Notes from FooCamp:
Saturday, I sat in on a session in corporate blogging. What is good blogging for, and what is blogging good for? What happens when bad blogging happens? How does one establish policies for organizations?
And what does it take to make bloggging work?
(The answer is "search", but you should follow below the fold...)
The session gave a lot of thought to internal blogs and what they can be used for. Google has a thriving culture of internal blogs. In contrast, people from Microsoft, SUN, and Yahoo each reported that while their external blogs are working fine (blogs.sun.net; scoble and the microsoft bloggers), the internal ones tended to fall apart.
The crucial difference, as far as I can tell, is that Google has really good intranet search: that is, you can put something on an intranet blog and it won't get lost. When a Googler puts something on an intranet blog, other Google employees can find it with a quick search. This is very different from many other intranets, which are fairly hard to search.
If you joined this blog yesterday, you probably haven't read back more than a dozen posts or so -- tops. Maybe three or four. I discussed this a little bit when talking about "getting the data out.": http://drzaius.ics.uci.edu/blogs/danyelf/archives/000348.html
You might have found me through someone else's blogroll, or someone elses' link to an entry. But really, most anyone gets anywhere by searching for it. If you trust your search system, then you can find lots of good stuff. But intranets often have poorly-designed search tools, and so an intranet blog is likely to be kind of ignored, especially if there are other ways to get information about. The blogs at Google apparently are used as searchable repositories: engineers discuss problems they've run into, and then can use search to understand each others' problems.
I do the same thing and use Google to find entries on this blog.
What I'm thinking, I guess, is that there are a lot of ways to get data in to the net, both ones that exist and ones that don't yet. There are far fewer ways to get data out. In fact, I'm going to suggest that the answers are, basically:
1) RSS feeds (and reading the newest stuff)
3) Everything else.
But wait! you say. What do you mean there are many ways to easily put data into the web? And why do you think there are so few to get it out?
Let's break this into two pieces. I'll take a look at a few paradigmatic easy publishing systems. I'd emphasise blogs, in which users individually write paragraph-plus posts, with comments; wikis, in which users (roughly anonymously) write anything from a single-character change to a new page; and del.icio.us and other link systems, in which users contribute a link, possibly with some annotation
|Atom||Paragraph||text||link + annotation|
|Voice||One voice1||Many authors||one author, but interlinks|
|Newest information||RSS||Recent Changes||RSS|
Now we can imagine a lot of possible ways of reshuffling this table. We could imagine, say, a Wiki-like tool where individuals wrote (blog-like) paragraphs at a time, for example, and oculd only shuffle around or move those units. Delicious' RSS feeds manage to act like link-oriented blogs, both individual and collective.
But I'm still only seeing a small number of ways to get infromation out. If you are reading something very nicely organized like Wikipedia (and alphabetical order really does make things easier), you can dig around in the obvious alphabetical way. If you are dealing with a less-well-organized system, you'll have a little more work:
I'm not trying to caricature here, although I realize that I'm oversimplifying a bit. The question I'm wrestling with is whether we can find a better way to find stuff that might be out there: there's a lot out there, and a great deal of it is desirable (at some point or another).
The answer to this is not particularly obvious to me.
Because it needs to be put down for Google to find: yes, a dynamic-search API and file system is a very good thing. And yes, it would be very cool if it talked NFS or Windows File System or a half-dozen other things.
The Longhorn File System (now delayed) was going to contain a full relational database that allowed documents to be linked to all sorts of nifty properties, both explicit (such as labelling, delicious/flickr/gmail style) and implicit (such as size and last touched).
It's worth skimming over the presentations to see some of what worked, and what didn't.
I was invited (by proxy) to FOOCamp this year. This makes me an (indirect) Friend Of (Tim) O'Reilly, publisher of books that people put on their wishlist based purely on the series1. He gets together a bunch of people together in a room, and has them talk about stuff that interests them.
Word has gotten around, and apparently crowd control wasn't where it might otherwise have been. And so the place is pretty filled to (and, really, past) the brim.
In theory, we decide what we do with a large collaborative message board. We all stand up, and announce ourtopics, and a schedule evolves: "Sure, I'll show up at that." "Hey, can you merge these sessions?" "Ok, if not enough people want to talk about THIS, let's try THAT."
There's a technical name for this sort of meeting -- my "organzational development':http://resourcesforchange.com mother -- has mentioned it a few times, but I'm blanking at the moment.
Unfortunately, it needs fairly careful control--and the meetingm with so many people, didn't get fairly careful control. In theory, then, the board was negotiated and evolved. In practice, it was staked out, claimed, and then -- defaced? edited, anyway. But no group consensus ever attempted to figure out what will work.
I'm curious how the sessions will go--will there be a convergence of topics? Will we run about between empty sessions, trying to decide where to show up?
(more below the fold...)
1 Um, not Harry Potter. Or Harlequin. They're cute little techie books with technicolor spines and animals on the front.
Anyway, there's some cool people out here--lots of fifteen-minutes-of-famous hackers, a number of Very Clever People, and some folks who ordinarily get paid a great deal to Think Aloud. I'll try to drop names only when the names, you know, say something relevant.
(I'm here largely as the Voice of Marc Smith, who is caught up with other travel right now.)
Last night, the chief engineer of GMail (and the Google Toolbar, and involved in Froogle) kicked us off with an impromptu midnight session and a long thinky discussion of what this GMail thing is about and how it's done. We also got a chance to talk about some of the people-aware features that I'd argued for before. About half of them are on the list, while the other half are things that he doesn't particularly plan to get to. But I'll try to blog about that separately in a little bit.