Open-Standard Deja Vu at Harvard's Berkman Center

Back on Thursday (October 27, 2005) -- only one week ago, today, but what seems like years in OpenDocument-time -- a group of standards experts met in a room at the Harvard Law School's Berkman Center for the Interenet & Society to talk about file format standards and the present situation in Massachusetts.

I'm offering a selective transcription of some of the high points in the first hour of proceedings and welcome others to do the second hour somewhere. Certainly anyone interested in this topic should be encouraged to listen to the recording -- if you have 2 hours to spare.

dramatis personae

Steve Bratt
COO of the World Wide Web Consortium (W3C)

Doug Levin
CEO of Black Duck Software

Tim Bray
Director of Web Technologies, Sun Microsystems

Bob Sutor
Vice President of Standards & Open Source, IBM

John Palfrey
Executive Director, The Harvard Law School's Berkman Center for the Internet & Society


Opening remarks of the panelists came first.

Steve Bratt:

Dr. Bratt describes how the Internet developed out of open standards.

“In our case, for the things we do – the foundational specifications of the Web – that's a good business model, that by spreading out and reducing the encumbrances on specifications that we do you enable the more rapid expansion and growth of those technologies and then enable commerse or communication on top of those things, whether it's commercial software or open source.”

“We also care about interoperability...to make sure our standards work together...accessibility...internationalization...”

Doug Levin:

Gives a concise description of Black Duck's business model.


Tim Bray:

Desktop has 2 problems:

1) it costs too much; and
2) it's not very innovative

Need to open up a truly competitive marketplace; we need completely unencumbered file formats.

“The lifespan of information is immensely larger than the life of any particular iteration of software technology. And particularly in the public sector, where the relationship between a citizen and a government is measured in decades -- and not a small number of decades, either -- this is just immensely longer than any software has been observed to live and so we shouldn't count on history suddenly changing itself.”

Bob Sutor:

“We're at an inflection point.”

How can you tell well you look around...tells the story of overhearing a young lady at Walmart telling her mom about OpenOffice: 'Well there's the package...you pull it down from the Web and it's free and it does all the stuff: wordprocessing, spreadsheets & presentations...'

“Massachusetts is not an argument between branches of government. There's something fundamental people are questioning about the way we always used to do business and, maybe, need to do in the future.”

Q&A

John Palfrey:

"The Topic: What's the hard question what is in fact the – forget the politics – what's the hard question here in Massachusetts?”

Morgan Reed
Association for Competitive Technology (ACT)

“Morgan Reed, Association for Competitive Technology. We represent about 3,000 small guys in their garage who haven't seen their wives in six weeks. There have been a lot of good points. When this policy was initially announced, the biggest concern they had was that it seemed 1) to be an elimination of choice; uhm that from their perspective – and these are my friends on the open source side too – was allofasudden you went from being in a position where you were able to argue the technological merits of what you were doing if you were an integrator if you'd built a product for, say 30,000 government employees, that glued different pieces together, uhm, you were now restricted from doing a value-add to say, 'I can take this format and this format.'"

"Uhm, there was a question of loss of choice, was the first merit; the second part was a confusion as to what it meant in terms of uhm why is PDF, you know, what, how does this work because there was this feeling there was a continuum of intellectual property rights, you know, what does this mean for me, what if I innovate on it, what if I create this new idea; then the last big complaint was a feeling that uhm – and you said 'process' – if you want to identify what are the thee things: Choice, Innovation and a feeling that it was untested. Uhm there were concerns that the format was completely untested that none of the integrators had yet even rolled out this solution in terms of marrying various document formats together. There was a feeling that they had no proof of concept; there were some enourmous training costs that were going to be, that they had to deal with. And again you pointed out earlier, you know this question of guys complaining that their business model might be out, that's easy to say and just wing off but it's actually, when it's peoples' jobs and it's people's livelihood, I think it's flip to say that, but it actually has some practical, there are some practical considerations that you have to take into consideration.”

