Alex Barnett blog : SOA

Why OData Matters (IMHO)

alexbarnett — Fri, 19 Mar 2010 15:24:00 GMT

Earlier this week I was in the MIX10 crowd as Douglas Purdy announced the Open Data Protocol (it was a great presentation - summarized here).

I want to share with you why I think OData could be a very big deal…But before we go there...let's start with the basics...

What is OData? Where Did OData Come From?

To understand the history of how OData came to be, you need to understand how project "Astoria" came to be...I won't go over that again as this is already pretty well documented. ~~Astoria~~ OData has come a long way since.

In a nutshell: Today, OData builds on a few conventions, popularized by AtomPub (see OData AtomPub Format), to using REST-based data services. These services allow resources, identified using Uniform Resource Identifiers (URIs) and defined in an abstract data model (see OData URI Conventions, to be read and edited by web clients using simple HTTP messages (see OData Operations).

An Open Data Protocol for the Web

OData offers a standardized way for programmable data to be made available across the web and in turn allowing "consumers" of that data to rely on a set of conventions to be followed that in turn allows many interesting things to happen if widely adopted...

...And to this end:

As announced, OData has been released by Microsoft under the Open Specification Promise (OSP) "to allow anyone to freely interoperate with OData implementations" . Since then, the W3C has invited the OData team to Bring OData to a W3C Incubator (I haven't seen a public response yet but I urge the team to do so.).
OData is not a Microsoft-only thing and it won’t succeed if it is. The originating philosophy was about transparency in the design process, with an Open end-point as the goal - not a .NET lock-in play (“agree on standards and compete on implementation”). This approach has already yielded an initial set of clients, servers, services, and tools. Today, a number of OData SDKs and libraries are available for .NET, Java, PHP, iPhone (Objective-C) and more – and there’ll be more coming.
There is a small but growing ecosystem of non-Microsoft "producers" and "consumers" - (where OData "producers" include Netflix’s catalog as OData and the VanGuide (a social map of Vancouver Open Data) and Public data from the city of Edmonton as OData) and OData “consumers” - standard web browsers, RIA "data explorers" – such as Sesame OData Browser, and the client libraries mentioned above – p.s. somebody build a javascript-only data browser please!)

Where can OData take us?

The clue is in the OData icon (next to the RSS feed icon. Can you see the similarities?

The big idea here is that in the same way we have the "RSS" feed icon, we'll get used to seeing the "OData" icon on commercial and non-commercial websites everywhere (especially for government-related data). So in the same way you know today that the RSS icon means "get an XML feed for this content", the "OData" icon means "get this web data" - you'll know (and your client will know) what to expect in terms of how to read in, and navigate through and query structured web data sets - and in many cases write against them - using a common syntax.

Q: Right, But So What?

A1: Open Government OData. From Open Government Data Principles

The Internet is the public space of the modern world, and through it governments now have the opportunity to better understand the needs of their citizens and citizens may participate more fully in their government. Information becomes more valuable as it is shared, less valuable as it is hoarded. Open data promotes increased civil discourse, improved public welfare, and a more efficient use of public resources.

That’s great, but it needs to be practicable…And number 5 of the 8 Principles of Open Government Data sensibly states that the data should be (via David Eaves):

5. Machine processable

Data is reasonably structured to allow automated processing.

It'll be down to each government agency (and local government) as to how they decide to implement this principle, but wouldn't it be great if they agreed to a standard (and a powerfully simple, web-oriented one at that)? This is what Jon Udell concluded here:

"The open data movement, in government and elsewhere, aims to help people engage with and participate in processes represented by the data. When you publish data in a fully articulated way, you build a framework for engagement, a trellis for participation. This is a huge opportunity, and it’s what most excites me about OData"

A2: To ODatarize your data is to RESTify your data.

As more data-oriented web APIs come online, each team responsible for the design of each web API is confronted by the same kinds of questions, and each team answers these in their own particular way. Increasingly, “RESTful” is a design goal of web APIs. Great…but what does that mean? How do you expose the data, the relationships between the entities inside the model, and what should the querying syntax look like? Unfortunately, there are as many answers to these questions as there are RESTful web APIs. And there needn’t be.

For to ODatarize your data is to RESTify your data. Do spend the time at the value layer - figure out the way your developers / consumers want to see the data and expose it that way. Do make it easy for devs / consumers to learn / navigate about the data and use it. Do not make them learn about the unique idiosyncrasies you’ve built into your API (or those that leak out of your originating store) :-)

From a developer’s standpoint, OData is ultimately about productivity.

For the OData “Production Developer”: Point at your data store – define your entity model and map it to the data model you already have (so your developers consume / program against the data that makes most sense to them – effectively ORM’ing) and expose as an OData service, inheriting: all the REST characteristics; entity relationship self-discovery; and querying goodness.
For the OData “Consuming Developer”: If you know the web API is OData…great! Pick up a client library, get to the API end-point (data.foo.org/blah.svc). Point and Shoot: Traverse the data model, query it (and bookmark as needed – it’s a URI)…play!

(see links at the bottom of this post to technical content that provides details on all this)

A3: Since the announcement, I’ve seen quite a bit of excitement around the web (especially Twitter) by developers who see the potential here…there is plenty of experimentation going on. At Intuit, my team is also experimenting with ODatarizing some of our data services, exploring how it might be applied across a number of our cloud based data services. And when our team’s architect Tweets that “Looks like #odata is going to be a good fit for our data services”, I know there’s something interesting going on here…

So I encourage you to find out more about OData and get involved.

