Kernel Trap Interview with Theo de Raadt

An anonymous reader writes "KernelTrap has an insightful interview with Theo de Raadt, creator of OpenBSD. The wide-ranging interview focuses first on the past few years of OpenBSD development, then moves on to the recently released OpenBSD 3.9. De Raadt talks about how binary blobs threaten free software, and how OpenBSD developers work to reverse engineer them. He also talks about the future of OpenBSD, his views on Linux, and why developing truly free software is so important to him."

Comments

Theo

[Anonymous Coward]

Weird... was Theo having a bad day? He's always seemed like such a nice guy, but in this interview he really comes off like a total a-hole... very un-Theo-ish.

Re:Theo

[MerlynEmrys67]

Here is the problem with Theo. He is smart and opinionated. Having these two things in common make him a very difficult person to get along with if you are either Smart, but hold a different opinion because you come from a different set of assumptions - but especially if you are NOT smart and opinionated.

I have had discussions with Theo about trying to get my current employer (at the time) to open up documentation so OpenBSD could write drivers for our hardware. Lets just say I failed (Sorry Theo - I really tried, to the point that my annual raise was affected by it). However I found Theo to be very supportive and personally agreeable to me - I assume he realized I was trying to help and doing the best I could.

I can imagine people that are fighting against things he is trying to do could see him in a negative light - but again... I see the same kinds of things said about all of the great ones.

Re:Theo

[ArbitraryConstant]

I've witnessed him being an asshole.

Having to deal with him regularly might not be fun, but sometimes it takes assholes to get things done because they're prepared to piss people off to do what needs doing. If the goal were to make OpenBSD into another Ubuntu or Gentoo, his attitude probably wouldn't be that helpful, but for the goals they have it seems to work.

FCC Rules

[jusdisgi]

I sure wish he had taken a better position on the wifi "FCC Rules require Binary Blobs" issue. He basically agreed that the FCC does require that the consumer not be able to change the frequency, but claimed that it should be dealt with in hardware, not the driver. This line is particularly poorly thought out: "Let the FCC go after the vendors who made the flawed devices."

See, here's the thing...the people he needs to convince here are the hardware manufacturers. You aren't going to get them to release open drivers by suggesting that the FCC should "go after" them. In fact, it serves to reinforce their binary-blobs-only position; after all, that's their current protection. But worse, by tacitly agreeing with their position about the FCC rules, he cedes the important part of the argument...the part where he could have won it. That's because while the FCC does indeed require that the consumer not be able to change the frequency to licensed spectrum, they have never taken the position that changing the source code is normal consumer operation. After all, consumers can change the frequency on many other chipsets (even in Windows) with binary patches. This is simpler than changing source code and recompiling it. I have never heard anything from the FCC that says you can't distribute source code with this functionality. Which is good, because the current mainline Linux kernel does distribute code that does this. If FCC rules actually forbade this (as the hardware companies are claiming) then it would be illegal to distribute the Linux (and presumably OpenBSD) kernel in the USA.

There was a wonderful discussion of this on the LKML recently in context of Intel's binary blob driver.

Re:FCC Rules

[TigerNut]

As a current and past employee of several companies that make wireless transceivers subject to FCC licensing, I can tell you that there is no cost effective way to limit a device to FCC restrictions purely in hardware. Example: A cellular radio or any other modern RF link uses a synthesizer to set the transmit frequency. The output frequency of the synthesizer is a function of the reference frequency and the programmed divide ratio, and the total span of achievable output frequencies is dependent on the VCO that the synth is controlling. The maker of the synthesizer is not usually in a position to dictate the exact reference frequency, nor the VCO that it's hooked up to. The VCO vendor doesn't dictate the type of system that it will be installed into, and therefore can't strictly limit the frequency that it will tune to - and even if they did know exactly where it was going to go, then production tolerances dictate that you have some tuning margin in the design to allow all parts to hit the specified span. That means that individual parts will be tunable outside of the specified span on either the high or low side, and if the micro that controls the synthesizer commands a frequency outside the FCC limits, a lot of the time the hardware will have no problem doing it.

The same thing applies generally to power output levels. Sophisticated radios have some spare margin in the transmitter power output, and the actual output power level is calibrated at manufacturing time and then set in a FLASH based lookup table. The output power is then controlled using the embedded micro, driving a DAC. In this system, having open code on the embedded micro means that an uncaring individual could just crank the power output without regard for the FCC requirements.