Dan Bricklin
Software Garden

“What spec are they complaining about since the ETRM has a broad scope and the document [file format] piece only came in late? ETRM has been out for many many months, so what specifically are they complaining about.”

A: Morgan Reed: “It was pages 18-21 of ETRM [v3.5].”

Palfrey asks David Berlind to cover his article which Microsoft approached him to write on the Process. See “Microsoft: We were railroaded in Massachusetts”

David Berlind concludes his review by seying that, even if Sun had a unique opportunity to address some IP issues in ODF which MS did not receive [the Sun non-assertion statement was not released until after ETRM v3.5 was finalized], all sense of impropriety goes away because Peter Quinn's group still holds the door open for MS to introduce ODF. PQ: 'We'd be delighted to include MS in procurement if either they opened up their schemas fully or if they adopt ODF in Office.'

David Berlind, returning to Palfrey's initial request:

“The hard question here is 'Why would a vendor be concerned about anything but its right to sell their product to the customer?'”

Berlind: Microsoft's claim of illegal procedure does not address this question as long as Peter Quinn's proverbial door remains open.

Berlind: “Microsoft doesn't need to lift one engineering finger. If they want to check about an illegal procedure in the process, all they have to do is announce that they're going to support ODF 'cause by doing that, if Massachusetts was really against procuring Microsoft's products, it would be forced to scramble and come up with some other way to block Microsoft's products.”

Andy Updegrove
Counsel to OASIS

'We all made the transitions [in the early 1990's] from WordPerfect to Word.'

“Two years from now – if ODF transitions go smoothly – we're going to be wondering, “What were we talking about?”

Morgan Reed

“This is Morgan Reed, ACT. I feel like I'm the only one here who has some concerns. Two follow-ups that I think I wanna go back to: You hit the nail on the head when you said, 'This is the end-all, this is the end. Uhm, this is the end decision. Once and for all,' I think is what you said. That, in a nutshell, from the small innovators technical side is what I see to be the largest problem.”

“Frankly I don't care about Microsoft. If Microsoft chooses to support ODF, fine. But when you say that you want to adopt a standard that is admitedly not state-of-the-art technologically [Sutor interrupts: “NO!”] -- okay but we'll come back to that – you're going to adopt a standard and you're going to say, 'That's it!'”

"Well, my CD player doesn't support wax cylinders for listening to music. Technology will move. Everybody moved from Wordperfect to Word. Why? Well, probably because WordPerfect didn't step up on certain issues that we wanted out of Word -- in some cases – now you'll notice that I'm using a Mac, and I use Pages because I like some of the feature-sets on it. You should not have a situation where the state is locked in – locked in, talk about vendor-lockin – locked in on a standard when a small company might innovate something better. Their choice should be based on, 'What gives us our need?' And the discussion as to whether or not they have access to the data, that's a separate vendor-to-client discussion. And I think that one, and I think the state has an opportunity to say, 'We need – in purpetuity – access to our data.'

Tim Bray

“So. And. I've seen this movie before, and this movie was in the 1980's when you wanted to go buy a network and you could get a Prime network from Prime or an IBM network from IBM or a DEC network from DEC and then ISO was coming along and TCP/IP was coming along and they all said, 'Don't lock us into a standard, you're going to cut off innovation...you know 'cause innovation happens inside...you know...' and that was just completely WRONG!”

“And you know now if you went to the market and introduced a new networking product and said, 'Oh this networking product is incredibly innovative but of course it's only single-vendor.' Well, it's crystal clear to me that a few years from now anybody who comes to market with any product that says Oh, by the way, the file format is proprietary will be locked out of the market for the very same reasons.”

“Now you raised a bunch of issues, a lot of which I think are actually material and worth investing time in. You said one thing: 'By standardizing on a common file format, we're going to reduce choice.'”

“Excuse ME! Choice comes from STANDARDS! Choice comes from COMPETITION! Choice doesn't come from MONOPOLY!”

“I mean you look at the Internet. Look at the Web. Look at the electrical system. Look at the telephone system. These things are based on standards down to the core. Now, compare that to the world of office document software, which is not based on standards and, I ask you, which of those is obviously the least innovative?”