More Resources

An easy way to set up an OData feed from your SQL Azure database - Stephen Forte
OData interoperability with .NET, Java, PHP, iPhone and more - MSFT Interop team
Microsoft, Bring OData to a W3C Incubator - W3C blog -
Silverlight 4 + RIA Services - Ready for Business: Exposing OData Services-4-ria-services-ready-for-business-exposing-odata-services.aspx - Brad Abrams (MSFT)
Open Data for the Open Web - Douglas Purdy (MSFT)
Producing and consuming OData feeds: An end-to-end example - Jon Udell (MSFT)
OData for collaborative sense-making - Jon Udell
The three laws of open government data – David Eaves
8 Open Government Data Principles
Open Data Protocol (OData) with Pablo Castro - Hanselminutes podcast interview with Pablo Castro (MSFT – Architect for OData)
OData, a RESTful contender for your API - Andy Thompson
Announcing Sesame Data Browser - Fabrice Marguerie
Retrieving Data with the OData SDK for PHP - Brian Swann
Got SQL Azure? Then you've got OData - OData blog
Implementing only certain aspects of OData-only-certain-aspects-of-odata.aspx - Pablo Castro
Every SharePoint 2010 server is a Data Services server - Pablo Castro
Microsoft Creates New OData.org Website for Open Data Protocol – CMS Wire
Microsoft delivers updates on OData, Houston, Dallas - ZDNet
Timeline of Project Astoria - Pablo Castro
How ADO.NET Data Services came to be (formerly known as Project Astoria) - Alex Barnett
previous Astoria / OData posts - Alex Barnett

Federated Apps on the Intuit Partner Platform

alexbarnett — Wed, 03 Jun 2009 13:52:00 GMT

So what have I been working on since joining Intuit? A bunch of stuff, but something I wanted to shout about is the release of “Federated Applications” on the Intuit Partner Platform.

Working closely with five new developers to Intuit's Platform as a Service (PaaS), VerticalResponse, DimDim, Rypple, Setster, and ExpenseWare, we've got some kick-ass apps federated into the platform.

Some nice coverage so far this morning, including:

Ben Kepes, CloudAve Intuit Launches the IPP Version 2

"After the briefing I have to say that it looks to me like IPP is finally offering to fulfil the promise of the end-to-end integrated small business software platform that I’ve been evangelising for a few years now – my catch cry of late has been that application integration should only be seen as the very first step in building a SaaS ecosystem. Much more important is the aggregation of applications. This may be data aggregation, UI aggregation, sign-on integration or billing integration – and ideally users and vendors would determine what parts of the integration they bought into."

Leena Rao at TechCrunch - Intuit’s Partner Platform Goes Multilingual With Federated Apps

“The “Federated Applications” functionality lets developers who have existing SaaS applications that are built with any programming language, database or cloud computing platform publish their apps on Intuit Marketplace. Applications won’t have to be rewritten to conform to QuickBooks but will instead go through a minor configuration process.”

Phil Wainewright, ZDNet: Intuit makes two-pronged PaaS and SaaS push

“The significant element of Intuit’s PaaS announcement is that it is a land-grab to capture mindshare among developers on other cloud platforms, who can take their AppEngine, Amazon Web Services or self-hosted applications and make them available using Intuit’s single sign-on, billing and QuickBooks integration infrastructure. Market reach being one of the key attributes developers look for in a new platform, perhaps the most appealing factor is that applications will be showcased within the Intuit Marketplace, with a potential reach to the four-million-strong installed base of QuickBooks accounting software customers and their estimated 25 million employees.”

What I've found very cool about this new Federated Apps capability for IPP is that fact that these different SaaS apps are developed on a variety of stacks and hosted outside of the IPP platform.. One of the apps is built on Java. Another is built enirely on .NET. Another is a mix of RoR and LAMP. Another built of Flex (on their own hosting environment - not IPP). If an app was running on EC2, that would work too, as would an app built on Google's AppEngine. It doen't matter - the integration points for Federated Apps are just that and pretty lightweight (one of the partners was able to turnaround the work with 1 developer in less than two weeks, including time for the technical review of the app). We made a deliberate decision to make IPP agnostic to the technology that developers want to use. Yes, we have a "native" stack also, but the options we are providing developers now means there is no technology lock-in to speak of.

Here is some more information on the details of federation and how you as a developer can get started.

Oh, and a quick plug on the talk I'll be giving on how to federate applications to IPP at the Startups and the Cloud event next week in the Boston area.

Joining Intuit

alexbarnett — Sun, 21 Sep 2008 22:02:00 GMT

Firstly - thanks to everyone who has reached out to me in the last three weeks via email, phone calls and comments since I shared the news of my pursuit for the next adventure - I have really appreciated everyone's support and interest in my next steps.

The great news is I'll be joining Intuit as a Group Manager working in a fast growing start-up team responsible for leading the Intuit Partner Platform. Below is a snippet and some links to blog posts and articles that should give you a fairly good idea about where my focus will be a week from now.

Although I'll be working from the Orem office (Utah) to start off with, the plan is to ultimately move to the greater Boston area - another new adventure. I've visited Boston three times in the last year or so and have loved it more and more with each visit, so watch out Boston!

This is a fantastic opportunity for me personally - the team has ambitious goals and an amazing set of existing assets to leverage (see some of the numbers below) in becoming a significant player in the cloud computing space. I look forward to sharing stories of my new journey with you.

Intuit Partner Platform Opens to Developers (Sept 15 2008)

"The Intuit Partner Platform not only gives developers the opportunity to build Web-based applications, but successful SaaS businesses by taking the complexity out of managing infrastructure, hosting, user management, integration and billing. Now developers can focus on developing innovative on-demand software solutions that solve unique and important problems for the four million small and mid-market businesses across the U.S. that use QuickBooks and the 25 million employees that work in those companies.

The platform-as-a-service offering allows developers to combine the powerful open source Flex framework and Adobe Flex Builder and the proven database of Intuit QuickBase to build rich Internet applications that work with QuickBooks data. They can also leverage Adobe AIR to provide additional desktop-like functionality in their applications, such as pop-up notifications, local file system access, local data storage, and the ability to create a fully branded user experience.

"We have now accepted more than 1,000 developers into the program and it is exciting to hear their ideas and energy about what they want to build," said Alex Chriss, business leader for the Intuit Partner Platform. "Customers will benefit greatly from the imagination and expertise that developers use to solve problems facing their specific industries." "

Web 2.0 Expo NYC - Intuit Connected Services (Sept 19 2008)
Intuit Makes SaaS Play Guns For Salesforce (Sept 18 2008)
Intuit’s flexy PaaS (Aug 14 2008)
Intuit enters the PaaS wars (April 28 2008)
Intuit’s Radical New Flex + QuickBase Cloud Platform (April 16 2008)

Open Source in a SaaS World

alexbarnett — Fri, 16 May 2008 03:01:00 GMT

About a year ago, I took part in a meeting where the question: "What does open source "mean" in a SaaS world?" came up in conversation.

A year later, that same question is becoming increasingly pertinent as the IT industry's move to Software-as-a-Service (SaaS) and cloud-based computing accelerates.

