Solipsism Gradient

Sébastien Paquet wrote “Personal knowledge publishing and its uses in research”, an excellent paper on weblogging.

Link courtesy of Der Schockwellenreiter.

This weblog is now syndicated by Syndic8.

Also, NetNewsWire Pro now lists my RSS feed in its default sites drawer. Thanks, Brent!

I posted some more comments about the recent site update. A good excuse to learn lots of new stuff and buzzwords – PHP, MySQL, RSS, RDF, XML, XML-RPC, and so forth 😉

Matt Croydon’s postneo asks a profound question:

The effect of RSS On Design

Dane Carlson redesigned his weblog. (…) This begs the question:

If a redesign happens and everyone is reading via RSS, did a redesign happen?

More or less simultaneously with my discovery of RSS, phpBB version 2.0.4 came out.

So this was a good excuse to simultaneously update and implement RSS. It turned out that this wasn’t as cut-and-dried as I first thought, since I had already modified the standard phpBB code quite a lot – mostly to force descending order for my main weblog topic and to do several cosmetic and functional enhancements. There are still several bugs in less-used areas, mostly related to administration and moderation, but I’m working on them…

In any event, BBEdit helped a lot with its “compare folders” feature. I even found some hidden bugs in my previous code. I was up and running in about 24 hours (not counting a week of learning details about RSS).

Nevertheless, there’s still lots of thing to do. My RSS code now runs OK through the RSS Validator, but most of my pages still fail the HTML Validator, and the CSS stylesheet still needs a few tweaks. Even so, everything seems to work fine in all major browsers…

Besides some added functionality for phpBB, the major next thing to do here is to implement trackbacks. Stay tuned…

Well, I’ve somewhat belatedly found out about RSS. This stands for Really Simple Syndication (or Rich Site Summary, depending on who you’re asking). This is a XML-based file format that is becoming popular for automatically distributing news or weblog posts. As with nearly everything on the Internet, there are two sides to RSS: server (generator) or client (reader).

Let’s look at the RSS reader side first. I’m using NetNewsWire Lite for Mac OS X, written by Brent Simmons at Ranchero Software. This sort of software is also known as a “news aggregator”.

Basically, you go to a news site or weblog you’d like to read daily (or even hourly). If you’re lucky, you’ll see a red tag, just like the one at the top of this page; some pages have a simple “Syndicate” link instead. Copy the tag or link’s URL from your browser, switch to NetNewsWire, use the “Subscribe” command, and paste the URL into the subscription dialog. From now on, NetNewsWire will check that site (and all others you’ve subscribed to) every hour or so, and if it finds new news items or weblog entries it shows you a list. You can read summaries right there or have the full items opened in your browser.

To make a long story short, I’m now subscribed to over 60 news sites and about the same number of weblogs, and save at least an hour every day! No more sitting at the browser and selecting dozens of bookmarks, one after the other… or at least, the number of sites I do this for has shrunk considerably. I find myself forgetting to check sites which don’t have a aRSS feed.

Now let’s jump to the generator side of RSS. In my case, I’m interested in making it easier for people to read my weblog. Many important bloggers don’t have time to read non-RSS sites, so generating RSS is a big step in getting to play with the big boys.

Ready-made blogging software which runs on your own server, like Movable Type, Radio Userland and Blosxom (to name a few), usually have RSS generation capability built right in. Centralized systems like Blogger usually don’t, although I hear Blogger Pro has an experimental RSS generator.

In my case, I’m using phpBB which is more bulletin-board-oriented; I’ve hacked it to use it as a weblog generator, too. So now I’ve spent a week or so putting in RSS generation capability (and updating to 2.0.4 while I was at it). There still are some nits to pick, but it works now. At least NetNewsWire reads my feed with no problems, and the RSS Validator says I’m generating valid RSS/XML.

RAILhead Design has a great tutorial about making “favicons” – those little 16×16 icons which Safari may show next to the URL. (Yes, I know my own isn’t all that great… I’ll be changing it, but it’s not a priority.)

There’s also an entire website devoted to the subject. This article talks about best practices.

In my own experience, making a single 16×16 icon in 256 colors will cover 99% of the cases where the icons appear… the 32×32 size is apparently only used when dragging an URL to the desktop. The articles recommend also making 16-color versions, which usually suck.

And of course, other browsers like Mozilla and Opera (not sure about Chimera) also show favicons.

Thanks to Bryan Pietrzak and Jack Small, who posted these links on the Carbon Developer’s List.

I fixed a bug that was, errhh… bugging me for some time. Linking to individual posts in the main weblog topic now works, and it returns to the correct place after posting, too.

I’m also checking out Apple’s new Safari browser. Practically my whole site seems to work OK, except for button placement which sometimes is off a few pixels… no great problem.

Dave Sifry wrote:

When you wrote, “they don’t necessarily conflate different URLs on the same site”, I’m not sure I understood your criticism. Technorati will list all of the people linking to URLs that are at or below the URL you provide. In other words, if you put in “http://www.sifry.com/”, it will show you everyone it its database who is linking to http://www.sifry.com/, http://www.sifry.com/alerts/, and http://www.sifry.com/alerts/foo.html.

Hello Dave. Thanks for dropping in. I didn’t mean it as a criticism, but thought it would be a limitation of having to allow for many blogs on the same site.

On second thought, it seems it was my mistake. Cory linked to my main page, and this link appears when I click http://www.technorati.com/cosmos/links.html?url=www.brockerhoff.net/index.html. I asked a friend to link to my main weblog page, and checked by clicking http://www.technorati.com/cosmos/links.html?url=www.brockerhoff.net/bb/viewtopic.php. At one point, it appeared I could see distinct search results, neither of which appeared when I searched on my general URL with http://www.technorati.com/cosmos/links.html?url=www.brockerhoff.net. I must have mistyped one of those URLs…

Your explanation makes sense, and it seems to be working fine now. Apologies.

And thanks for setting up TechnoRati!