Slashdot Mirror

SJVN is, of course, the well-known nickname and abbreviation for Steven J. Vaughn-Nichols, who has been covering technology as a journalist since... since longer than he cares to admit... and has been covering Linux and FOSS since the 1990s. This was basically a one-question interview: "How has reporting on Linux changed in the last 10 years?" After that, except for a couple of words requesting clarifications, we just let the webcam roll. (Note: if you know someone who would make a good Slashdot video interview victim, please put us in touch with them. Thanks.)

How great is Stephen Satchell's chance of being named to a court-ordered Microsoft oversight committee, assuming such a thing actually gets set up? And how much influence will a Slashdot interview have on the people who make those appointments? Probably not much, but Satch sure did a thorough job of answering your questions about how he'd behave if selected, and why he feels he's qualified.

Satchell:
Before I get to your questions, I want to say something. Given my druthers, I would not be "campaigning" for a seat on the Technical Oversight Committee -- I'd rather be writing code, publishing articles and books, and testing the hell out of computer products. I like what I do today. Unfortunately, I have to act as a grown-up and recognize that there are some things more important than my personal comfort. I feel that, given my skill set, the TC is where I need to be. If they'll have me. Bob Cringley wrote that I "wanted" the job. I don't blame Bob for the slight inaccuracy, because having been a magazine columnist I know what it's like to write to a specified "news hole." Sometimes nuances get dropped when you have exactly 500 words to get your thesis across.

I apologize for correcting spelling and punctuation in the questions submitted by the readers. Blame the writer in me. Besides, Slashdot has enough mispelt words as it is.

So as Art Baker used to say on television every week, "You Asked For It!"

1) What makes you the best man for the job?
by Binestar
Do you have any special agenda to get across or have anything for/against Microsoft that would make it so you were not impartial in your oversight of any federal rulings? As much as most people hate them this needs to be done in a fair and impartial way; will you be able to be fair and impartial?

Satchell:
No, I have no special agenda or strong feelings about Microsoft pro or con. And I don't love or hate the company. Microsoft is neither "good" nor "evil."

And (as the lawyers would say) even if I did have some bias, I've proven to the satisfaction of the computer industry, and to the satisfaction of several judges over the years, that I can be fair and impartial in my evaluations even in the face of bias. (Most of the companies who suffered my reviews add the word "tough" to "fair" and "impartial" in their description of me.)

As I read the explanatory text in the Competitive Impact Statement (CIS), the Technical Committee performs investigations and evaluations of Microsoft's compliance with the terms of the Final Judgement as issued by the Judge, using software expertise -- which is a requirement for being on the Technical Committee -- to catch those things that might otherwise fly by a legal beagle. In other words, the TC is a fact-finding body, and as long as it finds the facts, all the facts, and nothing but the facts it will discharge its job fairly and impartially. The Proposed Final Judgement (PFJ) doesn't do as good a job of explaining it all, but will most likely be the controlling document.

Brian Kendig asked a side question that bears answering: "How will I be able to work with Microsoft without appearing to be biased?" Another question asked about how I would deal with the monster egos at Microsoft. I'll kill two birds with one stone; let me tell you a little story:

Over the years I have worked with the Telecommunications Industry Association, a "standards provider" recognized by ANSI. A number of years ago, a TIA Standards ballot crossed my desk, SP-2812, that described a method of encoding commands and multiple data streams in computer-to-modem connections for enhanced fax and voice-over-data support, among other things - the cellular people started using it, too.. Most of the work on this Standard was contributed by a Microsoft representative (MR), who was also the editor for the Standard. (If the MR in question wants to identify himself, he can do so - he reads Slashdot.) The Standard was virtually unreadable as balloted. After working a full two days to understand what the Standard was trying to say, I was able to construct a description that was clearer, covered all the corner cases, and left room for future expansion. I then voted "No" and faxed my vote and comments to the TIA. The committee chairman asked if I would meet with the SP-2812 editor - the Microsoft representative - as soon as I could. "Mr. MR has a huge ego," said the chairman. We met, we argued, we discussed, and we came to an understanding. I prepared a "contribution" that was a rewrite of the contents of the ballot. The result was EIA 617, better known to the rest of you as ITU Recommendation V.80.

