WikiWeblogPIM

This is my page of thoughts and links about...about...well, there's an elephant i'm after. I have many perspectives on it. One is that if three blind geeks were hanging onto it one might say, "it's a wiki!" and another might say, "it's a weblog!" and another might say, "its a personal information manager!" (someone else's visual, information architecture perspective). Excellent venn diagram. There are a few other things that serious geeks might say about it.

Against the constant tug of far-out visions, this page tries to stay focused on actual software, that is easy to share (ideally cross-platform, free or cheap, and source code available). Additional relevant links show up in my wiki AbbeNormal. Once the elephant actually exists, everything important from this page will be part of it, because that will be the most sensible way to maintain/publish it.

Wait for the file system (and file browsers) to incorporate needed metadata & extensibility (and improved interfaces) or use databases (and create custom interfaces)? My gut instinct is to go with the file system. blosxom - a weblog system - takes this approach, and one of the nice things about many keep-it-simple wikis is that they keep pages in plain text files, with little or no metadata beyond what the filesystem provides. The more the filesystem can handle, the less tempting it is to add complexity to the elephant. Today (2004, still!) file systems and user tools for organizing stuff in them aren't very good. If they were, creating entirely separate apps such as iTunes or iPhoto just wouldn't make sense. You'd almost think Apple and Microsoft had a financial incentive to hold back :-). Even as things are now though, any user has some familiarity with manipulating their files, and we can leverage that. Two specific areas for improvement would be a good GUI for hard links and more smarts about name spaces (Plan 9 anyone?, or more simply just more thinking along the lines of happy name collisions in wikis, and among wikis with SisterSites).
* elephant * programming language * file system (using plaintext or some other format well-understood by & integrated into the filesystem - XML?)	* elephant * programming language + database (e.g. Perl+PostgreSQL, Python+mySQL) * file system (database - SQL tutorial | ACID)	* elephant * integrated database / language system (e.g. Zope/Python, Frontier/UserTalk) * file system

(from this Mathemagenic post)

