Volume Number: 21 (2005)
Issue Number: 6
Column Tag: Interview

Business Unusual

by Dean Shavit

Open-Source With Teeth

Editor's Note: The following discussion touches on many controversial and potentially sensitive areas. We want to make it clear to our readers that, while we find this discussion compelling, enlightening and important, the opinions expressed herein are those of Michael Bell and Dean Shavit, and do not necessarily reflect the opinions of MacTech or its parent, Xplain Corporation.

Dear reader, if you've been reading the Source Hound, you'll know that I'm a big proponent of Apple's use of Open Source in OS X as well as using Open Source software to enhance the OS X experience. You might think that I'm a bit unusual in this business, but I do have a kindred spirit whose passion for Open-Source has taken him out of the comfortable confines of a writer's office, to directly into the line of fire. The person I'm speaking of is Michael Bell, a.k.a Drunkenbatman, (we'll call him DBM for short), proprietor of Drunkenblog (http://www.drunkenblog.com), a web site which has raised more than a few eyebrows by its rigorous defense of those targeted by Apple in the Tiger leak lawsuits, in which even Steve Wozniak weighed in with an email to Drunkenbatman as well as 25 other major Mac developers.

Taking a Bite out of Open-Source

Most recently, DBM has spent hundreds of hours deconstructing a software company called Maui XStream, and their CherryOS product. His investigation into Maui XStream is both interesting and technically enlightening, I encourage MacTech readers to check it out. As you will see, DBM has declined to talk directly about the CherryOS/Maui XStream situation.

I, the Source Hound (TSH) had a virtual sitdown with DBM (over iChat) to discuss Open-Source software (OSS) intellectual property (IP), and other issues as they relate to OS X and OSS in general. While we agreed on some points, we diverged on others. One point we both agreed on is that the world in general and OS X needs OSS, and that OSS transcends the interest of any one company, person, or platform.

Drunkenbatman in the House

DBM: drunkenbatman in the house.

TSH: Just a little background to get things going.

DBM: is this going to be verbatim or cleaned up?

TSH: I'll clean it up with your approval.

DBM: w00t

TSH: I've been reading your blog "Drunkenblog" http://www.drunkenblog.com for a couple of years now And I've been amazed at the steady progression of effort that's been going into your articles.

DBM: Um, are we doing the interview right now? Lol you have to say go.

TSH: OK, let me get us warmed up, go! How did you get started with Drunkenblog, why Drunkenbatman as an identity, what about the cow?

Figure 1. The Drunkenblog Cow

DBM: Erm, its a little weird to me that I've actually been doing it for a few years. There's kind of a cut-off point between when people were actually reading, and when I actually had it up. I'd have to check, but I think the first post is "So, I'm doing this because everyone is and you're supposed to, but don't really know what to say" and then a month of nothing. And then a terminal tutorial or two, and then things started to go bang. Back then, it was originally DrunkenBatman.com, because I used the nickname 'drunkenbatman' on a lot of places I hung out. 'My Blog' sounded lame, and so did 'Drunkenbatman's Blog'. Then I got a weird letter from Time Warner, and while they eventually backed down, I decided it was good to pick up DrunkenBlog.com and here we are. The Cow is just there because it cracks me up. Are you asking if I actually drink while I write?

TSH: No I don't think you drink when you write.

DBM: But I do sometimes. If you check the time stamps, I'm often posting at 3am or 5am, so you often need things to keep you awake or things to settle you down. You can generally tell when I've rounded the horn with the rum during the making of a post, as typos go way up.

TSH: I know how hard it is to write interesting, well-researched technical articles, so I am sure you must be sober the majority of the time. So run-ins with lawyers are nothing new for Drunkenbatman?

DBM: They're pretty new, and the thing with Time Warner wasn't really a big deal. They just wanted the domain, and I had to explain that for a variety of reasons that wasn't an option, and they agreed to walk away as long as I wasn't using it for anything commercial. I can kind of get where they're coming from -- they might not care that much about what I'm doing personally, but if it's their trademark and all, they have to defend it or they can have problems when someone starts hurting the image of their "hero."

TSH: You had mentioned a while ago that you came from the user interface side of the computer business, but for some reason, reading Drunkenblog convinced me that you were a developer.

DBM: Yeah, that is a misconception I get a lot, which is actually kind of cool. I can certainly code some, but no I wouldn't call myself a developer.

I think a lot of that misconception for people is rooted in the fact that I know a lot of technical things, and about the developer side of the process of creating software. And when you're making an application, there are certainly sides, even though someone might have to be wearing multiple hats.

TSH: You know a whole lot about OSS (Open-Source Software), as much as anyone I've spoken to. How would you characterize your role in the OSS great chain of being?

DBM: God. Or a minnow. Sorta depends on the day. Lol sorry. Um, can I ask why you're asking?

TSH: I want you to tell me if you're a hobbyist, journalist, aficionado, spy, etc. How do you characterize yourself?

DBM: Honestly? Reluctant poster boy for a whole slew of causes. When it came time to do some interviews, I ended up deciding that a lot of the commercial guys are good at getting their name out, and I wanted to pick people who were slaving away--they don't generally ask for money, but maybe I could shine some light on the kick ass work they're doing.

TSH: You know that's exactly what the Source Hound Column is all about, bringing attention to Open-Source projects! However, you're the first interview I've ever done!

DBM: So. . . CherryOS came about because Pieter Van den Abeele, the head of gentoo-alt (they do gentoo macos, gentoo for powerpc, etc.) forwarded it onto me.

TSH: The amount of effort you put into researching CherryOS was astonishing.

DBM: It was an off-the-cuff email from one of the developers where he asked "what's this about you getting ripped off", and he just thought I could maybe do something to help, since I'm known for putting together some big things

TSH: Big things such as. . . .

