In recent days I’ve been thinking of JavaOne, as we kicked it around and decided we just couldn’t send speakers; and of Oracle OpenWorld, to which JavaOne will now serve as an appendage. It reminded me of a conversation I had last year about Oracle.
[Update: I reported this conversation, which I thought was instructive, carefully avoiding any conclusions. The commenters on the piece, however, drew lots of conclusions, which I enjoyed reading, and you might too. In particular, I thought some of the guesses as to my un-shared opinion on all this were quite illuminating.]
The conversation involved myself and a person with a convincing title who, as they’d say in the paper, was “familiar with the situation”.
My question was: “OpenWorld is this totally all-about-business conference. The Oracle Develop meeting is just a second-rate sidebar. Where does Oracle go about building developer mindshare?”
I’ll try to reproduce the answer in full as best as I can remember it:
“You don’t get it. The central relationship between Oracle and its customers is a business relationship, between an Oracle business expert and a customer business leader. The issues that come up in their conversations are business issues.
“The concerns of developers are just not material at the level of that conversation; in fact, they’re apt to be dangerous distractions. ‘Developer mindshare’... what’s that, and why would Oracle care?”
Comment feed for ongoing:
From: Eric (Aug 31 2010, at 16:28)
This is the attitude at most business-as-the-business companies, I would wager, based on my experience: the developer is just a technician that makes things work.
[link]
From: JimDesu (Aug 31 2010, at 16:47)
As a former Oracle employee, I assure you that developers are the last on Oracle's mind, except in so far as the marketing folks might find a short-term competitive advantage in some feature-ette. They've never understood the benefits of making things easy for developers, nor do their incessant internal turf-wars allow those who do get it to influence the status quo. Developer productivity & mindshare are completely unimportant to Oracle's modus operandi, and we can expect the same "good things" to happen with Java that we've already seen with Oracle's other [abominable] tool-sets.
[link]
From: John Cowan (Aug 31 2010, at 16:54)
Quite so. People use Oracle databases because they're told to.
[link]
From: Nando Sola (Aug 31 2010, at 16:58)
In Spain, we'd say that this is "throwing a stone and hiding the hand". In other words, I have no means of believing this source works for Oracle, or is maybe a disgruntled ex-Sun. Please, just either stop the FUD or share all the facts. No half measures.
[link]
From: Ric (Aug 31 2010, at 18:13)
I have been known to say in a number of forums that Oracle is NOT a technology company, it's a sales company. If there were better margins in used cars (and the sort of free money they get for maintenance contracts) they'd be out of software in a heartbeat.
[link]
From: Pete (Aug 31 2010, at 18:15)
Remind me again about how the Schwarzian policy at Sun worked in bending over backwards to build communities and gain developer mindshare. Yeah, millions of developers were attracted to download and use Java, Glassfish, OpenOffice, and VirtualBox but none of them ever spent a penny in doing so. All that handwaving about monetization of the "community" never materialized, did it? Why bash a company like Oracle for being honest in admitting how they make money?
[link]
From: Andy O (Aug 31 2010, at 18:33)
So, I'm a developer who has been dealing w/ Oracle for the last 12 months...in anger. But after reading this, I thought, what if this guy (and Oracle by extension) is right? I mean, Oracle is an even more powerful company now. What if developer mindshare is overrated?
Please someone help me come to my senses!
[link]
From: Parveen Kaler (Aug 31 2010, at 18:48)
How is this a bad thing? It's all about building the best applications for your customers.
In the enterprise space the customer is other businesses. Apple does essentially the same thing in the consumer space. The user is more important than the developer.
This is a good thing.
Imagine if airlines treated their relationship with the flier as the most important. Imagine if politicians treated their relationship with constituents as most important.
[link]
From: Rob Heittman (Aug 31 2010, at 19:22)
Well, it's hardly FUD when you could ask any Oracle executive and they would say exactly the same. Oracle puts business before all else. It's not a bad thing.
But there are other metrics than financial success. Joy. Fun. Ease. Love. These are not words in Oracle's corporate vocabulary. None of them make obvious and direct money. And, really, this is why Oracle does so well, despite a contempt for mindshare. There is a lot of room to make money with that kind of thinking. Go, Oracle, go.
