James Governor's Monkchips

SOA flatlines: BRAINS!

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Everything is dead.

Its something I have known for a long while. When I came into the business in 1995 as a junior reporter I got the mainframe beat because “MVS is dead”. I got the IBM beat because “it used to be important”. 13 years later and that frigging mainframe still ain’t dead – well certainly not if death is a function of revenues. IBM had a bumper year last year.

Windows- Windows is obviously dead. MacOS and Linux have killed it stone dead. Yeah 80%+ market share is a terrible place to be. And don’t even get me started on IE. Because it. is. dead.

Unix is dead. Damn straight. Oh yeah don’t forget the relational database. Dead. Cobol – totally dead. Hundreds of millions of lines of new code each year. That’s dead right? Dead, dead, dead.

Everything is dead.

Roland Barthes has LOT to answer for…

The latest technology to bite the dust is SOA. Totally dead. IBM, Oracle and SAP haven’t committed anything to SOA. Neither have their customers. Lets chuck that shit out. Right… Lowering ERP integration costs by an order of magnitude- why would we want something like that???

Honestly – I don’t need to defend SOA. Anything that has garnered that level of investment just isn’t dead. RedMonk has long since aligned ourselves with the REST development and integration style. We saw developers getting on with it while SOA sales cycles for architects and CIOs were taking 18 months or more. But that is not to deny SOA’s value. SOA is about Enterprise Architecture, which has a value. Architecture, Discipline, Freedom.

It turns out that some of the Web Services standards so beloved of a certain constituency that wanted to “do Corba right” are indeed flatlining, but SOA is actually in bloody good health.

The Web is SOA. Put things in, Get Things Out, Post them or maybe even Delete them.

So the dead stuff runs the Enterprise, and there is no money there of course… Because the enterprise is dead. 99% of enterprise transactions run across dead platforms. 90% of Offices still run on dead software. Its all dead.

Dead dead dead. BRAINS….

40 comments

  1. So, basically, “dead” is something to aspire to. Excuse me, I have to go adjust my plans a bit…

  2. In other words $OA is dead, but the original concept lives on.

    I’m happy to see the money making (i.e. marketing) side of a concept pass on… it means that we can get back down to doing the work and making the solution work better as opposed to worrying about what the marketing brochure says were supposed to be doing.

  3. SOA is dead, long live services!!

    Yah, please sir can I create the acronym this time (after all the HaaS/FaaS/SaaS/IaaS/PaaS/DaaS and other *aaS debates).

    I suggest we call the new paradigm an architecture oriented around services (whether it’s RESTian or some other approach). The key thing we want to promote is the idea that it’s services and provides all the benefits of agility through componentisation of lower order subsystems and processes.

    A Service Built Orientation (ASBO)?
    Architecture of Standard Services (ASS)?
    System Architecture of Aggregated Services (SaaS)?
    Holistic Enterprise of Architectured Processes (HEAP)?

    Hold on, …. I’ve got it … how about a Service Oriented Architecture? (http://blog.gardeviance.org/2008/04/more-oa-with-your-aas.html)

    P.S. Paper’s dead too and everyone is going to use e-readers.

    P.P.S I’m bored of the world Cloud, can’t we call it Floating Vapour or something else instead?

    Great post James.

  4. Brilliant! That is all.

    -Jesse Robbins

  5. I’d also like to be the first to declare Mashups, SaaS, Apple and Cloud Computing as dead. In 18 months, I want to point to this comment to prove that I was right.

    Good post James. 🙂

  6. There’s a false dichotomy that many in the IT world succumb to: whatever is new and cool quickly and automatically invalidates what was old. Only one paradigm can exist at a time. “Everything will run in the cloud.” “SOA is dead.” “REST rules.” Etc. As you correctly point out, technologies overlap, sometimes for a very long time. Of course, the backlash against SOA now somewhat equates with the hyperbolic promises made about SOA in the past. Many SOA boosters have succumbed to this same fadism.

  7. The term zombies just took on multiple meanings here. Excellent post.

  8. Extremely dead: OpenVMS. And whilst we’re at it, PDP-11s are *totally* dead. As for PDP-8, that architecture is _fossilised_.

    Well, that is if you ignore the fact that many/some businesses still depend on these platforms. Yes, there are still PDP-8s (born 1965) in production.

    Define dead.

    Not cool/trendy/enterprise wet dream du jour != dead.