DBM: I had Steve Wozniak give me a quote for the site and twenty-five plus other big-time Mac developers. None of that happened by accident, I went to all of them and asked. I knew Woz from ages ago in a minor way; then I took his quote and went to the others.

TSH: Your technical knowledge seems eclectic and granular, you sprinkle your articles with it, adding flavor. Where does this knowledge come from? Is it part of your day-to-day work environment? I was a big Mac fan even before the OS X days, how about you? Are you willing to date yourself? I am!

DBM: Lol, that's a lot of questions at once. I know a lot of things I don't say, but a lot of it just comes just because it's something I happened to be interested in and went nuts researching. And really, when it comes to technology, it's like Mark Twain said: "History doesn't repeat itself, but it rhymes." There are patterns in most of what we use, and once you grok the patterns, you can have an idea of how things will or could go. For example, a lot of the technology we use right now is really an extension of stuff that was around in the 70s and 80s. You pull someone here with a time machine, and they'll understand the algorithms and equations being used, even if they didn't have the horsepower to use it as pervasively as we do. If you grok one protocol, you usually have the patterns in the rest. You grok one object oriented programming language, and you know the patterns of what's up the others, the rest is just implementation.

TSH: Of course, it's the same with scripting languages. A loop's a loop; if you've seen one subroutine you've seen them all. . .but I'm more interested in finding out how you came to write about OS X and the Mac. . .

DBM: As for dating myself, my first Mac that I actually owned was a Quadra 660 AV. I loved that machine, and lost approximately one man-year to playing Bungee's Marathon. I loved that machine. It was so starkly different from anything else on the market at the time that the only thing I had to lust for were some of the machines SGI was putting out, or its big brother, the 840AV. I'll never forgive Apple for the GeoPort, but I loved that machine.

TSH: Yeah, I had a Geoport telecom adapter too, it wasn't one of Apple's best efforts. . .

DBM: I started using Macs because our high school used them for the school newspaper, and I was really interested in the arts at the time. I had a really supportive family, and my dad basically used his earnings from his summer job to pay for it, which was a really big deal. If I recall, it was about $3,500 with the monitor, plus the GeoPort adapter that shipped over 6 months later, which was a huge chunk of cash to lay out for a kid in his bedroom.

TSH: OK, so now having dated yourself, I'm going to hit you with this question. . . Do you remember COS?

DBM: I'm not the best with acronyms. COS? CherryOS?

TSH: Not CherryOS, but a predecessor, both in opacity and form. . . CryptoOS.

DBM: Lol, no, I honestly can't say that I do, although the name sounds familiar. Emulators in those days were hardware cards you'd slap into your computer; even Apple was selling one. Was someone trying to do it all in software then?

TSH: Not exactly. It was a hoax perpetrated by some folks in Germany, who claimed they'd written a new OS for the Mac that was encrypted throughout. This was when we were waiting for Copland, 1996 or so.

DBM: Oh yes, I do remember them.

TSH: I remember they kept it up for quite a while, until one day, they actually asked on their website for a copy of Copland. They soon disappeared after that.

DBM: They never actually produced anything--it was all just a press release here and there with a simple website, right?

TSH: Yup, they generated a lot of buzz. We were so desperate for something "new" for our Macs, that we would've taken just about anything that wasn't System 7.5! I couldn't help but make some sort of strange connection between COS of the 1990s and the COS of 2004 and 2005!

DBM: *grins* I actually think hoaxes are kind of cool for just that reason. They can show what we want to believe. The fine line between a hoax and a fraud is when they're asking for your cash.

TSH: Ah, that is the line, isn't it . . .so it seems our Mac lives intersect somewhat, though I'll have to admit that the computer I used in High School was an Apple II, and it was brand-new.

DBM: Hey, I used an Apple II in school too, because our schools weren't really rolling in it and once you have Number Munchers you're really just adding nicer graphics. The nicest computers were in the media lab and newspaper, which were primarily LC II's and III's and one Quadra. Oregon Trail, all the way.

TSH: Do you work in a Mac-related field now?

DBM: No, I don't really work in a Mac-related field, I just try to work in fields where what I'm using doesn't matter.

TSH: That's really easier said than done, so I'm guessing that you must be a writer or wordsmith of some sort.

DBM: In some ways, it gets harder and harder as time goes on, even if it's not a fun subject to talk about. I have friends who do nothing but web work, which you'd think would just require a text editor and web browser, and at worst VirtualPC in order to test stuff in IE for Windows. However, it's all those little vertical apps that just kill you. You'll be consulting for someone, and they use Source Safe to version their code, which has no Mac client to speak of, so you're just screwed. I guess what I'm trying to say is that what I do for a living isn't that important.

TSH: So some of your passion for Open-Source Software is work-related? I know mine is an outgrowth of wanting to avoid the verticals you mentioned.

DBM: A lot my passion about OSS comes from using and working with the apps, and I was originally fascinated by the idea of giving your work away. When you consult for someone, I think your job is to do what's best for the client, even if that means you end up picking up a PC. I don't have any problem with people using Windows or locking themselves in, even if it annoys me.

TSH: Interesting. "Giving it away" versus "sharing." There's a big difference there.

DBM: Yes, there is, but when I first got into it, like most, I didn't really get what was going on. All I knew was you could download this stuff, and the source, and use it for free. Peripherally, I knew there were things going on with MySQL and KDE that were causing a stir, but I wasn't really aware. And that's probably most people's exposure to Open Source. They'll be using something proprietary, and something comes along that's 'free' so they download it and use it. Sometimes what's 'free' is actually better than what's proprietary. And then they're not making excuses for what they're using to save a buck, so they become loyal. After a time, they might even become a believer. My passion for it now is primarily for practical reasons. We need software. Lots of it, piles of it. More people are being born than are going into programming, and while new coding paradigms and tools can act as force multipliers, we need to improve what we have and create new things. More and more often, those new things are being based on OSS, because reinventing the wheel just gets old.

