about

who
what
where
when
why
how
comments

feeds

ATOM

RSS

categories

identity
jabber
language
literature
music
personal
philosophy
politics
public domain
society
technology

archive

current
2007-04
2007-03
2007-02
2007-01
2006-12
2006-11
2006-10
2006-09
2006-08
2006-07
2006-06
2006-05
2006-04
2006-03
2006-02
2006-01
2005-12
2005-11
2005-10
2005-09
2005-08
2005-07
2005-06
2005-05
2005-04
2005-03
2005-02
2005-01
2004-12
2004-11
2004-10
2004-09
2004-08
2004-07
2004-06
2004-05
2004-04
2004-03
2004-02
2004-01
2003-12
2003-11
2003-10
2003-09
2003-08
2003-07
2003-06
2003-05
2003-04
2003-03
2003-02
2003-01
2002-12
2002-11
2002-10
2002-09
2002-08
2002-07
2002-06
2002-05
2002-04
2002-03
2002-02
2002-01
2001-12
2001-11
2001-10
2001-09

2004-07-31

Desert Island Redux

My favorite music, again.

Because of a special deal that Apple is running right now, my new PowerBook came with an iPod (that's a really sweet, portable music player for you non-techno-geeks). The thing has "only" 20 gigs of storage (compared to the 20 megabyte capacity of my first hard drive, at about one-twentieth of the size -- progress in storage capacity is even more amazing to me than progress in processor speed). So I figure this thing will hold about 100 CDs' worth of music. Back in January I tried but failed to come up with a list of only 10 "desert island discs", but the prospect of whittling my CD collection down to only 100 CDs is less daunting -- in fact, it's forcing me to recognize which music I like best and which is nonessential in my life. Perhaps I'll post again with my top 100 (I wonder if iTunes has an export function?).

Posted on 2004-07-31 at 21:11. File under music.

~ link ~

Beauties

Making good things.

Matt Miller (a.k.a. linuxwolf) loaned me his copy of Paul Graham's Hackers and Painters the other day, and in the last 24 hours I've read about the first half of it. Graham's analysis of American schooling from about age 12 to age 18 seems spot on to me (at least, that's how it felt to me growing up), and I appreciate his historical perspective throughout. His call to beauty -- that is, to cultivate one's taste regarding the things one makes (whether that be writing, coding, art, or anything else) -- struck a chord with me. While I am far from one of those who claim that beauty is in the eye of the beholder, I would have liked to see him delve more deeply into the nature of beauty. For example, I think that there are different kinds of beauty in any field of endeavor, and in nature itself. The beauty of a desert (such as I experienced in Tucson earlier this year) is different in kind from that of the high mountains here in Colorado, the rocky coast of Maine where I grew up, the rolling hills of the Shenandoah, and so on. In music, the austere purity of Bach (such as the Well-Tempered Clavier, to which I'm listening now) is far removed from the stormy passion of Beethoven, the quiet fervor of Dvorak, the melancholy joy of Ellington, the striving spirituality of Yes, and so on. The same is true of math, physics, biology, painting, sculpture, architecture, writing, coding, athletics, and countless other pursuits -- there is not one beauty in each field, but many varieties of beautiful experience. So it's not all that helpful to challenge people to strive for beauty -- one who would so strive needs to learn more about what makes a certain kind of production beautiful. Part of that is learning the art behind its creation -- and, as Graham says, that often involves a great deal of imitation and putting yourself inside the mind of the person who created the beautiful thing (in painting, this comes in part from copying the works of the old masters; in music, from learning to play the works of Bach and the other masters; in software, from reading good source code). In the past, that learning also came in part not just from familiarizing oneself with the works of the masters, but from working with the masters themselves through apprenticeship. But essential to that learning process is the ability to appreciate many different kinds of beauty -- to analyze beauty into its manifold aspects, such as purity, economy, elegance, coherence, integrity, cleanliness, rhythm, flow, naturalness, symmetry, difficulty, depth, significance, individuality, suggestiveness, and timelessness. The word "beauty" includes as many aspects or meanings as "good", and it's important to understand those aspects rather than to accept each term as an unanalyzable whole.

Posted on 2004-07-31 at 20:31. File under philosophy.

~ link ~

2004-07-30

Oh, S X!

A dearth of Jabber options.

I'm getting awfully spoiled with my new PowerBook. It's as if I've been driving a Yugo or Pinto for years, and suddenly I'm behind the wheel of a Jaguar.

Sadly, this does not extend to Jabber clients. Because I do so much work via Jabber, I need a solid, powerful, simple Jabber client in my life, but so far I have yet to find one on OS X. Herewith some first impressions:

• Nitro -- The kind of spare interface I like, but it can't handle my roster of 500+ items and development has stopped, so the issues probably won't get addressed. However, I can use it to log into one of my stealth accounts and at least join some groupchat rooms, which other clients can't do.

• Gush -- A nice interface (if a bit big for my tastes) and the roadmap looks good, but it can't handle my roster, either (just hangs). Perhaps I can convince Dudley to send me a debug build of some kind so that we can figure out what's going on.

