Wednesday, October 28, 2009

Monty's Almost Certainly Looking For Investment

Remember how I predicted that Monty was attempting to create such an untenable situation for Oracle that they had to dispose of the MySQL IP and put it into his hands? Remember how all of Monty's protestations were based on the fact that he doesn't have the money to buy it? Remember how I said that he might not have it now, but could probably raise it?

Yeah. He's trying.

From Reuters, Florian Mueller is touring Wall Street:

EU strategist who is former MySQL shareholder and adviser announces Silicon Valley press conference (26 October) and New York City analyst briefing (27 October) -- Florian Mueller wants to "explain positions of critics of proposed transaction in the lion's den" -- New York event due to "strong Wall Street interest in the matter."

There haven't been any protestations from Wall Street that I've seen that they're unhappy with Oracle acquiring MySQL. Rather, this has been a European affair as Monty tries to get the EU to interfere in industrial policy for his own personal benefit. So why in the world would you send your sockpuppet to Wall Street to explain any positions? What interest would Wall Street have in the matter other than to provide funding?

You might argue that he's going there to talk about the position they're taking with the EU competition commission because there are lots of people who own Oracle and/or Sun stock, and might be interested in the matter. But I think a much simpler story is at least as likely: Florian is attempting to drum up a capital raise to acquire the MySQL IP to make the problem go away for Oracle, and to convince Oracle and Sun shareholders that Monty and Florian will do whatever it takes to block the acquisition so that they'll tell Larry to let go.

Add to the Look at the Balls on that Guy category: Florian Mueller.

Monday, October 26, 2009

Java APIs and Ivy: Click-Through Agreements Suck

We've been using Ivy for project dependency management at OpenGamma since day-one. It was one of the earliest decisions we made, and I'm quite happy that we made it. Furthermore, the work that the Ivyroundup guys have done to package up Ivy-specific files has been fantastic.

That being said, there's one big, honkin problem with it: Sun and their click-through agreements to download almost any of the "interesting" APIs that you need for enterprise software development. These packages (like the JMS or JavaMail APIs) are necessary for pretty much anything that you might want to do on a real system, and you can't legally put them up for download. And unfortunately, there are a lot of them.

The workaround, of course, is that someone in your organization downloads them all, clicks-through on the agreement that is completely and utterly unenforceable, and then sticks them on a private Ivy repository and you configure your ivysettings.xml file to refer to that location, but that's not particularly scalable. And for open source projects, it just doesn't work at all.

Yeah, we've got this Ivy thing so that it automatically identifies all the dependencies and downloads them so you don't have to do anything but have Ant installed. Except for the Java stuff, which you have to manually download and put into these particular locations, and that's a 20 minute task minimum.

Kinda takes the shine off the whole experience, doesn't it?

Now given that everybody seems to know that you can handle licensing restrictions in different ways (by a disclaimer in each file and a copy of the license included), why does Sun still live in the past? Why do they require a clickthrough for something that is essentially just an explicit disclaimer of warranties and liability? Why can I download Eclipse or NetBeans without a clickthrough but not the JMS 1.1 API interface files?

Man, I hope the EU finally lets Oracle buy Sun and put this type of stupidity out of its misery.

Wednesday, October 21, 2009

Monty, Stallman, MySQL, Oracle, and Sun: Open Letter Wars

I've tried to confine my ranting about the current state of the Sun/Oracle/MySQL debates to my Twitter feed, but I think I need to do more than the 140 character limit allows.

Background On Recent Moves

In case you haven't been following the state of play, we've got two recent open letters sent to the EU competition commissioner: If you've been following either of my posts on the subject, you'll know I'm not a dispassionate observer in this matter, particularly where Mr. Widenius is involved.

Competition and Acquisition

First of all, let's directly address the core matter at hand, which is that Monty, RMS, and the various others appear to believe that the Database market is hopelessly consolidated and were Oracle to get its hands on the copyright to the MySQL source code that would be bad for competition.

