But even though adoption has been slow, more and more applications are gradually getting Gears treatments.

Today Google announced that Gmail can now be taken offline with Google Gears, meaning that you can now read and reply to your Gmail messages while you’re disconnected from the interweb, and have all of the email synchronize when you’re back online again.  There are still some bits of functionality that aren’t available in an offline mode – like the calendar, for example – but it’s a great start and about time.

Google Reader has offered offline support through Gears for some time now, and so has Google Docs.  Wordpress, although not an application developed by Google, has featured gears support for some time now.

It’s exciting to see so many powerful and rich applications available for offline use with Gears.  This could be one of those little things that Microsoft hears about and gets them a little bit nervous about the long-term future.  It’s probably far too early to make predictions about how much further web applications will go in replacing traditional desktop applications – and certainly there will always be some applications that will probably never be practical as web applications, like Adobe Photoshop for example.  But traditional “knowledge worker” applications – office productivity applications – seem like prime targets to be replaced by rich web applications.  Reliable and usable offline versions will likely accelerate the pace of innovation in these types of applications, as having an offline version will likely speed up end user adoption.  In other words, I think we’re hitting an inflection point where Google may consider a heavier investment in engineering and development of these tools.  Plus, Microsoft is planning a full web interface of their Office suite with the Office 14 release.  This could be the start of a new software war: the offline office suite war.

But I could be wrong.  As I said, I overestimated how quickly Gears would be adopted in the first place.  So we’ll see, I suppose.  My money is on Google for now, though.

Well, it just so happens that Symantec announced the beginning of a beta program for their new bilkoid engine, dubbed GoEverywhere.  Symantec is positioning GoEverywhere as a SaaS web security proxy solution.  There aren’t many details posted yet on their beta site, but GoEverywhere looks like a giant single-sign on web proxy that allows customers to stitch together web applications from several locations and include SSO, dual factor security, etc.

This doesn’t look quite as strong as the solution that BEA had, AquaLogic Ensemble.  It looks like GoEverywhere relies on a bunch of IFRAMEs to stitch together multiple sites onto one page.  BEA AquaLogic Ensemble, now Oracle WebCenter Ensemble, can perform the same type of functionality, but does it without using IFRAMEs.

It’s interesting to see Symantec position GoEverywhere as a security tool first and foremost.  This makes sense for Symantec, since their reputation is web security tools.  It’s also ironic, I think, that most Ensemble customers were also interested in Ensemble for the same reasons: web security (and policies, single sign-on, etc.), and less so about mashing up pieces of disparate applications.  Does this mean that mashups are really dead?  If so, long live the bilkoid.

According to the company press release:

LotusLive is designed to help companies work smarter by making it easy for them to connect and work together — with an emphasis on simplicity and ease of use. LotusLive’s online services give businesses of all sizes access to Lotus’ rich collaboration tools without requiring an up-front investment in IT support resources or infrastructure.

What’s most interesting about this news is IBM’s plans to partner with several key SaaS vendors:

• LinkedIn will provide LotusLive with searchable contact information.
• Salesforce.com will integrate LotusLive services within its CRM application.
• Skype announced plans to integrate voice and video into LotusLive.

Right now only LotusLive Meetings (Lotus Sametime) and LotusLive Events along with the hosted Notes application appear to be available for purchase, with the other solutions still showing up as being in beta.  I wasn’t able to find details on when IBM plans to make the additional functionality available.

The dashboard page for the collaboration component looks a lot like what you might expect from a "web2.0” site, and is actually pretty slick looking.

Even without the details about when it’ll be available, it’s exciting news from IBM, and it stacks up nicely against Microsoft and their plans for more SaaS offerings with the upcoming Office 14 release, as well as Google’s recent push to start selling Google Apps through their recently announced reseller program.

More details at LotusLive.com, full LotusLive press release at IBM.

Team Hosted combines the enterprise features of Confluence with the convenience and affordability of SaaS. It lets users:

• Edit pages using Microsoft Word
• Utilize free third-party plugins like EditGrid spreadsheets and Gliffy diagrams
• Attach and share documents, photos and video (up to 10GB)
• Start immediately. No installation or payment is required

You can get a free 30 day trial of Atlassian Confluence, which might be a good way to test drive it before you decide to buy into a monthly Enterprise hosted or on-site license for Confluence.

In order to achieve that, Google wants to hire you to become a reseller of their business applications, and set up small and medium businesses with Google’s SaaS offerings.  The premise is simple and brilliant, and it’s the type of thing that Microsoft has been doing for years with their Microsoft Certified Service Professional program.

Sounds great, so what’s the catch?

The one “gotcha” part of it is that it’s up to you to bill the customer for your services and time, and Google in turn will bill you directly.  But the bonus for you as a reseller is that you can start building a value-add proposition for customers: they trust you to convert them to a lower-cost SaaS offering (or deploy productivity applications in their environment for the first time), and in turn the company continues working with you as a trusted IT services vendor and architect.

Perhaps one possible criticism of Google’s plan to create a reseller program: it seems to be violating one of the principles of the evolving SaaS market.  Specifically, SaaS offerings should be so simple and easy to use that any employee of a company with little to no IT experience should be able to purchase any software-as-a-service direct from a vendor’s website and start using the product immediately.  This perhaps shows a sign of weakness in how easy it is at the moment for companies to transition to Google as a SaaS provider.  But, that criticism aside, Google’s reseller program seems to be a smart move.