    Is Cloud Computing dead, yet?

  9. […] James Governor’s Monkchips » SOA flatlines: BRAINS!So the dead stuff runs the Enterprise, and there is no money there of course… Because the enterprise is dead. 99% of enterprise transactions run across dead platforms. 90% of Offices still run on dead software. Its all dead. […]

  10. @Andrew Back : Yep, cloud computing is so last year, just like utility computing is so 1961 (see John McCarthy, MIT). The future is all about Floating Architectural Vapours and what’s your FAV?

    Hopefully 2009 will be the year that smoking on the crack pipe of computer terminology will become a punishable offence.

  11. This, sir, is a brilliant article.

    Our industry feels very involved in a media-like need for innovation. It’s new, it’s the future, I’ll invest in/learn/supply it.

    Not that Innovation isn’t good; my job is in innovative fields. But we are building on previous innovations, still using what works. In fact, a huge principle is reuse, right?

    So, is the need for constant novelty to do with a mentality? A method of marketting (blogs, memes, and folksonomies)? Or, is it just that we pay too much attention to our own echochambers?

    @Simon Wardley: Brilliant, sir.

  12. […] A quick fire response came on a number of fronts, like this one from Duane Nickull at Adobe, and then this from James Governor at Redmonk, where he charismatically claims, “everything is […]

  13. The best hourly rate that I ever made as an independent consultant was $300/hour in 1983 on “obsolete” technology. This really was obsolete – 1410 autocoder running under emulation on an IBM 370.

    I resolved there and then either to work on the very trailing edge of technology – or the very leading edge. The leading edge is more fun, but the trailing edge often pays better.

  14. How can something be dead that was never alive. A soluton concept that was overmarketed in an attempt to get customers to pay for moving in the direction that the investment was made. (aka IBM style)
    Try to tell a Bank or Insurance company that providing data for a consumer to make a decision is called “cheese” and they will be off to the cheese shop.
    The real question in cloud services is why I can’t watch the eagles beat the giants on my PC and be a potato with no decision to be made other than second guessing the coach?
    BR/MCR
    This comment was originally posted on Adventures in systems land

  15. […] Books, White papers etc. « Is SOA dead? […]
    This comment was originally posted on Adventures in systems land

  16. Bloody well written. 🙂

  17. […] they are all on their own. [UPDATE: HP’s Tim Hall has some reactions to the SOA debate. Dell’s Mark Cathcart also has SOA thoughts.] Instead, IT should be seen as a set of problems to be solved by the best means, and common means […]
    This comment was originally posted on Adventures in systems land

  18. @nathan gilliatt – Aspire To be Dead. sounds like a good goal for 2009.

    @paul comeau ++

    @swardley the perils of working in a fashion industry

    @jesse robbins thanks!

    @ian skerrett – Cloud is definitely DOA. (dead on arrival or a Dead Oriented Architecture)

    @john bennett at redmonk we have long railed against binary thinking http://redmonk.com/sogrady/2004/11/03/the-trouble-with-binary-arguments/

    @andrew back i remember writing a lovely story about the ongoing use the old ICL George 4 platform…

    @zach beauvais its all about the hemline, baby

    @chrisbird – good point. work on “dead” tech = well paid

    @cathcam I love you man. You’re a great reason to blog more, because you write awesome riffs on the subjects i blog on!

  19. […] great read in this general category comes from the insightful James Governor’s assertion that 99% of everything in software is “dead&#8221…, a rant I had heard in person but had not seen as a blog post… until […]

  20. […] RSS awhile ago, but Marshall is onto something when he recently threw the zombie dust on it (the usual retort applies); and I’m a big nut for “enterprise RSS” from way […]

  21. […] From James Governor (Monchips) “So the dead stuff runs the Enterprise, and there is no money there of course… Because the enterprise is dead. 99% of enterprise transactions run across dead platforms. 90% of Offices still run on dead software. Its all dead.” […]

  22. Old IT never dies. IBM customers are still paying maintenance on punch card readers.

    There is also a great deal that few people under the age of 35 (everyone now in the tech and investing world?) understand about technology like MVS, RelationalDB and Windows.

    One element is support. If you are running a big company and the system gets a hiccup at 3am posting transactions for the day, you can’t just put a fail whale up on the doors of the company at 9am the next day.