This, to be honest, completely and utterly disregards the actual history of the database market, which has always been one of consolidation and benefits to the consumer:

  • Illustra, a Berkeley spin-out, was bought by Informix
  • Informix was bought by IBM
  • RedBrick was bought by IBM
  • RDB was bought by Oracle

While this consolidation has reduced the number of vendors in the market, as of 2007 there was still pretty hefty competition with Oracle even then only with a 44.1% share of the paid database market. As someone who has had to work professionally with Oracle, DB/2, Sybase, and Microsoft SQL/Server, I can say this is almost certainly because it's the best overall product.

Furthermore, those numbers in terms of the database revenue are completely suspect (with the exception of Microsoft's). There's a huge amount of revenue for IBM and Oracle which are tied to services and software sitting on top of the database (such as Oracle applications and IBM services), and realistically the CFOs of each company can tune the percentage of the deal that goes to the underlying database based on what numbers they want to report. The overall deal may be $1MM, but the sales person has a lot of discretion on how they price the database component.

Is there so little competition in the market that it's hurting consumers? Hardly. The recent squabble over Oracle's TPC-C pseudo-announcements indicates that the vendors actively compete with each other. Furthermore, the rate of feature expansion has been truly dramatic. Finally, the ability of firms like Vertica to rapidly jump into the market indicates that this isn't a market that requires significant levels of competitive concern on the part of regulators.

The technology industry is based on larger firms buying smaller ones. Competition authorities should rightfully be concerned only if it harms consumers in general, not whether it harms a particular subset of users.

But MySQL Is Special

With all due respect, no it isn't.

Let's consider the pseudo-market for Open Source databases. We've got:

And that's just considering the relational ones. When you consider the NoSQL movement as potential competitors (which I, for example, most certainly do), MySQL just isn't that special anymore.

While it is possible that an Oracle acquisition might be bad for MySQL consumers, it doesn't follow that MySQL is so special and perfect and pure that it harms any general category of consumers. While databases aren't perfectly replaceable, if someone found that Oracle's stewardship of MySQL was so onerous that they wanted to move off of it, it wouldn't be impossible to move to another database, either commercial or Open Source.

What that means is that you're in a classic case where the acquisition of a particular company might be harmful to consumers of that company's products, but it doesn't generically affect the market in a negative way. IBM and Microsoft will continue to compete in the commercial space, and PostgreSQL, Ingres, and LucidDB will continue to compete in the Open Source space. There's no net harm to consumers as a whole from an acquisition, even if the result of the acquisition was the complete shutdown of all commercial support for MySQL.

Oracle Is A Bad Acquirer

First of all, let's get the obvious out of the way: Oracle bought BerkeleyDB, and continued to enhance it; Oracle bought InnoDB, and continued to enhance it. At no point did they crush them to drive Oracle database revenues, or change the licenses, or stop forward momentum. So when you look at the actual track record of the company, they're in the clear.

But they might do, because they're an evil, scary corporation that MySQL turned down once before (from the Stallman piece):

Oracle made an earlier effort to buy MySQL in 2006, but the management rejected Oracle's offer, in part because Oracle would not disclose its plan for MySQL, and some members of the MySQL management team were concerned that Oracle was only acquiring MySQL to curb its advances in the marketplace.

I know a number of people involved with MySQL when it was an independent organization. While there were people who worried about that fact, senior management wasn't. More importantly, Monty was willing to sell MySQL to Oracle in 2006 for the right price. The use of the words "in part" there are telling, because the primary consideration that MySQL's senior management had wasn't some happy-clappy love for the Libre Software Movement, it was money.

I'm sorry, but I fail to see what's changed in between 2006 and 2009 except that Monty is a whole heck of a lot richer. Why in 2005 and 2006 were offers ultimately rejected from Oracle based primarily on money, but now Oracle is an evil corporation that can't be trusted with MySQL? Larry's the same guy he was then, Oracle has bought BEA but they don't compete in any way with MySQL, it's the same company. Why would Monty trust Oracle back in 2006 but not now?

Force Oracle to Sell MySQL

This is Monty's solution. And it's cunning. It's particularly cunning that he says repeatedly that the obvious Monty-connected acquirer, Monty Program AB, lacks the funds to do such a purchase. Again, a half-truth.

MySQL was worth $1Bn in early 2008. Since then markets globally have tanked, but MySQL has had some good commercial strength recently within the Sun organization. So let's conservatively say that it's still worth $1Bn. Let's then say that Oracle values the acquisition of Sun highly enough to let MySQL go for less, and do a 20% haircut to $800MM. Who's got that kind of money to acquire?