For Bungee Labs (I work there), where

we provide an entire platform-as-a-service (PaaS)
developers create, share and re-use code and deploy apps in the cloud
developers "consume" and program against third party web apis and will create their own

...the "meaning" of FOSS is central within these different contexts and has many possible answers with many non-trivial implications...Three dimensional chess as it were.

(pic source: Memory Alpha, the Star Trek wiki)

For this post, I want to share some of the considerations relating to # 1) above: the context of open sourcing Bungee Labs' own system (Bungee Connect). Last month we stated that:

"Bungee Labs is evaluating several Free and Open Source Software (FOSS) licenses for the software components that comprise the complete Bungee Connect system. However, the task of reviewing the various FOSS licenses, and then identifying which of them best aligns with the software components and subsystems created by Bungee Labs–as well as ensuring compatibility with third-party components upon which Bungee Connect relies–requires considerable review and source code preparation. And we want to do this right, with the community’s involvement."

Since and before that announcement, Ted Haeger (who runs the Bungee Connect Developer Network) has been discussing some of the issues at hand and some of the options we see before us with some very "FOSS savvy" communities at events such as Socal Linux Expo, LugRadio Live USA and LinuxFest Northwest and of course with Bungee Connect's own growing developer community.

Today there's an interesting conversation going on between Ted and Simon Wardley, ex-COO of Zimki / Fotago who resigned last year over the company's decision not to open source their platform (the video of his announcement at a OSCON 2007 talk he gave "Commoditisation of IT and What the Future Holds" makes for entertaining and informative viewing all of its own...Simon discusses open source in a SaaS context. Update: Simon let me know of this video which also includes the slides

Anyway, back to the thread:

Simon wrote a post this morning providing his thoughts on the some the FOSS options available to Bungee Labs
Ted wrote back responding to Simon sharing his point of view
Then Simon responded to Ted

All three posts (and more to come no doubt) make an informative and interesting read, but I want to highlight one of the key issues in discussion.

The SaaS Loophole

The issue goes back to the question: "What does open source "mean" in a SaaS world?" and specifically the licensing issues. I'm going to quote and edit from Ted's post somewhat liberally (Ted owes me a Sushi, so we're quits now :P ) and isolate an (if not "the") open source licensing issue in the context of SaaS (my emphasis):

"Personally, I think that GPLv3 is the wrong license for freeing any SaaS or PaaS offering. The Free Software Foundation has a better license for this purpose.

GPLv3 is inadequate because it does not mandate that modifications that others make be opened. Originally, GPLv3 was planned to close up the “SaaS Loophole” (a.k.a. the “ASP Loophole”) in GPLv2. However, as I understand it, several large companies pressured the FSF to remove the key clause that would have closed the loophole.

What is the loophole? It’s this: if you take free software and offer it as a hosted service, then you are not conveying the software, and are therefore not obligated to reciprocate your modifications to the original code. In the context of service providers, GPLv3 is effectively the same as the BSD license. Many companies, Google among them, live inside this loophole. (For now, Bungee Labs is also in that camp.) Some remain there deliberately. Others are in it simply as a matter of course…that is, where they are in their business development process."

So that's the "SaaS loophole". Where's the loophole now? Ted explains:

"Perhaps the argument could have been made in the age of GPLv2 that the SaaS Loophole was an oversight, but now that GPLv3 has the loophole by design, it’s really no longer a loophole. The latest version of the license supports the practice. (And just to be clear, I am not advocating this for Bungee Connect.)

...Say Bungee Labs opens Bungee Connect under GPLv3. Is there a danger that small companies could replicate our offering? I don’t think that’s the case. But could a well-funded company do the same, fork the code, and then fund an engineering team to outpace the original inventors?

...The Free Software Foundation also provides the GNU Affero General Public License version 3, or AGPLv3. AGPLv3 specifically closes the SaaS loophole. Instead of being triggered by conveying the software, AGPLv3 is triggered by accessing the service. This helps to reduce the risk that a company could not branch the code and then out-engineer the originators, as the vulture company would be obligated to share-alike terms with their derivations."

So, is the AGPLv3 the right license for Bungee Labs to pursue? Is it the right license for SaaS providers? Is it enough on its own? (back to Simon Wardley's point in his post). Each company has their own unique circumstances and they each need to think through the 3D chess game. We're still working it out at Bungee Labs.

For us at least, I think some of the potential answers are becoming clearer, and others not yet. But it is the kinds of discussions that Ted is having with Simon that are a critical part of Bungee Labs' decision making process around FOSS. It cannot be an insular process.

Designing Web APIs - Twitter Learnings

alexbarnett — Thu, 01 May 2008 11:08:00 GMT

Although I made it to Web 2.0 Expo in San Francisco last week, I didn't make it to a session Matt McAlister blogged about by Twitter’s Alex Payne and Michael Migurski of Stamen Design who presented learnings from the perspective of an API provider.

But I can see the slide deck discussing the Twitter API and so can you:

| View | Upload your own

More Web 2.0 session slides available here. Recommended:

Videos of sessions here. Check out Clay Shirky's session, author of Here Comes Everybody: The Power of Organizing Without Organizations (a good read btw).

Sync Google Calendar with Outlook and more with WideLens

alexbarnett — Thu, 06 Mar 2008 05:37:00 GMT

Google has just released a very cool utility (.exe download for Windows) providing users with the ability to synchronize their Google Calendar with Outlook.

Some nice features in their 0.9.3.0 release:

schedule the sync frequency: every x minutes
define directional flow: 2-way, and 1-way (either way)

A bit of a buzz going on about this...

Now, what if you could do the same over the web - no download, just use your browser (IE, FF, Safari)...? And not just Google Calendar <> Outlook, but others too...?

Well, it's certainly possible...First, watch this screencast I put together tonight (apologies for sound quality...done from home equipment):

About WideLens

A couple of weeks back Bungee Labs released a reference calendaring application, called WideLens, designed to show off some of the power of the Bungee Connect platform, from the kind of rich AJAX UI experiences delivered through to the high level of functionality developers can create by wiring up and integrating multiple web services and distributed web data sources into a single web app.