2) Do you feel it is possible to have a unified MS?
by petree
Do you feel that it is possible to have a unified monolithic Microsoft exist in the market without being anti-competitive? Specifically, if the United States government leaves Microsoft as-is (no "break-up") do you feel it is possible to regulate a company that in the past has shown no respect for government intervention?

Satchell:
Yes. Interestingly, it's the TC portion of the PFJ that lets me make that affirmative. Let me explain my take on it.

The effective sales life of a version of commercial software is now one year. The time required to get redress for grievance via lawsuit is around four years. Four years is more than enough time for a commercial software company to crash, die, crumble to dust, and blow away so that there is absolutely nothing left of it. Even if the software company wins its case against Microsoft, it's a Pyrrhic victory because the company will have lost where it counts most, in the marketplace.

The key to making the Final Judgment work to the short window dictated by the commercial-software market cycle is that the TC and Microsoft's Judgement Compliance Officer can solve a problem informally, rather than the complaining party and Microsoft taking years to build and litigate a case. It costs everyone less money, too.

In the academic and non-commercial software market, the key bottleneck isn't time, it's money. Very few non-commercial software efforts can raise the money required to take on the 360-kilogram gorilla - most of the time the leaders of such projects just sigh and go to Plan "B." With the TC (and some changes, hopefully, in the PFJ to allow academic, government, and non-commercial software projects to have standing) the cost of resolving a disclosure difficulty drops dramatically. That means that researchers, government employees, and non-profit developers can spend time getting the job done, not scratching their heads wondering how to do the job.

3) Asking slashdot?
by heyetv
Are you concerned that tying yourself to Slashdot, a known haven for us *nix freaks that are generally hostile towards Microsoft's actions, will harm your chances of obtaining this position, as it would require that those appointing the position perceive you as "objective"?

Satchell:
No. I welcome this opportunity.

I was asked by Slashdot if I would consent to be interviewed, and I agreed. This is no different than if ABC, CNN, InfoWorld, or The Wall Street Journal had called and asked for an interview.

Let me be blunt: I believe the Slashdot readership extends far beyond the boundaries you suggest. If you look at all of the questions that earned at least a +3 score from the reader-moderators, I believe you would see what I see: a broad cross-section of thoughts, opinion, and concern about the effects of the Microsoft Final Judgement. Slashdot questions are different than those that might be posed by a panel of journalists, or a panel of lawyers, or a panel of businessmen. I welcome the questions that Slashdotters ask because I suspect, frankly, no one else will ask quite the questions you have. And those (these) questions deserve answers from each and every candidate.

I would just as readily and as eagerly answer the ten most-insightful questions from, say, a community of Microsoft Certified Engineers or the Fraternal Order of Windows OEMs.

Slashdot isn't exactly a private club, either. My answers to your questions will be read (dissected?) by a wider audience than just those who have slashdot.org in the Bookmarks file of Netscape on their Linux or BSD systems. Count on it.

4) Restrict What?
by JJ
In what areas/functions must Microsoft be restricted in order for it not to violate anti-trust rules in the future? Endless loop:(n); see Loop, Endless.

Satchell:
That's part of the Great Debate between the Microsoft lawyers and the gaggle of Plaintiff lawyers. As you may have read recently, the Plaintiffs have divided on the issue over the number of yards to penalize Microsoft. The DoJ camp feel that ten yards is enough. The California camp feel that fifteen yards plus loss of down is more appropriate, and grumble that perhaps splitting up the Microsoft team into multiple "independent" teams shouldn't be taken off the table. It's now up to you fans to voice your opinion about the call. Unlike your federal elected representatives, the Department of Justice is required by law to respond to your comments, and the judge has to take that response into account.

The Technical Committee will then work within the framework of what the judge decides. I suspect that the rules of play following the pronouncement of the Final Judgement will require some tweaking and amplification once it gets going. The Court writes the rule-book. The TC will be the refs.