TSH: In my opinion, free is always better than proprietary. It makes me a little ill to see software being sold (even in the OS X domain) that isn't as good as the OSS alternative. Yet many find a comfort in paying for it, even if they know they shouldn't. I'll agree that we're going to need more software, but we also need processes and quality control and testing and new means of version control and interoperability. The wheel is really all we have right now, it's often about the hubcabs and how shiny they are. . .

DBM: *grins* Many people have that opinion, but there's a difference between proprietary and free. I don't have a problem with closed-source software, but I get entirely annoyed at software that locks you in. Our data is our life, and I've lost count of the apps that I've bought that died and all that data was locked away within them, because I couldn't upgrade the app when there was a new OS or something. And yes QA (Quality Assurance) is certainly in real trouble, and yes much of it comes from companies having to meet the requirements for the quarter, but those types of things come up in OSS too.

For example, pressures are just often more indirect. The more users you have, the more things they want. If you choose not to put them in, you run the danger of your app being forked. Hell, you just may need to get something out because you'll be going on vacation. One OSS app I know rushed a release a bit more than they probably should have, just because they wanted to make the deadline for Apple's awards at WWDC.

But really, much of the great software I use is 'closed source', but none of it is proprietary. Most of the great RSS readers are closed source, but they can all import and export your data.

TSH: OK, we have slightly different definitions of proprietary.

DBM: Probably. But hey, why do you use a Mac? It's proprietary, yes?

You could pick up an X86 and run Linux and know everything there, with the exception of perhaps a video driver, is Open Source.

TSH: To me, proprietary means "closed source" and not necessarily proprietary in the sense that the data can't migrate out of the app. I've been over that ground before in my columns.

DBM: Ah, right. You're talking about formats primarily then.

TSH: Basically, I do have quibbles with some of Apple's proprietary practices, especially in the area of QuickTime and Fairplay. Also in the area of hardware that's unsupported by OS upgrades.

DBM: And their ROM?

TSH: The ROM is a sword that cuts both ways. At least there's less and less of it now. Man, who's wagging who here?

DBM: Yeah, some of those situations are a little uncool. Apple and what they do and don't support isn't helping their warm and fuzzy image. Wagging is fun. Share.

TSH: Well, let's just say that I'm on the record as supporting OSS on OS X and Apple Hardware for several reasons. First, while I am comfortable with Linux, I prefer Apple Hardware and the consistent GUI that OS X offers. Like yourself, I grew up on Apple hardware and the Mac, then made it my professional choice. I think OSS on OS X offers the best of both worlds, a cliche, I know, but it's really true. And the degree to which they protect certain things is really silly, especially when it comes to the lawsuits over the Tiger "leaks" you wrote about. . .of course, being a developer, I had Tiger, but didn't see what all of the fuss was about. In some respects, I think a lot of it is smoke-and-mirrors. Today, I'm using Tiger and it's great, don't get me wrong, a real advance in many ways, but like you said, some of the QA has gone out the window and it makes me wonder if the lawsuits were just a way for Apple to let the world know "we've got something really good, and it's a secret."

DBM: 10.4 probably never should have shipped in the state it was in, and most people know it. The fact that some real hardcore Mac people are making some rumbles is telling, as it's starting to freak some people out. The truth is, 10.3 never should have shipped in the state it was in either. Almost everyone who was a developer was saying "This is really going GM? My god, this isn't ready," yet ship it did. And then it blasted data off of people's Firewire drives, and FileVault is still one of the scariest and flakiest technologies around.

TSH: FileVault is actually pretty simple. But I know what you mean. Well, quality assurance issues aside, it was the "secrecy" surrounding Tiger that irked me somewhat. I mean, I'm a Select ADC member, I've paid a fee, and when it went GM, I couldn't download the GM build until the night Tiger was released. There were some real surprises waiting for me too! Anyway, we're getting off topic.

I wanted to explore proprietary v. OSS.

DBM: Understood. So, let me swing back around to why I love OSS.

Mac OS X really shows off exactly why I've fallen in love with OSS, as there's an old expression "Standing on the shoulders of giants." I doubt we'll see another major kernel rewritten in our lifetimes. Apple proved they couldn't even really do it, at least not with the current management and requirements they had. Copeland didn't just die, it fell on its sword spectacularly. In fact the only kernels we really have, minus something like a PDA, are Tron, BSD, Linux and NT.

Now that software is at the level it is, the cost of just improving what we have is so large that to start from scratch is becoming more and more of a non-starter, for both OSS and closed-source software. In a lot of the closed-source stuff, this comes in the form of libraries Apple ships with routines developers can draw upon, because it's just stupid for every developer to have to implement spell-checking. However, more frequently it might well be an OSS library too, and Apple may well be incorporating an OSS library when it doesn't make sense for them to reinvent the wheel.

Even the best OSS stuff we have is often built on shoulders of giants. An example might be AdiumX (www.adiumx.com) or Fire (fire.sourceforge.net), where they wanted to create a kick-ass native chat client for the Mac. If there wasn't an OSS solution available for them to draw on--LibGaim--(encycl.opentopia.com/term/Gaim) we'd probably still be waiting for the first version to be released. Because it was there, they were able to write a smaller percentage of new code to do what they needed, and their efforts were multiplied by the available library and the fact that they didn't have to reinvent the wheel.

TSH: Aren't you forgetting Mach in your list of kernels?

DBM: I am of sorts, but Mach would be pretty lame without the BSD userland, so I just sorta lump them and BSD together in my head, as minus a few weirdities it feels like BSD to a user. Still out of all of them, if we're not talking about handhelds and such, NT is really the only major closed-source kernel around, and I doubt they'll be rewriting it anytime soon, especially since they were smart at the beginning and made the code very portable.

TSH: Who are the giants in the OSS world now? Which projects are the ones to keep an eye on, especially for OS X users? If you've been reading my column, you'll know that Patrick Luby is certainly at the top of my list.