• Adium -- It handles my roster! But its protocol support is borked (sending invalid XHTML-IM causes other clients to show empty messages from Adium users, for example) and I simply cannot stand this cutesy duck icon who flaps his wings down in the dock every time I get a message in a window that doesn't have focus (which for me is about 80% of the time). No, I'm not into cutesy.

I have not tried Gaim on OS X (what I was using most recently on Linux), Psi (sorry Justin, but I just can't stand ICQ-style interfaces), Fire, Proteus, BuddySpace, or Tkabber. I've tried some of these on Linux and they either choked on my roster or were too big for me to use (I keep a lot of windows open at any one time, though at least I've discovered the beauty of tabs in Mozilla and Safari).

The search continues...

Posted on 2004-07-30 at 21:29. File under jabber.

~ link ~

Running on Empty

The state of my inboxes.

I'm finally getting caught up on my voluminous correspondence. As of this moment, my inboxes stand as follows:

• jabber.org (traditionally my main account) -- 6 messages
• gmail (certain mail redirects to this account) -- 0 messages
• gmane (Jabber and IETF discussion lists) -- 7 messages
• saint-andre.com (personal mail) -- 0 messages

In my recent experience (say, the past 2 years), this is quite simply unprecedented. It feels good to clear the decks, because I have a lot of work to do.

And of course it helps that I'm asking people to contact me via Jabber, and that I've switched all possible discussion list reading to Gmane. Both of these measures have really cut down on my email (the vast majority of which is now spam, even with SpamAssassin's Bayesian filters doing their best).

Speaking of which, today Stowe pointed to yet another person who notes that the end of email may be nigh. Bring it on!

Posted on 2004-07-30 at 20:53. File under personal.

~ link ~

2004-07-29

Position Paper

My position paper for the Jabber Council.

Peter Millard has posted his Jabber Council position paper as a weblog entry, so here I am following his lead. I wrote this a month ago but I figured I would finally post it now so that folks can read it over at Planet Jabber and such...

Background

Back in college, I majored in philosophy and ancient Greek. Somewhere along the line I got involved in the Internet, first working for several years as a writer and then a systems analyst for Logical Design Solutions, where I helped create award-winning HR intranets for large companies like Lucent Technologies. In 1999 I moved to Colorado and worked as a developer (mainly on XML/XSLT) for Webb Interactive Services, an early supporter of the Jabber community and eventually the parent company of Jabber, Inc. In my first weeks at Webb in late 1999, I was introduced to Jeremie Miller since I was told he also had an interest in XML. From that time forward, I have slowly but surely become more heavily involved in the Jabber community. In October 2000, I transferred from Webb to its Jabber Inc. subsidiary, contributing as a tech writer, systems analyst, customer liaison, and even for a while in marketing. In May 2002, Jabber Inc. made me a "sponsored employee", which essentially means that I don't do anything directly for the company but instead focus on standards work, protocol development, and the affairs of the Jabber Software Foundation.

Contributions

Early in my Jabber career, I mainly contributed to documentation, such as the Winjab user guide and the jabberd 1.4 admin guide. It seems to have been a natural path from there to protocol documentation and eventually protocol development. Even now I am mainly a glorified tech writer, not a coder. My main contributions to the Jabber community over the last 2 years have been:

• Editing the XMPP specifications and shepherding them through the IETF.
• Developing protocols such as the following (a small representative sample -- I have probably written or co-written over 30% of the existing JEPs):
  • JEP-0030: Service Discovery
  • JEP-0045: Multi-User Chat
  • JEP-0065: SOCKS5 Bytestreams
  • JEP-0071: XHTML-IM
• Defining and managing the JEP process and the Jabber Registrar.
• Distilling protocol design guidelines.

Through this experience I have gained significant insight into Jabber protocol development as well as the best practices of standards development organizations such as the IETF, W3C, and Liberty Alliance.

Priorities

The Jabber Council needs to do the following in the 2004-2005 term:

• Advance a number of Draft JEPs to Final, including Multi-User Chat and the File Transfer suite (JEP-0096 and its dependencies).
• Advance a number of Experimental JEPs to Draft, including Authenticating HTTP Requests via Jabber, XHTML-IM, the "extended presence" protocols (Geolocation, Mood, Activity, Address), Entity Capabilities, HTTP Binding, SHIM, and Service Administration.
• Finalize the basic and advanced protocol suites.
• Advance a number of older but solid protocols from historical to standards-track (such as Roster Item Exchange).
• Hammer out consensus solutions in the areas of multimedia (voice, video, whiteboarding, file sharing, etc.) and security (including sane, usable end-to-end encryption methods as well as a complete threat and security analysis of Jabber technologies via a dedicated Security JIG).
• Work actively with the developer and user communities to identify and fill in gaps in the Jabber feature set, such as avatars.
• Improve its processes by meeting at least once every two weeks.

Why I Want to Be on the Council

Lord Acton once said: "Power tends to corrupt; and absolute power corrupts absolutely."

I have always been leery of concentrating too much power in the hands of any one person within the Jabber community. This is one reason why, since taking over from Michael Bauer as Executive Director of the JSF in 2002, I have not sought to be on the Jabber Council.

