Interview with Simon Phipps

Simon Phipps defends the open-source roots of Sun and the GPL-ization of Java.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

this is worst I've seen in hijacking Open Source

Dan Farrell's picture

I couldn't get past the line-

"I regard Sun as the original open-source startup company."

I'm sorry, but once you carve out your own perspective on OS, you can pretty much lay claim to it. I can't wait until MS lays claim to the Open Source world.

I'm an OpenBSD zealot... so you tell me what I'm thinking when some officer from Sun comes out and claims they are the original open-source company... it's fraud, plain and simple. Of course, if he came out and conceded that Sun really hasn't been the darling of the Open Source world, he be fired. Basically, as far as I'm concerned, he's in a position where he is forced to tow the line, no matter what. How can believe anything he's saying?

If Theo de Raadt can be claimed as a company, or the OpenBSD project itself (which I believe he owns or controls rights to ultimately), then THAT would be the first truly open-source operation.

Or can anyone say "Free Software Foundation"? "GNU"? Buehler... Buehler?

Richard Stallman long predated Sun--or Theo--in this

Anonymous's picture

You're partly correct; Sun cannot lay claim to this at all. Simon had no business trying to claim that.

However, Theo is not the first either, not by a longshot. Years before, in 1985, Richard Stallman made the offer, "for $150, I'll send you a tape with Emacs on it," and that's how he made money for a while. Yep, he actually made a living doing this. The FSF later took over this operation, and they still do it to this day, but the tape is now a CD-ROM, and there's a lot more than just Emacs on it. This predates OpenBSD, which I love and use, too, by ten years.

Furthermore, RMS would sell support services for Free Software that he had written for a fee of $200/hour. Lots of outfits took him up on it. Why? He could do work on the software (modifications, whatever) a lot faster than a lot of other people could. Thus, these organizations would spend a whole lot less than if, say, I were to do it for $50/hour (it'd take me way longer than 4x as long, believe me! Richard is truly a master hacker).

So, RMS himself was the first "F/OSS company," if you will--specifically, a sole-proprietorship. He showed us all how to do it.

Now, to Simon's comment: Simon, you're trying to do some good things here, but please, stop trying to give Sun credit for that which it does not deserve! It does indeed make it hard to believe you sometimes.

Actually Sun & RMS are roughly contemporaries

Simon Phipps's picture

My assertion about Sun being one of the first open source startups rests on the fact that the company was founded by a bunch of geeks who used a free operating system with off-the-shelf hardware to start a company. The geeks included Bill Joy, the free operating system was BSD Unix, the off-the shelf-hardware was components used to make workstations and the year was 1982.

That sounds like Apple's claim to "open source"

Sum Yung Gai's picture

And that operating system that Sun ended up producing was SunOS...a *proprietary* OS based on BSD UNIX. Yes, I remember that, too; I actually remember the Sun 386i. But that's exactly what NeXT, and later Apple, did with NeXTStep and Rhapsody/Mac OS X, respectively; both of these are also based on BSD UNIX. As Apple has emphatically shown, having your OS based on BSD UNIX doesn't make you an "open source company." Rather, it makes the Sun of 1982 a *user* of Free Software (the term "open source" didn't yet exist until 1998), just like virtually every other UNIX vendor was with X11 and the BSD TCP/IP stack.

So no, it doesn't appear that Sun was the first "open source company." If anybody was, it was indeed RMS. If you have a way to refute this, I'll listen...but please do make sure it's accurate.

Semantics, my friend

Simon Phipps's picture

Well there you go - I'm sure whatever I say you'll define away :-) Tough crowd round here, I must say.

More than just semantics

Steve Walker's picture

Hi Simon,

I've been following this with interest. Actually, I'm "Sum Yung Gai's" co-worker--he got me into Linux a few years ago. Sure, he's a little tough, but yeah, we are a tough crowd. You ought to know that it's because we have a high expectation of honesty from our members, including you. We engineer-types are like that. And that can't be just "defined away" by anybody. Yup, including you.

These people have a point. Just be honest, bro. That's all most of us ask for.


