charlie nesson, from the berkman center, is introducing dave winer's keynote on the second day of OSCOM. he speaks of being able to "see the shape of the cathedral and the bazaar" emerging in our futures; concerned with the forces of law and politics creating new orders or things. so here they travel on the borders between closed and open; offering legal support to poorly resourced open projects - a resource for defiance, perhaps. the "singularity of major powers" owning the structure of our lives - technical, social, societal - and why the original OSCOM made such sense to him - "here was the point of the spear, the experience of this quite extraordinary coding movement". a brief and pleasing valedictory address, and on to the show; "dave, as far as i have come to know him, is a force of nature".
here is the expression of dave winer's theme; that we must "find ways to work with each other, not to build walls" between the open and closed source worlds; even between different open projects and standards. he objects to the use of 'proprietary' as a denigratory term; the conceptual positioning in words of 'open' and 'closed' source. parenthetically, pointing out that he is no longer an active software developer, points to his own source release approach, which sounds rather like sun's; a look-but-don't-touch approach to open view, but not GPL, code. "let's start working with each other", he exhorts a roomful of the hardest core open software hackers, gesturing towards the closed world; "we have a lot of the same issues, we are the same people."
"think of it as a live blog," dave declares, gesturing his superior officer off round the room with a microphone. joe reagle, w3c/MIT, stands to point out that "proprietary", at least, was once a positive label, a badge of honour, a company's pride of product. "they get to market themselves, position themselves; can't open source do the same?" dave asks him if he's ever written proprietary software. no, he replies after a few seconds reflection. what big companies do well; shrink wrapping for the masses; user-accessible featuresets, 'ease of use'; its "harder to write something that's easy to use" (or perhaps, less interesting?) - an inverse proportion of usability between code reading and writing.
"halley, from halley's comet" remarks on her dissatisfaction with open source self-positioning from a sales and marketing standpoint; in user land, not knowing what linux looks or feels like, despite the hype. "it's like there's a gigantic non-sales-and-marketing effort going on".
dave invokes the linux advocacy howto - "a masterpiece... a very humble document." urging the recapitulation of a theme of his own; "our software sucks, and we know it, but we're trying to make it better" - not something one could imagine oracle or sun descending to admit. a nice anecdote about managing complexity by consent in the protocol design process, a proud story about the invention of xml-rpc. including new features and suggestions in the design by default; so instead of the contributor arguing vociferously for their own idea, the contributor is almost arguing against it, self-questioning; do i need to be responsible for this having been included? small practically working groups, as opposed to 500-strong consortium committees, a sideways glance here.
'tony from cmswatch' interposes briefly a statement about the "ceiling on the open source CMS movement"; question about momentum run, where next. dave states, the key is being a user of one's own software. "the open source community", he declares, "hasn't done bleeding-edge things that nobody is doing". contrasts a fine ideal of active, controversial discussion in academia.
"i don't think you actually want my source code... i think you want something else, and havent found a way to articulate it." so we proceed round the audience, asking the second row what they want when they say they want open source. larry rosen decides; freedom is important; code is an enabler for it; the "productive creation of derivative works" is what drives our commitment to an OS license. with data in open format, source seems less necessary; a brief obligatory round of microsoft-bashing. shared concerns about maintainability, vendor bankruptcy and bug fixing; code self preservation - a source escrow idea is mooted about. the internet explorer case; the software is not evolving; dave declares "they should be obliged to give it up" as part of the settlement of the antitrust case.
bruce tarkosvsky [sp?] maintains his need for a well-defined, open and stable API (leading questions on wanting or needing source in this situation; here it works as documentation). the 'motivators' for code activity: "i wrote this software because it solved my problem, published it because i like what other geeks think about me... somewhere at number 47 in the list of motivators is accessiblity and ease of use"... postulation of a geek selfish gene. the newton twist, "standing on the shoulders of giants... in software developement, we are standing on each others' toes"
dave maintains, one should never use software *because* it is proprietary or open source... "software is a tool, not a religion" (the ends justify the means?) an unrecognised audience member asks - "do we need specific reasons to live in a democracy, or a free market? open source is an ethos that has to be internalised." the themes of communism and capitalism echoed from charlie's too-brief intro.
we as hackers, points out brian someone, suffer from "an aversion to marketing oneself", an aversion to self-promotion (do i see this in the world?) sam ruby states, "i want software that will still be around 2-5 years from now, because somebody will need to fix it" or even 10 to 20 years, suggests dave, stretching the timescale - what now do we expect to be around then? apache, possibly. larry: users "don't want to be beholden to a company they perceive as charging them too much and behaving in a monopolistic fashion".
a radio userland user complains about the lack of support and interface documentation available in radio userland. well, you get what you pay for. "there is no money in software; the software industry doesnt exist... let's support the users, and give them what they want."
the conversation drifts inexorably towards RSS, the "fundamental plumbing in changing things". if enough people, dave opines, re-read their history wrt the evolution of RSS, then the sequence of events would have come out very differently; we wouldn't be where we are now. "we have one final shot at getting this thing right," gearing up "other big w3c members, dealing with these people pretty soon" with a de facto standards consensus. (sam ruby briefly objects to this, having heard a 'one final shot' story many times over during 20 years at IBM) dave presents the compelling case to "follow the trail blazed by the person before you" in the case of de facto standards, citing his own adoption of the trackback spec, of the blogger API, in radio userland; "would i rather have interop, or a better API?" well, "aggregators should have a vote," maintains sam; the owners of applications needing a voice.
at this stage it seems appropriate for bill kearney to stand up, suited and booted, at the back. "i'm bill kearney, from syndic8.com in washington." though he seems to be speaking clearly, dave asks him his company name again; then his name again. "oh god, *you're* bill kearney". the audience laughs resoundingly, if a little nervously; dave's on his front foot now. "well," launches bill, "i hear a lot of platitudes being served around..." the RSS issue's "suffered from a lack of democracy, in the guise of a benevolent dictatorship... a lot of us would like to collaborate, participate in the rising tide that raises all boats."
dave inhales, and attempts to sidestep this, quite overtly: "i don't want to hear this, don't want to discuss this now." charlie nesson tells him, "dave, this sounds like complete bullshit. explain; what is the dispute between the two of you?" "it's probably not contributive to this conference," mutters bill. "get the microphone away from him!" retorts dave, "...it's my keynote... whatever you think about me personally, i've made a lot of contributions"... a plaintive wobble in his voice, self-justification in tone. "i have heart disease, okay", digression wrt gregor asking him to keynote oscom, thinking of "cultural interop before technical interop", ("no, that ain't gonna happen"), dave proceeds, "this asshole over there wished me dead!". "that's absolutely not true!" screams the now unamplified bill from the back of the auditorium.
larry rozen tries helpfully to interpose: "there are a number of people who hate each other. it's sort of unfortunate." in civil defense, in a formal legal model, explains charlie, this would be arbitrated by an un-self-interested moderator. "this is a dispute that you can't moderate," he informs dave... attempts to move the debate sideways. michiii pipes up, "now i understand why the jerry springer show was invented here in america". there's a brief audience attempt to drag the discussion back to topic - the concept of provenance, of credit for others' work in open source, going back into the roots of academia, and the currency of ideas - publishing, peer review, nothing real until subjected to show and tell, nothing settled until fully discussed, all voices listened to. but dave is too choked to add more of his voice; after a brief thanks, he walks away, voice cracked, visibly shattered.