Fortunately, as developers, we each get to vote with our feet, as it were, and opt for values that might contribute less predictably to our bottom line, and more to the thrill of creation, and the opportunity to do something really great. This is where innovation and change come from. This experimental frontier is where we get GMails and Waves; crazy ideas that work and crazy ideas that fail.
There's room for both Google attitudes and Oracle attitudes in the world. I do respect both, though I know which house I sort into.
I don't really care that Java One is becoming a developer bolt-on to a enterprise conference, when there's a also Google I/O to go to, that has a bolt-on Enterprise event or two :-)
[link]
From: Bryce Kerley (Aug 31 2010, at 19:27)
> How is this a bad thing? It's all about building the best applications for your customers.
It's not even that (I'm sure lots of Oracle customers would like to spend less in developer and operations), it's about building a business that will make shareholders rich. Building applications that customers will continue to buy just happens to be a great way to do that.
[link]
From: Andrew (Aug 31 2010, at 20:30)
As some of your other commenters have pointed out, why is this a problem? Clearly it works.
Maybe what Oracle developers do isn't as "cool" as all you hipsters building pull my finger apps but they are just as important to the world's overall productivity. Plus just because the central relationship is based around financial metrics like ROI and TCO doesn't mean that there isn't a developer relationship in addition to the business relationship.
And no, I'm not an Oracle developer and neither do I want to be one (I do embedded firmware on products you've never heard of). I can however see how such an engineer can have a happy, productive professional life without going to conferences that launch t-shirts into the crowd.
[link]
From: JulesLt (Sep 01 2010, at 01:04)
John - Oracle didn't get to be where they are without having a database product that was once recommended by software engineers (over jokes like Access, DBase, etc, and against the serious competition of DB2, Ingres).
And I'd agree with some of the comments about 'developer mindshare' - the fact that the self-styled 'development community' has no interest in solving the business problems of logistics, manufacturing, insurance, banking and retail, is exactly WHY companies like SAP and Oracle make the money they do.
The fact that all the cool people want to be 37 Signals is part of the problem. The other part is that the cool people also presume they are the only clever people, which is where they start to get patronising. I'm quite sure that Hugh Darwen is far more intelligent than David Heinemeier Hansson for starters - but it seems that solving hard problems is out of fashion.
[link]
From: Paul Morriss (Sep 01 2010, at 02:11)
I'd be interested in hearing what you said in answer to the last question, Tim.
[link]
From: Daniel Sobral (Sep 01 2010, at 05:20)
So Oracle makes a ton of money because "the cool people" (by which I presume the author meant developers who care enough about their profession to be politically engaged about it) don't care about hard problems?
Funny. I thought the cool people created MySQL and Postgres, not to mention dozens of NoSQL products, because they couldn't stand Oracle products.
[link]
From: Hynek (Pichi) Vychodil (Sep 01 2010, at 05:51)
Isaac Newton knew that he stays on the shoulders of giants. Oracle seems doesn't.
[link]
From: Tony Fisk (Sep 01 2010, at 06:34)
Th dry content of this conversation reminds me of a similar exchange between Robot Monster and his glorious leader (with a few name changes for relevance):
Ra-Man: I cannot - yet I must. How do you calculate that? At what point on the graph do "must" and "cannot" meet? Yet I must - but I cannot!
Great Oracle: Earth Ra-Man, you violate the laws of plans. To think for yourself is to be like the hu-man.
Ra-Man: Yes! To be like the hu-man! To laugh! Feel! Want! Why are these things not in the plan?
Great Oracle: You are an extension of the Ra-Man, and a Ra-Man you will remain. Now, I set you into motion. One: destroy the girl. Two: destroy the family. Fail, and I will destroy you!
[link]
From: Dave Rodenbaugh (Sep 01 2010, at 08:14)
Hi Tim,
Love the quote...I was about to post my own Oracle story when I saw your post and I had to add the quote because it's so relevant:
http://www.lessonsoffailure.com/companies/how-i-learned-to-love-new-evil-empire-oracle/
Because Oracle doesn't get it, I'm hoping for some great new stuff to come of it...