You can say what you want about the motivations and ethics of the OpenBSD team members - if the source is out there, there will be others that take advantage of any "gains" they could make by tweaking some tuning parameters beyond the design or regulatory limits.

Ask Theo de Raadt how long it took him to get from his buffer-overrun Sun console hacking days to where he is now - almost everyone goes through a phase where "Just because I can" is sufficient justification to do poorly thought out things.

Re:FCC Rules

[TigerNut]

Filters limit the frequencies that a system can broadcast or receive, but they also have an insertion loss penalty. This reduces the efficiency of the system significantly - if a given filter has 1 dB insertion loss (which would be pretty good, implying that the filter probably costs a decent amount of money) then it would impart a 20 percent reduction in power output. Therefore it would cost you 20 percent more current, at least, to get the same RF range. That would (a) decrease the battery life and (b) increase the heat load in your system.

Wireless system designers use filters already to limit out-of-band emissions, but the problem is that no practical filter has a 'brick-wall' response where the passband ends exactly at the edge of the allowed spectrum. In a typical 2.4 GHz wireless network system you could probably go outside the band by 10 MHz before the filter rolloff became significant. With that freedom, an enterprising wireless LAN operator could set up his own little playing area away from everyone else's interference - but he'd be tromping on some unsuspecting folks.

Re:FCC Rules

[Kadin2048]

I'm not sure if you're being intentionally thick or what. FCC regulations cover more than just how a device can be used, they affect every stage of its design, and the market that's controlled by the FCC is a pretty big one. You over in Europe may think that what the FCC does isn't relevant to you, but I can guarantee you if you turn over a few peripherals you have on your desktop, that you'll see "Tested to Comply with FCC Standards: For Home or Office Use."

Because hardware and device manufacturers don't want to have to make multiple versions of their product if they can avoid it, chances are they're going to make it compliant to the largest number of regulatory bodies that they possibly can. Hence why my mouse is manufactured in China but approved according to regulations in the U.S., Canada, Germany, the E.U. (separate from Germany), and a bunch of Asian ones I can't read. And that's without even counting the non-governmental certifications (UL, CE, etc.).

An FCC regulation that changes something fundamental about how electronic devices have to be made is almost sure to affect people everywhere in the world, just like the E.U. RoHS rules are going to change the stuff I buy here in the U.S., even if we as a country didn't give a damn about how much hazardous substances were in our electronics. (We do, we're just taking our time about it.)

So while the FCC doesn't have any direct authority outside of the U.S., it affects how lots of things which end up on the world market are made, and you'd have to be pretty naive to just ignore that.

Financing?

[AltGrendel]

...I swear I will never get over how incredibly much money a University acting as a middle man between DARPA and us can bleed the flow of financing.

Any idea who he's refering to?

Re:Financing?

[kevin_conaway]

About 15% of the funding, awarded in mid-2000, had remained unspent, de Raadt said. According to de Raadt, two days before the funding was cut off, Jonathan Smith, the computer science professor in charge of the project at the University of Pennsylvania, phoned de Raadt. Smith told de Raadt that several people at the university and DARPA were uncomfortable with de Raadt's antiwar comments, which appeared in The Globe and Mail of Toronto in early April.

Source

We don't buy hardware that OpenBSD doesn't support

[linuxbaby]

Though we only use OpenBSD on a few of our servers (we have about 150 servers) - we NEVER buy hardware that OpenBSD doesn't support, because to us that's a good test of whether this hardware is going to last or not.

If a hardware company is so proprietary or secretive or locked-down that OpenBSD can't (or chooses not to) support it, I don't believe that company will last in the long run.

Great Interview...

[link915]

This was an excellent interview and Theo seemed fairly down-to-earth. I actually agree with many of Theo's POV's but don't always agree with how he conveys them. This interview seemed to show his *softer* side :)

Honestly though, he is right...the big Linux vendors really needed to step up and donate to the project. I am a FreeBSD user and certainly understand the need for funding to keep these projects going. OpenSSH is an amazing piece of software that we all use quite a bit. I can't say that I give all of my money to these projects but I do purchase CD sets and can only hope that the rest of you do as well.

I guess sometimes we are all dicks when we really believe in something. Although Theo can come across as a dick sometimes he really does stand for a good cause. Software should be free!

The reason companies do not open up their drivers

[Anonymous Coward]

The fundamental reason why companies do not open up their drivers is because the average end user considers it a Linux problem when Linux doesn't have proper support for a given proprietary piece of hardware, instead of a problem with the maker of the chipset in question.