Video: WideLens User Experience
Video: Developer Overview

WideLens connects to Microsoft Exchange calendar, Google Calendar, Salesforce.com, Facebook, MySQL and iCalendar feeds, representing a variety of protocols and authentication schemes. MS Exchange is accessed through WebDav, Google Calendar through gData, Salesforce.com via SOAP, Facebook through REST and MySQL connectivity is based on client libraries provided by MySQL (integrated directly inside Bungee Connect).

WideLens is an uber-mashup.

WideLens connects to each of the sources in real-time, presenting the user with live data. With the exception of Facebook and iCalendar, users can create and modify events and those changes are immediately posted back to the source. MySQL pulls double duty, serving as both a WideLens native calendar source and as the persistence layer for all kinds of application data including user preferences and credential information for each service.

Developers: Have At it!

As mentioned above, WideLens has been released as a Bungee Connect reference application where we're encouraging Bungee Connect developers to customize the WideLens application as much as they want, deploy their own version of the app as their own app - to their own domain, at no charge, branded however they want and with whatever features / cuts / modifications / extended they want - the WideLens code is released under a BSD licence (read more here).

To get going with Bungee Connect and develop your own vision of what WideLens could do, sign up for your Bungee Connect account.

Astoria at MIX08 (REST in Vegas)

alexbarnett — Wed, 05 Mar 2008 02:11:00 GMT

As much as I'd love to make it to MIX08 this week, time will not allow me...But if I were, then I'd be going to the following three sessions related to the Project Formerly Known as Astoria:

Wed, March 5th - RESTful Data Services with the ADO.NET Data Services Framework
Fri, March 7th - Accessing Windows Live Services via AtomPub
Fri, March 7th - Building RESTful Real World Applications with the ADO.NET Data Services Framework

Time to Define "Platform as a Service" (or PaaS)

alexbarnett — Tue, 19 Feb 2008 09:27:00 GMT

Before joining Bungee Labs last year, I knew they were on to something big. I mean, really big.

A big idea, an ambitious vision: to provide developers with end-to-end development, testing, deployment and hosting of sophisticated web applications as a service delivered purely in the cloud.

Since we announced our private beta back in May 2007, we've had over 1,500 developers sign up. In January alone we had over 400 developers kicking the tires - not just signing up and disappearing, but 400 returning developers, learning, building and deploying out increasingly sophisticated apps on a fast evolving developer platform, requiring no install of anything on their machine - all through the browser.

And since May 2007, the trend to delivering software as a service (SaaS) has been moving at terrific pace. New web APIs are being made available every month and new announcements by start-ups as well established big players are reinforcing and fueling the acceleration to the inevitable world of cloud computing.

As we announce our move from private to public beta today, we've also tried to articulate the new category of product and service we believe Bungee Connect is at the forefront of defining, the category of Platform as a Service, or PaaS, and our big bet is that PaaS is the next big thing.

So what is a "Platform as a Service"?

In September 2006, Marc Andreessen posted his thought provoking "The three kinds of platforms you meet on the Internet" and it got a fair level attention from the web industry. And we took note. We thought what Marc was describing in his Level 3 definition where:

"A Level 3 platform's apps run inside the platform itself -- the platform provides the "runtime environment" within which the app's code runs.",

...was right, but only partly right. Given Bungee Labs' ambition and vision, we felt there was a lot more to Marc's definition of the highest level definition of an "internet platform", a definition more holistic and comprehensive than a runtime.

But we kept focused, kept working on what we were hearing our developers telling us we needed to fix and improve on Bungee Connect, to give what developers are telling us what they really want - a Platform as a Service - to provide everything required in the lifecycle for the development through hosting of full-on, sophisticated and highly interactive web apps, not just widgets.

As we were readying for our next phase -our public beta - we thought it would be a good time to put a stake in the ground and actually define what we mean when we use the term Platform-as-a-service, and thereby describe the comprehensiveness what Bungee Connect has to offer.

So early this morning, our CTO and Founder of Bungee Labs, Dave Mitchell posted a definition describing PaaS in concrete terms.

What follows is a summary of Dave's post, with a selection of my favorite "soundbites" and ideas, but I suggest you read the whole post for yourself - there's a fair amount to consider:

1) Develop, Test, Deploy, Host and Maintain on the Same Integrated Environment.

"It’s time to stop developing “here” and running “there”. Today, most applications are coded in one environment (usually custom-built for that project by a developer), then tested in another, and redeployed to yet another for production...In a completely-realized PaaS, the entire software lifecycle is supported on the same computing environment, dramatically reducing costs of development and maintenance, time-to-market and project risk. A PaaS should let developers spend their time creating great software, rather than building environments and wrestling with configurations just to make their applications run — let alone testing, tuning and debugging them...Also, an end-to-end PaaS should provide a high productivity Integrated Development Environment (IDE) running on the actual target delivery platform, so that debugging and test scenarios run in the same environment as production deployment.

2) User Experience Without Compromise

"A Platform-as-a-Service must deliver compelling user experiences, with all the richness and live interactivity that consumers have been conditioned to expect....Hiccups like software downloads or plug-in installations, browser dependencies and inconsistencies, or local executables break the web model, and are inherently less secure, less maintainable and less user-friendly. In order to be relevant and popular, PaaS must deliver the best user experience available on the web, comparable to or better than conventional approaches.

3) Built-in Scalability, Reliability, and Security

"Developers should be free to build applications with the comfort that the security of customer data, network traffic, source code (intellectual property) and even server hardware is maintained automatically by the platform through-out application development and delivery."

4) Built-in Integration with Web Services and Databases.

"Applications need to leverage existing software investments in databases, and internal or external third party web services, requiring that the platform offer a wide variety of connectivity options."

5) Support Collaboration

"A PaaS must support both formal and on-demand collaboration throughout the entire software lifecycle (development, testing, documentation and operations), while maintaining security of source code and associated intellectual property."

6) Deep Application Instrumentation

"With instrumentation, organizations can see exactly how users are using the application, the type of performance they are experiencing and any application crashes. This information can also be leveraged to create new business models where costs are tied to actual utilities, rather than flat-rate subscriptions or licenses."

Over the next couple of years we expect to be hearing a lot more about PaaS and how "Y announcement" by "X company" is now providing true a PaaS offering to businesses and developers.

But saying you are providing a Platform as a Service has to mean something, and we think the above definition sets a high but reasonable standard that must be met for any company to claim they are providing a "platform-as-a-service' and legitimately describe themselves as a PaaS player.