The biggest area of restriction that I see is Microsoft's use of NDAs to keep secret things that would block a non-Microsoft program from interoperating with Microsoft products. Much of the PFJ waxes long on this particular subject. Information is power.

5) What Would You Do With Passive Committee Partners?
by UberOogie
What would you do if you were saddled with two other do-nothings on your committee?

Satchell:
For the purpose of this answer, I'll assume that you do not intend to imply that I would be a do-nothing. Otherwise, what's the point? There would be too many lawyers watching for a slacker to get away with it for long, I would think.

If you have never bid on a Federal contract, you wouldn't believe the amount of law that covers the behavior of contractors, and the PFJ makes clear that each member of the TC would be a contractor of the United States Government. When I prepared a bid for a Patent and Trademark Office contract, the applicable law as described in the Code of Federal Regulations ran for more than 150 pages.

The TC members had damn well better care, and work hard and fairly.

6) How tough?
by silicon_synapse
Microsoft is sure to test their boundaries and see how far the overseers will let them go. How much would Microsoft have to stray from the new regulations before you make some noise? Would you be tough and bring to attention the most minor of infractions? Or would you be more lenient and use your judgement to make sure the intent of the regulations are observed?

Satchell:
Good question. One of the nice things that the CIS stresses is that the minor stuff is intended to be settled informally. That implies to me that "the most minor of infractions" can be cleared up quickly and without any attention needed by the lawyers, let alone a judge. That's good for Microsoft, that's good for the TC, that's good for the lawyers who should have better things to do, but most importantly it's good for the outside party with the complaint because, with the dispute resolved quickly, that outside party can get back to the business of developing and out of the business of bellyaching.

Not stated anywhere (yet) is what happens when there is a legitimate difference of opinion about what the Final Judgement really means. Legitimate differences of opinion regarding the meaning of clauses in the Final Judgement might be "kicked upstairs" for resolution or clarification; alternatively, if the TC staff includes a parliamentarian a proposal could be prepared and negotiated at the TC/Compliance-Liaison level and kicked upstairs for an up-or-down decision, again quickly. That's one of the many details that would need to be determined once the Final Judgement is in place.

More importantly and implied in the PFJ is that the TC's six-month report would be able to show any pattern of tendency toward non-compliance and edge-skating, a defect in the 1995 decree that made micro-infractions, to coin a term, almost impossible to track.

7) Playing the devil's advocate...
by BOredAtWork
While the methods Microsoft has used to become an industry giant are questionable, to say the least, the fact is, they are THE industry giant now. Microsoft is responsible for a great number of jobs, conducts research that would be too expensive for almost anyone else, and MSFT is a staple of a great many investment portfolios. Assuming you would become partially responsible for ensuring their compliance with federal regulations, part of your job will inevitably become spin control.

To break Microsoft's chokehold on the industry will send their stock into a tailspin, cause their R&D cycle to slow, and cause a chaotic move for power in various niches by everyone from giants such as IBM to various smaller companies that most people have never heard of. This will cause ripples (or shock waves) in everything from the Dow Jones Industrial Average to unemployment figures to the number of dot-coms that show up and fail at trying to corner a niche to the price of new computers.

My question for you, then, is the following: If you do assume a role such that you oversee Microsoft's compliance with federal guidelines, how will you keep the ripple effects caused by your enforcement in check, and how will you justify the ripples that inevitably are created to the American people?

Satchell:
First, let me disagree in part with your first statement. Microsoft, from the beginning, had a set of axioms of operation that I believe were well-suited to the company as a start-up, served as well during its mid-growth, and led them into anti-competitive action when they became a monopoly. The axioms: "if you use it, you pay for it"; and, "we don't support products from other companies."

Some of the anti-competitive behavior described during trial was proven to be intentional and not related to company axioms, so I'm not trying to excuse Microsoft. That doesn't detract from the fact that some of their acts were good intentions coupled with bad choices, the usual paving material for the road to damnation.