  • Microsoft. You think Stallman and Monty would be happy with that? No.
  • IBM. #2 in the database market. Erm, raises same issues that Oracle would.
  • Sybase has the market cap (super-recently) but not the cash.
  • Red Hat has the market cap, but not the cash.
  • Novell lacks the market cap and the cash.
  • Computer Associates has the market cap and the cash, but is the place technology goes to die. They also have Ingres to work with.
  • VMWare has the market cap and the cash and an acquisitive streak, but would MySQL really fit into their product strategy? I can see Spring driving people to vCloud, but can't even fathom the same kind of strategic benefit for MySQL.
  • Symantec has the market cap and the cash, but their storage work has been pretty solidly focused on backup and low-level storage these days.
It doesn't look to me like there are that many companies out there that could really buy MySQL in cash and make Stallman and Monty happy.

But you don't actually need to have the cash yourself: you can use private equity money. It's happened before: BEA was funded by private equity originally to consolidate the Tuxedo market. That's why Monty's protestations ring hollow: his statement is explicitly "we don't have the money." But I think he could probably come up with it, and if he doesn't, then he needs to work with better financiers.

Finally, let's assume that Oracle really wants the rest of Sun, and considers carefully Monty's open statement that Sun is hemorrhaging $100MM of cash per month. Wouldn't it make sense for Oracle to actually just donate MySQL to pretty much anybody to make the EU issue go away? Oh, and lo and behold, Monty has two of those ready to go: Monty Program AB, and the Open Database Alliance.

To me, the current situation amounts to blackmail: we'll keep blocking your acquisition of Sun until you do what we want.

Consider The Sources

So let's look at the motivations of the major current players.

Stallman is irrelevant to any commercial discussion. His press release essentially says "I don't like the GPLv2 anymore, even though I wrote it, and it would be better if MySQL was under the GPLv3." Tough. Furthermore, RMS has no commercial experience of any kind. I fail to see how someone who has never even worked for a profitable commercial enterprise could be considered knowledgeable about how an acquisition would affect the marketplace in an anti-competitive way that harms consumers.

Furthermore, RMS' press release completely belies his previous positions regarding the possibilities for commercialization of GPL projects. He's stated in the past that offering dual licensing is only one of many ways that you can make money with the GPL being the dominant licensing model. Why all of a sudden does he believe that this is the only possibility for MySQL? Why is he so adamant that without that ability, there's no ability to derive commercial revenue from MySQL?

Monty has been slinging FUD about this acquisition for months. He was such a disruptive element inside Sun that they released him from his Non-Compete just to make him go away. Given that he's an extremely rich, disgruntled ex-employee and project founder, he has personal reasons and financial ones (under the Monty Program AB umbrella) to cause as much disruption to this deal as possible.

I've said it before and I'll say it again: Monty has been playing a long game here, and I think he'll be obstructive to any potential move that Oracle would make with MySQL until the IP is under his control.

Personal Opinions Should Not Drive Competition Policy

Ultimately, you can sum up the entire argument against the Oracle/Sun acqusition due to the MySQL situation as:
  • We don't like Oracle owning the MySQL IP
  • Therefore, don't let Oracle own the MySQL IP
Unfortunately, saying that you personally dislike something doesn't provide a valid reason to block an acquisition on competition grounds. Saying that you don't trust Oracle doesn't alter the marketplace in a way that disadvantages customers as a whole. Saying that nobody else could make money by selling commercial licenses for MySQL doesn't mean someone else must be allowed to.

The moment the MySQL founders, who have been handsomely rewarded, took VC money they turned MySQL from being a hobby project/company, and into a major technology company and an asset. The change happened years ago, it's just that they're only starting to wise up now.

But it's happened. It's done. It's no longer anybody's pet project; it's an asset that can and should be used by whomever is willing to pay the most for the IP. As a customer, under the GPLv2, you still have rights, including the right to fork. But don't go whining that a company that made massive amounts of money for its shareholders by commercializing a technology is no longer under your control.

Thursday, October 01, 2009

OpenGamma Is Hiring

People who have been following the blog for a while, or my twitter feed, will know that I've co-founded a startup called OpenGamma, that we've received a significant equity investment, that we're working in the financial services technology space, and that we're so stealth we don't even have a web site. And now I'll add another nugget of information into the mix: we're hiring.