The amazing thing is, for me at least, is that Bungee Connect is delivering all of the above, today. From our point of view, delivering PaaS - the real deal - is not statement of Bungee's intent, it's a statement of fact. It's bold, but so is our vision. Yes, we've still a lot to do before we're commercially ready and we think that's coming soon, but so much is already there. Try it out.

Video demo: REST Describe & Compile for WADL

alexbarnett — Tue, 22 Jan 2008 01:08:00 GMT

In March 2007, I naively asked if REST needs a WSDL and if yes, was WADL the one (Web Application Description Language). The conversation goes on, but I thought the video below and pointers might be of interest to those who have an opinion one way or the other.

The video Google Chalk Talk was published last week and shows Thomas Steiner demoing REST Describe & Compile, an online editor and a compiler for REST Web services based on SUN engineer Marc Hadley's Web Application Description Language (WADL).

(huh? video not available as embeded??? wtf? Here's the link to it)

"ABSTRACT

REST Describe & Compile is an editor and a compiler for REST Web services based on SUN engineer Marc Hadley's Web Application Description Language (WADL). REST Describe & Compile is implemented using the Google Web Toolkit (GWT) and is split up in two sub-apps:
REST Describe:
This component takes a (set of) URL(s) as input and tries to analyze the parameters regarding parameter types and Web service structure. It then generates a WADL representation for the given URL(s), represented in an editable, tree-like form. Typical input would be GET
http://webservices.amazon.com/onca/xml?Service=AWSECommerceService=0ZMFM7HPBYYY0SPERCR2=ItemSearch=Enterprise+Integration+Patterns=Books
or GET http://picasaweb.google.com/data/feed...%7BuserID%7D/albumid/%7BalbumID%7D/photoid/%7BphotoID%7D?kind=kinds=0=1{userID}/albumid/{albumID}/photoid/{photoID}?kind=kinds=0=10

REST Compile:
This component allows for the WADL representation of a Web service to be "compiled" to working programming code in various languages (for the moment these languages are Java, PHP5, Python, and Ruby). The idea is thus similar to WSDL2Java, however in a more general WADL2Anything way.

The application can be tested online:
http://tomayac.de/rest-describe/latest/RestDescribe.html

There are some screencasts available on YouTube: http://youtube.com/profile?user=tomayac

Speaker: Thomas Steiner
Thomas Steiner works as a Customer Solutions Engineer from Barcelona, Spain, and has joined Google full-time only since October 1st. However, he has been with Google first as an intern and then as a temorary contractor since February 2005. During his time at Google, Thomas has created the Google APIlity PHP Library for the AdWords API, as well as the APIlity Agua Ajax application, which can be seen as an Ajax GUI for AdWords. From January 2007 till July 2007, Thomas did his Final Year Project with Google, where the REST Describe & Compile application was to be developed, hosted by Googler Patrick Chanezon."

Related, last week Assaf (a true web pragmatist) wrote in his post REST IDL, substance over style

"So far most of the suggestions I saw are turf wars in disguise, arguing for this syntax or that, but never for what it should describe. WADL so far looks like one of those best-for-my-platform choices, the constant mentioning of Microformats, which are not optimized for this task, is another danger sign.

It has to start somehow, but it better start with substance, not with style."

I agree with the last statement...

Now, I don't know if Assaf has seen the above video or not, and if not, whether the video would further sway him one way on WADL or not...anyone else have thoughts on what's in the video or tool demo'd, or WADL itself?

Podcast interviews - smart people in the world of the web

alexbarnett — Sun, 20 Jan 2008 15:41:00 GMT

One of the fun parts of my job at Bungee Labs is to partner up with Ted and interview some smart people in the world of the web. We publish these as a podcast series (the Bungee Line - podcast feed here) over on the BCDN blog.

If you have ideas about someone you think we should interview, let me know! We're focusing on topics we think web developers might be interested in the worlds of software as a service and web app development, in particular profiling web apis. Related topics are good too.

I've listed out below our most recent podcasts below...plenty more in the works (previous podcasts are listed here). Hope you like :-)

Alan Lewis on eBay Desktop and eBay APIs

"As product manager for eBay Desktop, Alan Lewis relies on the same web APIs that eBay makes available to all developers. In this edition of the Bungee Line, Alan tells us about what the eBay Desktop is, how it came about, and various details about eBay’s developer program and web APIs. We ask Alan about eBay’s position Oauth and on open source."

Toby Segaran on “Programming Collective Intelligence”

"Since the publication of his O’Reilly book Programming Collective Intelligence: Building Smart Web 2.0 Applications, Toby Segaran has become well noted for his ability to explain easily-understandable algorithms for the kind of deeply complex problems involved in social applications. Toby joins Alex and Ted to discuss some of the high-level concepts that he tackles in his book."

Jon Aizen of Dapper.net

"Jon Aizen joins Alex and Ted to explain how Dapper.net provides a no-fee tool for making almost any structured web site data accessible via a REST API. In a past life, Jon was involved in creating The Internet Archive. Jon also helps the Bungee Line introduce romantic intrigue into the podcast.

Punditry Alert! At the end of this show, Ted and Alex speculate a bit about Android, Google’s open source mobile device platform, the Apache License, and whether Robert Love is involved. Please consider this as another demonstration of Ted’s idiocy, brought to you by the Bungee Line."

Jeff Barr on Amazon Web Services (Part 2)

"In part 2 of our interview with Amazon Web Services evangelist Jeff Barr, Alex and Ted ask Jeff about Flexible Payment Service, virtual user group meetings in Second Life, the Startup Project, and pry at Jeff’s views of possible futures of technologies that developers might anticipate."

Jeff Barr on Amazon Web Services (Part 1)

"Developer evangelist for Amazon Web Services, Jeff Barr tells Alex and Ted about how he became a native Amazonian, his recent visit to “The Business of API’s Conference,” and a bunch of stuff on Amazon Web Services, including: Mechanical Turk, EC2, and S3. Additionally, Jeff explains the newly announced S3 Service Level Agreement*."

