Adventures in home working

OneNote 2003 already has one of the most comprehensive implementations of “flags” that I have seen in any software,  and it’s getting even better in OneNote 2007, this post gives some details.  It’s worth reading for three reasons:

1. If you use OnNote you can learn a lot
2. If you are interested in the whole concept of how to use flags to catagorise the content your personal knowledge management system, OneNote can provide some inspiration, although tags also take some beating
3. The most important reason to read this post though - because the OneNote blogs (here, here and here) represent a superb example of how a product development team can use this new media to inform, build a relationship with and take feedback from their community of users and prospective users!

After reading about my rebuild nightmare you are probably thinking that I would be migrating fast to web 2.0 sites.  Wrong!  I really like some of the web 2.0 sites and work extensively in the browser - but I like to use the best tool for the job and right now I think that’s:

1. 2003 server
2. Driving 3 screens
3. Running most applications local but with background synchronization of data onto servers where it makes sense

I just love the customizability, the great integrated experience and the fact that from one desktop I can run local apps, smart apps, web apps, terminal served apps, virtual machines etc. 

I also love the fact that I can leave my desktop running and grab my Tablet and have everything synched but have a completely different environment optimized for a very different type of activity and different form factor.

So for me web 2.0 is a complement to the desktop, not a replacement.  My desktop has been rock solid for 2 years and largely because of my own messing around has probably cost me 3 hours of downtime to fix.  Not a bad trade off for the productivity benefits and fun I have had over those 2 years.

I switch to a new machine and Vista in a few weeks, I am pretty worried that I won’t get the rock stability I am used to but here’s hoping!

VMware is not fast at the best of times, here are some basic tips to make sure its not slower than it needs to be:

1. Make sure that AV on access scanners exclude .vmdk .vmem and probably .iso files as well
2. Store VM’s on a separate partition, preallocate disk space if you can and schedule a daily defragmentation
3. Add MemTrimRate=0 and sched.mem.pshare.enable=FALSE  to your .VMX file provided you are not short of host memory

More details of the rationalle for these recommendations can be found in this article.

I may be stange but I scan all of my feeds online and then read a small subset - generall about 20-30 a day offline.  I do the scanning while watching TV on my laptop and the reading on my Tablet.  I scan the feeds in FeedDemon and then read the associated web pages in Maxthon as a series of 20-30 tabs.

I noticed recently that there was a real buzz about Google Reader, so I gave it a go.  Since I scan online it worked pretty well, I really liked the single keys shortcuts for navigating between posts, however for me it was too slow,  when scanning it just could’nt keep up with me :-(.  So I went back to FeedDemon, the speed was good but I missed the shortcut keys.  FeedDemon does have keyboard shortcuts but they are all multi-key combinations, which I find a real pain.

I started to get annoyed,  in fact I was going to write a post asking for people to describe how they managed to efficiently scan thousands of posts a day in FeedDemon because it seemed to me that no end-to-end working practice had been developed that worked for people who scanned on one PC and read on another. 

Then I discovered Control-K which opens a dialog that leys you define single keys for anything you want to do in FeedDemon,so here is what I did:

N - takes me to next unread post,  I just press NNNNN until I see a post I want to read and then I press

L - which opens up the link in Maxthon.  If Maxthon is minimized and set to “not activate” when a newpage is opened it just sits on the task bar opening up tabs in the background

so here I am NNNNN and then my brain catches up and says you missed one so I press

P - which opens up the previous unread post, this works because I don’t mark posts read as I read them,  I wait until the end and then I press

R - which marks all posts Read.

So this is great however there are two minor issues - which hopefully Nick can fix:

1. I could not see a way to define a key to download a PODCAST
2. When I press L to open a link in an external browser I need to click back on the FeedDemon page to return focus to it before I can start to press NNNNN again.  The same thing happens if I click a link in a page or download a podcast.  This takes my hands off the keyboard - not desirable

So now I have loads of tabs open in Maxthon,  all I do is click Groups > Save Windows as Group > I have a group for my Laptop so I pick that (3 clicks).  Then back on my Tablet (via remote desktop, 1 click) I open that group (2 clicks) and I am done.  I can now read away on my Tablet closing tabs as I go.  If I don’t read them all I just save the few I have left back as a group and open that group up on my laptop before I start my next scanning session.

Actually my workflow is slightly more complex

1. I open up tabs in Maxthon because many feeds don’t contain the full text
2. After I have finished scanning I scan the tabs, because quite a few web pages have articles spread over multiple pages so I need to open up the other pages as well or show in print view
3. Quite a few web pages turn out to be a few lines of comment and then some links,  its often the article or articles then link to I want to read so I click to open up those
4. Sometimes I come across new sites I want to subscribe to, so I generally subscribe to them while online.
5. Sometimes I have opened links to files I want to download,  so I sometimes initiate the downloads there and then,  or if I am offline when I come across the download then I add the site to a Maxthon group called Action and come back to it later.