DBM: What project is Patrick Luby on?

TSH: NeoOffice/J (www.neooffice.org).

DBM: Oh my gawd. Duh. I have an email from him in my inbox, that's how far behind I am. I feel so sorry for that guy. Many people have bitten off more than they can chew, but he and his partner really have the proverbial elephant. It's literally like two guys and 300 megs of code.

TSH: Why would you feel sorry for Patrick? He and Ed Peterlin are really the only ones who are positioned to do what they're doing. Patrick used to work at Sun on the OpenOffice Project and is a J2EEE wizard.

DBM: Simply because of the requirements they've put on themselves to do what they're doing--it's a Herculean task. Absolutely Herculean.

TSH: I think what they've done is spectacular. Absolutely spectacular. If Apple isn't paying close attention, it should. If they don't win something at WWDC, I'll be very very sad. All of the hacks that make NeoOffice/J work should be codified as libraries for XCode.

DBM: They have a project with a massive code base, doing a ton of interrelated things, that was most certainly not intended to be on a platform like OS X, so the amount of work they've set out for themselves--their final goal--would take a team of 20 a year to complete if not more. And they're two guys, with a project so large only they really understand it, so few people are willing, let alone able, to roll up their sleeves and pitch in.

The biggest 'projects' I keep an eye on are the force-multipliers, which have a way of sneaking up on you. Growl (growl.info) is a very good example of this, which I'm keeping an eye on for a few reasons.

TSH: I've done the best I can for NeoOffice/J from the sidelines. Rah rah rah. If Patrick and Ed could "bottle" what they did, it could bring a lot more applications to OS X.

DBM: If you've never heard of Growl, it's basically a unified event system for Mac OS X. You can set any up to throw its messages through it, instead of bouncing the dock or throwing up an alert. It's like the iChat bubble, except it fades away, so you don't have to switch to your mail client just to see what the new message in your inbox is.

TSH: Yes, I know what it is, but as far as OSS goes, it's for OS X only.

DBM: It's a really valuable service, so it won't stay OS X only for long. I'm told others are looking at implementing it almost verbatim on other platforms, as people are really digging it. They've gone from 19 apps supporting Growl when I first talked to them to over 120. People like Skype are dropping in support, now.

TSH: That's certainly going to help!

DBM: One of the things that fascinates me is what a force multiplier it is. Here's an app showing a great service that people love. Normally, those 100+ apps would all have to implement their own, instead they all get the functionality by tying into Growl. If it was closed source, they probably wouldn't have their app depend upon it unless it came from Apple, and Apple doesn't have anything like it.

B>TSH: OK, so a shared library=force multiplier in your view?

DBM: Well, all OSS code is a force multiplier in the sense that someone is saving time by using it instead of reinventing the wheel. In the case of OS X it's their kernel and userland, the CLI stuff, and in the case of Growl it's a way to throw messages on the screen.

It's actually ingrained into the idea of OSS, and why we have licenses like BSD and LGPL instead of just the GPL. There are cases where it just kicks ass if more people are using something than if they aren't, and if the TCP/IP stack wasn't BSD, who knows what Microsoft or Apple would have used and how screwed up it could have made things.

Another thing I think is notable about Growl is how their developer system is setup, and I'm kind of fascinated to see how it'll work out as it contrasts to heavily with say, the NeoOffice/J situation.

They have bounty of help, well over 20 developers from various other OSS projects, all pitching in on this one service doing things they want to see in it because their app will be making use of it. Normally in a situation like this, there's a tendency to go to lead developers, choke points who keep what goes into the software sane. They basically opted for a different approach, which is practically socialism in development with no real "no's" going around.

TSH: Growl is interesting to me in the sense that it was never intended to be an Application, but an extension to other applications. Sort of like the OSS version of an OS X Framework. I wonder what would happen, say, if Apple open-sourced one of their Frameworks, like Coreimage. I wonder if the code would lend itself to picking up by others, or would be considered a proprietary mess. It's an interesting hypothetical question.

DBM: Probably a little bit of both. I know a lot of the KHTML developers are not big Apple fans right now, or rather they're very annoyed with parts of the fan base who have the wrong idea about their relationship with Apple and Safari. One of the biggest problems with anything like that though are prior considerations.

TSH: Much of the same speculation's revolved around the source code for Hypercard, which people have been asking Apple to release. I say "be careful what you ask for, you just might get it."

DBM: Most of the licenses out there require you to give up the patent rights to whatever you release, so you can't start beating people over the head later, and there could well be things Apple doesn't want to give up. And truth be told, you never know what Apple actually owns. What I'm saying is, Apple just might not really have the option with some of their stuff. If we go back and look at NeXT, Apple was using Display Postscript for their display layer, which they'd licensed from Adobe. Assuming NeXT wanted to open source their OS completely, I am going to assume NeXT would have had a hard time getting Adobe to allow them to Open Source their Display Postscript stuff.

TSH: Well, yes, it's widely known (and even trumpeted by Tiger itself) that OS X (and QuickTime 7) contains royalty-bearing technologies.

DBM: Yes, that it does, many of them surrounding codecs.

But most of Apple's software is a black box. We have no idea what's really in it, so we can't assume that there's nothing in Quartz they haven't licensed from other companies or don't own the patents to, so they can't give it away.

TSH: The "Black Box" analogy of Apple's software is more than a little interesting to me. In your recent deconstruction of Maui XStream's CherryOS, you used various techniques to compare memory dumps of CherryOS and PearPC (www.pearpc.net) to prove your case.

Of course, knowing in advance what to look for made your job that much easier.

Yet if we were to apply the same techniques to OS X, or Windows 2003 Server, and had the luck to know what we were looking for, do you think either would come up clean?

