Adventures in home working

I have always looked upon InfoPath as a example of a product that needs to be part of the infrastructure of the Longhorn platform.  At its simplest it’s a product to render forms defined in XML, allow them to be completed offline, validated, and then submitted them to web services. 

If you think of WinFS as effectively an XML store, which manages sometimes connected interactions with server side data sources (especially web services) then InfoPath type capabilities are a natural part of the WinFS shell.  So I was interested to see this MSDN paper on Submitting forms in InfoPath 2003 because of the potential implications on how Microsoft is thinking about WinFS and Synchronisation and sometimes connected operation.  These new adaptors allow:

1. Submitting to a Web Service
2. Submitting to a SharePoint Site
3. Submitting through E-Mail
4. Submitting to a Database

These new capabilities are interesting but the ability to complete the form off-line and then, when connected, send it to the server is still way to clunky (but likely to be a key area the Longhorn team will need to make slick).  

I was also disappointed that they did not include submitting via email to a SharePoint Site.  This to me would be a fabulous feature, allowing business to business transfer of XML data from a rich client, to a (fairly) rich XML store right through firewalls with no need to establish an internet facing web application.  As anyone within a typical enterprise who has tried to get a new application hosted on an Internet facing web site will tell you, it’s HARD WORK!  So this email adaptor would be especially powerful for short lived collaboration scenarios, or to support high value transactions between relatively small companies who don’t want the overhead of building web applications.  

Of course the real draw back right now is that InfoPath is part of Office Enterprise and Office Enterprise costs money.  Built into the platform however you can just image how this sort of technology could transform the ease with which structured and semi-structured data would flow between companies.

I have been asking myself the following question:

If Linux begins to capture significant desktop market share what options are open to Microsoft? 

In this article I have listed my initial ideas, and it would be great to get some feedback on the technical feasibility (some of them maybe plain crazy) and political acceptability of the options within Microsoft.  It might also be interesting to get feedback on how the Open Source and business communities might respond.

Here is my headline list, with each option described in more detail later:

• “bet the company” on strategies to retain the consumer market
• Make Windows a better host for Linux applications
• Make Windows a better server for Linux Desktops
• Make Linux a better host for Windows Applications
• Make .NET the most attractive Linux Development Platform
• Make Windows a better client to Linux Servers
• Make Windows appeal to Open Source developers
• Win the TCO and Security debate
• Reduce the cost of Windows and Office

Retain the consumer market

1. I have already blogged on this here

Make Windows a better host for Linux applications

1. Purchase an existing X Server product to integrate into Windows Services For Unix, continue to give SFU away for free
2. Extend SFU to create a LSB 2 compliant sub system for Windows.  I don’t know enough about LSB 2 to know how practical it would be to achieve this, but my guess is that since LSB 2 leans heavily on POSIX and SFU is a POSIX sub system it should be possible.
3. Release SFU as Open Source, possibly merging it with Cygwin and Cygwin-X and sponsor/support the Open Source community in achieving option 2

Make Windows a better server for Linux desktops

1. Make Windows 2003 + SFU a viable NFS file server, release a Linux client for its Volume Shadow Copy functionality (previous versions client)
2. Develop its own RDP client for Linux or sponsor and support the development of rDesktop.
3. Work with Sun on other interoperability areas including authentication and authorization, directory enabled policy based management of Linux.  Including management of their own LSB 2 subsystem.
4. If Avalon is ported to Linux make sure that a next generation Windows Terminal Services client that uses Avalon works on Linux as well

Make Linux a better host for Windows applications

1. Sponsor and support the implementation of the .NET Framework on Linux.  The Mono team are doing a pretty good job on their own, but official support from MS would be worth a lot.
2. Support their own .NET applications on Mono

Make .NET the most attractive Linux development platform

1. Invest in making .NET on Linux as functional as it is on Windows and vice versa
2. Provide other WinFX services like Indigo and Avalon on Linux as well
3. Release .NET development tools that support both platforms
4. Extend developer support services, (MSDN), to .NET development on Linux
5. Support Linux on Virtual PC and Virtual Server

Make Windows a better client to Linux servers

1. Include NFS client capabilities in Windows, (part of SFU)
2. Include an X Server in windows for access to GUI applications running on Linux servers
3. Work with Sun on other interoperability areas including authentication and authorization
4. Port Indigo to Linux

Make Windows appeal to Open Source developers

1. Make the user experience as attractive as possible
2. Create tools that can target .NET development on both platforms
3. Support Linux on Virtual PC
4. Provide a great command line shell
5. Prove that developer productivity is higher on the Windows platform
6. Use Microsoft’s marketing expertise to make Windows cool again, learn from Apple J