So why do I want to be on the Council now? For one, while overall I have been happy with the work completed by the first three Jabber councils, we need to move faster. There are too many JEPs in the queue and too many additional problem areas that need to be addressed. Past Councils have moved too slowly in some areas, have not met often enough, and in general have suffered a bit from a lack of strong leadership and direction, especially after an initial burst of activity in the first few months of the term. If elected to the Council, I will seek to be the Council Chair and will push hard to complete all of the priorities listed above. I will do so by motioning for countless Last Calls, holding regular meetings, demanding attendance and prompt voting from all Council members, and insisting on strong and consistent productivity from the Council over the entire term. I will also seek to streamline existing processes where appropriate so that the Council and the JSF can work faster and more flexibly while retaining due oversight over the JSF's standards process.

IETF approval of the XMPP specifications has only increased interest in the JSF's output. We need to move faster, work smarter, and produce consistently intelligent protocols that address real needs in the developer and user communities. There is a lot we need to accomplish. Let's get busy.

Posted on 2004-07-29 at 20:49. File under jabber.

~ link ~

Powering Ahead

Getting back up to speed.

It's taking me a little while to get everything I need set up on my new PowerBook. Thanks to Joe and Diz, I now have all the developer tools I need, including Fink (which provides the goodness of Debian's apt-get on OS X). I just installed a newsreader called MT-NewsWatcher so that I can start posting again to all the lists I follow through Gmane. And I've been adding lots of music to iTunes. :-) But I think I may need more memory -- 256 megs is not quite enough...

Posted on 2004-07-29 at 09:14. File under technology.

~ link ~

2004-07-28

Hacking the Organization

Some thoughts about the JSF.

I'm not a big fan of elections, which is one reason I've never been fully satisfied with how the Jabber Software Foundation is set up (though in no way do I mean that as a criticism of Michael Bauer, who did the main legwork in setting up the JSF -- we've simply learned a bit about what works and what doesn't over the last 3 years). The JSF just held elections to choose new and returning members. Soon we will hold elections for the JSF Board of Directors and the Jabber Council. It seems that all we really do in the JSF is hold elections, which IMHO are just as wasteful in a small organization as they are in large nations (yes, I am a market anarchist, radical decentralist, contractarian voluntarist, localist democrat, and what have you). As recently noted, the ancient Athenians randomly selected the members of their governing council by drawing lots, not running campaigns and asking everyone to vote for their favorites (who will usually end up being those rich and powerful enough to buy influence). So I've begun to think that perhaps there are better ways organize the JSF than our current regime of endless electioneering. Herewith are a few tentative thoughts.

My preferred approach would be for all JSF members, without exception, to take turns serving on the Board, Council, and other teams (Infrastructure and Trademark come to mind). Let us imagine that every three months, a new Board and a new Council are chosen at random from among the members (in such a way that everyone will have to serve at least one quarter a year on each body, but never on both at the same time). This gives people long enough to make a difference, but the terms are short enough so that people don't lose interest (which traditionally has been a problem with both the Board and Council, through no fault of those who have served in these bodies). This also gives us a transparent method of choosing the Board and Council (and other teams if we add them), but dispenses with the bother of elections. (Whether this meets legal requirements for our organizational status as a non-profit membership corporation is something I need to investigate.)

So much for the Board and Council elections. What about JSF membership elections? Here again it seems to me that the current elections process is waste of valuable time, and in any case is not very effective in choosing worthy members (usually, almost everyone is elected a member, despite sometimes rather weak credentials). But if everyone is expected to serve on the Board, Council, and other teams -- and to make important decisions on those teams such as which protocols to accept, how to structure our relationships with other standards development organizations, and the like -- then I think we will need to maintain higher standards about who is accepted as a member. Those ancient Athenians were notoriously selective about who could become a citizen, and I think the JSF would do well to be selective, too. Several JSF reforms discussed in the past have included requiring a current member to nominate new members (not accepting applications from just anyone), which I think is a good idea. Naturally, once current members know that they will have to do something other than vote once in while (and that not contributing to the Board, Council, or another team after having been randomly chosen to serve is simply not an option), they may well resign their membership. Let's face it: some folks like to be JSF members because they love the technology and it sounds cool to be a member of the JSF, but they don't necessarily want to contribute actively to the work of the JSF. (To be fair, we haven't necessarily given people a lot of opportunities to contribute; random team selection will change that.)

This sounds elitist, and it is. Open source communities -- or at least the ones that aren't dysfunctional -- are traditionally meritocracies. If you work hard and contribute value (not necessarily in the form of code, I might add), then people will pay more attention to what you say. But they'll also expect real work from you, not just yammering on.

So there you have it: an elitist meritocracy on the outside, a radical democracy on the inside. Would it work? I doubt it's perfect (what is?), but I think it would work better than what we have now.

Let the flames begin!

Posted on 2004-07-28 at 21:07. File under jabber.

~ link ~

Switch

My fifth machine.

Last night I went to my local Apple Store and bought a 15" PowerBook. Excuse the mess while I convert from Linux to Mac and get this new machine set up.