Your question suffers from a touch of tunnel vision, methinks. Heikkile makes the point that Microsoft's actions go far beyond the borders and citizenry of the United States. I believe that the enforcement action that the various settlement documents contemplate will be sufficient to help everyone, not just the "American companies." As a TC member, I can put forward that anyone should be able to put forward a complaint...as long as it's in English. You can put that forward yourself, in your letter to the DoJ about your feelings about the proposed settlement.

The goal of the DoJ is to craft a Final Judgement that will stop anti-competitive behavior, reverse any gains Microsoft may have gotten from anti-competitive behavior, and keep Microsoft more on the straight and narrow in the future. The goal of the Technical Committee, as I see it, is to reduce the pain level to all involved when Microsoft strays over the line, to get Microsoft back in line with a minimum of fuss, delay, and fireworks.

I opine some more: by keeping the company intact, the PFJ doesn't upset the millions of business arrangements already in place, from developers to packagers to retailers to OEMs to volume customers. By DoJ's losing on the bundling/integration issue, Microsoft is saved from having to rebuild its operating system products and creating even more skews of its product line, which in turn saves on technical support by Microsoft and, more importantly, retraining at the OEM, retail, and IT level. In short, the DoJ/Microsoft PFJ already works to keeps the ripple effect down to something manageable, which says to me that the DoJ learned some things from its experiences with AT&T.

I repeat: the sole function of the Technical Committee is to keep Microsoft honest and to help them keep to the agreement they made with the Plaintiffs and the judgements of the Court. By doing my job to the letter, by the book, and according to Hoyle I would be minimizing the ripple effects while permitting Microsoft a chance to continue to do what it has done, bring product to market, and let the marketplace decide thumbs-up or thumbs-down.

8) To boldly go insane....
by jd
This may seem like a really obvious question, but how do you propose to oversee an organization the size and complexity of Microsoft, by yourself and maybe two others?

Microsoft has managed to avoid scrutiny by companies, courts, governments and even users. Many allegations made in the trial, such as "knifing the baby" remarks alleged by Netscape, would simply not be visible, by simply looking at Official Policy Documents. In fact, probably very little actual policy DOES appear in their Official Policy Documents.

In short, you can't hunt ghosts with an electron microscope. You need knowledge of what the right job is, and then you need the tools to do it.

Do you even remotely imagine that this is even possible?

Satchell:
It's obviously impossible. You're right, policing at the level you describe is a task too large for even 300 people, let along three. I doubt it's desirable, either. That would be like having your very own cop sitting in the back seat of your car every time you drive, writing a ticket every time you rolled a stop sign or cut a left turn (right turn for you Brits) too close.

If you read the Competitive Impact Statement, you will see that it contemplates that the Technical Committee will be complaint-driven. Because the process for the complaint (not yet defined, by the way) should be less onerous than filing a lawsuit with the 800-pound gorilla, people who feel that Microsoft has violated the agreement with respect to them will file complaints.

As suggested by the CIS, the TC will have to do some kind of triage and prioritize the complaints, and group like complaints together. Then they will investigate the complaints and work with the Microsoft Compliance Liaison Officer to see if there is a simple fix for the problem. If there is, the fix is put in place and the file closed and logged.

And if it can't be fixed? The TC reports to the Plaintiff committee, providing all information available about the complaint. It's then up to the Plaintiff committee to decide what to do. It could be as simple as a lawyer-to-lawyer phone call and directions to the TC and the Compliance Officer. It could be as complex as a special court hearing. Who knows? But all that stuff is above the TC's pay grade.

The PFJ anticipates that the TC may well need a staff. This wouldn't be a large staff, and would be sized according to the load. No, I'm not taking resumes... :)

One reason I suspect that the members of the Technical Committee must be "experts in software design and programming" is to minimize the "staff effect" that cripples the decision-making process in many oversight committees. If the members themselves have the background to understand the underlying issues, to judge the clues found in the source, books, and memos, and to assimilate the Microsoft technical and non-technical response, you eliminate the mini-trials that are the hallmark of the operation of many government regulatory boards. You also eliminate the exclusive judgement of staff, people who do not take the heat for their decisions like a member would or should.