The two faces of Sun

Nicholas Petreley's picture

For what it's worth, I don't think you're giving Simon enough credit. I can't read his mind, so I'm not going to defend any particular argument he makes. I'd just like to point out to those of you who have those "videos" and evidence that Sun was anti-GPL, anti-open-source, etc., that Sun has long been two-faced about all these issues.

I don't think that has been an intentional deception, but the result of the fact that people at Sun have (internally) been split as to their attitudes. And that came through pretty clearly in what people said (they'd contradict each other) and what was published on their web site. I even recall writing a column about how Sun had (and may still have for all I know) two sections on their web site: One talks about how Linux is the best solution for business, and another talks about how it's a lame hobbyist solution compared to Solaris!.

Instead of nit picking at Simon for all the double-talk we've gotten from Sun over the years, I think it's wiser to encourage where they're going NOW.

You might be misunderstanding some of us a bit here

Sum Yung Gai's picture

Hello Nick,

I'm not sure that you're understanding where I'm coming from. We have certainly seen Sun's two-faced behaviour, just as you have. You're quite correct; some at Sun have said, "'Linux' can be a good business complement to something like Solaris", and others have said, essentially, "Down with 'Linux', THERE CAN BE ONLY ONE!", referring, of course, to Solaris. That's not what I was criticizing here, and it doesn't seem that the first poster was, either.

Rather, I was criticizing--and correcting--two things that Simon said. He made factual inaccuracies and misrepresentations. Whenever you do that, you must expect to get called on it, especially in our crowd. Just tell the truth; don't try to obfuscate, sugar-coat, or otherwise spin something. We can't read Simon's mind any more than you can; like you, we are limited to reading what's written. Had Simon been a bit more honest in his comments about why Sun has often been viewed dubiously by the Free Software community, I wouldn't have said a negative word. Rather, I would have said something like, "Yes, you're correct, and thank God you realize that and are helping Sun head in a different direction now!" I would've thanked him for that honesty. From what I read of the first poster's remarks, I infer the same from his/her post.

In short, it's not Sun's two-facedness. It's Simon's attempt to sugarcoat it here. He doesn't need to do that with us.

The only attempts at

PhoneyWar's picture

The only attempts at 'sugarcoating' I see is Sum Yung Gai's blatently untrue assertion that there was no such hostility. Like Simon, I have also personally experienced that hostility and it was solely about the language, nothing whatsoever to do with licensing issues.

As for Sun's so called 'two-facedness', you will find a certain degree of that in any large company. Every employee brings their own opinions and they rarely, if ever, fully coincide with those of directors and/or senior management. When some of those are in a position where they can speak for the company and are permitted to give honest answers and/or opinions then you are going to see this apparent division.

Compare this with certain proprietary software companies where only PR clones or carefully vetted senior management are permitted to discuss anything at all with the press. They are all saying the same thing but are they telling the truth or simply saying what the Board want you to hear? (Psst! Wanna buy some Enron shares?) :)

I have only one very brief comment to make about Clifton's assertion that copyleft forces multinationals to share code. It's certainly a nice idea. :)

Slashdot users don't count :-)

Anonymous's picture

I agree with Sum Yung Gai. Don't hang out on Slashdot. :-) Just 'cause Wil Wheaton's on there doesn't mean that everybody else there uses their brains.

Personally, everything I've ever read on the Java kerfluffle was about the licensing, not the language. That goes for that DebConf video, too. Even the long-haired stoner dude from Debian Legal (I can see where he could be a real jerk if he wanted to) kept going on about the license issues, not the technology. Just what kind of hostility did you see regarding Java itself?

As for the Microsoft comparison (it's pretty obvious who you're talking about), who cares? They don't own Java, they don't own Solaris, so it's irrelevant. I don't use their crap anyway. Neither do I care that Sun was two-faced back in the day. To borrow a little from Ross Perot, you can make a mistake, and if you're honest about it and you work to fix it, time passes. But even though you shouldn't live there, neither is it good to just whitewash over the truth about the past. Yeah, I think Simon was doing a little of that here. All corporate big-wigs do to some degree though, don't they? I'll give him this: I think it was pretty gutsy of him to even do the interview. But yeah, he's doing a little butt-covering there.

