Catching Up With ZohoCatching Up With Zoho

Those of you curious to see how ambitious a Web-based application suite can be while still being built entirely on open standards need look no further than Zoho. The other day I spoke with company evangelist Raju Vegesna about what they've been doing in their ongoing attempt to beat both Microsoft Office and Google Docs at their own games. The short answer: quite a bit.

Zoho's been gussied up since I last looked at it for my information article where Barbara Krasnoff and I compared and contrasted it with Google Docs. Among other things, they've added spellchecking in 43 different languages, the ability to export documents in Office 2007's OOXML format (if you use that), a pivot-table function for Zoho Sheet, a complete revamp for Zoho Show, and ways to create and execute macros. Some new things also are on the horizon, and will be released next month. I can't talk about them here just yet, but from what I've seen they will be more than worth waiting for.

How is Zoho built on the back end? I wasn't surprised to learn that it is built entirely on open source software and open standards. The server farm they use runs a highly customized version of Linux (CentOS), recently migrated to the 64-bit edition. They originally ran Debian, but were disappointed with the glacial pace of security updates and other fixes, and so switched away. Security on the back end was hugely important for them, and they weren't comfortable with what they got from Debian in that regard. The rest of the stack consists of things that should be familiar to any open source guru -- like MySQL and Tomcat.

One of the big issues that had been on my mind while looking at Zoho was how the product was most hemmed in by the fact that it was an Ajax-driven product. There are limitations to how much you can do reliably in Ajax (although I'm constantly surprised at how clever programmers are working around those limits a little more every single day). Raju agreed, but at the same time made it clear that Zoho was going to stick with Ajax despite its limitations, precisely because it's an open standard, and not switch to something proprietary like Flash to drive what they do. They have used, and will almost certainly continue to use, custom browser add-ons to accomplish certain things, but from what I've gleaned those things (especially on the Firefox side) are actually received fairly well by users as long as they're optional and not mandatory for basic functionality. More so with Firefox than IE, since Firefox's plug-in structure is a lot cleaner and more straightforward.

Another interesting thing that came up as an adjunct to questions about their support for OOXML: Most Zoho users treat the system a bit like a wiki. They create and edit documents directly in Zoho, and then share them with others directly in Zoho as well, instead of re-exporting them to a common document format. I don't figure that means they'll de-emphasize export and import in the future (it's too useful), but it points strongly to how Zoho is used. (What's easier, to export and e-mail someone a document, or simply shoot them a link to a shared version that's already on the Web?)

That, in turn, brought up another major question that's been bubbling in my mind for a long time: How do you plan to win the trust of people who are leery of putting their documents on a server that's controlled by someone else? They're exploring that as well in the form of either client- or server-side encryption (or perhaps a combination of both), where the user can define their own encryption key and make their documents inaccessible to anyone else -- including Zoho's own engineers. The downside of doing the encryption on the server side is that it does slow things down; hence talk of offloading it to the client side, possibly through one of the aforementioned browser plug-ins.

The final big question was how they saw themselves shaping up against Google Docs. "Breadth and depth" was how Raju characterized the Zoho approach. "We see ourselves as the IT department for companies that don't otherwise have one." Their ambition is to create applications that cover a lot of different kinds of ground -- breadth -- and which also have a lot of useful features within them -- depth. This he explicitly contrasted with the Google approach, which seems to revolve more around simplicity and approachability, but which to him meant they wouldn't be as useful to companies. "We think businesses need more functionality [than what Google offers]," and it's hard to see how he's wrong. My hope is that even in as little as a year's time people will be able to go to Zoho and see -- and use -- the fruit of that ambition.