9) The obvious question...
by Stonehead
Where lies in your opinion the boundary between anti-competitive functionality and "improving the users' experience"? By now, everybody is used to bundling a browser with the OS. But what about video-editing software? The (Sun) Java VM or the .NET Common Language Runtime? Passport? etcetera..

Satchell:
As a member of the Technical Committee, my opinions in this matter are, frankly, not applicable. Other people decide the policy, the TC just implements the policy as set by the judge. But it's an interesting question, and I'm happy to take a stab at it. To wit, my opinion:

Richard M. Stallman asked the right question a long time ago: "If it's broke, how do I fix it?"

There is a BIG difference between "bundling" and "integrating" user-land applications into an operating system. If Microsoft wants to bundle a DVD-ROM full of stuff with Windows 2006, I say more power to them...as long as I don't have to have any part of Windows Recording Studio in order to install Syntrillium's CoolEdit as my multi-channel wave editor.

On the other hand, Microsoft has to be able to test and support functionality that the users want, or say they want. In the enterprise environment, the ability to install patches from a central facility is a real boon to the Information Technology department. Integrating everything together reduces the technical support headache for Microsoft. If it reduces the time for a tech support call, that's money saved, either for Microsoft for warranty support or the user if he or she is buying by-the-incident support services, or for the IT department if it provides its own support. Fewer variables.

And that is where "competition" and "user experience" clash. The optimal solution from Microsoft's perspective is also anti-competitive, not because they are worried about sales but because they are worried about tech support cost. The optimal solution from the perspective of Big Business is to standardize on a single solution that can be deployed, and if Microsoft's solution doesn't do the trick then Big Business will find someone else who can do the trick. The optimal solution from the perspective of the home user is that something happens magically. The optimal solution from the perspective of the SOHO user is to have none of it, because the SOHO computer is an attractive target for the Black Hats and the SOHO user doesn't want to have to spend $2K on a super-tricked-out firewall appliance between his computer and the outside world.

If it's broke, how do I fix it? Any solution MUST answer that question, for all users.

10a) Corruption?
by jamesidm
What is in place to prevent Microsoft from potentially bribing you or other members of the committee? Would you turn down 7 figure offers for the good of the computing community?

-- plus --

10b) Re: Corruption?
by Odinson
Perhaps I could elaborate with my intended question.

Do you think board members should make themselves available for financial audits?

If so...

How deep into the board members lives can the audits go and how long after their stay on the board should their financial records be reviewed?

Satchell:
The Technical Committee has a report it has to file every six months (every month if I have any say) that details what it's done and the rationale it used for its decisions. An experienced lawyer is very, very good at seeing BS in reports like that.

And now for something I don't expect anyone here to believe: I've turned down bribes while I was a product reviewer. I turned down money. I turned down sex. I turned down neat toys. Several of my colleagues said I was a fool to turn down all that stuff - indeed, I have a long-running feud with a well-known writer over the importance of the appearance of being ethical. He tells people I commit "fiscal suicide." (Don't worry, I have equally nasty things to say about him. It balances.)

Ever been fired from a job because you couldn't lie? I have. Now you know. (Hint: it wasn't a magazine that wanted me to lie. Not that I would.) I'm just not good at lying. In order to be successfully corrupt, you have to be able to lie convincingly. I don't have that skill.

I expect that as a TC member I would have to file financial disclosure forms, that indicate major sources of income plus investments. I filed these as a journalist and as a reviewer, so I don't have a problem with that.

As for audits: people, please! This is the UNITED STATES DEPARTMENT OF JUSTICE we are talking about. If DoJ suspects ANY problem, they have an in-house investigative arm that you may have heard of: the Federal Bureau of Investigation. With the power to probe into any aspect of my life that they choose.

===

That's it for your ten questions as selected by Slashdot editors. There were other good questions that I think deserve a response, so I'll lump answers to them in the next handful of paragraphs.