Win the TCO and Security debate

1. Prove that Microsoft’s integrated solution stack has a lower TCO and is just as secure as Linux (not easy)
2. Demonstrate that the integrated nature provides a better user experience and allows them to extract more value from their investment 

Reduce the cost of Windows and Office

1. These are Microsoft cash cows, and it would certainly hurt Microsoft badly if it needed to reduce their profit margin on these products.  However the potential is there if retention demands it

A colleague of mine recently asked me what my IT research interests are.  It got me thinking; I am interested in lots of things, but only a few that I am prepared to put serious time and effort into.  Here they are:

1. Personal Knowledge Management
2. Team communication, collaboration and co-ordination, especially ad-hoc inter-business scenarios, but also more formal team, project and programme collaboration
3. Mobile working
4. Home working
5. Workplace design and workstation ergonomics
6. Office tools and associated office systems, especially the transition to XML enabled office tools interacting with collaboration services and client/server XML aware databases for unstructured and semi-structured data
7. Client technologies, and platforms, especially as they relate to the above
8. Information lifecycle management
9. Personal productivity, especially as it relates to the above
10. Best practices and processes associated with the above
11. Client application development and delivery technologies and associated user interface approaches
12. Digital rights management and evolving trust and assertion based security models
13. Consumerization of technology and the impact of consumer provisioning experience on enterprise service provisioning
14. Enterprise infrastructures and especially how these need to evolve due to extended enterprise and flexible enterprise pressures as well as some of the above
15. Enterprise infrastructure programme management and programme design
16. User satisfaction criteria as distinct from budget holder satisfaction criteria
17. The totality of the end user experience, rather than vertical optimisation of individual processes
18. DCO, TCO and TVO modelling and business case development, especially as it relates to the above

Michael; who writes the Shared Spaces blog has recently written about the challenges of collaborative editing of documents/presentations.  The problems not too difficult to solve within an business (Net Meeting etc) but solving the problem between businesses (through two sets of firewalls) in a secure fashion is another problem altogether.  Its even more difficult if like me you want to do it on an ad-hoc basis, during a telephone or IM conversation, rather than using a multi-user conferencing system like Lotus Same-time, Oracle Collaboration Server or MS Live Meeting.

These are the options that Michael came up with, as these all cost loads of money, require a client installation and lots of coordination between collaborators they don’t fulfil my ad-hoc need, any other areas would be welcome:

• Groove Virtual Office Professional 3.0. Staff put the document to be shared into a Groove Document Review shared space. Internal and external participants are invited. Using the “Co-Edit” function of Groove, one person in turn can have edit rights of the document, and once they save their changes, everyone in the co-edit session gets to see the changes immediately. It’s not true real-time co-editing, but it’s extremely close … and it definitely solves the problem of having to email the changes around again. See www.groove.net
  
• GoToMyPC Corporate. The document owner invites the other party to a real-time co-editing session by enabling controlled screen sharing. The document would be opened directly from the document management system, and checked back in immediately upon finishing the edit session. Both participants can share editing of the document. No extraneous copies of the document are left lying around, thus solving the content security problem. See www.gotomypc.com/corp
  
• Authentica Secure Office. Authentica Secure Office enables staff to set access rights on documents—such as the right to read, the right to edit, the right to print, the right to copy text to the clipboard—for named individuals, as well as global rights for document expiration regardless of physical location. Secure Office does not deal with the problem of real-time co-editing of a Word document. It does, however, significantly reduce (or perhaps entirely eliminate) the risk of unauthorized disclosure of document contents. Hence, the document could be emailed as normal (or put into a Groove shared space), but no authorized access would be possible, since it was tied to an explicit rights policy. There are other companies that play in this space too. See www.authentica.com

This is a great series of articles on different blogging styles, it includes hints and tips on when to use each one and how to use it to best effect.  Well worth a read even if you don’t blog, as some of the insights are useful for any type of written communication.

I tend to focus types 1 to 6 myself.  Although I often promise to get around to doing a series, for some reason I seem to loose interest – definitely any area where I need to try harder.

After reading blogs for awhile, I’ve come to see some patterns in the ways postings in text-based blogs are formatted. As I see it, there are seven basic formats for blog postings. Each serves a different purpose for bloggers and their readers.

The format of a blog posting, if chosen consciously and carefully, enhance communication – particularly the delivery of certain types of content. Consequently, some formats work best for commentary or explanation, others for alerts and references, etc.

Here are the seven basic blog posting formats:

1. Link-only
2. Link blurb
3. Brief remark
4. List
5. Short article
6. Long article
7. Series postings

Keep in mind that these types represent points along a spectrum. A specific posting might blend aspects of two or more formats.