James Governor's Monkchips

IDE as a service: DaaS hawt! and some Enterprise OpenSocial thawt

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

I was chatting to Atlassian founder and CEO Mike Cannon-Brookes recently when I asked him what comes next. You know… the “plastics” question. It was Mike that first turned me on to OpenSocial’s potential as an enterprise technology – as far as I know Atlassian was the first enterprise vendor to adopt the technology, to allow smooth flow across different elements of its product Suite: you can pop in a context from another app into a smooth user flow. Today IBM, SAP and Jive are onboard (OpenSocial apparently made a bit of a splash at IBM Innovate recently – given how many tools are in the Rational portfolio its no surprise IBM would see the advantage of using the technology). Lotus is also playing with OpenSocial containers.

OpenSocial has even hit the desktop. Xobni, the Outlook integration specialist, now supports OpenSocial, which means web apps can be easily integrated with Outlook. Using Xobni Atlassian now gives users to access Jira issue and project tracking via Outlook. Xobi is also supporting DropBox, Evernote, Salesforce.com and Webex.

As ever I’m madly off topic in the intro. Mike said to me the next Big Thing would be online IDEs. Given Everything as a Service its no surprise IDEs would come up (indeed, the idea of online IDEs has been kicked around, then kicked into touch, a bunch of times.)

But Mike just said:

We’re watching a lot of the emerging amazing tools. People say “an IDE will never go to the web”…. Well people say tools will never go to the web and they have been proved wrong every single time.

He has a point. Web performance is frankly staggering these days. Javascript is super fast in modern web browsers like Google Chrome. So why not an online IDE? Everything else is moving to the web. It was no surprise then when I got a press release this week about Atlassian and Accel Partners investing $5.5 million in a series A funding for Cloud9IDE – a web-based IDE. It currently only supports Javascript and HTML/CSS, with highlighting support for Coffeescript, Ruby, PHP and others. The investment should mean more language support sooner rather than later.

Code9 isn’t alone. Our client the Eclipse Foundation now has an IDE project in play called Orion, led by IBM (client). “Tools For The Web, On The Web”. The big concern people have about online IDEs is that desktop apps should in theory perform faster. On the other hand what’s going to parse Javascript faster than a superfast web browser? In our first briefing about Orion, one thing that really struck me was that the developers told me that for some large files at least, Orion is faster than the Eclipse IDE. I am sure some haters are now thinking this is about Eclipse performance, but that really isn’t the issue at hand.

Bespin is a Mozilla experiment to build an online IDE. Ecco is another.

What is DaaS? Development As A Service – which really goes beyond IDE as a service. You see developer toolchains are changing dramatically. Where IDEs traditionally needed to integrate with heavyweight systems like Maven, the game today is all about Github and distributed version control systems enabling forking and rapid developer-led innovation. Developer services are moving online, and so are the developer tools.

Its early days for the trend, but I think we’ll see major development in the next 12-24 months. I am sure some of you are thinking “an online IDE could never do x”. But as Mike said, we’ve said that about the web before.

Final point before I go. In terms of immediate developer experience the Eclipse guys seem to have the edge over Code9. Orion’s web site is beautiful and you don’t even need to register to start using the editor. Its always great when clients do impressive work.


  1. Web-based IDE efforts will surely keep grabbing headlines and driving additional convergence, but here’s a quick reality check on the topic, especially as it pertains to enterprise developers: http://tasktop.com/blog/eclipse/prediction-7-web-ide-pipe-dream

    1. @mikkerten – “especially as it pertains to enterprise developers”. last refuge of the scoundrel ๐Ÿ˜‰

  2. James,

    Just so you know Bespin is no longer an active project. The code code merged into ACE, which is the editor used in Cloud9.

    You also missed the eXo Platform IDE http://www.cloud-ide.com/.

    btw, thanks for the nice comments about our web site. In the old days we were known for our ugly web site. ๐Ÿ™‚

    1. ian- i knew bespin was orphaned- i should have made that clear. perils of posting last thing on a friday when the only thing between family time and a glass of wine is a post.

  3. Orion is a great initiative! I am still wowed by the performances and the concept. However I really look forward to see how it can go beyond Javascript, CSS and web only code. I don’t think Java code needs to stay always on the desktop, let’s be ambitious!