My first two computers (purchased in '86 and '92 or thereabouts) were Macs -- the first one with no hard drive (an external 20 Meg hard drive cost me $750 and weighed more than my new PowerBook). My third machine ('96) ran Windows so that I could dial in from home while working for LDS. In '99 I switched to Linux, first on that old P100 and more recently on a machine put together by a friend of mine. After 5+ years on Linux, I'm back on the Mac (which nowadays is really just Unix with a nice GUI -- I'm still using vi to edit my weblog entries, so don't worry that I've gone off the deep end). My big challenge now is to find a Jabber client that I can live with. So far I've tried Nitro (which didn't like my huge roster) and Gush (pretty sweet, but took up quite a bit of screen real estate and froze up on me, perhaps also a roster-related issue). I'd try Psi but I really do loathe that ICQ-ish interface, even if it is quite standards-compliant. Eventually iChat may be the client I settle on, once it releases full Jabber functionality. I wonder if they'd take me on as a beta tester. ;-)

Posted on 2004-07-28 at 15:54. File under technology.

~ link ~

2004-07-27

The Numbers

Slow but steady at the IETF.

People keep poking me as to when the XMPP Internet-Drafts will get RFC numbers. I won't be sure until the RFC Editor publishes them with nice shiny numbers, but I can tell you that we're getting closer because the XMPP-E2E draft has been approved by the IESG and XMPP-IM depends on the e2e draft. Furthermore, several non-XMPP drafts on which the e2e draft depend have recently been published as RFCs (3850, 3581, and 3852). However, we're still waiting on four IMPP WG documents to emerge as RFCs:

• draft-ietf-impp-im-04
• draft-ietf-impp-pres-04
• draft-ietf-impp-cpim-msgfmt-08
• draft-ietf-impp-cpim-pidf-08

These four are all in "Author's 48 Hours" and I plan to inquire about them further at IETF 60 next week in San Diego, so stay tuned!

Posted on 2004-07-27 at 20:13. File under jabber.

~ link ~

Googlenym

One-click searching?

I'm sure I'm not the only who, when asked how to find me online, says "go to Google and type ______". In my case, you fill in the blanks with stpeter, for which the first two results are:

1. http://www.saint-andre.com/blog/
2. http://www.jabber.org/people/stpeter.php

David Weinberger calls this a Googlenym. Mine happens to be easy to remember, since it's the same as the nickname that I use in Jabber and (before that) IRC.

Posted on 2004-07-27 at 13:47. File under technology.

~ link ~

2004-07-26

From The People

Putting the demos back in democracy.

Doc wonders: what is a democracy for? His take is that "with a connected citizenry, we are in a position to practice Real Democracy, or something closer to that ideal, for the first time". I'm perhaps more skeptical. The ancient Athenians didn't have elections, which they thought would lead to an oligarchy as the rich bought their way into power (sound familiar, Messrs. Kerry and Bush?). Instead, people were randomly chosen to serve in the governing council. Ah, but I hear you say that we could never run things that way today because political entities such as the USA and the EU are no longer small city-states but instead large conglomerations that contain millions and millions of people. That's true, but what do we conclude from that? That power "from the people" (as Doc puts it) is unrealistic and that we need to accept an oligarchy as the only way to run a modern state? Or that the only remotely legitimate power is as local as humanly possible? Democracy contains an inherent tension because it means literally "people power". Today we have raw, naked power, but very little governance that is people-oriented, personal, or humane. And in large measure the reason can be summed up in one word: centralization. Without deep structural safeguards, power trumps people every time. Legend has it that as he walked out of the Constitutional Convention, Ben Franklin was asked what kind of government had resulted from the deliberations; supposedly he answered: "A republic, if you can keep it." The ancient republics were small and, ideally, governed by the people themselves, who were chosen by lot (not governed by elected oligarchs and unelected bureaucrats as in the modern nation-state). Toward the end of his life, Thomas Jefferson's mantra was "divide the counties into wards". This vision of keeping power as small and local as possible might result in a much safer and saner, and certainly a more humane, world. What would the world look like today if almost all power in, say, America resided at the level of the block or neighborhood (where the "wards" did not exceed, say, the effective range of human group sizes as reflected in Dunbar's number). Here's a clue: in the War of 1812, American militas refused to cross the border into Canada as the power-hungry national government wanted because they knew that their job was to defend against attack, not to invade foreign countries. Are you listening, Mr. Bush?

Posted on 2004-07-26 at 16:52. File under politics.

~ link ~

XHTML-IM Update

Refactoring in progress.

Finally last Friday I had a long IM session (through the Yahoo gateway, ick!) with someone from the W3C about XHTML-IM, a proposal for formatted Jabber messages using XHTML 1.0. Most of the changes he suggested are a matter of documentation and being politically correct with regard to XHTML Integration Sets and such. The only really substantive change is to change the "wrapper" element from "html" to "xhtml" to prevent any possible overlap with XHTML itself (which of course has "html" as its root element). Hopefully my correspondent will get a chance to review my changes Real Soon Now and I can publish the next version of the JEP. Stay tuned!

Posted on 2004-07-26 at 15:59. File under jabber.

~ link ~

2004-07-23

Hypocrisy

A few observations.