DBM: Feel like you're treading on dangerous waters? *grin*

See, here's the deal. Companies like Apple and Microsoft are basically becoming intellectual property companies. It's their lifeblood now, it's just everything. People think of Microsoft as not being very innovative, but in some ways that's old-school thinking, and doesn't take into account just what they spend on research and just how many patents and such they're pumping out, and it goes the same for Apple.

TSH: Maybe. However, before I went into the computer business I wanted to be a poet. I wrote poetry, and studied for my Ph.D. in creative writing for six years. You read other people's poetry and you read. And you forget it, but it creeps into your writing. T.S. Eliot once said, "bad poets borrow, great poets steal."

Now you can get away with that in the arts. You can hear an echo of someone else's voice inside your writing and often that's desirable, yet there's no mechanism to make a byte for byte comparison of that. In the software business, there is. Especially when you know what to look for.

The reason I guess I'm bringing this up as Apple and Microsoft rigorously defend their IP, is the question of how well they'd fare if put to the metal?

DBM: Right, and there's that element, but not here. Their IP is their treasure-chest. Their current product might die, but their IP is what they live and die by, and to endanger that would be the height of stupidity--the potential cost is just not worth the risk. People say that the GPL has never been tested in an American court, but it's written so tightly that no one wants to try. And the larger the company, the larger the payout. It's hard to go after small to medium business, but lawyers start smelling dollar signs when you hit multinational and the words class-action are used.

TSH: Yes, the GPL's a work of "spare genius" (quoting myself)

If there was any type GPL code in their product, and it was proved, who knows what would happen or what they'd have to do or how devastating it could be. They could lose all the patents that are in it, so there's just no point in taking that risk.

DBM: Now, you could well find code if you knew exactly where to look, but it probably wouldn't be what you think.

TSH: There's no way that Apple or Microsoft can control their code that closely. All it takes is one lazy programmer that wants to make it to their son's baseball game, and it's in there. No one is the wiser. OSS is so prevalent, that it has to get inside some programmer's head.

DBM: Exactly, and that's pretty much where I was going. If a programmer has a problem or deadline, and access to Google, they may well come across a function or algorithm somewhere and drop it in and no one is even aware. That stuff is going to happen, sure, and there might even be some commonality to it, but it's not the end of the world.

TSH: I have seen enough of Tiger to know that Apple isn't spending a whole lot of time scrutinizing every aspect of its own work. Tiger changed the output of diskutil and system_profiler, for example, which broke some of my scripts.

Really the question in my mind becomes how many proprietary software products could stand to be scrutinized the way you scrutinized Cherry OS and pass the test. I wonder how common it is to cobble together OSS code and sell it as your own.

DBM: I can't talk specifically about CherryOS and such, because of lawyerly-things, but while I think its a good question the answer always comes down to: quite a few, but rarely anything big.

TSH: If, for example, I were the CIO of a corporation, and someone came to me with a streaming video solution that worked quite well, and priced well, could I safely assume that it wasn't a fraud?

This seems to me to be the biggest question your CherryOS investigation has spawned, at least in my mind.

DBM: Right.

TSH: Shouldn't software be "checked out" before it's bought or sold? And who is going to do the checking?

DBM: If it were me, one of the things I'd look at was how well the software was known. This stuff doesn't really occur in the mainstream, simply because someone has too much to lose and it almost always end up coming out if someone mainstream tries it. As the exposure of an app goes up, and people become aware, they'll get suspicious if things aren't on the up and up.

TSH:I'm going to WWDC in a couple of weeks, and I'll be surrounded by many folks who I admire from the OSS world while simultaneously surrounded by those from Apple who code for the black box. In this information-driven world, wouldn't cross-pollination be the norm, rather than the exception?

DBM: OSS is starting to become so powerful because of economics, not ideals. The ideals are awesome, but long term it's just economics. Copyright law goes through the congress, they are the ones who would be legislating such a thing, and OSS doesn't exactly have a major lobbying arm.

TSH: Well you and I are forming a very weak arm, but maybe it'll get stronger.

DBM: I know, I've just had to learn so much about trade secrets and IP that its just a lost cause in this way, because to me the real problem is that OSS is actually very well-protected, it's just that so few of those can make use of those mechanisms.

TSH: I honestly believe that ripping off OSS is more common than you'd like to think.

DBM: Oh, I think its common and happens, just more on the fringes. As awareness of an app goes up, and more eyes are on it, suspicions are there and it's not hard to poke around--and the consequences can be devastating to a company.

And there are great mechanisms for protecting OSS, it's all built into the copyright law, but because it's a civil thing the resources that are often required to exercise them are just huge. Here's an example, the vast majority of OSS developers never go out and register the copyright for their code. This costs all of $30.

TSH: Well, they should!

DBM: So let's say they find out a company is basically taking their code and reselling it, and want to go after them. A good IP lawyer costs $150 to $400 per hour or more, and most individuals simply can't afford that kind of cash. However, if you've registered your copyright before someone is distributing stuff with your code in it, it triggers all kinds of things like "willful infringement" and damages are starting at $100-150k and up. Now it starts making sense for an IP lawyer to take your case on contingency for getting his cash from the damages. . . but no OSS app developers do this, because this is fun for them, not a business.

TSH: So it costs an OSS developer just $30 to copyright their code, but how much does it cost Apple to patent something? A whole lot more! So I understand exactly what you mean. But the OSS programmer has another recourse, one much more influential than an IP attorney. . .they contact Drunkenbatman and cry tears of grief into the White Russian on your home page. . . .

Figure 2. The Drunkenblog White Russian

DBM: Lol There is that option, but it's not one I'm encouraging. Lol sorry, I'm cracking up here.

TSH: But let's take this example. Not only is there the legalities of "willfull infringement. . ." I would, as the developer, take my case directly to the customers of the folks hawking the product with my code in it . . .I think they might be a little concerned, at least as far as their image goes.