    Another is training. The single biggest cost many companies have is the sunk cost in training thousands of (in some cases very busy and highly paid) staff. The key factor is being able to use the technology, not how “great” or “cool” it is.

    Doesn’t mean the guys at Microsoft are smart, just installed.

  23. […] I am not of the opinion that Service Oriented Architecture (SOA) is dead, clearly as an industry we got carried away with the attempt to establish a specific, common set of […]

  24. The problem with SOA, pretty much from the getgo, was that it was pushed by the anti-RDBMS folk, largely from the java/web crowd. What they tried to do was: 1) reinvent the era of siloed COBOL programs with siloed data, albeit with a lower case syntax, and thus 2) undertake to implement ACID in each and every application, just like was done in the 1960’s. Dr. Codd knew better, which ticked off the IMS folk, and was rewarded with Chamberlin writing SQL. SOA, xml, java, and such were reactionary attempts by coders to establish a coders’ permanent employment regime.
    SOA was all about putting GUI lipstick on 30 or even 40 year old mainframe applications; at least in the Enterprise systems I worked on since 2001. It was not, nor could it ever be, a solution to any real problem. The Emperor’s New Clothes critiques began to appear very soon, just not from the vendors who wanted to profit.
    Those who believe that control of transactions belongs anywhere other than the database are just wrong. Full stop. Those that try it will either fail miserably or get sued, or both.
    For simple read-only queries, MySql is just fine. One needs only a sql parser fronting the file system, and MySql was designed to do that. IBM may well get Sun and replace all that DB2 infrastructure with a database called DB2, but just re-labeled MySql. Poetic justice that would be.
    This comment was originally posted on Adventures in systems land

  25. Apparently Biggie Smalls was right: you’re nobody until somebody (in tech journalism) ‘kills’ you.

    For my part I am sorry rumors of COBOL and Microsoft’s death have been apparently greatly exaggerated.

  26. […] ComputerWorld’s article on the anti database movement reminded me of James Governor’s “SOA flatlines: BRAIN!” post.  I always love going back to James’ post when someone proclaims that Y is so much […]

  27. […] article on the anti-database movement reminded me of James Governor’s “SOA flatlines: BRAIN!” post. I always love going back to James’ post when someone proclaims that Y is so […]

  28. […] article on the anti-database movement reminded me of James Governor’s “SOA flatlines: BRAIN!” post. I always love going back to James’ post when someone proclaims that Y is so […]

  29. i love this post

    so true

  30. […] of most Enterprise IT is that “dead” software actually lives a long time. In fact dead software (nice post by James Governor) runs 90% of the economy. Another word for “legacy” is IT projects that worked. The word for IT projects that […]

  31. […] of most Enterprise IT is that “dead” software actually lives a long time. In fact dead software (nice post by James Governor) runs 90% of the economy. Another word for “legacy” is IT projects that worked. The word for IT projects that […]

  32. […] Rumors of the death of Enterprise Software have been greatly exaggerated (nice post by David Hornik). The thing people find hard to understand about the longevity of most Enterprise IT is that “dead” software actually lives a long time. In fact dead software (nice post by James Governor) runs 90% of the economy. […]

  33. […] Rumors of the death of Enterprise Software have been greatly exaggerated (nice post by David Hornik). The thing people find hard to understand about the longevity of most Enterprise IT is that “dead” software actually lives a long time. In fact dead software (nice post by James Governor) runs 90% of the economy. […]

  34. […] some have variously said SOA is Dead, or we need to rethink everything and move to Web Oriented Architecture… IBM had just […]

  35. […] of most Enterprise IT is that “dead” software actually lives a long time. In fact dead software (nice post by James Governor) runs 90% of the economy. Another word for “legacy” is IT projects that worked. The word for IT projects that […]

  36. […] great read in this general category comes from the insightful James Governor’s assertion that 99% of everything in software is “dead&#8221…, a rant I had heard in person but had not seen as a blog post… until […]

  37. […] card Hugh made for me speaks my language. After all, Everything is Dead. But the conversation that the card really kicks off in my mind is […]

  38. […] I have argued before, in IT everything is dead. That is- if its in production, its “dead”. When a commentator says Technology A is dead, they generally just mean its not an Apple or Google […]

  39. […] I have argued before, in IT everything is dead. That is- if its in production, its “dead”. When a commentator says Technology A is dead, they generally just mean its not an Apple or Google […]

Leave a Reply

Your email address will not be published. Required fields are marked *