First, in my lab I run 14 systems variously populated with Windows (98SE and 2000 Pro), Linux (Slackware and Red Hat), and MacOS (8.5) on a 6400. I have a couple of Sun boxes loaded with Solaris, too, but they aren't running right now. All but one Windows system dual-boots Linux. I have an early BeOS CD-ROM, but it's been a while since I loaded it. My 286 box has QNX on it, but that's been powered off for a long time. The last MS-DOS box bit the dust last year. I switch between operating systems frequently, choosing the one that is best suited to do the job that needs to be done. The general-purpose Linux boxes all have SAMBA, and one of those SAMBA boxes is my Windows domain controller. (I'm trying to get HP to give me Linux software to make my Network Scanner 5 work with my Linux domain controller - no luck. Anyone at HP listening?) I may be forced to load the copy of Windows NT Server I have on the shelf, or I may put the newly-acquired HP Kayak into that service. The house is wired with 100-base T, and there are a couple of ethernet switches to break up the collection into logical groups.

I have written software for resale that runs on Ultrix, SunOS, Windows, MacOS, Linux, MS-DOS, and Hunter & Ready's VRTX. I have written operating systems for embedded products, and worked on operating systems all through my 30-year career.

Some high points: Worked on ARPAnet while at University of Illinois. Software and system design for embedded-computer products at Rockwell, Recognition Equipment, Addressograph-Multigraph. Benchmark writer at InfoWorld. Technical editor at InfoWorld. Built modem test lab at Ziff-Davis Labs. SPECmark member via MacUser magazine (did SPECmarks for early 680x0 Macs). Editor of the modem testing standard for TIA (now EIA 3800). Reviewed and wrote about more than 600 products in over 400 articles. Wrote Linux IP Stacks Commentary for Coriolis Books. Work at the annual Glenn Tenney THINK conference. Cameraman for local PBS station. Test Manager at Motorola ISG on the soft-modem project. Ran BBSes, BIX Telecom Exchange, assisted with CompuServie IBMSIG. Wrote, sold, and supported OTTO suite for analog modem testing.

I have never worked for Microsoft or any operating system vendor in any capacity, technical or non-technical. Microsoft and its direct competitors, directly or through PR representatives, have never paid me a dime, given me a T-shirt, or gifted me with a logo-bearing yo-yo. All Microsoft products that I've gotten through the years were purchased, bundled with systems, or were evaluation units (clearly marked "not for resale") provided to me in my capacity as a member of the working press and obtained through [Microsoft's PR agency] WaggEd .

Some yahoo bitched about how obtuse my contribution to W3C was. My training as a writer taught me how to write to an audience. That contribution - a formal response to a W3C position paper - was written in Standard-ese. I hope the bozo sees the difference between that paper (written for stuffed shirts) and this little tome. Don't like it? Write me.

Remember the Cyberporn story Time magazine ran in 1995? A bunch of us on alt.internet.media-coverage who work in the press fumed and fumed about that story. After much discussion, and many complaints from others who fumed that we were taking over the newsgroup, we decided to form The Internet Press Guild as a resource to mainstream press people who got an Internet beat without knowing much about it. So far, nothing as bad as the infamous "Rimm Job" has hit the mainstream press since we started.

I was an OS architect a number of years ago, so I understand the mind-set all too well. Grew out of it, but I remember the feelings, the attitudes, the arrogance. It helps to understand the organization you are about to oversee. Microsoft is one of the few organizations I have ever encountered that exhibits the mind-set, behavior, and mannerisms of the OS architect.

If I don't get the job, I'll just continue doing what I'm doing today. If the TC that is selected does a bad job and the computer industry craters, I'll seek out a new career. It's that simple.

That should do it. Now some of you know just a little more about me than you did before.

PS: To be complete, I should mention I'm not the only one with my hat in the ring. John Dvorak announced his "candidacy" in his November 2, 2001 column in PC Magazine. You need to talk with him about his qualifications for sitting on the Technical Committee.