[link]
From: SteveP (Sep 01 2010, at 10:07)
This seems accurate based on my experience with the company and interaction with Oracle employees.
A business is a business, so I don't see anything wrong with this approach if it is successful for them. As a developer though, you have control over what skills you want to invest in and this can have a significant effect on your career. Understanding how important (or not) the development community is to a technology producing company is an important part of deciding whether you want to bank your career on them.
[link]
From: JulesLt (Sep 01 2010, at 12:05)
It's nothing to do with 'hard problems' - there is little that is conceptually difficult in most 'enterprise' application.
But they are 'big problems' that don't motivate/interest 'cool' people; in contrast mySQL or postgres are actually technically interesting projects - far more so than building a stock control system using mySQL.
There are relatively few successful open source projects operating at that next level (industry verticals, rather than technical components) - even when it would make huge amounts of sense for the industries in question.
As for being 'politically engaged with your profession' - the question is to what degree you let your politics affect your professionalism. From a professional point of view you should be putting the needs of your users first - and as we know that can mean dealing with things (Windows, IE, etc) that we'd rather avoid.
(And we shouldn't blame our users - we all almost certainly have an ethical blind spot in our purchasing habits)
[link]
From: vlad (Sep 02 2010, at 16:33)
well, history has decided it:
- Oracle : business for the reasons of business
- Sun : art for the reasons of art
Absence of the driving force of a business behind R&D at Sun allowed for the proliferation of parasitism and BS beyond body survival threshold (for example Sun's Six Sigma - that was a crossbow bolt right into the liver) As result Sun's engineering was failing more and more despite massive investment (which thus was more like waste actually)
Good bye, Sun. The once great company which was wasted by Schwarz, Yen and their groupies.
[link]
From: AnthonyG (Sep 03 2010, at 07:15)
So i call BS - making up fragments of conversations and attributing it to a philosophy of an entire multi-billion dollar company that employes 50,000+ people is - well absurd. At best it is one persons opinion of a methodology probably espoused one night after a few beers - at worst it is complete fabrication. Statiing it is official policy ... ludicrous.
Checkout oracle technet - always seems helpful to me. Full community involvement, free downloads of pretty much anything, free tools, good articles.
but hey bash away...
[link]
From: R. Hamilton (Sep 03 2010, at 08:40)
Oracle's primary interest is in a relationship with the money of big-spending entities; small businesses and independent developers don't seem to offer enough return on investment for them; witness how all their OS maintenance contracts are premium-level only.
As much as some Linux or BSD advocates criticized Sun's relationship with open source, compared to Oracle, they never had it so good as with Sun. RIP, you'll be missed...which is not to say that those engineers and programmers that haven't left aren't great folks; but Larry, while he may have business acumen that Schwartz lacked, doesn't have a _clue_ about building communities.
[link]
From: len (Sep 03 2010, at 12:17)
When you manage processes on both developer and business sides for a contract the customer signed and will litigate if not satisfied including bits they expect but did not contract for then you understand Alec Guinness as Faisal and why getting Lawrence out of the Middle East was so important to all.
If a developer becomes a self-obsessed ideologue, they are a danger to everyone. Like it or not technology is just stuff.
[link]
From: BigRed (Sep 03 2010, at 15:44)
Sad to say that after spending so much time and effort on its Java community, Sun never really figured out how to monetize Java. It sort of supports the argument that Sun was pushing Java as a loss leader to sell more hardware.
While Oracle doesn't have the strongest or most loyal developer community, it is one of the most profitable software companies on earth. One might conclude that Sun's lack of business relationship focus led to its demise.
The cruel fact of the business world is that Oracle had the right model and Sun didn't. While communities are very important to open source projects, they are not as important to businesses, entities which are focused on the bottom line, not the "l33tn3ss" of their developers.
[link]
From: Robert Young (Sep 03 2010, at 21:25)
The analog to Oracle: Microsoft. In both cases: a competitor failed to execute in the beginning, ceded the playing field, and the winner became a "relationship" organization rather than a productive one.