Simon, you got guts. Keep up what you're doing there at Sun. You're helping a lot of folks. I don't use Gnome, but I do use OpenOffice. And keep pushing OpenDocument! Lobby those state legislators, especially in Texas. We've gotta teach Microsoft that you just don't mess with Texas.

RE: Slashdot users don't count :-)

PhoneyWar's picture

Well, at least that's one thing we can agree on. :)

Seriously, Most of the hostility is from the C++ fanatics of course but there is still far too much from others about performance, interface 'look & feel', etc. All of which were genuine concerns several years ago but Java has moved on since then. Unfortunately most of it's critics haven't.

I see the same nonsense at the initial planning meeting of almost every new project. Someone will suggest Java and someone else will object, sometimes quite loudly. I used to have one project leader who would actually shout at anyone daft enough to suggest Java in front of him. He once sent a client's specification document back to me with 'Preferred Implementation Language: Java' scribbled out and "No F**king Way" scrawled underneath. After that I had no choice other than to ask for his resignation.

I never did find out why he disliked Java so much :)

As for Microsoft, I wasn't referring to them specifically because they are not the only offenders in this area, although they are certainly one of the worst, and who cares? Anyone running a mixed shop or using Java, etc. Microsoft may not own Java but that has never stopped them in the past. They've already made one attempt to 'Finlandise' Java as they have with so much else. Fortunately Sun stepped up to the plate and demonstrated that you can stand up to the monster and survive the experience.

I'd have canned him too. But there is a valid issue here....

Anonymous's picture

Well, I've heard that Java's slow, too. There's some truth to that, though. Java bytecode definitely is slower than natively compiled C or C++ code, assuming programmers of roughly equal skill. That's just the nature of executing bytecode. That doesn't mean that Java isn't useful, though. You were right to ask for that guy's resignation; that's just straight-up insubordination there. That, and when your client specifies that they want it in a certain language, then that's what you do. But yeah, there is a speed difference, and it's true of any interpreted language. That's part of the cost of "write once, run anywhere."

Here's a case in point: Azureus, the BitTorrent client. It's fantastic, and I used it to get (and seed) the CentOS 5 DVD torrent. It's also how I downloaded (and yes, seed) Ubuntu Feisty Fawn. Yes, I use it for legal stuff only. :-) Well, Azureus is great, and I love it! And yes, it's noticeably slower than some other BitTorrent clients I've used. This is probably due at least in part to the fact that it's a Java app and therefore requires a JRE to execute that bytecode. Do I care in every-day, practical terms? No, not really; my 2GHz chip isn't bothered by it one bit. Matter of fact, I'm kinda glad the Azureus team used Java, 'cause it's probably easier to run it on anything with a Java bytecode interpreter than if you had to use a compiled language. But I'm also sure glad that The GIMP is in a compiled language, 'cause we do need the speed there. Same for ffmpeg and the libtheora transcoding programs. The point is that C or C++ (hell, even compiled Pascal!) will be faster than Java for most things, if not all things, so if someone does state this as a reason for choosing C++, that's a valid technical point. Still no excuse for that insubordination, though.

A little revisionism at work here on Simon's part

Sum Yung Gai's picture

Simon says:

"What's more, the people who were doing the open-source stuff were really pretty hostile to Java. Their hostility wasn't moderated by a recognition that Java came from what would now be recognised as an open-source company."

Wrong-o, Simon. The "hostility", if you can call it that, was simply a refusal to deal with a restrictive (i. e. proprietary) license on Java. *That* was the problem. PERL is Free; PHP is Free; Python is Free. These three examples are in Freedom. Java, as a language, did not get the hostility. Your license terms at the time did. Fortunately, you are fixing that now, if much, much later than you should have.