I think one reason for this is because there are a zillion consumer devices out there and no real place to be able to look up a given piece of consumer hardware and see who is making the chips for said hardware, and whether the chipset in question has a Linux driver. More importantly, if a given chipset doesn't have a Linux driver, the documentation should tell us whether this is because the chipset in question is closed, or if it is because no one has had a chance to write a driver.

If this information is out there, when people give the usual "Linux sucks because it doesn't support X piece of hardware" flame, the reply can be "blame the makers of X piece of hardware, not Linux". If this mindset catches on, companies will start supporting Linux better. For example, I bought a Creative Zen Nano instead of an iPod Nano because the Zen had full Linux support; the iPod doesn't.

The problem with making this online database is that someone will need to be motivated to make such a database; this is a non-trivial task. The wiki model is perfect for something like this. Indeed, someone has a wiki-based database like this for IBM Thinkpad computers

Re:We don't buy hardware that OpenBSD doesn't supp

[idontgno]

If a hardware company is so proprietary or secretive or locked-down that OpenBSD can't (or chooses not to) support it, I don't believe that company will last in the long run.

OpenBSD confirms it. Adaptec is dying.

Re:So petulant and arrogant.

[Anonymous Coward]

Oh my, you really don't have a fucking clue, do you?
The OpenBSD project's recent funding problems have absolutely nothing to do with licensing; zero, zip, nada. The problem is not companies (Linux vendors, Cisco, Sun, etc.) modifying OpenSSH and without releasing changes publicly. The OpenBSD/OpenSSH project doesn't care about that, they expect it to happen. The problem is with said vendors using, redistributing and profiting from OpenSSH without making even a modest monetary donation in return. Given this, please, enlighten me as to releasing OpenSSH under the GPL would have any impact on this? Where in the GPL does it state that all redistribution and/or modification requires supporting the software's developers financially?
You think expecting a little money for something you poured blood, sweat, and tears into is "arrogant"? How about including open source software in almost all of your products (Cisco, Sun), and not giving a penny back for being given the opportunity to do so? Of course you have no obligation, but given the fact you're profiting off of this software, wouldn't it be wise to donate something (money, hardware) to the developers so that the software you're profiting from can continue to be developed? Some companies/projects have: GoDaddy and the Mozilla foundation. And hopefully more will in the future.

Oh, and whoever modded the parent up as insightful needs to be hit with a cluestick.

Re:So petulant and arrogant.

[hahiss]

Agreed; this analogy is utterly awful! Not only is there this unhealthy response to prostitutes (someone needs to get some therapy. . .), the *ENTIRE* analogy doesn't work:

A prostitute is someone who gives what they otherwise wouldn't (sex) in exchange for cash. Theo gives his software away for free, to anyone, to use as they wish.

Now maybe you (GP) think the Free Software isn't a sound business strategy, and maybe you think Theo's a jackass---and heck, maybe you think he's getting what he deserved because he didn't demand that corporations leave their cash on the nightstand ahead of time [THAT'S how you make a prostitution reference!] but holy crap son could you find a way to say that without invoking repellant examples that contradict your point completely.

Re:So petulant and arrogant.

[mobby_6kl]

>A prostitute ... gives what they otherwise wouldn't ... for cash. Theo gives his software away for free, to anyone, to use as they wish.

So, he's a slut?

Re:NDAs are a big problem?

[J.R. Random]

The very fact that an NDA is used means that the manufacture knows that the writer of the driver needs facts that can not be determined by looking at the source of the driver itself. Typically this involves the use of various magic constants that must be loaded into device registers at appropriate times. The manufacturer knows what the magic constants mean. Hopefully the writer of the driver does too. But nobody else does, and the author of the device driver can't tell them. So if there's a bug (maybe because the magic constant wasn't quite the right one to use in certain circumstances) there's no way for another person to fix it. Likewise if there's a desire to expand the functionality of the driver there is again no way for a third party to know what the magic constants should be.

Re:So petulant and arrogant.

[OttoM]

The problem is not the other open source projects. It's the commercial Linux and Unix vendors (and other as well) that use all the benefits of OpenSSH, but do nothing in return. To name a few: IBM, HP, Cisco.

Re:You are a little confused.

[Homestar Breadmaker]

Unfortunately, its not so simple. Many of the optimizations required serious recoding of gcc, making it MUCH slower to compile code, even when you don't have any optimizations turned on. Notice how gcc3 is twice as slow as gcc2? Notice how gcc4 is even slower?