Software for more than one person, aka groupware, aka... ... CSCW (Computer-Supported Cooperative Work), aka social software, whatever. These are some of the terms that apply to the part of this vision that's about people communicating and acting (even programming -- from CVS to TinyWiki) together. There's a history of our humanity getting lost in the focus on the technology or the money, so please keep your eye on the ball - culture & community count. Decent examples: Wikis, Kuro5hin, Recently "social networking software" (like the 1990s' sixdegrees.com) has become a buzz-thing. Everyone's trying to build the elephant, or enough of it to establish a new ubiquitous thing. Hopefully we'll see an extension of Internet sensibilities (something that hasn't happened - yet - with Instant Messaging) and get an Inter-SNS (FOAF?), and later Inter-elephant, which connect existing systems and perhaps eventually replace most of them. (There is an element of this that is not just about groups or community but simple sharing contact information without everyone having to maintain it manually - Micorosft wants their Passport to be the one central thing - not happening! See XNS; OpenLdap) Group-forming weblog | Emergent Democracy | links on Group-Forming Networks | Reed's law | Dunbar number SubEthaEdit (nee Hydra) lets multiple people write on the same document at the same time, over networks. 2004 is looking to be the year of local wiki clients. See VoodooPad, wikidPad, WikiBook. Expansion: Wikis could send/receive requests to mirror/copy/append/etc. wiki pages between sites (see XML-RPC above). JSPWiki and a few others implement APIs for wikis to automate page-interaction. (see post August 2003? on AbbeNormal) Peer-to-peer vs. client/server conversations are not "only" about inner architecture, they relate to how humans use the system. Open/free collab. projects | Open Groupware Standard Internet Groupware paper by Jon Udell; Discussion (wiki, weblog, rss, xmlrpc!) Newsgroups: Usenet2 (/. discussion) folk computing (report/links) stigmergy On-line presence A personal server which is on-line all the time doing your bidding independently, with whatever amount of input/interaction from you, from none to plenty. This is not a new idea. Somehow, we just managed to get to the 21st century without it being available as a mass product. What is it doing? Besides serving your pages and reading your chosen RSS feeds, it could be applying complex filters to the feeds, bringing you the information you wanted as soon as possible after it exists. And with whatever degree of your oversight republishing any results (see AutoBlogging). A server in every home (wrist, pocket, ...) that wants one.	Development I wish to learn as few different tools as possible, and accomplish as much as possible. DevShed looks useful for development in general. In thinking about being lynx-friendly, consider also links (OS X version), a text-based browser which handles tables & frames. Does emacs' browser? guiplatforms | XDND | GNUStep | GNOME lang kits | KDE lang kits? Parrot (->GUI&HTML!) | Diff. project same name: April 1 & Real (!) Parrot Bridge the user/programmer gap The elephant wants much more power in the hands of "users", e.g. a far smoother transition to programmer powers, either not having to get into programming at all, or a path there with far less than today's quantity and scale of insurmountable-seeming cliffs. The user should have a sense of a thing under whatever interface they're interacting with, something that's the same whether they interact with it from their hand/wrist device, talking to it by telephone, etc. Miscellany Hm. Tivo is just handling feeds, isn't it...? Part of the vision i guess is to fill in the line between "programming" your VCR or Tivo, and programming your computer with a little script, and programming a full interface for something. Multimedia came too early for the good of hypertext methinks. The place i'm aiming for, everyone programs, and it's no big deal. and/or, more graceful degradation of computer illiteracy? "Better" languages and interfaces is the easy way out, but doesn't really say much.	On-line vs. off-line? We can talk bout computers being on-line all of the time all we like. For now and probably forever, it makes sense to design most things so that they can do something useful even when not connected to the entire globe-spanning network. Vint Cerf has already been talking about the interplanetary Internet for years, and it's a useful thought experiment for these issues (NASA has already worked on Internet in Mars probes if i recall correctly). Organizing stuff is one thing computers can be really good for whether they're on- or off-line, but few operating system developers have successfully pushed this past limits common since the 1980s, except for Be. Palm deserves mention for some serious integration thought that went into their early operating systems, but that focus was lost somewhere along the way. From a user perspective, some time after you find one or three or more tools you really value, you realize that you keep learning different, limited ways of organizing stuff. I think we're quite past the point when the common needs for sorting music, pictures, personal notes and many many other things would be best served on a common platform that helps people learn to build and organize things as they like. Plucky is an RSS reader with an iTunes-style search box in the corner. Why isn't such interface implemented in the Finder, with enough other customizability to basically have iTunes in the Finder (with visuals as screen saver). If John Robb is right (need exact Klog archive link), it's simply not in Microsoft or Apple's competitive advantage and they'll hold it off as long as possible. (another chapter for the unwritten book, Inefficiencies of Capitalism) Where will it come from then? It seems likely we are approaching a sweet spot or critical mass of some kind.

What serious geeks might say about what i'm after: "It's a file system!" "It's an object database!" or "It's an integrated development environment!"

Ward Cunningham is probably too humble to say it's a wiki, but i would. In the end, we won't formalize and categorize every last bit of information. Much will be a mush, and we'll partly wade through those parts, and partly depend on code to extract formalities and categories for all the wonderful things that makes possible.

These are not quotes, but they probably could be:

It's a memex. --Vannevar Bush

It's a human augmentation system. --Douglas Engelbart

It's the singularity. --Vernor Vinge

It's the continuing evolution of the noosphere and humans' relationship to it. --Pierre Teilhard de Chardin

It's Xanadu. --Ted Nelson

My software can do that! (--many other software developers)

----

Cruft

Wikis: Wikipedia policy | (!wiki Nupedia lic) | GreenCheese | AndStuffWiki | MeatballWiki | PracticeMatters | PortlandPatternRepository (original wiki)

HandheldsWiki || html2PDoc Pippy (Python for Palm) |

Web journals: suburbs (categories) | Diarist.net (guide) | Open Pages archive |

(SF wiki)

MailToTheFuture (Frontier/Manila-based)

Amusing but not directly relevant

DadaDodo (based on CutUp theory)

Horizontal weblog: Kreuzfahrt

Not exactly

Infocetera propr. ?-> free

----

go to the InterComm conference September 2001? Will this be a sort of WikiConference? I know i would enjoy hanging out with a bunch of wiki-ers; maybe movement on some standards could even percolate.

Seems to have not happened. We will have a wiki conference, just a question of when where and how. Or maybe it's happening everywhere already.

----

Yes, i know, i could use CSS everywhere. and XHTML. It will happen. Don't hold your breath.

<==--- back to John's home page