Bill Clinton was in town the other day signing copies of his book and keynoting a fundraiser. I was pleasantly surprised to hear that he was not raising money for the Democratic Party, but for a private effort to fund a memorial to those killed in the shootings at Columbine High School in 1999. This got me to thinking: why doesn't Michael Moore contribute some money to this effort? He's probably earned enough money from Bowling for Columbine to pitch in, no?

This train of thought led me to reflecting on some other instances of hypocrisy I've witnessed of late. Like, oh, people driving past me while I'm biking to work and they are riding in their cars with Ralph Nader bumper stickers. (Why aren't they getting around town in a more eco-friendly manner?) Or local U.S. Senate candidate Pete Coors touting his conservative principles. (Have you ever seen the scantily-clad young women in a typical advertisement for Coors beer? Nothing traditional or conservative about that!) Or John Kerry bemoaning tax cuts for the better-off (as if anyone else pays many taxes in the first place) and asking for donations even though he could fund his entire candidacy from his personal fortune. Or George Bush saying he values freedom while he and his minions have eviscerated what's left of the Bill of Rights. Or -- well, the list goes on, and this is getting depressing.

Posted on 2004-07-23 at 21:47. File under society.

~ link ~

Ralph Rocks

Some recognition.

I would just like to say that Ralph rocks. He recently helped me fix the mailman troubles on jabber.org, he started Planet Jabber, and he's taking my mantra to heart by rewriting his pubsub implementation (Idavoll), maintaining his fun and famous Jabber World Map, and most recently coding up hooks into Rhythmbox so that Jabber users can publish their "now playing" information (as defined in JEP-0118: User Tune). Tune in and turn on, because Ralph is on a roll!

Posted on 2004-07-23 at 21:29. File under jabber.

~ link ~

Free Text

CC not so free?

Chatting in the jdev room yesterday, talios said he was mentioned in an NZ Herald article on weblogs. Following some links, I came across this post on the undue restrictions in Creative Commons licenses (it seems that the Debian community has concluded that the CC licenses violate the Debian Free Software Guidelines). While thinking hard about content licensing last year, I flirted with the CC licenses but just never could get comfortable with them, which is partly why I ended up putting all of my writings and music in the public domain. CC < PD.

Posted on 2004-07-23 at 21:17. File under publicdomain.

~ link ~

Loony

35 years ago.

Thanks to RocketForge, I found this heartfelt post from Rand Simberg about the thirty-fifth anniversary of the Apollo moon landings. I especially appreciate his hopeful thoughts on the future:

Thirty-five years after Neil and Buzz walked on the moon, we have neither the NASA Mars base, or the huge spinning space colonies. But we're finally seeing new progress on a front in between those two visions. Forty years after the end of the X-15 program, we're recapitulating some of the early NASA program privately, and diversely, with the efforts of Burt Rutan and the other X-Prize contestants and suborbital ventures. They won't be diverted down a costly dead-end path of giant throwaway rockets. Instead they'll slowly and methodically evolve capabilities and markets, creating the infrastructure for low-cost access to space. Once we can afford to get, in Heinlein's immortal words, "halfway to anywhere," we'll finally be able to return to the moon, to complete the job begun by those first voyagers, and this time we'll be able to stay.

The key word here, I think, is "evolve": pursuing space in an organic, sustainable fashion through private initiatives, leading not just to exploration but to permanent settlement.

Ad astra!

Posted on 2004-07-23 at 21:09. File under society.

~ link ~

More Specs?!

Still clearing the decks.

Yeah, I know: more code, fewer specs. But I'm still clearing the decks. Last night I finished defining protocol flows for all the use cases in JEP-0133: Service Administration, and today I performed my usual JEP Editor / Jabber Registrar review on JEP-0122: Data Forms Validation (authored by Matthew Miller). I also reworked the interaction between pubsub and disco in JEP-0119: Extended Presence Protocol Suite, clarified a few matters in JEP-0078: Non-SASL Authentication and JEP-0114: Jabber Component Protocol, and advanced JEP-0030: Service Discovery to a status of Final in the JSF's standards process. Oh, and I've been trying to finalize the XMPP-URI spec, too. I need to work through all this protocol stuff to free up more coding time. I started delving into the Jabberzilla code a bit during the evenings this week but there's more to grok there, so I need to keep busy. But at least I finished updating the README for Proxy65 so that DizzyD can release version 1.0...

Posted on 2004-07-23 at 20:53. File under jabber.

~ link ~

2004-07-21

The Book Machine

The future of book shopping?

Hey, my birthday is coming up soon, maybe I'll ask for one of these ;-)

Posted on 2004-07-21 at 11:49. File under personal.

~ link ~

Learning to Share

A protocol spec that's actually needed.