Immediately, we have head count for one person, and I'm describing what I'd like for that one hire below. However, we're always open to finding someone good, so no matter when you find this post, if you think you might be a good fit for the company, send your CV over; if we can't hire you right away we'll let you know right away, but our headcount requirements are changing constantly. We might not be able to hire you right now, but if we think we might have room later on, we'll let you know.

About OpenGamma

Let's completely ignore whether you're a good candidate for the role that we're looking to fill. The #1 question you should be asking yourself, without even knowing what we do, is do I want to work with these guys? Hopefully this section can fill you in about who we are.

OpenGamma was founded by three people: your humble author, Elaine McLeod, and Jim Moores. Of the three, I'm the only without a PhD, and Elaine and Jim remind me of this on a regular basis. I'm the CTO and Acting CEO [1]; Elaine is our resident quant; Jim works on our core engine. We've known each other for years now, and bring a combined 30 years of technology experience (13 in finance) to the table.

The team has already grown beyond the three of us, and by the time you read this we'll have one more person working on-site and one more who's agreed to join. So you'd be coming on board when we're still quite small, but that's one of the major attractions for a lot of people. We believe that since you'll likely be spending more waking hours with us than your family or friends, it should be an enjoyable, social atmosphere (one of the earliest decisions we had to make was which local pub to adopt [2]).

We're well capitalized, having just closed a Series A equity investment by a major international Venture Capital group [3]. We have an office in the London Bridge area [4], from whence we can easily make lunchtime trips to Borough Market [5]. We have fast computers and big monitors and good (unfiltered) network connections and tons of Diet Coke and all the types of things that mean nobody's banging his or her head against "if only I had X, I could code better/faster."

What we don't have is meaningless policy, bureaucracy, procedures, politics, or other stupidity. We don't have a dress code. We don't have any paperwork other than expense claim forms and what we're required by law to fill out. We believe that the only policies we need are "do the right thing." We believe the only procedures that are worth having are the ones that help you, and us, execute better. We believe in hiring the best, enabling them to perform, and trusting them.

About You

I probably don't know you. But I can guess a lot about who you are, if you're going to be a good fit for the team at such a critical early stage in OpenGamma's life.

First of all, you know something about Front Office or Risk technology in capital markets. Ideally, you've worked for a broker-dealer or investment bank in your past, and might even be working there now. While this isn't critical for every hire we make, it is for this one. You get a thrill out of working under the constraints that modern trading software requires, and you like producing software not just for its own sake, but to solve business problems.

However, you're not 100% happy working for a Big Bank. You've beat your head against policies, procedures, internal audit, external audit, anything that stops you from doing your job effectively, one too many times. You're sure that it doesn't have to be that way. Maybe you started out trying to change things, thinking that one person could make a difference to the organization, and gave up. Maybe you just thought "that's how it's got to be" until you discovered the wide world of technology outside your bank. But you know that you don't want to work there anymore.

You like technology; it's not just a 9-5 job with you. You read tech blogs [6], user-contributed sites like Proggit, Hacker News, and DZone. You post answers on Stack Overflow. You research new technologies even though you know your employer would never let you use them, because you find them interesting. You go to events like CloudCamp, Pub-Sub, and BarCamp. You probably code in your spare time.

You live within a reasonable commute of Central London. While at some point in our future we'll be happy for full remote working, we're not there now. You might need reasonable accommodation for your personal or family situation (and we're not just legally obligated, but happy to do so), but you have to be able to make it to the office on a regular basis.

Most importantly, you've gotten this far and thought to yourself "Wow; this sounds great! Kirk, tell me more!"