8 Trends in Software as a Service Platforms

alexbarnett — Sun, 13 Jan 2008 17:40:00 GMT

To kick off the new year, I presented to around 40 or 50 members of Utah Technology Council (UTC) last week. The title of the topic they asked me to speak about was "Trends in Software as a Service Platforms". I searched around for some ideas and came across two recent posts predicting trends in SaaS for 2008, one by Phil Wainewright "Eight Reasons SaaS Will Surge in 2008" and Jeff Kaplan's post "Top Ten Reasons Why On-Demand Services in 2008". I decided to borrow liberally from these (thanks Phil and Jeff) and mash these two together (along with a couple of thoughts of my own) and present "8 Trends in Software as a Service Platforms" to an audience made up of CTOs and VPs of engineering and development for software companies in the Utah area.

In preparation for the presentation, my boss (Martin Plaehn) at Bungee Labs suggested I write up my presentation as notes blog them afterward, so here they are.

8 Trends in Software as a Service Platforms

SaaS is just part of the web mega-trend
Mainstream opinion says “Yes” to SaaS
Software vendors stampede into SaaS
All is being virtualized
Explosion of Web APIs
Economic factors favor SaaS
Enterprise and SMB IT embraces SaaS
SaaS platforms proliferate (PaaS)

1. SaaS is just part of the web mega-trend

Most of us have witnessed and many of us have been a part of the transformation in the way goods and services have been digitized, virtualized, delivered and consumed. Software, the data behind that software and the functionality that software provides is no different - software is subject to the very same transformational forces.

Just think about how even a class of product that is natively digital - such as software - has been transformed in the way it is delivered and consumed. For prosperity's sake, I've still got a few of those ZX81 software cassettes stashed away somewhere, gathering dust, looking ever more antiquated with each passing year. How will today's mode of software delivery and use look to us in a few years from now?

The web wants to connect things, and that's interesting. But connecting and interacting with "live" data, information and remote functionality make things more interesting.

At the fundamental level, the web connects things. It connects people to people, businesses to businesses, and people to businesses. Since the early 90's, the web has enabled the connection of so many things to so many other things at an ever accelerating rate, and yet we crave even more connectivity. But we increasingly also want the ability to interact with those things.

And it is the nature of these connected things that have changed since the early internet. The early web was good at connecting to static views of information and accessing limited and rigid functional services, very much a read-only mode. Then, as we learned a) the ability to read more dynamic-type information - at least regularly updated, and b) access richer remote functionality, we created whole new opportunities for ourselves. Next, we grew our ability read and write against dynamic, near real-time data and information and to program against remote functionality to create a new class of web applications leveraging those capabilities - and hence a new order of business and experiential opportunities have emerged. Some label this as "Web 2.0".

At its essence, it is the "liveness" of these real-time read-write data, information and functional sources available as "always on" services and the increasing ease to connect to, interact with - specifically change those resources available as live, programmable services that allows us to create new value out of those resources, opening up brand new market opportunities for businesses and the compelling, rich "live" end-user experiences of tomorrow.

2. Mainstream opinion says “Yes” to SaaS

Not surprisingly, Wall Street loves the the predictability of subscription services. It's good for cash flow, forecasting and business planning.

The venture firms also relish the opportunities that are opening up in a software as services-oriented economy. The ability to circumnavigate the incumbent software players with new disruptive technologies and propositions that are significantly easier to try and access for prospective customers compared to traditional software evaluation, along with usage and subscription-based business models verses the old licensing model makes investing in services-based software companies very compelling propositions from the venture firms' point of view. We should also see healthy M&A activity based on these similar opportunities in the coming year.

And then there's the trend for offshore / IT business process outsourcing. These providers will surely get in the game and make their plays through investments in and acquisitions of SaaS vendors that align well with their current core businesses.

Add to that the excitement we're reading about the SaaS space from the IT Analysts, journalists and bloggers, plus the new book by Nick Carr (author of “IT Doesn’t Matter”) - delivered by Amazon to me last week: “The Big Switch: Rewiring the World, from Edison to Google”. I think there's little doubt Carr's excellent analysis of the computing industry as an analogy to the electricity industry's shift to a utility model will be on business bestseller list for much of 2008. His messages resonates with corporate executives and end-users agree with him:

IT is a needless hassle,
it should be as easy as electricity and
be as reliable as a utility

3. Software vendors stampede into SaaS

The Big Software Players are following the early SaaS successes

CRM as a case in point. If you've been following the CRM software market, you'll know about the noises Oracle-Siebel, SAP and Microsoft started to make in the 2007 about what they are are lining up for the 2008 in terms of CRM as a service. Their efforts to emulate Salesforce.com's success delivering CRM as SaaS will be key strategic bets from the incumbents' point of view - and loud, price and functionally competitive propositions from the point of view of their existing and prospective customers.

CRM is just one of the multiple horizontal solution categories to transform from on-premise with traditional licensing model to a service-based delivery and subscription-based revenue model. ERP, supply chain, e-commerce, HR and many more...the horizontal solution list goes on. And then there are the vertical solution players...

Here's another data point to consider regarding the move by traditional software vendors to a SaaS model:

“15-20% of application ISVs have already either begun new skunk works initiatives or gained access to SaaS assets and development experience through M&A activity”
(Source: Key Trends in SaaS: 2008 and Beyond, Saugatuck Technology)

4. All is being virtualized

Virtualization is a technology trend.

Virtualization enables hardware as a service. The demand for virtual machines met by hypervisor software (VMWare, Xen, Hyper-V) and the success of Amazon's Elastic Compute Cloud (EC2) in the last couple of years point to a continuation of further virtualization of applications and hardware. Virtualization is accelerating the move from traditional on-premise software to services.

Virtualization is a business trend.

We continue to become a mobile workforce. The younger entrants into the workforce in service-oriented economies expect and want to be always connected. It's very hard work, if not impossible to get your traditional on-premise applications and centralized servers sitting behind a firewall to serve today's mobile workers. SaaS and managed services meet the needs square on.

5. The explosion of Web APIs is upon us

According to ProgrammableWeb.com, there are 559 commercial and public APIs available today, most of these are new and there are plenty more to come. How many will we see go live this year? And how many private web APIs are there and will be developed and consumed in the coming year?

Data from ProgrammableWeb.com

6. Economic factors favor SaaS

On-premise software requires upfront capital investments
To lower costs, many companies hold back on their capital investments to mitigate their risks, especially in recessions
Adopting on-demand services on a pay-as-you-go basis will be a perfect sourcing strategy for businesses seeking greater cost-controls and flexibility – the utility model

All well and good, but the real economic value of SaaS is that fact that it unleashes new value of previously isolated data silos and functionality.

