I guess part of what you're getting at is that you'd like to see JUST the stuff you want to show off to people, without the distracting other entries before and after? Anyway, we're at the risk of turning this into an issue of semantics, we're pretty much in agreement besides.

I purposefully made my primitive search engine use GET instead of POST, so you could link to the search results. Which are excerpts (it also searches over my old Palm-based journal) that tend to escape out the HTML, because I couldn't get my act together to close unopened tags and all that.

"Brian, I disagree that permalinks are the same as anchors, since permalinks imply a convenient archive infrastructure, while anchors only are how to get to link to something at an arbitrary point on an HTML page."

But a permalink is usually nothing more than an anchor on an html page. The real problem, IMO, is that you need a way to disaggregate posts on the same day -- permalinks are one way to do that, but not a very elegant solution IMO (i.e. each post should have its own URI rather than each date having its own URI as is now the case in many blogging software programs).

I do see what you mean, however, with the way you are doing permalinks.

"even if there weren't any embedded anchors, you could link to something like this page ala 'http://quicktopic.com?blahblah#""

That is a very cool idea. That would be helpful in a number of contexts.

You might find this interesting. I don't ever really use this method because of the way I organize my weblog, but the software I'm using can take a URL like this:

http://brian.carnell.com/index?body=boing!

And then return a dynamically generated view of the weblog that only shows the days in which the text "boing!" occurs. It would be preferable to show only the posts that mention "boing!" on those days, but it's a start (if I had a calendar for the blog, the calendar would also let you go back further to see older "boing!" entries, but I find those displeasing aesthetically).

Wondering, you're right, permalinks aren't, in and of themselves, amazing. The blog format *is* interesting (to me, YMMV) because of the its pattern of frequent updates, and all permalinks offer is a way of getting back good old-fashioned regular linking INSIDE this new dynamic format. (so that two blogs can grow old gracefully together, and not just have no-longer-apropos links to each other's front page.)

Brian, I disagree that permalinks are the same as anchors, since permalinks imply a convenient archive infrastructure, while anchors only are how to get to link to something at an arbitrary point on an HTML page. (Lore of Brunching Shuttlecocks fame (I think that's who came up with this) has a brilliant idea, he longs for a URL syntax that would jump to the first appearance of a phrase on a page, so even if there weren't any embedded anchors, you could link to something like this page ala 'http://quicktopic.com?blahblah#"permalinks were just as cool"' and go right to your entry. This might be easier than the usual link plus directions, ala "scroll down to the second column, do a text search for "just as cool", etc)

I'm a little more sensitive that permalinks != anchors 'cause my site http://kisrael.com has the former without the latter. (I have an atypical structure in that "a day" is the typical unit of linkage, even though I generally have multiple subupdates each day)

Incidentally, re:searching, Google does sort of an ok job (I wonder if it helped that I changed my archive structure so it looks static, not dynamic) but I usually use my own homebrew site specific search, which is extremely primitive (case-insensitive, full phrase search ONLY) but I can usually get to what i need to get to.

Permalinks were just as cool the first time around when they were called anchors (in fact, many bloggers -- including Boing! Boing! -- use anchors that are invalid HTML since the fragment identifier is supposed to begin with a letter A-Z).

That said, the UI for most blogs is still largely unusable if you want to do anything serious with their archives. One of the things that continues to frustate me is how difficult it remains on most blogs to track topical information over time. Reverse chronological order of posts is great for some uses and lousy for others -- and permalinks are generally only of use if you bookmarked or linked to that entry in the first place (I have spent considerable time, for example, trying to locate articles I know were posted to Boing! Boing! in the past, but I have to try to think how Cory or Mark would have written it up to find words that might work in a Google search -- and then sometimes that's pointless since the Google server I'm connecting to is only reporting a partial index of the site because of the way they handle re-indexing).

Actually, it's NO BIG DEAL. (via designweenie.com)

I mean, are we patting ourselves on the back because we have an ability to list a bunch of excerpts on one page, excerpts which each individually link to individual pages? Woohoo!

Permalink, shmermalink. It's all links.