Late last week I wrote up a proposal for enabling "shared roster groups" over Jabber. I discussed it a bit with maqi earlier today and he has some comments, such as that it would be awfully nice to do this in a way that did not require any changes to existing clients (my proposal is built on top of pubsub, which no clients implement yet, but it does have the virtue of re-using the old Roster Item Exchange protocol). Zero client modifications is a good goal, but I'm not sure it's possible here (especially since lots of clients don't support roster item exchange yet, either). One approach would be to have the server (not a pubsub service) send the roster items to clients. It's not clear to me how this would address the need for cross-domain roster groups, which is one of the requirements I was attempting to address in my proposal. But I suppose we'll discuss this at length in the Standards JIG once this proposal emerges as a full-fledged JEP (probably next Monday). In the meantime, maqi has created a wiki page for further discussion of the topic.

I know, I know: fewer specs, more code. But we really need this one!

Posted on 2004-07-21 at 11:43. File under jabber.

~ link ~

Patents Amok

More nonsense from the USPTO.

As far as I can tell, this application is attempting to patent Internet Relay Chat, for which the original RFC was published in May 1993. Silly.

Posted on 2004-07-21 at 11:33. File under jabber.

~ link ~

IANAMOTSM

Playing with security.

I Am Not A Member of the Security Mafia. However, I know that the Jabber community can do more to address security issues. One step in the right direction would be to define a Jabber threat model, similar to (but more detailed than) this browser threat model, which hopefully would also result in a Jabber security model down the road. Will Kamishlian and I have authored a proposal to form a Jabber Interest Group devoted to security so that we can define the threat and security models in an open forum. Now that more and more people are using Jabber, it's time for us to really investigate, define, and improve the security profile of the underlying protocols and technologies. A number of people have told me that they want to contribute to this effort, and some of them even know a thing or two about security. For those who don't, there is always time to learn. Two good blogs on the topic are Financial Cryptography (see this post) and Educated Guesswork (see this post on the same topic).

Posted on 2004-07-21 at 11:19. File under jabber.

~ link ~

2004-07-19

JIGs and JEPs

More specs???

Yes, even though my mantra is to write fewer specs and more code, I'm still writing protocol documentation. Will Kamishlian and I have written a proposal for a Jabber Interest Group that will complete a security analysis of Jabber protocols and technologies. Friday night I also finished a preliminary proposal for shared roster groups (e.g., so that a system administrator could centrally defined and distribute roster groups within an organization). The proposed solution uses pubsub rather than a server-side protocol + module, which means it could be used for cross-domain groups (e.g., all JSF members).

And speaking of JEPs and clearing the decks, I recently updated my .plan to reflect all the JEPs that need a change in state (Retracted, Final, Last Call, etc.). It's a long list, and the next Jabber Council will have its work cut out for it (don't forget, position papers are due by July 30!). And here is fair warning: I plan to run for Council and crack the whip so we can clear out the queue and finalize a boatload of specs that are Experimental and Draft. Furthermore, said whip shall be cracked throughout the entire term of the Council. No more lollygagging!

Posted on 2004-07-19 at 21:59. File under jabber.

~ link ~

Process Improvement

Clearing the decks.

As noted, mantra #2 is now to write more code and fewer specs. The code I've been working on for the last few days consists of Python scripts that will help me work more efficiently. The first script handles all the niceties involved in publishing a Jabber Enhancment Proposal; based on metadata in the JEP XML file format, the script does the following:

1. automatically generates a message to the Standards-JIG list (done)
2. inserts or updates JEP data in a database, whence the JEP list and content feeds (ATOM | RSS) are generated (in progress)
3. publishes an event, with payload, to pubsub.jabber.org (planned)

Other scripts on the drawing board will automagically gather the information used to populate the public servers page and detail pages for the Jabber clients page. Also on my list are a new version of memberbot, a similar bot for Jabber Council votes, and a bot that will handle manual updates to all the software listings via x:data.

After I finish a few of these scripts, I'm hoping that my time will be freed up a bit to work on some less utilitarian projects, first among them Jabberzilla.

Fewer specs, more code!

Posted on 2004-07-19 at 21:39. File under jabber.

~ link ~

2004-07-15

My Campaign

A better image.

Sweet! hildjj made me a shiny new image for my campaign to kill off email. What a nice guy. Now I can start marketing in earnest. :-)

Posted on 2004-07-15 at 16:01. File under jabber.

~ link ~

Shared Groups

Bringing people together.

While folks were jabbering away in the jdev chatroom this morning, ralphm mentioned that his Jabber World Map has "filters" so that you can, for example, see all the JSF members. Kinda cool. That got me to thinking again about the need for "shared groups", about which I recently posted to the JADMIN list. Cross-domain shared groups would be especially useful, so that (for example) Ralph could subscribe to the JSF group and automatically update his world map whenever someone becomes a JSF member or drops out of the organization. Sounds suspiciously to me like an area in which pubsub might prove useful. Expect a JEP from me soon that will solve at least some of what people think of as shared groups...

And hey Ralph, how about a Planet Jabber filter? :-)

Posted on 2004-07-15 at 15:51. File under jabber.

~ link ~

2004-07-13

Escape Artist

RSS feeds updated.

A few readers of this blog have pinged me of late asking why my blog entries don't have links and such when appearing in their news aggregators, but show up fine when sourced from Planet Jabber. The answer is that Planet Jabber uses my ATOM feeds, which until this evening were more triumphant than my RSS feed. But I think I've fixed the problem now, thanks to some XSLT madness suggested in a post by Wendell Piez of Mulberry Technologies. Let me know (via Jabber, of course) if you experience further difficulties.

