Women in Technology

Hear us Roar

Weblog:   Software Choice vs. Sincere Choice
Subject:   Points of Agreement, Points of Disagreement
Date:   2002-09-29 18:02:54
From:   adamsj
I'm largely in agreement with what you say, Tim, but the points on which I disagree are important.

You say:

> It is true that I disagree with the idea of
> legally mandated open source, because I do
> believe that decisions should be made on a
> product's merits.

Bruce Perens makes many good points in the article you reference in your post. Let me add one:

What's you don't mention here is that open source is a merit in and of itself. I'm not referring here to technical benefits generally, but specifically to security benefits.

Reasonable observers of the Microsoft anti-trust settlement have suggested that there's been a quiet deal between MS and the US Government to put backdoors into Windows. Here's part of what Dan Gillmor had to say on the subject:

What if this means Microsoft is party to a U.S. government back-door into Windows, for spying and surveillance purposes, that the government understandably wants to keep secret?

I'm torn between whether I think this is unjustified paranoia or reasonable suspicion.

(As an aside, the conspiratorially-minded might note that the judge in the anti-trust settlement, Colleen Kollar-Kotelly, is also the presiding judge of the Foreign Intelligence Surveillance Court.)

Given the desires of some to have US intelligence involved itself in industrial espionage (presumably in a defensive, counter-intelligence mode) and the NSA's mission of making sure nothing anywhere is impervious to their snooping, I think Gillmor's suspcions are credible, and thus I think non-US governments would be well-advised not to use any Windows OS in any secure system.

I might extend this caution to state and local governments inside the US. Here's an example:

Several state governments are going toe-to-toe with the federal government on medical marijuana. Given that, I wonder whether it's safe for the State of California to keep relevant prescription records in MS-SQL databases. No jury in California will ever convict on a medicinal marijuana case, but if you're a doctor, you don't get a jury trial if the DEA pulls your presciption privileges.

(Normally, I'd dismiss this as a paranoia. However, I live in the court district where the current head of the DEA was a US Attorney. Later, he was my congressman. I think he's a scary guy.)

You also say:

> But I'm all for the softer European version of
> some of this legislation, which gives open
> source a strong boost without mandating it.

This doesn't go far enough.

I'd rather not see a flat mandate for open source, but do want to see preferential treatment. Unless there's an up-front overweighting for open source, salesmen (more so than lobbyists--the Davis/Oracle episode in California is the exception, not the rule) will unjustifiably beat open source. Open source has no salesmen: is this a bug or a feature? That depends on who you ask.

An instructive, if inexact, analogy here is to affirmative action: Given an institution with pervasive historical bias, preferential treatment is often required to correct imbalance.

> That gives the technical people in government
> some cover if they choose open source, and gives
> a reason for those who aren't considering it to
> take a closer look.

I don't completely disagree with what you said above or with your reasoning here, but I think you don't reach far enough, either in what should be mandated or in the effect such mandates can have.

What should be mandated: Open source technologies proposed for governmental use should be given an evaluation by the agency involved, if a competing closed source technology is in competition.

Why? Because open source products don't have the marketing infrastructure that supports sales motions. Without this support, open source can't compete on a level playing field.

I acknowledge this is problematic, given the desire to let open source compete on its merits. It could be said that a merit of proprietary software is that the revenue it generates is used to provide information about the product. While some of that information is pure marketing fluff, other information is genuinely valuable information. Open source doesn't have this advantage.

Is this truly a merit of proprietary software? I argue that it is not. The effort required to sort out the wheat from the chaff when going through sales support information is signficant. I'll claim (with no hard evidence) the effort required to evaluate open source software is roughly equivalent to the effort required to evaluate information provided about proprietary software.

Now, as to the impact of this mandated evaluation:

In the business world, open source software often doesn't get a level playing field, for the same reasons noted above (among others). With the government-sponsored evaluations described above placed in the public domain, open source advocates in business have a fighting chance. Of course, this is a double-edged sword--if an open source product is found to stink, the reasons will be documented.

While I agree with you that, at least so far as the US Government is concerned, I don't want to see open source mandated, I think the argument you give against it is insufficient.

It's true that

> If legislators can be persuaded to mandate open
> source, they can equally well be persuaded to
> mandate against it.

But I don't agree that

> Once that battle is joined, I'm pretty clear
> who will win. Most modern governments seem to be
> far too much under the sway of special interest
> money for me to want to see them getting
> involved in any issue where the other side has
> the dollars and the access to get their way.

I'm not so sure about this, for two reasons.

The first argument is on practical grounds. As you accurately point out:

> Of course, now that the battle has been
> foolishly joined, the ISC is not going to go
> away, and open source in government advocates
> are going to have to sup at the table they have
> set.

Fair enough--but by the same token, now that the cat is out of the bag, withdrawing from the fight is not a winning tactic. I doubt very much that the ISC and its backers will quit if we do.

But more importantly, my second reason is that I'm not so sure this is a losing issue.

You've put up, I believe, a false dichotomy of "special interests" against the rest of us.

I claim that various groups of citizens--Common Cause, the AARP, the NRA, to name three with varying beliefs--have successfully banded together to become "special interests". I put "special interests" in quotations because I think the phrase itself is misleading. Rather, these should be viewed as groups of citizens acting in an organized manner to express their _common_ interests.

(There is more to be said in criticism of the phrase "special interests", but not here and now.)

You are partially right to say:

> You have only to look at some of the absolutely
> insane legislation introduced on behalf of the
> RIAA and MPAA to see whether the public benefit
> or the power of special interests has more sway.

This is true at the moment, for a variety of reasons.

Primary is the delusion, usefully debunked by Larry Lessig in Code, that the Internet has a nature. Of course, it doesn't--what human creation can usefully be said to have a nature in the sense that the word is usually used? The idea that the Internet has a nature lulled people into a false sense of security. After that illusion left, a false, needless sense of despair arrived.

The idea that the Internet was immune to censorship and control and the idea that censorship and control of the Internet is inevitable are two ways of making the same mistake.

Just as human creations are mutable, human events (particularly political events) are not determined. Accepting (or denying) one tends toward acceptance (or denial) of the other.

Refusing despair is the first and most important step in avoiding the events we dread.

In particular, given a more realistic goal than "All software used by the U.S. government must be open source", such as what I've outlined above, we are in a winnable fight.

I'm not certain how to organize to win such a fight, though I have some ideas. What I am certain of is that many less probable successes are now facts of history.

Forty years ago this month a pitched battle was fought in Oxford, Mississippi. Dozens of federal marshals were shot. One reporter was killed, as was another man. It took thousands of soldiers to restore the peace. But at the end of the school year, James Meredith--a man no less eccentric or principled or brave than many in the free software movement--was the first black graduate of Old Miss.

This was not a likely outcome.

Truly remarkable things have happened in America over the last half-century, some good, some bad.

I don't think it'd be that far out of the run of recent events that citizens, acting through an existing, more general advocacy group or through a newer, focused group, could change current policy.

My belief is that a specialized advocacy and lobbying group--not a special interest group, but a common interest group--which covered many of the issues of technology and freedom (not all computing-related--pervasive video surveillance and DNA screening by insurance companies are two which come to mind quickly, and there are surely others) will be a great success.