Simon further says:

"The Open Source movement is busily accepting grace from IBM to promote Linux-shall we say, not entirely in isolation from the fact that Linux isn't Solaris. And, an engine of bad feeling was busy humming away nicely, between 2000 and 2003, with all the players doing their utmost to make sure that understanding and cooperation didn't break out."

No, we accepted--and continue to accept--"grace from IBM," as you put it, because Solaris's (totally proprietary) license was totally incompatible with the GPL, the license for the Free Software OS, born in Freedom, that most of us were already using. IBM, by contrast, backed the GPL'd GNU/Linux, so we welcomed them. Then, you came out with a license that, to quote one of your own former comrades, Danese Cooper, was *BY DESIGN* written to be GPL-incompatible. Don't deny it, Simon; I have the video. Now, if the parts of Solaris that *could* be released under the GPL, actually were released under the GPL (dual-license, perhaps, similar to or Mozilla code?), then we might've embraced those parts of Solaris as well. Not doing that was Sun's mistake. Make no mistake; we have no problem with Solaris. We do, though, have a BIG problem with the license that you've chosen to attach to it.

Has Sun done good things for Free Software? ABSOLUTELY!!, which I use every day, is but one of them. But please, don't try to sugar-coat or go revisionist on Sun's previous GPL- and GNU/Linux-hostile behaviour.

Revisionism cuts both ways

Simon Phipps's picture

I'm sorry, Sum Yung Gai, but I personally felt that hostility. I agree that there were plenty of thinking people whose issues were with Java licensing (although again there the issue is actually deeper since OSI set its rules after Java already existed, with no attempt to mediate the issue, and since Java shipped with full source in 1995 there was clearly scope to attempt a mediation). In fact, I was one of them and I joined Sun in 2000 expressly to fix that problem. But on Slashdot, on debian-legal, and in plenty of other places, there was a clear hostility to Sun that did not stem from any of these practical issues. Pretending otherwise is the real revisionism.

As for Danese's comments, I was there too (at DebConf, check that video) and I reject her assertions - she is reflecting the views of a few bigots, not of Sun. The license is just a bug-fixed version of MPL and was chosen because it created a set of freedoms calculated to start an effective OpenSolaris community, with no regard for or against Linux (maybe that's the real problem, huh? The Solaris and Linux kernels are separated by differences of design that make a license barrier unnecessary anyway). The hostility to Solaris that folk in the Linux kernel community have exhibited is not my imagination, I have the forum postings ridiculing Solaris, DTrace and ZFS. And yes, there was and remains Linux kernel hostility in the OpenSolaris community. Neither help any of us, nor does GPL-only-ism.

The real story of Sun's long-term engagement with Free software is the actual good news. You mention, but Sun's engagement is much wider. The accessibility code in GNOME came from Sun. Sun also worked on the I18N in Mozilla. Those and more are what lead to Sun being the origin of 26% of Debian. And they all flow from Solaris, regardless of the license its kernel uses. The good news is that, whatever the competitive positions various corporations prefer, Sun's engineers have actually been working on real F/OSS code for years and years.

I'd suggest that focus on what divides us is the heart of the matter. All the time those disputes have been in play, Sun has been diligently embracing the F/OSS model elsewhere. Today, by bringing together as many of the people actually working on the code as it's been possible to do, and by allowing past slights to be set aside, we're seeing a unity developing in the F/OSS communities that is finally giving Microsoft and the proprietary world cause for thought - witness Microsoft's increasingly desperate actions. That unity and change is my goal and I'm still amazed and delighted Sun asked me to work on it.

Good to see you here, Simon

Sum Yung Gai's picture

First, thank you for responding. You could've just left things as they were, but a reasoned discussion is never, in my opinion, a bad thing, so I'm glad you didn't. I do believe that you're trying to, as you said in that video, "do good."