7. Enterprise and SMB embraces SaaS

When it comes to IT, who doesn't like

Low-maintenance?
Low cost?
Low-resource profile?

IT and business folk like these things, and externally delivered SaaS applications deliver these benefits.

8. SaaS platforms proliferate (PaaS)

The more mainstream SaaS becomes the more the large vendors will be forced to offer effective platforms for ISVs, enterprises and SMBs.

If the move by the software vendors from traditional on-premise software to a services model is to be successful, they will need to provide programmable interfaces - not just end-user interfaces - to their services for their customers. Customers need and want the ability to access, intergrate and create new value out of live, programmable data, information and functionality living in the cloud. And in turn these same customers will want their custom-developed composite applications and integrated data available as programmable services - yet more APIs.

Customers want to unleash new value of previously isolated data silos and functionality through the development of their own applications programmed against those resources. And in turn these same customers will want their own custom-developed composite applications and newly integrated data available as end-user interfaces and as programmable services - yet more APIs. These customer needs will drive the software market to provide platforms to provide businesses and developers with with end-to-end:

programmable services and data integration
application development, testing and collaboration tools
deployment and scalable delivery

...all as a service with a utility model.

(hey...I needed to mentioned Bungee Connect just the once ;-).

2008 will mark a the proliferation of such offerings as "platforms as services" (or PaaS) through 2009, where then the consolidation will begin. Interesting SaaS and PaaS times ahead.

Update 2/20/2008: see "Time to Define "Platform as as Service" (PaaS)

The presentation seemed to go down pretty well and we had lots of interesting discussion throughtout. One of the topics we discussed was data security in a SaaS world. Don Kleinschnitz (VP, Development at Symantec) followed up with a mail linking to his blog covering Security 2.0 topics.

Again - thanks to Phil Wainewright and Jeff Kaplan for their post and to Martin for suggesting I blog this.

PaaS, more than SaaS

alexbarnett — Wed, 18 Jul 2007 15:01:00 GMT

In May 2007 I attended Salesforce.com developer conference where Salesforce SOA was announced as an add-on to the Apex platform. Industry analysts focusing on the area of SaaS considered the move as significant, but not surprising (see Phil Wainewright's take on the new in May 2007 as an example).

The game is moving on. PaaS, or "Platform as a Service" could be the acronym that defines a new web-oriented model where more than just specific vertical "services" are delivered as SaaS (e.g CRM, ERP, etc).

PaaS speaks to a more generalized services platform concept. If the "web as a platform" is the notion of multiple services in the cloud, then where does the "composition" happen? There are several Mashup design patterns to chose from. These patterns are along a continuum - from pure browser presentation mashups to client-side mashups to server-side services and data mashups (composite applications). If architecturally speaking you land in the composite applications pattern for delivering your SaaS apps, then this is where PaaS comes in.

This last week there is further evidence that Microsoft and Salesforce want to benefit from the opportunity in the power of PaaS. From SaaS out, PaaS in, by Angela Eager at CBRonline.com: Sub heading is:

"The Salesforce.com Summer '07 release is all about the platform, with software-as-a-service evolving to platform-as-a-service, a move that will antagonize and incentivize big league players Microsoft, SAP, and Oracle."

Angela's analysis includes a comparison to Microsoft's own forthcoming offering, including "Titan"

"Salesforce.com might be the first to deliver an on-demand platform but it certainly will not be the only one, and the on-demand specialist is starting to feel Microsoft's breath.

Releasing snippets of information about the forthcoming Titan offering at its partner conference last week, Microsoft chief Steve Ballmer said Titan referred to a CRM application-development platform on which developers would be able to build customized applications that could run on their servers or on Microsoft's own servers.
Although the Microsoft offering is not yet available, and even after launch it will take time for some of the planned functionality to be available such as running custom user applications within the Microsoft data center, it will be a direct challenge to Salesforce.com. With similar platform offerings, users will have a choice for the first time, which has the potential to cut into Salesforce.com's growth."

Maybe "Titan" is the news I was looking for last week, but we're seeing hints at more than just Dynamics APIs as SaaS from Microsoft. There is still no real hard news on this today, but watch out for posts by the likes of Phil Richardson, a Microsoft Program manager in the Dynamics team (and my ex-neighbor in Redmond) like this where Phil hints:

"Soon (can't say exactly when) I'll be starting to blog about Titan features....Once marketing gives the go-ahead you can expect to see boatloads of Titan content on this blog. I'll be focusing primarily on features which are targeted at VARs & ISVs - but I'll make sure to throw in some end user features."

And then Phil hints a little more this week...

"Some vendors have the 'heads in the sand' and refuse to talk services seriously (with no offering or a token offering). Other vendors only offer SaaS and trash talk those who aren't pure services players. There is no question that our industry is moving towards services and Microsoft's CRM has a strategy which is sensitive towards these industry changes. As we announced: we are using our Titan Project to create a Microsoft Hosted version, a Partner Hosted version and an On-Premise version. I use the term 'version' loosely as the final number of SKUs is up to marketing. I believe we are entering an time in the industry when customers will start moving away from self hosting 'non-unique' business functions. Anything which can be achieved by configuring on the shelf apps (like Microsoft CRM) will eventually move to hosted services (if I knew exactly when 'eventually' was I'd be a rich man). Our strategy understands that customers and partners will find themselves in varying degrees along the on-premise to hosted spectrum. Some will want everything 'in the cloud' and others will want everyone their own datacenters. Some might want to prototype in the cloud and move to on-premise etc etc. I believe vendors need to interpret these changing times appropriately for their business."

(hey Phil....trackbacks? comments? common dude!)

My take is that Phil hints to more generalized services platform concept from Microsoft. PaaS, more than SaaS.

Delivering on PaaS is hard though. Yesterday, Dan Farber's post Salesforce.com transitions to platform as a service highlighted one of the greatest challenges in this PaaS space - uptime at scale:

"Collins said that salesforce.com consistently provides above 99.9 percent availability. “Three nines is the best quality of service you can get today from any on demand player,” he said, claiming that the best deployments of Siebel and Oracle are closer to 96- to 97-percent available.
In any case, as salesforce.com scales up the enterprise and companies have more dependence than for just CRM on its platform, three nines, which doesn’t include times when the service is up but suffers performance problems, won’t be sufficient for customers who can’t afford 8.76 hours of downtime a year."

