Microsoft lacks a roadmap for its Exchange Server product (although one is due next week apparently), which I discussed last month. Chris Williams of Sirana Software, who previously worked on the Exchange product team for six years and was a senior product planner for Exchange Server 2000, has published a two-part manifesto for Exchange Server (short-term and long-term). I applaud Chris for publishing his thoughts, and in this post I want to consider the points he raises in light of my ruminations on this topic.
Response to Short Term Proposals
Proposal 1. Address the Issues Raised in the Newsgroups
Chris proposes that up to 50% of time spent on the next release of Exchange should be focused on addressing the issues and problems administrators and users complain about, eg, the lack of customizable warning messages. He calls this “delighting your customer”. Although Chris doesn’t concede it, I see this as validation of my original proposition that as a mature email and calendaring platform, much of the future for Exchange is merely “tweaks around the edges”.
Chris advocates that if this “small time stuff” was resolved, then customers will see the value of the newer versions of the platform and upgrade. However, none of these small issues would solve a fundamental roadblock to adoption, that being the forced embrace of Active Directory from Exchange 2000 onwards. With 40% of the installed base still on Exchange 5.5, that’s a real hurdle.
Proposal 2. Better Product Planning Disciplines
Perhaps as a throw-away line, Chris says that he sometimes thinks that “Microsoft builds its product plans based on whomever showed up in the Microsoft Executive Briefing Center last year”. Unlike Chris I haven’t worked in product planning at Microsoft, but if that’s true then indeed Microsoft needs a better approach to product planning that considers the needs of the small and medium-sized business markets, not just the top-end enterprises.
Proposal 3. Resolve the Spam Issue by Buying Brightmail or Cloudmark
The anti-spam capabilities of Exchange are delivered via its Intelligent Message Filter, which Chris notes as being “pretty decent”, but lacks the surrounding policy management capabilities that commercial products such as Brightmail or CloudMark deliver. I agree that a robust resolution of this issue in the short term is critical.
Proposal 4. Enhance the Disaster Recovery Capabilities
Chris advocates that Exchange requires better disaster recovery capabilities, particularly for the “casual recovery” of individual items, and notes a couple of vendors who deliver much better offerings in this space. Again, I agree that more is needed, and since the underlying data store is here to stay for the next 5-7 years, Microsoft should make a concerted effort to raise its game in this area.
Response to Long Term Proposals
Proposal 1. Integrate New Communication Technologies into Exchange
At the beginning of his discussion on the long term, Chris suggests that Exchange could serve up blog content, just like Domino Server can. He references an existing blogger who “already uses a form of Domino” for blogging. (Aside: the phrasing of that is wrong … the blogger uses a Notes blogging database that runs on the Domino Server platform).
Some thoughts:
- A different product from Microsoft, Windows SharePoint Services, can be used to deliver a blog today.
- The comparison to Domino illustrates the key architectural difference between the two platforms: Exchange tightly serves email and calendaring, whereas Domino tightly serves collaborative applications, of which email and calendaring are two examples. Blogging applications are another. Web sites are another. Collaborative workspaces are another. Bespoke databases are another.
- Microsoft has abandoned the “Exchange do-all” strategy of 2000, in favor of a series of server products that work together to deliver communication and collaboration capabilities. I don’t see how it could do a U-turn on this without looking really silly.
Proposal 2. Re-Brand Everything as Exchange
Chris’s essential proposal for the long term strategy for Exchange is to re-brand the other servers in Microsoft’s collaboration lineup as Exchange Server add-ons, and to offer a graduated level of integrated offerings for the small, medium, and large enterprise markets. Eg, a base offering of email only, through to an enterprise edition with all of the features. However if it was only a re-branding issue, it doesn’t solve the complexity of having multiple servers that require individual care and attention. That’s why I think IBM has a better product strategy in Workplace—a single install with check-box provisioning of capabilities to individual users.
In terms of his proposal, I think Chris misses the point about the small and medium enterprises … they need all of the capabilities (email, IM, meetings, collaborative workspaces), but the number of servers to install and manage is too many. A product like Groove Virtual Office, with its integration of all of these capabilities into a single client, makes much better sense for these two markets.
Some thoughts:
- It allows for more focused software development efforts by the individual Microsoft development teams, rather than the creation of a monster process for coordinating all of the different intended capabilities into a single server deliverable. On the other hand, if Microsoft is unable to coordinate the development of such a platform, it pushes a higher total cost of ownership onto its customers to coordinate the implementation of such a platform.
- The various products are at different stages of maturity. Exchange is very high on the maturity scale, whereas Portal Server, Live Communications Server and Live Meeting are lower down, and Windows SharePoint Services is even further down. Microsoft shouldn’t push for technical integration until there is a more even mix of maturity across the different capabilities.
- Having a multitude of servers enables large enterprises to design a technical topology that optimizes the unique requirements of each class of offering. The email topology will differ from the presence topology, which will differ from the portal topology, for example.
- Microsoft can gain revenue for each server and a user CAL, rather than bundling the newer capabilities directly into Exchange, and then finding it difficult to raise the price.
However, re-branding the various products would deliver a simpler go-to-market marketing strategy, but it would require much greater internal collaboration by the currently disparate product development and planning teams.
Finally, if Microsoft did re-brand everything under a single banner, I suggest going for SharePoint and dropping the Exchange name altogether. It would signal that something big and new is on the market.
What was your reaction to the points Chris raised? Please drop me a line to discuss (michael.sampson@shared-spaces.com), or leave a comment below.