Posted on 2004-07-13 at 22:08. File under personal.

~ link ~

Make Your Own Music

An RIAA Solution?

Embedded in this screed on phishing, I found this little gem:

As it happens my take on the whole RIAA thing is to stop buying CDs and learn the guitar.

My philosophy exactly, which is why I make my own damn music (as yet unrecorded, though...).

Predictably, my take on phishing is, yet again, abandon email and use an IM system in which people can't spoof 'from' addresses. Jabber, anyone?

Posted on 2004-07-13 at 13:39. File under technology.

~ link ~

Kudos

Just call me I-D master of the universe. ;-)

I was Jabbering with Lisa earlier today and she said:

(10:51:05) lisa: You're the best I-D writer I've seen.

Isn't that just the sweetest thing you've ever heard? ;-)

Posted on 2004-07-13 at 13:29. File under jabber.

~ link ~

2004-07-12

Mantra #2

Fewer specs, more code!

Today I revisited my .plan for the first time in a month, and was struck by how non-essential some of the entries seem. Distributed calendaring? Multimedia extensions for voice and video? Whiteboarding? Hmm. Are these what Jabber is all about? Right now it seems to me that the answer is no.

Jabber was born on January 4, 1999 when Slashdot featured a story on a new, open-source instant messaging system developed by Jeremie Miller. The early days were highly productive, leading up to the release of jabberd 1.0 in May of 2000. Since then the community has witnessed more code, more contributors, and more process. We formed the JSF in mid-2001 to manage the Jabber protocols in a more open and measured manner, which of course required more process. In 2002 we submitted our core protocols to the IETF -- and, as I can vouch for first-hand, that meant a lot more process and formalization.

No one especially wants to go back to the early days, which were exciting but involved daily changes to the protocols. However, we need to find a balance between process and innovation. The first step, I think, is to back-fill a bit. Sure, voice, video, whiteboarding, and calendaring are all fun applications. But IMHO it's a bit premature for the Jabber community to go off gallivanting in search of cool new applications when we have a plethora of Jabber clients that are at best incomplete and at worst pieces of junk. There is no really good Jabber client for Linux or Mac (though future versions of iChat may address the latter failing). In server-land, the venerable jabberd 1.4 codebase is stable but showing its age, and it's doubtful that it will ever include the XMPP advancements such as SASL authentication, TLS channel encryption, and whitelisting/blacklisting. The jabberd 2.0 codebase contains all those features but is less than stable (perhaps because it's mainly been developed as a heroic solo effort by Rob Norris). We have a powerful publish-subscribe protocol but only one open-source component (thanks to the great ralphm) and no client implementations that I'm aware of. And so on.

All those specs look nice on paper, but they're meaningless without the code to back them up. I see little point in continuing to push forward on more fancy specifications when 90% of the clients out there haven't even implemented pubsub or file transfer yet. So personally I plan to focus only on specs that we really seem to need (a good protocol for "shared groups" is one of them) and to do what I can to contribute to the open codebase, starting with the recently-resuscitated Jabberzilla client, which I hope can be a fully standards-compliant and cross-platform Jabber client.

So you can add a new mantra to my battle cry of "email delenda est": fewer specs, more code!

Posted on 2004-07-12 at 21:03. File under jabber.

~ link ~

Joining the Band

More on Dunbar's number.

I finally had occasion to re-read Christopher Allen's post on Dunbar's number and the size of effective human groups. It seems that 150 people is something of a maximum limit, although (according to Dunbar) a group of that size would require "42% of the total time budget to be devoted to social grooming". Dunbar sees boundaries at much small sizes: 4 is the most effective size for a group of people having a conversation; a person can maintain very intense relationships with only 10 to 12 other individuals at a time (e.g., a small, focused team that lacks hierarchy or lines of reporting and is based on strong mutual interest and interpersonal trust -- such as, in my experience, the early Jabber developers); 30 to 50 people can function as a small living group such as a "troop" with well-defined leadership and a clear division of responsibilities (this seems to have been the usual size of pre-tribal bands throughout human history); larger groups up to 150 can work, but only with increasingly large investments in socialization and management. Beyond that, you need to "divide and conquer" by forming workable sub-groups (early human tribes were probably formed for short periods from individual bands and did not work together constantly).

I'm especially interested in the smaller groups. One analogy that organizational psychologists and motivational speakers like is sports teams: the nine members of a baseball team, the five members of a basketball team, etc. (such teams have more members, but only a handful are fielded simultaneously). Another analogy I'd like to see pursued is that of musical groups. What's interesting to me about such groups is that they often sustain highly creative output for a long time (e.g., I noticed the other day that my favorite rock band, Yes, is currently on their 35th-anniversary tour). The most stable size of such bands seems to be five, but three (Rush), four (Beatles), and six (Grateful Dead) also seem workable. More than six seems too much, but that may be more for musical reasons than for organizational reasons. In general, Miller's insight from 1956 still seems to hold true ("The magic number seven plus or minus two: some limits on our capacity for processing information.", Psychological Review 63: 2-5). In my experience, though, it's more like five plus or minus two: I've found that it's hard to hold more than seven items in my memory at once, and that groups larger than seven become unwieldy.