Like you, I don't see the benefit in criticizing Sun's past behaviour. You're right; that's in the past. What I do point out is that you are sugarcoating that history here, and you don't need to. It's like with slavery; we know it happened, and I'm not upset with whites or anyone else about that historical fact, but when someone tries to sugarcoat it (and a few have), *that* is when I say, "whoa, hold up there!" Sugarcoating can, and will, breed suspicion, and (sometimes unfortunately), engineers don't generally set corporate policy, executives do. For that reason, when a top Sun executive speaks, we must take that as official Sun policy. I'd love, for example, to see an executive-level response to Clifton's point, below, regarding Jonathan Schwartz's comments about GPL'd software in poorer countries; Clifton appears to have a very good point there. You're not sure Jonathan was wrong there? I am. Most businesses fail because they fail to execute a good, solid, long-term business plan, not because they choose to develop or use GPL'd software. In fact, it seems to be working out pretty darned well for Red Hat, JBoss, MySQL, and several others.

The only serious criticism that I ever heard about Java was indeed regarding the licensing at that time. I don't count the idiocy that I see at Slashdot. Heck, licensing was the only serious criticism that I myself ever had of Java! You mentioned the OSI. The OSI may have set its rules after Java's birth, but the FSF had put out the Four Freedoms--and the GPL v2--years *before*. So "the rules" actually already existed.

However, that's history. After I read about the GPL+Classpath relicensing last year, that same week I went out, bought a Java book, and got busy (my other "Web language" is PERL). With that licensing change, my objections to Java immediately went away. Same with Richard Stallman, Eben Moglen, Mark Shuttleworth, and many others who matter. Heck, even the company who Sun often vilified not so long ago--Red Hat--celebrated Sun's move here, and they were right to do so.

As for Danese Cooper's remark, I gave you a fair shot and watched the video again, in case there was something that I had missed. I do note that you did look a bit like you got blindsided by what she said. I also did some Googling and found a statement by you making it clear that you were quite ticked off at her for that. I gotta tell you, her saying that didn't help Sun one bit, and I don't envy your position there. What's important to remember is *why* what she said was damaging. Isn't that part of the reason why Sun wisely choose GPL+Classpath, even though the LGPL effectively accomplishes the same thing?

You're right; let's move forward, as there's a lot to do. Sun is doing a lot of it. Let me make it clear; I applaud what Sun is doing now, as well as the contributions that Sun has already made; GPL'ing Java actually got me to start spec'ing v40z servers instead of those from HP (I like your use of the LSI Logic RAID controllers in particular due to their open specs). And as we continue to make progress, let's be fully frank--yes, including about history--as we make that progress.


Simon, don't defend the indefensible.

Clifton Hyatt's picture

"GM: In a speech a couple of years back, Sun's CEO Jonathan Schwartz argued that the GPL is "IP colonialism", because he claimed it imposed on poorer countries "a rather predatory obligation to [give back] all their IP to the wealthiest nation in the world". Why did he change his mind?

SP: Well, you know, this is an interesting thing to contemplate, because I'm not sure he was wrong. The GPL does require you to set aside commercial protections for your software. And, it is possible that the use of the GPL for the indigenous software industry, for example, in Brazil, might harm the Brazilian economy. If you actually read the argument that Jonathan was making at the time, it's a good academic argument. What made it controversial was that it was the Chief Operating Officer of Sun saying it."

No Simon, what made it controversial is that it doesn't square with reality, and an otherwise intelligent man (Jonathan) seemed to expect that no would notice, it was a bit of Orwellian doublespeak.

I called out this argument when Jonathan made it, and this is a pretty vaporous answer on Simon's part now.

I submit the truth is that proprietary software in combination with sw patents are the "economic colonialism" in the software industry. Simon or Jonathan needs to square the copyleft provision if they want to continue to defend this argument.

Copyleft forces a programmer to share their code with powerful multinationals, but what they won't factor in is it also forces the same conduct on the multinationals as well.

The multinationals can't outproduce you, because their production becomes your product as well, and service of that code is generally preferred locally (witness the outsource backlash).

That is the deceit of that argument, the GPL is a *two* way street. unlike proprietary and (potentially) the BSDs.

The "economic colonialism" argument is a rather shallow straw man when applied to the GPL, but actually quite valid when applied to proprietary and potentially some Open Source licenses.

Thanks for at least asking the question, maybe next time Simon will offer a more appropriate answer.