Uptime is just a small (but critical) component of the whole PaaS vision. Developer tools for PaaS is another kettle of fish (see Dana Gardner's post on Software Development and Deployment as a Service, or "SDDS" - on this topic). It's how Bungee Labs sees the future. But that's for another post.

RESTpectful comparison of Microsoft's Astoria and Google Base Data APIs

alexbarnett — Fri, 13 Jul 2007 06:36:00 GMT

Dare has written up RESTpectful comparison of the programming models of Microsoft's Astoria and Google Base Data API for querying relational data over the web (RESTful data services).

Both these APIs also happen to support operations for changing the data - inserting, updating, and deleting - Google's details here and Astoria's in this .doc, where in addition Astoria allows associations between entities to be created and deleted). These aren't mentioned in Dare's analysis (I'm sure he knows of the ability to change the data via the APIs, not just query), however the article is worth a read nonetheless.

What does open source "mean" in a SaaS world?

alexbarnett — Thu, 12 Jul 2007 13:19:00 GMT

In March 2007, this article by Bryan Richard discussed the the implications of Software as a Service (SaaS) with respect to the GPL licence.

"Conservative estimate: In the future 75% of your applications will be delivered over a computer network and you will interact with them with no "transfer of a copy." For evidence of this trend just look to the popularity of Gmail and the creation (impossible 10 years ago) of Google's document and spreadsheet apps; analyst suggestions that the importance of operating systems is fading; the success of web apps like SalesForce; wiki culture; the popularity of Web APIs and application mash-ups; Adobe building an online version Photoshop.
The future is networked. The GPL isn't."

In April, Matt Asay posted his recognition of the same issue after visiting Bungee Labs (as a member of our Advisory Board):

"I believe we're on the cusp of a hugely disruptive change in the way open source software is licensed. The OSI, of which I'm part, is against license proliferation. But the community and OSI will need to figure out how to make open source relevant to the online world, without stifling the creativity that has built it. This is a non-trivial problem, and it is a problem."

And I agreed:

"..the cloud-based future we're heading to will cause an invigorating and necessary debate within the industry on the topic of licenses in a purely networked computing environment:"

Today Tim O'Reilly chips in, concluding

"This is still an unsolved area for both open source and Web 2.0. I believe that there will come a time when we will need to rediscover for Web 2.0 the freedoms that led Richard Stallman to the GPL, but I don't think it will grow out of the current crop of free software licenses. It will be closer, perhaps to Wesabe's open data bill of rights."

And in response, Matt Asay picks up on the thread again this morning:

"There's no point in complaining about what might have been. The real debate should be over what we do now: we need a new crop of licenses to help make open source relevant for the web. It's actually surprising that we have almost nothing so far...."

The question is: What does "open source" mean in a networked, SOA / API-based world as SaaS becomes the dominant computing model?

It looks like this will be the main topic of conversation at this O'Reilly event ~~next week~~ in two week's time (I'm going to try to make it given the topic's pertinence to the future of our own business). My own (and maybe obvious to some) prediction is that the effort by the community to answer the "What does "open source" mean in a SaaS world?" question will become the center of a highly controversial and popular topic of conversation for the wider technical community in 2008.

And of Dynamics Live CRM SaaS APIs?

alexbarnett — Wed, 11 Jul 2007 04:18:00 GMT

Todd Bishop reported today of Microsoft's intended entrance next year into the B2B SaaS market with its announcement at the annual Worldwide Partner Conference in Denver, providing details on pricing and partner revenue-share plans for a hosted service version of Microsoft Dynamics Live CRM:

"Businesses use CRM programs to handle sales, service and other customer interactions. The new "Live" CRM option, to be offered by Microsoft as an online service, is part of its broader effort to expand beyond traditional software licensing into subscription- and advertising-based offerings.

...Microsoft announced plans for a subscription rate of $44 per user each month for the professional version of the new Microsoft Dynamics Live CRM, and $59 for the enterprise version.

Salesforce.com lists prices of $65 for the professional version of its Web-based customer relationship management service, and $125 for the enterprise version. List prices in the online CRM market are commonly in the ballpark of $70 to $100 per user each month, said Liz Herbert, a Forrester Research analyst. "

Phil Wainewright commented on the catching up Microsoft needs to do in this SaaS / CRM space (something I've commented on previously), hence the lower price-point when compared to Salesforce.

"Undercutting Salesforce.com is an obvious ploy for Microsoft to attempt with its CRM Live product, so the pricing it announced today shouldn’t come as any big surprise. Salesforce.com has always priced at the high end of what it could get away with, and it’s benefitted from five years of being allowed to get away with it, plowing the proceeds into an astonishing sales and marketing drive that’s blasted revenues past the half-billion dollar a year mark."

(screenshots here)

Although Microsoft will be offering Dynamics Live CRM at no charge during an early access beta starting this year, it is unlikely they will go "live" until the second half of 2008 (I'm speculating), presumably once the "Windows Live Data Centres" are ready to do so...a brand we are promised we'll hear a lot more of in the future.

Salesforce.com responded. In a phone call with Todd Bishop, VP of Corporate Strategy, Bruce Francis, took a typically aggressive tone (that is, typical of Salesforce.com, not Bruce) in reacting to the news:

"What it looks like is that Microsoft is just marking down an inferior product to what customers are actually paying right now. Also, one thing that I haven't seen is the url where I can sign up for a 30-day trial. ... I know a great multi-tenant on-demand service when I see one, and I see more of them every day. ... We could talk for hours about all the great on-demand services that are out there that I can sign up and use. Where is Microsoft? Microsoft has a price list, not a product."

But where is the news of Dynamics Live CRM web / SaaS APIs? No mention of these anywhere that I've seen, but if Microsoft wants to take on Salesforce.com they'll need a strong story here - the latter has a mature programmable surface today with a fairly healthy ecosystem of developers.

It's one thing to have on-premises software that can be extended and integrated with other software and services through SOAP APIs behind your firewall (Dynamics has a healthy story in this department today), but it is quite another (increasingly necessary) thing to provide APIs as part of a hosted SaaS service. I would have thought the provisioning of SaaS APIs would have been a central component of Microsoft's messaging today so am a little surprised by its omission.