DBM: Yeah, but while I'm not a lawyer, I believe there can be legal problems in doing that too. And really, the USA and such isn't where the scary stuff is going on. You want to be scared, drop into China or Pakistan or Russia. This isn't well known to Mac users, because no one there uses Macs, but it's just a nightmare there in terms of IP.

Intellectual Property laws have gotten a bit whacky here in the USA and such, but when you drop into one of these countries with almost no IP laws to speak of, all bets are off.

TSH: Microsoft especially has spent a fortune on trying to clean up Windows Piracy in China, and is succeeding, I hear.

DBM: Barely, and China is just one of them. Kyrgyzstan is a place I've learned a bit about, and you basically pop down to their markets and anything you want costs $5 on a disc.

TSH: The whole notion of IP is pretty much a Western ideal anyway.

I've done quite a bit of traveling, and trying to find "genuine" software, music, or DVDs in certain countries can be quite difficult.

Getting the rest of the world to subscribe to our views of IP is going to take generations, if we're successful at all.

DBM: There's simply no recourse in these countries, and while IP laws are whacky here, they're what allow open source in the first place. You drop in on some of these Windows shareware sites, and there are so many red flags around some of these little utilities. But again, Mac users don't really see this because hell, Macs are so fricking rare outside of a major urban area, let alone California. Half of their sales are to the USA, so half of their sales are spread out across the rest of the globe, which means they're spread pretty thin.

Oh, other countries will get there as they become more prosperous, they simply have to. It's economics. Corporations are starting to funnel a lot of their IP into countries like China, and they want that protected. And as they generate more of their own IP, they'll want it protected. They just aren't there yet, but they will be, and it'll happen faster than it happened here.

TSH: Here's a question: would open-source even exist if there were no IP laws?

DBM: Yes, but only in the form of "public domain software".

TSH: Well I think you and I agree that poaching OSS software should be met with consequences.

Whatever they might be.

DBM: OSS is basically someone saying "I'm going to put this on the shelf for you to use, instead of keeping it to myself, but only if you follow the restrictions in this license." If you remove the ability to put those restrictions in, you're just left with "I'm putting this up here, do whatever the hell you want."

And we've always had things like that in some form, but it would be much much less than the sheer mountains of code that are being funneled into OSS.

TSH: Essentially the only restrictions are (generally) 1) distribute the source with the product 2) all rights you have are everyone else's rights.

DBM: Right, the GPL is designed to keep what's free free.

I do think OSS should be protected, but it's basically about copyright law and our system. And there's no way in hell I think I'm smart enough to fix our system, or even offer suggestions. . .I just know it's very hard for an individual to protect his intellectual property, and that it's primarily the domain of corporations.

TSH: Yes, free in the sense of uncaged. . .

Not free as in the sense of cost

DBM: Yeah, that's free as in beer versus free as in speech. There's no problem with selling OSS code as long as you don't take away rights or impose additional restrictions. It's a remarkable document, and was put together by very smart people, although reading it can make parts of your brain ache.

TSH: I know, my February column was largely about the GPL.

DBM: The spirit of the GPL does match the legalese, but the spirit is about keeping free software free. However, a lot of people have never really read the GPL, but have the spirit of the GPL in their heads as something a little bit different.

DBM: This is where the problem with Apple and KHTML comes in -- they're following the legalese just fine really, but many don't think they're following the spirit.

TSH: Oh Apple and KHTML. Yeah, that's a good point. Do you watch The Bachelor?

DBM: LOL, no, I can't say I watch The Bachelor. There's a ton of good TV out there, but I generally go by what's recommended to me, and The Bachelor has yet to be recommended. Should I be watching?

TSH: Well, my girlfriend and I watched the last season of The Bachelor, and it seems that the guy picked younger, less mature women because they're less of a threat and will be more compliant. I see the relationship between Apple and KHTML as somewhat similar. I think Apple feels that the KHTML folks should be saying "thank you thank you Apple for selecting us, without you we'd be nothing . . . "

Many people questioned Apple's choice of KHTML over Gecko, I was certainly one of them. But if you look at the Mozilla Foundation (http://www.mozilla.org), that's something, not just a little open-source project. I think Apple felt it could have its way with KHTML, but now there's trouble in paradise. . .

In many respects, despite how far Safari has come, I can't help but feel that Apple's wasting its time because it seems that Firefox is rising, while Safari is simply bundled.

DBM: Really? Well, there are certainly wham-bam-thank-you-ma'am aspects to Apple, but most of that is related to your pocket book. In the case of KTHML, and why they chose it, you really have to go back to what their options were at the time.

TSH: They had another, more viable option. But they really needed a browser. Badly. And quick.

DBM: Here's the thing: Apple is really short on people. They give the impression of this large company, but they are really starved for resources--their Mac division could well be in the red for all we know.

TSH: I understand that partnering with an OSS group is a form of thrift, but it would've taken less to slap Camino into shape. I think Apple took the path of least resistance.

DBM: The Mozilla engines are massive. They're multi-faceted, and weren't just a browser engine, they were a platform engine. The entire interface is rendered in them, with technologies like XUL. The code base was just this huge, massive, spidered mess of code, and everyone on those projects knows it. It's similar to the NeoOffice/J situation--getting your head into it is hard. Wrapping your head around it with limited resources would be a nightmare, and my hat is off to Mike Pinkerton of Chimera and Camino fame in more ways than you know.

There aren't that many people working on Safari. If I recall, OmniGroup tried to take the spidermonkey code for the Javascript engine in their OmniWeb browser, and it was just a nightmare. KHTML has major deficiencies in terms of compatibility and tech, but from a time to market and resources issue, I wouldn't be surprised if KHTML was really their only viable option.

I think Pinkerton would say the exact same thing, that it's completely understandable why Apple went with KTHML over Gecko and friends, especially if they knew they were going to be embedding it into the OS and doing things like Dashboard.

TSH: For some reason, I even knew back then that something Mozilla-based would eventually become the dominant browser on OS X. As far as timing went, yes, I agree that KHTML might have been the best option at the time, and OS X users were truly desperate to see something other than IE.

