A bit of quasi-coherent quasi-incoherent rambling on this gorgeous Friday afternoon.
Years ago, I used Mozilla as my browser. Then it became apparent that IE was the better browser, so I switched to it. Ultimately, IE stagnated for a number of years, and I switched to Firefox. The fact that IE is now getting attention isn’t enough to get me to switch back.
For a number of years, I used Bloglines. Then it became apparent that it was stagnating, so I switched to Planet. The fact that Bloglines is now getting attention isn’t yet enough to get me to switch back.
What started as a small list of Atom 1.0 feeds that Bloglines couldn’t at the time handle well, has remained a rather exclusive list of Atom 1.0 feeds. They are not all perfect. But like most people, I can’t keep up with everybody I want to, and being able to either select a tool or being inclined and able to install a template or plugin that produce an Atom 1.0 feed has proven to be a reasonable — not perfect, but reasonable — clewfulness filter for me.
By now, my planet is much more to me than Bloglines ever was. It was visited by 738 unique IP addresses in the past week. This isn’t a large number. But the Atom feed it produced has been fetched by 160 unique IP addresses. It is this number that amazes me.
Despite the quick dismisal by Mr. Techmeme himself, I’ve found that I’ve grown increasingly fond of the MeMeme display in the top left. If you compare what you see there against what you see on Techmeme, you will see that there is a diversity in the tech world that Techmeme doesn’t quite capture.
If Bloglines is still hiring, I’d suggest that they look for someone who is looking to enable each user to build their own communities. I know Bloglines has the technical talent to pull this off.
It is in this context that I read the tale of apparent Shangri-La at Google. For well over a year, I’ve been told of massive rewrite is underway which will complete in a number of weeks and solve a number of Blogger issues. I’ve looked into using code.google.com for an effort that is underway, but I’ve now reconsidered due to the concern that it is entwined into this herculean rewrite and the people now working on it may chose to move on any time.
The idea of personal meme trackers is one that could become big, soon. The newest version of FeedDemon includes one which generates a personal meme page on the fly based on your feed subscriptions.
Is the MeMeme code available on Venus and does it require some special trigger to work that I overlooked? I’d be quite interested to try it out, and see how the list changes when I subscribe or unsubscribe from feeds. (My blogroll needs some updating.) Last I checked on your weblog, it seemed the code still needed some work.
Is the MeMeme code available on Venus and does it require some special trigger to work that I overlooked?
Here’s a current snapshot. On my todo list is to make it into a proper template and to make it use the cache (important for proper etag and timeout support). After which time, I’ll commit it to my branch of Venus.
How do you choose the title for the anchor tag that you use for the MeMeme content on your planet site? I assume it is the title of one of the posts that links to the item chosen at random?
Title is the entry title of the most recent post which links to the item.
That portion of the code is platform independent. Take a look at memes.json (which is served statically), and at the last block of code in personalize.js.
I must admit, however, to having a strong temptation to replace JSON and all that DOM wrangling with a simple:
Thanks for exercising your editorial controls, Sam.
You of course know that it’s not just to give you a warm fuzzy feeling about being standards-compliant when drifting off to sleep that should keep you ignoring that temptation to use innerHTML. I can currently view (a lot of) this site using my phone, but that would break if you go the innerHTML route, since you serve the content as application/xhtml+xml, and the browser will throw exceptions if you try to set an innerHTML property on an element in an XHTML document.
James: Odd. What browser is that? Firefox seems to handle xhtml + innerHTML just fine. (As I mentioned in a comment above, I’m looking to rewrite that code anyway).
The reason to avoid innerHTML in general is that it’s nothing more than client-side strcpy, and we all know what happens when you mix that with markup. The exact same concerns as elsewhere apply here, too.
Unfortunately, not only is DOM wrangling cumbersome (libraries can help that to a large extent), it is also slow compared with innerHTML (no way around that).
Well, the Blogger stuff is in Beta. So that “three weeks away” got rolled into a mammoth rewrite rather than a feature fix, and is now being rolled out.
The two feeds on code.google.com come from there, and team blogs aren’t available on the Beta, so we’re stuck with Atom 0.3. shrug ... I can wait for a while.
For our project hosting feature, we’ll be adding feeds. Those will be Atom 1.0 and RSS 2.0 since they’ll be based on the GData spec.
btw, got any ideas for this? - Planet Intertwingly includes a mix of many short posts and a few really long ones (I think in common with all the other Planets I’ve seen <include href="powerLawBoilerPlate" />). Net result is that it’s hard to skim. I’m not sure what the answer is (if there is one), maybe some Javascript collapsing in a feed filter..?