Google plans to weed out the flakey service providers by testing your IT skills before authorizing you to be a part of their reseller program.  It’s then up to you to go out and drum up business, and if you aren’t signing up clients, it’s no loss to Google.  Overall it seems like a win-win for Google: they are effectively building an extremely cost efficient sales force (which is almost free to them, aside from the training materials and reseller testing), and these resellers don’t actually make any money from Google.  It’s almost like all of those power eBay sellers who make a living selling their products on eBay.

This is good news for Microsoft, considering that recent research form Forrester shows that services like Google’s Gmail cost 1/3 that of traditional on-site email solutions.  According to Forrester, it costs around $25.18 per month per user for on-site, hosted email vs. $8.47 for Gmail to operate.

Email is one of those services that I still say is easier to convert to SaaS than others.  It’s one I’d really love to see more companies adopt.

Also of interest: cloud computing is expected to be less frequently used, with only 10% of developers using cloud services (although many of the SaaS vendors may ultimately leverage cloud computing vendors on the back-end to provide scalability, so who knows).  Development of RIA, or rich internet applications, was also expected by 68% of the respondents.

I still think this isn’t too surprising: cloud computing as a concept is useful for folks that want to take the halfway step.  It’s great if you don’t want to own datacenter maintenance, or if you want to deploy a piece of enterprise software on a scalable infrastructure, but there are more upfront benefits in jumping into a SaaS solution right away rather than taking the cloud computing interim step.

Also, with so many varied definitions of RIA, and so many easy-to-use frameworks (Microsoft Silverlight, Adobe Flex, Sun JavaFX), it’s not surprising to see such a high percentage of respondents say they’ll be working on an RIA project.

[c/o InformationWeek]

First there was VMWare and Microsoft Virtual Server (VHD) virtual machine images to demonstrate software.  The problem there was always: who wants to download a 600-900mb VMWare image and deploy it on their personal laptop?  And furthermore, who has the local computing resources for it? 

Now Adobe has taken it one step further, offering 10 hours of runtime on an Amazon EC2 hosted image that includes LiveCycle ES Solution Components for customers of the Adobe Enterprise Developer Program.

Additional details from Adobe:

• Provides pre-installed, pre-configured virtual instances of LiveCycle ES.
• Reduce time required to download, install and boot new server instances.
• Develop LC ES applications in a fraction of the time.
• Virtual instances can be used as to build proof-of-concepts

Using virtual machines hosted on the cloud is a good way to let customers play around in a sandbox environment.  If you’re not already a pure-play SaaS provider, and you don’t have a sandbox datacenter of your own that you want to maintain, this could be a good alternative.

[Adobe LiveCycle ES Developer Express c/o InformationWeek]

Here’s a round-up of three that I know of that focus on enterprise data (including Oracle, SAP, and Microsoft applications):

• Jitterbit
• Talend
• SnapLogic

Of the three, Jitterbit is my favorite so far.  They seem the most advanced in terms of capabilities, and their editor UI looks the cleanest.  I haven’t spent much time playing with all of them yet, though, so I’m not sure what the pros/cons are of each of them (yet).

It’s interesting to note that all three of the ones I mentioned are offered as open source.  Most of them offer free tools for smaller scale projects, and an enterprise version for larger projects.

And then there’s Gnip, which seems like a good solution for SaaS providers to create data endpoints that only get polled when data is updated, rather than periodic polls that often end up in empty sets.

On the first point about SaaS disillusionment: it is pretty clear that there’s a lot of hype and excitement about around SaaS right now, but that’s pretty normal.  SOA and BPM went through the same hype curves, and SOA in particular is currently at the tail end of what I would describe as a prolonged trough of disillusionment.  BPM, on the other hand, is finally mature enough to be providing companies with business value.

It’s certainly probable that some companies may get in over their heads and invest too quickly in SaaS without proper planning, but that can really happen with any kind of software initiative.  Just think about all the disaster stories you’ve probably heard about CRM or ERP deployments gone awry and how costly it ended up being.  But that doesn’t mean that all types of SaaS solutions and targets are going to end up being big disappointments.  As I wrote earlier, some target for SaaS conversion are inherently easier than others.  And smaller companies who are looking for new investments in CRM, ERP or other business productivity tools might find value right away in solutions like Salesforce.com, Zoho, SugarCRM or products from 37signals.

On the second point about SaaS being a platform: I think this is true for companies that still think of their business applications needing to be built all on a single platform.  The problem, as I’ve discussed earlier, is that a single platform almost always ends up creating a big data silo, where it’s easy to move data around on the platform but hard to write connectors to other applications.  Companies that have started to invest in SOA are probably less concerned about having a single platform – they may already be realizing the benefits of having loosely coupled systems that can interoperate regardless of the underlying development platform.

SaaS providers should certainly learn from what’s been through the hype curve before to try to prevent SaaS products from hitting an extended trough of disillusionment.  As I mentioned before, in part this can be done with data portability, but I think this also can come quite effectively by providing customers with tools to prove ROI.   For example, SaaS providers offering dashboard-like ROI calculators that can show a customer their SaaS usage.  SaaS providers could even potentially show aggregate data other companies of similar size and industry, that could give them an idea of how efficiently they are operating – akin to comparing site metrics in analytics tools like Google Analytics.

It’s true that some business software just won’t prove to be a viable SaaS offering at all, and some customers won’t be able to adopt it because of their business restrictions, but there are already SaaS success stories for both customers and vendors today, and there are likely to be many more in the future.