I think the KHTML choice was an expedient choice made at the right time. In terms of long-term viability, I don't think it's going to work out.

DBM: Has Gecko become the dominant engine on OS X?

TSH: As far as I'm concerned, it has.

DBM: I wrote about this a bit in "Gecko's Unsightly Ass" awhile back. KHTML may have been their real viable choice, but yes I'd agree that Firefox is a better browser than Safari--although now necessarily on OS X.

TSH: Not for home users, but certainly for the corporate world. Firefox has been a Godsend for IT managers because there's a PC version.

DBM: This is actually a really interesting thing on the Mac. Per capita, we have just tons of OSS around, but something like Firefox is just a bit deficient on the Mac because they don't have the Mac people working on it. Thunderbird kicks ass on the PC and Linux, but is just kinda weird on the Mac.

Windows has a ton of OSS code around, way more than the Mac. However, per capita, while it's growing, I don't think its as large as it is on OS X and Linux. If Apple hadn't shipped a free toolchain, I'd wonder if it would have the OSS community in the way it does now.

TSH: IT wants Firefox because it's not as insecure as IE, and because java stuff will work reasonably well cross-platform. Safari barely works with jsp or applets though it's getting better.

DBM: The key for Apple then is having to 'keep up' with the standards, and devoting resources to it. Firefox has nailed out a lot of these compatibility issues, so if they were using Gecko they could "ride their work", but again I think many think of Apple as having more resources than they do. Safari unfortunately doesn't work with a lot of things. It's important not to dismiss it, because Hyatt and others have done some massive and impressive work there. However yes, the onus now is all on them to improve Safari and bring it up to par.

TSH: I'd like your take on the VideoLan Project, whose developers I affectionately call the "Coneheads."

DBM: The VideoLan project is an awesome one, and ties into some of what I was talking about earlier in a major way. They're really screwed on OS X.

TSH: What do you mean, "screwed?"

DBM: They simply can't get the people with the expertise they need to work on the project. Remember, it's a cross-platform app, but there are almost always platform-specific things that need to take place in order to have the app kick ass. My understanding is that right now, they're having major, fundamental problems dealing with CoreAudio and other technologies. Which means they need people with real CoreAudio experience to help, and those are about as rare as little boys at Neverland right now.

TSH: Oh, you mean they're screwed as far as Tiger goes.

Not just Tiger, although that's a big stumbling block. There are issues with the current app on 10.3 they know about. Things that aren't in that they want to have in it, things that are in it that are flakier than they should be. However, VB developers are a dime a dozen on windows. Developers aren't a dime a dozen on the Mac, let alone people who really know what they're doing, let alone people who know what they're doing and know Apple's technologies, let alone people who are all of the above and have the time to contribute.

TSH: Well, I've been pretty impressed with VideoLAN on 10.3.

DBM: I didn't mean to imply I wasn't, just talking about the situation they're in. It's an extremely impressive effort.

TSH: I wouldn't be surprised if someone didn't step up to the plate to help them out. VideoLAN client saved OS X for many people, including myself.

DBM: It saved it for you?

TSH: In many ways, it's VLC that makes OS X tenable as a consumer OS.

DBM: There's always MPlayer. . .

TSH: Oh it's OK, but VLC's a thing of beauty

DBM: *grins* Just saying, you might be surprised at how few people are actually working on the OS X port of VLC, which is why it's so impressive.

And yes, QuickTime isn't exactly the draw it once was to consumers. And is especially onerous in 10.4, but perhaps that's just me.

TSH: I think I heard from all two of them when I wrote my first article "How the ConeHeads Could Save Apple" (http://www.themachelpdesk.com) in 2002. And hey, you get 100% of your daily "Royalty-bearing" technologies when you use QuickTime Player Pro. . .

DBM: I lurk on some of what they're doing so I can have a finger on their pulse, and while I can't say the exact number, I'm pretty sure I can count the developers on one hand while holding a mug. And yes, you do get your daily allowance. And they make sure to remind you of that, everywhere, while asking for your cash so you can play your movie full screen. It's really egregious and uncool behavior, and this is one of the things that can just set me off. They're theoretically in the middle of a fricking media player war with WMV and Real, yet they're making their player just suck. We saw this with Netscape and the browser wars too. Microsoft certainly did uncool things, but I'll go to my grave saying one of the reasons why Internet Explorer won, if not the main one, was that Netscape just started to suck.

TSH: Let me see, I use VLC for: 1) Streaming instructional videos for customers 2) playing back content archived form my ReplayTV 3) watching VCDs 4) Watching .asf files 5) ripping incompatible video formats to raw pixels to import into QuickTime. VLC's the Dutch Boy in the dike of OS X, as far as I'm concerned. Plugs lots of little holes.

DBM: I guess I'm a little bit of a newbie when it comes to VLC, I pretty much just use it for watching Doctor Who.

TSH: And I know from reading your blog that that's really important to you. . .

DBM: If VLC didn't exist, QuickTime does accept plugins so those codecs could be dropped in like 3iVX and DiVX are doing, but VLC just does it so much nicer.

Good entertainment is a thing of beauty, and should be acclaimed. Good entertainment with an element of random absurdity that only the British seem to be able to really do just makes me smile. Lol, QuickTime is, at least from what I've told, something you do not want to see open-sourced. The code is just monstrous. Ok, I'm starting to feel bad. It should be mentioned that I have great respect for Apple and much of their work, but just that they've been pushing the edge and letting lots of things pile up that are starting to come to a head.

Maybe we could talk about something positive. Like iPods or something.

TSH: Well, Tim Monroe, who's a columnist here at MacTech, shows bits and pieces of QuickTime's capabilities in each month's column. QuickTime is certainly a whole lot more than a simple video player --plug-- for example, if you wanted to develop a video interface to a hardware device, QuickTime has facilities for that.