In the case of Microsoft, it was DR which ignored IBM, giving M$ its opening. The software was acquired "illegally" from Seattle Computer. Then Lotus wrote 1-2-3 only for PC/DOS. The lock-in was complete.
In the case of Oracle, it was IBM which ignored the work of its own Dr. Codd. Oracle produced the first commercial RDBMS. Game over, the lock-in was complete; well, off the mainframe anyway. I've argued that the Sun acquisition was done in order for Oracle to build a stack which can compete to replace the IBM mainframe installations. We'll see.
IBM is the archetype of "relationship" company; has been since T. Watson, Sr. They've been been shucking off anything related to actually producing anything. Oracle continues to produce new releases of the database, but has been aggressive in buying up a portfolio of applications, often to the detriment of the clients of those applications.
It's the way of American capital to exit productive activities and into financial activities as soon and as fast as they can. The collapse of the economy is directly a consequence. For those interested, read up on the course of Uruguay in the 60's and 70's; their capitalists determined that being the "Switzerland" of South America was the proper route; deindustrialization just as American capitalists have done over the last two decades. What they got was urban guerillas and civil war. So will we.
[link]
From: Carl (Sep 04 2010, at 06:23)
The methodology used is similar to that of a classical Software Corporate Raider driven by sheer monetary interest only, its value for the software industry as a whole is questionable. This methodology was pioneered in the 80s and 90s by the likes of CA; see also Sanjay Kumar (for starters...).
Process is very simple and obvious: a software company is acquired, then all costs are brutally cut leading to massive exodus of top personnel, then that operation is put into harvesting mode. Harvesting mode in Software business means collecting exorbitant licensing and maintenance premiums while investing NOTHING. Often maintenance costs are at premiums and no actual maintenance gets done, this can easily be observed by prolonged release cycles that produce practically no kind of technical innovation whatsoever, simplest of obvious bugs are not fixed even after years. Some marketing yarn is then spun around the whole operation for stealth reasons to fool the unknowing.
Because many top corporations are more or less locked into their technology they have no way to escape and get off the carousel from hell such bestowed upon them. They can only pay up. While this may be a good model for finance and other “institutional investors” (hedge funds and the like) these practices are extremely damaging to the software industry as a whole. It ultimately leads to severe loss of credibility of the industry and other long-term detrimental effects in the market. It also leads to the rise of patent trolling.
Maybe I am being a bit hard but it sure looks that way... the extreme narcissistic personal styles of some leaders in the industry certainly do not help things.
[link]
From: Paul Downey (Sep 05 2010, at 04:04)
I guess if your business is Middleware Hell, there's probably more money selling instruments of torture to the devils than solace to the inmates.
[link]
From: Jay Gischer (Sep 07 2010, at 09:20)
* The behavior @Carl describes is what I would call "rent-seeking". I hate rent-seeking.
* I don't think that it necessarily follows that a company with a bottom-line focus must needs ignore developers. Microsoft doesn't. However, the success of their platform and of their business rested upon having a rich ecology of applications. Nobody gets a computer to play with the operating system. But they do buy computers to run their databases, and they don't really need that rich ecology.
* Sun's demise has little to do with Schwarz or management decisions and everything to do with datacenters filled with racks and racks of cheap intel boxes, running one O/S or another.
As an employee of Silicon Graphics we had an executive give us a speech about the WinTel threat. What he said was the joke about the guys running from the bear: "we don't have to run faster than the bear, we just have to run faster than you [our competitors]" I think we all know how that turned out.
Sun was successful at running faster than the other companies running from the bear. But they didn't run faster than the bear itself.
[link]
From: Robert Young (Sep 07 2010, at 16:49)
-- “The concerns of developers are just not material at the level of that conversation; in fact, they’re apt to be dangerous distractions. ‘Developer mindshare’... what’s that, and why would Oracle care?”
It's worth recalling that Oracle has been on a buying spree for the better part of a decade; buying apps, not tools. Theur intent, clearly, is to acquire (and hold the developers of) all applications that matter. To the extent that Oracle cares about employees, it cares about developers. Developers not of Oracle are, to a greater or lesser extent, The Enemy.
[link]