Morgan Reed

“I would just target two things that you said there. You said, 'Obviously.' You used the term of art, 'obviuosly least innovative.' Well, that's absurd on its face because you can't make a determination that it's obviously least innovative in the market-place."

"So let's just take that off the table as being hyperbole and say, 'You don't appreciate the features that they innovate.' Well I don't either; I'm not using it. But that doesn't mean that you can say, 'It is the least innovative,' ipso facto, it is not a fact, okay? And second of all when you talk about that it's interesting to note that you use the electrical system because there actually was some fights over what was the best versions of electricity to use. And when we standardized on it took time and it wasn't actually an open standard. It was a standard that was being pushed by one vendor. So the nation standardized on a single vendor's electrical delivery system, and it was a proprietary one.

Dan Bricklin

“The argument is about the definition of 'open' ...”

“When it comes to the innovation thing, this is not cutting down choice because first of all the particular standard is one that has a lot of latitude in it so there's a lot of room [to innovate].”

“And it just reminds me of the old days when we were doing EBCDIC versus ASCII and a big vendor had EBCDIC – IBM – and then ASCII came in and, you know, eventually ASCII took over by having ASCII and one form of data – now it's been extended to UNICODE – we have one format that everybody uses and it's made all sorts of things possible. But the thing is anybody could use it. So I think we're talking here about degrees of 'open,' about open standards and we should talk about that.”

Bob Sutor

“So as anybody ever raises concerns about these sorts of things, there's another format that's in play here: the new XML format in Office 12, the untested format that is not backward-compatible that will require everybody who has engineered around that base to do new things. Alright. So therefore at that level we can say, 'Well, we've got two formats here.' Alright? We've got a vendor-specific format, developed by a large vendor, admitedly with large market share -- a lot of desktops, if you will -- and you've got the other format which frankly is roughly equivalent, alright, in terms of functionality. What's happening is people saying, 'I'm gonna vote for the open format.' Ya know. Between the two of them, alright, here's the one that's developed by a community by people who have literally decades of experience in doing these things, alright, that's been open to everybody else, that's open in the future to be maintained by everybody else. Anybody can implement this thing. I mean speaking as a vendor, IBM will add support for this in Lotus Workplace. We're doing this right now. We're not going to be the only ones. And so what it's really coming down to is folks saying, 'Look, I've got two choices here.' Alright. One choice is more open, has all of the functionality, has a future that isn't going to be prescribed by just one vendor, shall we say. 'How am I gonna vote, here?'”

“And frankly, if I'm a small company – to be very honest with you – I'm going to vote for a format that, ya know, is done in common, a standard foundation that I can then go and build upon. If I'm a small company that's worried about staying under the wing of a particular vendor [thinking], 'This is where I'm most comfortable, I'm in their ecosystem,' well you're going to make those choices. I mean that's what you do. So there are larger questions here.”

Tim Bray

“You don't standardize what the software DOES. You standardize what the software SAVES. and what the software SENDS. And you would be hard pressed to find [standards] success stories that contravene that simple rule. The place to standardize is on the data.”

Web 2.0 is marketing papp...there's no significant new technology there.

The single thing that's driving innovation now is the standardization of the browsers and plugins around that.

Blogging & syndication IS significant...the Web is becoming a writable medium...RSS...HTML...those are the standards. This is a sociological, cultural change, not a technological change.

Bob Sutor

“If we think of blogging, wikis, now we have a new standard for document formats. Is there something someone can do that's clever on top of that? Does it increase the ecosystem there?”

“So that type of interoperability drives great thoughts on top because they know they've got the plumbing in place. Whether you call it Web 2.0 or whatever. I mean the type of cleverness we've seen on top of – as you said the HTML, the core Web standards – is exactly what leads to this. It's somebody with a bright idea who says, 'I don't need to reinvent the wheel.' Right? Nobody can say, 'I can't do it. And then I go and I do something interesting.'”