DBM: QuickTime is a platform, that's been built up over years and years, and as such is kind of like Mozilla--unless you've been working with it for years and years, your brain would just go into a coma if you had to deal with the code.

TSH: You mentioned that the VideoLAN Folks needed programming help. They might need legal help as well. It seems that the European Union is trying to shut them down. How do you feel about such a valuable (especially to OS X) OSS technology being subject to targeted legislation from a governmental body to uphold patents, even when it's obvious that no infringement has taken place? Is a format the be-all-end-all, or can a format be successfully arrived at from two separate code bases?

DBM: As to the legal issues around VLC, I'm assuming you're talking about the proposed and pending changes to the European IP laws to make them more in line with America's?

TSH: That's precisely what I am talking about.

DBM: It's a rough situation, and one I haven't really made up my mind on. See, there are different kinds of infringement at work here. There's copyright, which is an expression of something, and then there's a patent, which is an idea of something.

TSH: The VideoLAN folks seem to be quite convinced this might be the end of the road for their project.

DBM: It could well be. The problem isn't the code they've written, that's theirs. It's the ideas covered by the patents in many of the things they've written: namely the codecs.

I should say that people saw this coming, and it's why we have formats like Ogg, which were designed from the ground up to not infringe on any patents whatsoever.

TSH: Well, when something is distributed in a certain format, does the patent follow the product indefinitely? So what you're saying is that the actual video product can be irrevocably closed, so that it can never be legally viewed without a licensed codec?

DBM: Ogg is the container format, with various codecs, audio and visual, and something like Vorbis would be the audio codec. It's completely patent-free, but Ogg Vorbis hasn't taken the world by storm for audio files.

It's not about the format necessarily, its the ideas that go into creating the format.

Say for example, something like H.264. This codec was basically designed by a bunch of companies that got together and pooled their IP to create this killer codec. The key word is IP, and they own the patents to all of it, and setup a company designed to collect the royalties. Same for MP3. Lots of Linux distributions require you to download an MP3 player, or perhaps it's the encoder, because they can't do it legally. Apple went and bought licenses to all these things, which others aren't.

TSH: Well, that's a gray area to me. For example, Microsoft got away with FPNW (File and Print Services for Netware) and a Gateway Service for Netware, which helped Windows NT displace Netware as the dominant Network OS in the corporate world.

That was OK, right? Samba is OK?

TSH: SAMBA and such are different, that's clean-room reverse engineering.

And how is VideoLAN not reverse-engineering?

DBM: Because they haven't done any reverse-engineering. With something like SAMBA, they setup a Windows machine, sent it packets on the port, and tried to guess what it expected. After a ton of trial and error, they were able to work out -- with some bugs -- what NT was expecting and such. To my knowledge, patents weren't an issue.

TSH: And so being able to play back a format without the source code to the codec is done how?

DBM: What goes into creating that format is a published spec, what apple's calls an "Open Standard", but it's really a misnomer. This is usually in relation to Microsoft's codecs. It's not an open spec, it's one they've put up to use and use, provided you pay them what they're asking.

And really what it comes down to are things like method patents, and through them software patents.

TSH: What you're implying is that the VLC folks are knowingly NOT paying what they're supposed to?

DBM: Well, if they were in America they'd probably be shut down already, and that's the problem. Yes, if they are going to be using someone else's IP, under our system, they should be paying.

TSH: So what you're saying is that if the EU changes IP law to be more like ours, the VideoLAN project will have to move to China or pay.

DBM: Lol. Pretty much. I doubt that the code will die, this stuff is like playing whack-a-mole with a hydra. It'll go splat, spread, and reemerge all over the damn place.

Right now, the EU is lacking some things we have here in the USA around software patents, and that's really where the problem is.

TSH: Well, you've kind of shot down one of my favorite projects. Damn you Batman!

DBM: Once those IP laws go into place, there'll be a bullseye on VideoLAN in a big way.

No, like I said. You can rebuild it, and make it stronger. It may even make weird noises when it runs, it'll just be doing it in Norway or China or something.

Now see, lots of people think software patents are just wrong, and they may well have a good point.

TSH: Yes, I would be one of those, especially in those cases when proprietary software won't make nice nice with others, but that's a can of worms we won't open right now.

DBM: We have some really, really stupid shit going on. A good example is Amazon's "One-click" patent, which well, gave them a patent on being able to click a button and buy what you want. Apple ended up licensing it for $1 million dollars. Or someone ends up with a patent on a type of Window, or having something in a specific place. It doesn't matter how it was coded, or the implementation, it's just the idea itself, and people much, much smarter than myself think we should really be looking at how these laws translate to our software.

TSH: I'd really like to know where Drunkenblog is going to take us OSS fans in the next year or two.

DBM: Yeah, well I've set the bar kind of high now haven't I? Deconstructing MXS took well over 150 hours.

TSH: Yes, you have my friend.

DBM: my plan is back to business as usual.

Figure 3. Business as Usual

In Next Month's "Source Hound"

Well, after my long talk with Drunkenbatman, I think I need to retreat a little to something lighter, so I'm going to check out an Open-Source Project that everyone, including Apple, absolutely loves: SquirrelMail. That's right, the ubiquitous, fun and nutty piece of webmail magic.

---

Dean Shavit is an ACSA (Apple Certified System Administrator) who loves to use a Mac, but hates paying for software. Since he's not into breaking the law, his most common response to any cool solution is: "Does that cost money?" If it does, you can bet he's on the hunt for an Open-Source or freeware alternative. Besides surfing for hours, following the scent of great source code, he's a partner at MOST Training & Consulting in Chicago, where he trains system administrators in OS X and OS X Server, conducts large-scale Mac Deployment and Upgrade projects, and writes for his own website, http://www.themachelpdesk.com. If you have questions or comments contact him here: dean@macworkshops.com.