About The Role

So about this role. I can't tell you that much about it without bypassing the Great Shield Of The Stealth Startup, but here's your buzzword bingo section:
  • You need to know Java. It doesn't have to be your preferred language, but you need to know it.
    • +1 if you came to Java from a C/C++ background.
    • +1 if you got sick of Java-the-language and started working with other JVM-based languages.
    • +1 if you've worked with low-level APIs like the post-1.5 concurrency libraries and the IO and NIO systems.
    • -5 if it's confined solely to web applications.
    • -10 if you've only ever used a traditional full J2EE stack.
  • You need to be conversant with modern software engineering, architecture, and methodologies.
    • +1 for every agile technique you like to use
    • +1 if you understand the difference between agile and Agile, and prefer the former
    • +1 for IoC and use of Dependency Injection containers like Spring
    • +1 for rigorous use of automated test suites
    • +1 for dedication to continuous integration
    • +1 for every functional language you can still code in (Scheme from your SICP days doesn't count here)
  • You need to be familiar with a standard Front Office/Risk Technology stack. By this I mean message oriented middleware, enterprise-class databases, distributed systems.
    • +1 for every Trading System you've worked with
    • +1 for every source of market data you've worked with (Reuters, Bloomberg, Exchange Feeds, etc.)
    • +1 if you've setup your own MOM infrastructure
    • -2 if you used XMPP rather than a real MOM infrastructure.
    • +5 if you've worked with AMQP
  • You need to have some familiarity with analytic libraries for financial models. We don't need you to be a quant, but you should have worked with pricing models in the past.
    • +1 for every chapter in Hull you've managed to get through (and +2 for every problem you managed to do successfully without resorting to the solution book)
    • +1 for every pricing model you've seized from the quants to improve performance or stability
    • +10 if you understand why Correlation is a rubbish analytic for pricing credit derivatives (even if you can't work through the math)
  • You might be familiar, even if you haven't used them in anger, with some advanced technologies not every bank has rolled out yet:
    • Clustering and Data Grid technologies like Terracotta or Coherence
    • Hardware accelerated computation like GPGPUs and FPGA-based acceleration cards
    • Shared-memory or shared-flash clustering systems like Violin
    • Key-Value stores like Project Voldemort and CouchDB


Here's what OpenGamma can offer you:
  • The chance to work for an organization designed to allow you to produce great technology.
  • No bureaucracy.
  • An equity stake in the business.
  • The chance to learn about working for an early-stage startup first-hand.
  • A tech team filled with the smartest, best people we can get. In fact, if we can't find the right person, the role stays unfilled.

However, there are some things OpenGamma can't offer you:

  • Job Security. We're well capitalized, but we're still a startup. If you want/need a guaranteed job for several years, we're not for you.
  • The Chance To Hide. What you work on is going to be seen by everybody else in the firm on a daily basis. Moreover, you're going to interact with customers. You're going to interact with the public at large. No room for shrinking violets, and no room to coast on your teammates, I'm afraid.
  • Hedge Fund Cash Compensation. I believe in fair compensation for excellent staff; London is an expensive city, and if you're good enough to work for OpenGamma, your fixed expenses have largely grown commensurate with your banking salary income. We can offer good compensation, but we can't and won't match what you'll earn in a good year for an Investment Bank or a Hedge Fund: your equity stake is to make you whole over your employment at OpenGamma, and to make sure that everybody has aligned interests.

How To Apply

If you still think we might be a good match, send an email to jobs at opengamma dot com with your CV, and a description of why you think you're a good fit for OpenGamma, and we're a good fit for you.

Note: Recruiters/Agencies Not Welcome To Apply. We're not working with recruiters as of yet, so please don't contact us with a story about a candidate whose "profile" you're "exclusively representing." Any agency-provided CVs will be deleted unopened and unread.

Update 2009-10-28: We are now working with external recruiters, so if you're a candidate who just read that after a recruiter contacted you, ask to see the job specification. If they have it, they're definitely working for us.

[1]: I'm CEO during this, the most formative period of OpenGamma's lifetime; I'm not wedded to the role, and it will become quite clear to myself and our board of directors when it's time to bring in a permanent CEO. Until then, I'm it.
[2]: The Trinity. Best one around, even though it's farther from London Bridge station than the office.
[3]: Yes, I'm coy about who it is. No, I'm not going to tell you unless you get to the phone discussion stage. Yes, we have valid reasons for that.
[4]: If Old Street is Silicon Roundabout, can we be Silicon Bridge or something?
[5]: Shame that it's 1/3 shut down these days...
[6]: Like this fine screed. That guy's really good, if a little opinionated for my liking. He should calm down a little and go back to writing about tech stuff rather than politics and personal attacks on industry figures.