My renewed thinking about this was inspired by a brief chat last week with Mikael Hallendal of Imendio, creators of two well-regarded Jabber projects: the Loudmouth C library and the Gossip client for Linux. It turns out that Imendio is a consulting duo based in Sweden -- something like the Simon & Garfunkel of the Jabber community. ;-) I don't know if Mikael and Richard plan to scale up their little group, but the sense I got from Mikael was that they won't scale it up very high (perhaps to that sweet spot of 10 to 12, but no more). Personally I find the idea of working in a small "group" or "band" such as that quite appealing. Sure, it would require a fairly large investment of time and you would definitely need to know and trust the other people to an uncommon degree, but I think such a small band could be deeply creative and productive. (Paradoxically, I've always been a lone singer-songwriter-composer musically, but I think I could work well in a technology group like this.)

It seems to me that Dunbar's research also has political implications, but I'll save that for another day...

Posted on 2004-07-12 at 20:33. File under society.

~ link ~

Reading at Risk

On the decline of reading in America.

The (U.S.) National Endowment for the Arts has released a study of reading habits in America, and has concluded that Americans read a lot fewer novels, short stories, plays, and poems than they used to. Given the relatively narrow focus of the study, I suppose I fit the trend. Although I am re-reading the complete poetry of Walt Whitman these days, I mostly focus on history and philosophy rather than literature. Right now I'm reading Freedom Just Around the Corner (the first volume of a new history of America, in which author Walter McDougall argues that America is a nation of hustlers -- in both the good and bad senses of the word). I'm also working my way through the Complete Works of Aristotle -- so far I've finished all of the biological and psychological works except for De Anima. Expect a full report on these books soon...

Posted on 2004-07-12 at 16:53. File under society.

~ link ~

xmpp-e2e

Version 09.

I've just submitted draft-ietf-xmpp-e2e-09 to the IETF Secretariat. This version addresses the issues raised by the IESG. No further comment required. ;-)

Posted on 2004-07-12 at 16:23. File under jabber.

~ link ~

2004-07-06

Minding the Store

Another reason to use IM.

The technology behind email is called "store-and-forward": when I send you an email message, in general it is stored on your ISP's servers until you ask for your messages, at which point your ISP's servers forward it on to you. (The premise behind instant messaging is quite different: if you are online, your server delivers it to you immediately without ever storing it; if you are offline, it may or may not store it for later delivery.)

This may seem a bit arcane, but now the difference has not only technical but also legal implications: as reported in the New York Times, "last week a federal appeals court in Boston ruled that federal wiretap laws do not apply to e-mail messages if they are stored, even for a millisecond, on the computers of the Internet providers that process them - meaning that it can be legal for the government or others to read such messages without a court order."

In other words, your email messages are fair game for snoops both commercial and governmental. But since IM messages are not stored, they may be marginally safer from the snoops, at least under the laws of the government headquartered in the city of Washington. End-to-end encryption, anyone?

Posted on 2004-07-06 at 21:35. File under technology.

~ link ~

Lorem Ipsum

Real books.

My Jabber friend Matt Mankins has started a bookstore in Cambridge (Mass.) called Lorem Ipsum Books. Nice! I've always wanted to start a bookstore or a small publishing company...

Posted on 2004-07-06 at 15:12. File under personal.

~ link ~

The Number

Dunbar numbers and the realities of group interaction.

Joe Hildebrand recently linked to a fascinating discussion of the so-called Dunbar number governing the ideal size of cooperative groups (in fact there seem to be multiple numbers, or a range of numbers). One of these days I must re-read it and blog about it, but for now I merely provide the link.

Posted on 2004-07-06 at 13:49. File under society.

~ link ~

Some Observations

On taking a few days off.

I took Thursday and Friday off, and yesterday was a holiday in the U.S. (well, the 4th was Independence Day, but it was observed yesterday). Herewith a few observations:

1. Sometimes it's good to take a few days off away from the computer.
2. Email feels even worse when you come back from a few days away from the computer.
3. Is snail mail the answer? Some folks seem to think so.

Posted on 2004-07-06 at 13:29. File under technology.

~ link ~

identity...

my back pages

me
home
music
jabber
poems
journal
essays
dotplan
résumé
ism book
contact me
colorado blogs

my group blogs

albion's seedlings
extended conversation
floss foundations
microid development
planet jabber

jabberites

adam nemeth
daniel henninger
google talkabout
hal rottenberg
jeremie miller
kevin smith
mickael hallendal
ralph meijer
remko tronçon
robert quattlebaum

techies

barry leiba
bob wyman
eric rescorla
fred stutzman
future pundit
mike linksvayer
paul hoffman
the speculist
steve o'grady
stowe boyd

wonks

cafe hayek
chicago boyz
the futurist
instapundit
joel kotkin
marginal revolution
michael barone
rand simberg
rants and raves
samizdata

i use...

i support...

i listen to...

fighting censorship...

current threat level...

flying the flag...