My reply never appeared

Simon Phipps's picture

I actually wrote a lengthy reply & it never showed up here :-( I've not the patience to re-create it, but suffice to say that I disagree with you. All the stuff you say is obviously right within a GPL worldview but neglects the game that is being played around that worldview. I believe Jonathan's comments raised an important issue about how big corporations could game (in fact are gaming) the Free software system, and that instead of digging behind them people chose to read cluelessness into them. Maybe I'll pluck up the courage to blog about it some time.

Simon, please do elaborate

Clifton Hyatt's picture

I would love to hear your response, if there is some greater context around Johnathan's statements that make them make more sense, I would _love_ to hear it.

I hope you will blog about what you tried to post here, I think it would add to the dialog.

I hope if/when you do you will also explain this comment "within a GPL worldview", because I don't understand what you mean by it, and you have addressed it to me before.

It seems dismissive, though I assume/hope you don't mean it as such. But please explain, what is a GPL world view as opposed to a what, MS world view, proprietary world view, global world view, accurate world view?

I went through proprietary to get to FOSS, the stuff that the proprietary or enterprise vendors do in a business, political, or inter-vendor strategic sense is not some big mystery, it's not especially deep or complex. A FOSS supporter will _feel_ differently than a proprietary proponent about what those entities do but there should be collective understanding about _what_ or _why_ they did.

Yea sure, companies are attempting to game the FOSS system, we have been watching that unfold long before Jonathan's remarks in question. I think if we lay out the _full text_ of Jonathan's remarks you will be hard pressed to get someone to see how they apply to gaming of FOSS as opposed to FOSS itself.

I think mainly I am just getting tired of "leaders", either business or political saying whatever seems convienient to their goals irrespective of the truth; as if no one else in the world understands a cohesive argument, or how to use Google to to look something up.

You, me and Jonathan, by virtue of the internet get to address people that are smarter than all of us. I think we have an obligation to understand and respect that, and show up with our A-game.

Sun can choose whatever business strategy seems best to them, I hold no animosity toward them whatever that is. What I have a problem with is the pollution of honest candid dialog with half-truths, spin, or outright falsehood.

Thankfully, lately, Jonathan/Sun seems to have consolidated on a particular path and message, all I hope for is that he remains honest in his pursuit of _whatever_ that path is.

For what it is worth I _do_ think both you and Jonathan bring more to the dialog than many, so I continue to look forward to both of your contributions.


I look forward to seeing your rebuttal, then

Sum Yung Gai's picture

Clifton's made a good case here, and no, what he says absolutely reflects the game being played generally. Saying that it doesn't is--at best--naive. Your statement totally ignores the fact of the successes of Red Hat, MySQL AB, and several other vendors of GPL'd software, within a Microsoft-dominated IT world.

It's unfortunate that your first reply didn't show up, because I really would like to see your response to this. If/when you do, as you put it, pick up the courage to blog about it sometime, please do include Clifton's remarks here, for proper context.

As for the general issue of "the disappearing post," I too have been bitten by that in the past. It can happen occasionally. My solution is to first write it in something like KEdit, and then copy 'n' paste it into the post. This way, if something bad happens to it, you just copy 'n' "post" it again.

You said it, Clifton

Anonymous's picture

You're right on here. I'm rather amazed that Jonathan would say that. If he says that the GPL is "colonialism", that sounds just like when Bill Gates called it "Communism", or when Steve Ballmer called it "a cancer", and when The SCO Group called it "un-American" and "unconstitutional." Does Jonathan really want to be in that kind of company?

See, the two-way street nature of the GPL is also the part that Microsoft and Apple don't like to talk about. It's what the RIAA and MPAA don't like to talk about. Even IBM, at times, prefers not to talk about that. Software patents, by their very nature, create specifcations on which a certain individual or company gets a monopoly. That's what Microsoft desperately wants with their Uh-Oh-XML, and that's why it's not an "open standard" regardless of what their marketing department and lobbyists spew out. Yes, it is patent-encumbered. That's why it's no good as an international standard.