Solipsism Gradient

I think I’m getting up to a little speed with PHP and the phpBB source.

As I’ve said elsewhere, I can now receive trackback pings and they’re automatically incorporated as preformatted posts. Also, I’m including RDF comments which allow Movable Type-compatible weblogs to autodiscover the correct URL to ping.

Today I included an option to type all phpBB pages; you’ll see a line like “Page generated in 1.561 seconds, 12 queries executed” at the bottom of every page. This will help me get a measure of the load on the server.

Also, to help browsers and RSS aggregators preformat a page, images generated by [img] tags will now include the proper height and width values at a cost of a slight slowdown. As a side effect, non-existent images will spill several error messages over the page’s header… I’m still working on failing more gracefully in this case without slowing things further. Anyway, missing images should definitely not happen a lot.

Stigmergy and the World-Wide Web is an extremely interesting article by Joe Gregorio, author of Aggie, a .NET-based news aggregator.

Stigmergy, a term coined by French biologist Pierre-Paul Grassé… is interaction through the environment.

Self-Organization in social insects often requires interactions among insects: such interactions can be direct or indirect.

… Indirect interactions are more subtle: two individuals interact indirectly when one of then modifies the environment and the other responds to the new environment at a later time. Such an interaction is an example of stigmergy.

…The World-Wide Web is human stigmergy. The web and it’s ability to let anyone read anything and also to write back to that environment allows stigmergic communication between humans. Some of the most powerful forces on the web today, Google and weblogs are fundamentally driven by stigmergic communication and their behaviour follows similar natural systems like Ant Trails and Nest Building that are accomplished using stigmergy.

This is required reading for any weblogger or user of Google.

A pity “stigmergic” is such a cumbersome word. Gregorio himself misspells it several times. Curiously enough, his article doesn’t allow trackbacks or comments… two of the important new stigmergic resources.

Thanks to Sam Ruby for the link!

I found out what was wrong. My trackback:ping URL contains a ‘&’ character which was not being escaped to ‘&’, so the autodiscovery routine was falling back to the rdf:about URL, which contains the permalink instead of the trackback URL.

In other words, here’s what the corrected trackback autodiscovery comment looks like now:

<!--
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
         xmlns:dc="http://purl.org/dc/elements/1.1/"
         xmlns:trackback="http://madskills.com/public/xml/rss/module/trackback/">
<rdf:Description
    rdf:about="http://www.brockerhoff.net/bb/viewtopic.php?p=259#259"
    dc:identifier="http://www.brockerhoff.net/bb/viewtopic.php?p=259#259"
    dc:title="Re: Trackbacks are up - I think"
    trackback:ping="http://www.brockerhoff.net/bb/posting.php?mode=track&p=259" />
</rdf:RDF>
-->

This is copied exactly from the Trackback autodiscovery example. The question is, older versions of Movable Type apparently used the rdf:about URL for the trackback ping… so shouldn’t that entry contain the trackback ping URL for backward compatibility, instead of the permalink?

Update: Ben Trott of Movable Type informs me that the rdf:about field should contain the permalink. Apparently compatibility is not an issue…

Update: I was wrong… the & must NOT be escaped. See my post above.

Posted by NSLog();:
NSLog(); linked to this post

TrackBacks and AutoDiscovery

Ben has an entry here about TrackBack and auto-discovery. Overall, auto-discovery works pretty well. I don’t concern myself with finding TrackBack URLs unless I notice a site didn’t get pinged. What I dislike, however, is that trackbacks aren’t sent within my site. If I link to a previous article, why isn’t it sent a TrackBack? Why shouldn’t a reader who might stumble on to that earlier article not see that I linked to it in a later article? (Update: This is now working as Rainer has the answer – comment out the line next if $url =~ /^$archive_url/; in /lib/MT/Entry.pm.

Posted by Just A Test Blog:
Just A Test Blog linked to this post

Autodiscovery check

Let’s see if autodiscovery works now, by linking to here from MovableType this should generate a trackback ping.

Yesterday I worked on some more patches to the site’s code, but my provider had some sort of outage… supposedly they were updating to new versions of Apache, mySQL, php and whatnot. In any event, access slowed down to a crawl – or even dropped out completely – until very late in the night.

Anyway, I got up around 1 AM to get some water and couldn’t resist checking on my way back to bed; it was working again. That’s the downside of having “always-on” Internet and putting the computer to sleep instead of shutting down… it’s way too easy to get online!

Converting this forum into a weblog – and following the recent developments with trackbacks and comments on weblogs – convinced me that there’s a gray zone between forums and weblogs which just cries out to be filled.

For instance, Rael Dornfest just announced that he’s merged comments and trackbacks on his weblog. So his structure now is a reverse-chronological list of his posts, together with a chronological list of comments and trackbacks. The only actual difference is that comments are directly posted to his site (with no copy elsewhere), and trackbacks are excerpts of comments posted to somebody else’s site.

In a forum like this one, the structure is similar. I post stuff which appears in reverse-chronological order. People can post comments directly, or on their own weblog and ping me for a trackback. The major difference is that threads aren’t kept together in chronological order – but I’m working on that. Forum topics are equivalent to weblog categories.

So, a forum – based on phpBB or its competitors – can be viewed as a borderline weblog, with extremes of many collaborators and categories, and a weblog – based on Movable Type or its competitors – can be viewed as a borderline forum, with only one or a few collaborators.

Forum software often has many extra frills like avatars and private messaging, and necessarily invests heavily in collaborator control and moderation facilities. Weblog software is ahead in having trackbacks and RSS feed generation, and is more lightweight (less server-intensive). Both have similar facilities for archiving and searching posted stuff. Both can be seen to have evolved from 80s/90s BBS software – FirstClass, which I worked heavily with a decade ago, is one of the few still available.

Update: Sébastien Paquet is debating whether blogs really foster conversation:

Denham, Lilia and I are debating whether blogs are indeed suited for conversation or if, on the contrary, collective spaces are needed for true dialogue. Ton, you had something to say here, didn’t you?

I suppose when he says “collective spaces” he means forums…

Jeremy Zawodny worries about his input to output ratio:

I go through cycles of productivity like most hackers do. Some days I get a lot done while others are mostly wasted. Some of my productive days involve a lot of output like e-mail, code, discussion, debugging, and so on. Other times it’s a lof of input: reading, listening, etc. Once in while I manage to have a day in which the two seem to balance out and I go home feeling like I’ve accomplished three weeks wort of work.

Yes, these cycles happen to me too, and have in fact been getting more extreme. Before the Internet came up, I was buying 15 to 20 technical magazines per month, and would take every other day off to read all that stuff… even study the advertisements one by one, if you can believe that.

For the last years, information overload from the Internet has been increasing. I don’t buy any more magazines, but reading and responding to e-mail, browsing for news, and so forth has been expanding to fill most of my time. And now weblogging and reading RSS feeds is taking the place of e-mailing and browsing. (My advice: don’t subscribe to more than 150 feeds if you check them every hour, or you’ll never catch up ).

On the other hand, the amount of useful input – that can be converted into productive output – has also increased vastly, so it seems to boil down to a question of discipline. Inspiration doesn’t come by every day, so I usually slack off for periods that vary from an afternoon to a couple of weeks, and catch up again in frenzied bursts of creativity. Turning off the phone and ADSL – or making a trip to somewhere off-net – often works wonders.

Joel on Software also addresses this issue:

But it’s not the days when I “only” get two hours of work done that worry me. It’s the days when I can’t do anything.

…Maybe this is the key to productivity: just getting started. Maybe when pair programming works it works because when you schedule a pair programming session with your buddy, you force each other to get started.