll experiment for a bit and see how I like it.
Intended as a placeholder, the header image is a photo I took of plaster casts made of the Elgin Marbles. (The casts, on display at the Spurlock Museum, were made before the ill-fated cleaning attempt that so seriously damaged the originals. Scholars come from far and wide to study our plaster casts.) Now that it’s up there, I kinda like it. I may just leave it like that.
I’ve experimented with various alternatives to Google Reader for quite a while now. I used The Old Reader for a while, and then Hive Reader for a while. Both had limitations. (Hive is still in beta, and isn’t quite ready for prime time. TOR is closer, but had various issues, probably the biggest being that it doesn’t get feeds updated promptly enough.)
I had earlier tried using tt-rss, which also isn’t quite there yet, but has a different set of issues.
It requires a server. Steve had tried to cobble together an instance that ran on the server where we host our websites. It had just almost worked, but kept bumping up into the limits of running as a cron job, rather than a daemon. It eventually had several bad days in a row (which we later traced to an unrelated heavy load on the server), and we gave up.
Now Steve has installed an actual server machine in his house, and is running a tt-rss instance there, and has made me an account on it.
Running on an (essentially) dedicated server with a (reasonably) high-speed connection to the internet, it’s now doing a fine job of keeping all my feeds up to date. I’m having some minor user interface issues, but nothing that would keep me from using it as my rss reader for the foreseeable future.
So, I have officially switched over. You can follow the interesting stuff I share via a feed from that site, and have updated the “interesting stuff” item in my sidebar to draw from that feed.
Security expert Bruce Schneier wrote last week about some changes he was making to his blog to remove some anti-security features. Reading over his list of changes, I was pleased to see that I’d mostly avoided adding anti-security features to my blog in the first place.
- No offsite tracking. Although I’ve experimented with them a couple of times, I don’t have “like” or “share” buttons on my blog posts, so your visits here are not automatically transparent to Facebook, Twitter, Google, or other social media sites. It means you’ll have to copy the link yourself if you want to share my posts. I’d be delighted if you did, so I hope that’s not too onerous.
- No offsite searching. Similarly, the site’s search facility runs right on the site itself, just doing an SQL query of the database that holds the content of my site. Doing a search here doesn’t expose your query to anyone else. (I once looked to see if I was logging queries and couldn’t find them; as far as I know, doing a search here doesn’t even expose your query to me.)
- No offsite feed. I also run the RSS feed for the site right on the site, and always have. I thought for a while that I ought to use feedburner, but I never got around to it, and now it’s clear that laziness led me to the right choice.
Any attempt to keep internet activity private is probably hopeless, but that’s no reason not to try.
I do most of my on-line reading via a feed reader. For years I used Google Reader, without even really worrying about the risks. After Google ruined it, I experimented with several alternatives. I’m happy enough with a couple of the options, so I’m not so unhappy with how things have turned out (with Google having announced that it is canceling Reader). But the surge in interest has prompted me to think about how reading feeds is different from reading things via social media. Social media helps you find great posts. Feed readers are for when you’ve found a great writer.
I notice this whenever someone shares one of my pages (either here or on Wise Bread). I’ll get a surge of traffic to one post. Some of those people will read another post, or even a few. Only a few seem to become regular readers of my work—and fewer now than before.
Back in the old days—let’s say, five or six years ago—there was more of the latter, and I think it was because more people used feed readers. It was wonderful to find a great post, but it was much better to find a great writer. Then you could add their feed to your feed reader and read everything they wrote.
I still do that. Every time I find a great post via Facebook or Twitter (or whatever), I look at other stuff the guy has written, with an eye toward adding the feed to my feed reader.
I’m puzzled that more people don’t seem to do the same. Finding a great writer is way better than finding a great post.
I got email this morning from a schoolteacher who shares links to my story structure article with her young students, telling me that the links out from my sidebar under the heading “Recently popular” were going to porn sites.
Although somewhat dubious, I went and checked immediately, and discovered that it was true. Ugh.
Those links were generated by a widget that (when it hasn’t been hacked) uses site traffic statistics to identify the most visited posts and pages on my site over the past week or so and link to them. It was still doing the first part—that is, the items under the heading were names of my recent high-traffic pages—but the links no longer went to my pages. Instead, they were going to porn sites.
I immediately removed the widget from my sidebar and disabled the plugin that provided it. I looked at other links on the site and didn’t see any similar problems elsewhere, so I’m hoping that was the extent of the hacking.
I also sent email to the guy behind the plugin, telling him about the problem and asking if there was any diagnostic information I can provide.
The stats package that I use also tracks outbound clicks, so I can see that a total of 4 clicks today went to one or another of those sites. I looked at stats for the past several days and didn’t see any other outbound clicks to illegitimate sites, so I’m hoping that (with the help of that teacher) I managed to nip this thing in the bud.
Apologies to anyone who got directed to one of those pages!
I’m still investigating, and will add an update if I learn anything further.