    Also, I can’t help mentioning Nuxeo Studio http://www.nuxeo.com/en/products/studio/ Made specifically for the Nuxeo Platform, but definitely an online environment for developing Apps, in the cloud.

  4. James,

    Great to read you truly believe in the DaaS space. Our website is available at http://cloud9ide.com – saw the link in your article didn’t work. If you ever want to go more in-depth on this topic let me know. Love to provide you with insights on this exciting new way applications are built.

    CEO, Cloud9 IDE, Inc

  5. It is probably worth mentioning Compilr ( http://www.compilr.com ) which might be the most popular online IDE with 50k users.

  6. FWIW, IDEs follow the apps they are used to develop. So as apps move into the cloud, so will IDEs.

    1. dirk – point well made. i expect a coexistence, for sure. but some types of app, and environment, will certainly be natural for daas.

  7. I suppose such things will get press/air play but will be ignored by those who have come accustomed to the I in IDE not standing for Integrated but Intelligent aka JetBrains IntelliJ.

  8. Disclosure: I work at IBM, helped shape the original Orion proposal, and continue to support Orion.

    James: I enjoyed your blog entry here and share your enthusiasm on the topic. I also enjoyed Mik’s blog entry though I have a bit of a different feeling about the prospects for “web IDEs”.

    Mik said: “The current generation of browser-based JavaScript runtimes is not up to the task of running the IDE platform as we know it”. I agree with this but believe it’s a moot point.

    Please excuse me while I ramble a bit to try to explain.

    Early last year, quite a few months before the Orion proposal at Eclipse, I invited my buddy Rafe Colburn out for burgers to catch up and get his feedback on what would become Orion. I tried to explain to him that the goal was not to port Eclipse *as we know it* (same words as Mik) to the browser – the goal was to take a step back and think about how to help with the problems that an IDE helps with but assuming the browser and the web as the platform rather than a Java/OSGi runtime on a native desktop OS. My explanation to Rafe was about as articulate as that last sentence (i.e. not very) but Rafe is a smart guy and got it immediately. He even suggested a helpful analogy:

    “Sort of like how Gmail reimagined email rather than just recreating Outlook in the browser”

    To me the goal of these web tools should not be to replace existing tools, but rather to help do things that we can’t do well with traditional tools. People are going to be creating and maintaining Big Java Applications and Big .NET Applications for a long time. Eclipse and Visual Studio will continue to be excellent tools to support the creation and maintenance of those sorts of applications. But there is also (obviously) a huge amount of code that has the characteristic of “small pieces, loosely joined”. I believe that increasingly web-based tools (like Orion) will help with this sort of code and help to integrate small pieces of useful data and tools that are scattered haphazardly across the web.

    So to go back to the original premise: I agree that IDEs as we know them can’t successfully run in the browser, but I assert it’s a moot point because we will end up with a new class of tools that help with many of the same problems IDEs have traditionally helped with, but take advantage of the rich, distributed ecosystem of the web.

    1. Bill – a quite brilliant articulation. thanks! “To me the goal of these web tools should not be to replace existing tools, but rather to help do things that we canโ€™t do well with traditional tools”

  9. As I read my comment above, I feel a bit guilty and irresponsible that I quoted one sentence out of Mik’s very thoughtful blog entry without directing readers to read the entire entry for full context.

    I actually agree with most of the things Mik said but think the headline “Web IDEs remain a pipe dream” distracts from the interesting dynamics around new web-based tools like Orion, even if the actual statement is absolutely correct (which it is, if we define IDE as “something resembling Eclipse or Visual Studio”).

    Anyhow, cheers to Mik for yet another thought-provoking blog entry and if you haven’t already please read it in full so that you have the full context:


  10. Thanks for the pointer. You know, I never thought I’d use textareas in WordPress to edit PHP and CSS, but I sometimes do. More powerful web-based development tools will certainly lend themselves to greater use. A Web-based tool that were integrated with version control, automated testing, and deployment would certainly lend itself to use for a variety of tasks where IDEs or multiple applications are used today.

  11. If your are looking for a cloud based IDE please do check out our service http://www.codeanywhere.net

    And let us know what you think!


Leave a Reply

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