A post on the SharePoint blog on the potential for SharePoint to replace file servers has stimulated a good debate.  I replied to it here and Joel (the original author) replied to me here (thanks for taking the time and effort Joel!).  Initially I was left feeling less than convinced that Joel had addressed my concerns and considered a counter comment, however now I think a new post is required because of the whole I don’t think we disagree - I think we just have slightly different perspectives.

Rather than continue to throw around th pros and cons of the file server, maybe its better for me to describe the way I prefer to work, which blends the use of file storage (local and server) and SharePoint.  Neither is the total solution. 

Maybe that will help both Joel and I iterate to a better understanding of each others positions.

First off let me say that I have been using SharePoint since version 1 and have used Notes, Quickplace, WebSphere portal and various other collaborative workplaces over the last 10 years and my view of their role is simillar.

I consider SharePoint a great place to publish but not a great place to work,  so here are a few details of how I breakdown working with files and publishing them.

How I work with files:

1. I prefer working with files stored on my local hard drive.  Every application - not just Office - works great with the local file system and its quick, reliable and simple to use.
2. I have pretty much every file I have ever worked on and many reference files on my local file system and I can search the 30,000 files in a flash using X1. In addition I get fast high quality previews of the files I am searching.  In the last 10 years if I had saved files directly into collaborative spaces they would be spread across probably a 100 different workspaces and be impossible to find. 
   — Impossible might seem strong but remember these files would have been scattered across collaborative workspaces from several customers and partners and several of the systems that contained the files have now been decomissioned.
3. I re-use data from old files every day,  I don’toften go back 10 years but I go back 3 years every week.
4. I am really looking forward to Vista’s ability to combine search with tags and other meta-data and again I stress the importance of the ability to work with all my files, across customers and projects.
5. I love the fact that files in my local file system are available almost instantly with no network delays or reliability issues and that I can replicate them easily to my Tablet for mobile working.  I have found replication to be 100% reliable even with plenty of adds, deletes, renames and edits at both ends (obviously not at the same time)

How I backup my files:

1. I backup my files on-site using robocopy to mirror my master file system to another machine each night
2. I backup my files offsite using Iron Mountains connected,  which is fast and very efficient, especially when multiple users have many duplicate files (as it uses a single instance store).  There is a web interface to Connected as well to provide anywhere access (but I have never needed to use it).

How I publish files:

1. When I am ready to share files I need to publish them, if I am working with a small team all sharing a file server, this is really easy - just mirror (or copy) my files up to the file server.
2. Sometimes I use Groove as a way to share files between team members and it works well, but it’s expensive.
3. More often I publish the files to a collaborative workspace and the discontinuity always annoys me,  the fact that I have to browse to a document library and upload the file, or drag and drop it using web-dav, often needing to add document properties.  SharePoint makes this a bit better because it can use property promotion to populate server meta-data from document meta-data. 
4. Sometimes I copy files using web-dav but the poor web-dav integration always annoys me,  I particulary don’t like the fact that web-dav support rarely includes surfacing document properties in the shell or providing right click actions.  Now that most of the functionality in collaborative workspaces is implemented by well defined web services I wish shell extentions could be written that exposed this functionality via Windows Explorer as well as the Browser, much as SharePoint v1 did.

How I publish status:

1. When other people need to track what I am doing I don’t find uploading individual files very useful.  Instead I prefer to post  documents with embedded documents or more often links.  That way I can provide people with access to sets of related files with approporate comments.  Its even better if people can comment back to me.
2. Discussion databases can be used in this way and I often use Notes databases, but I think I have a preference for blogs to provide this function.  Now that SharePoint includes blogs this is a step forward, especially because I can use LiveWriter which allows me to save drafts in the file system and also easily edit previous posts.  Even better LiveWriter keeps track of all of my posts to all of my blogs and because all of its data is in the file system I can replicate it between multiple machines and easily back it up.

How I publish ideas:

1. Blogs are definately my favorite place to publish personal ideas, although I prefer discussion areas for developing ideas.
2. I also like wikis but prefer them for publishing ideas that need collaborative effort over time, for example a process or service definition that I know will evolve.  I am concerned though that wiki content is more difficult to reuse as it gets locked up within a particular wiki.  For someone who is working on many projects having my contribution to each project locked away in a wiki silo is an issue I have not come to terms with.

Team working:

Lots of the comments above touch on team working, but here are some particular notes:

1. When I work with a team on a file server every team member gets a personal working directory.  The owner of it gets READ/WRITE access and all other team members get READ.  This means there is never a risk of conflicts and everyone can easily track everyone elses activity.  This is especially true with search that makesit really easy to aggregate files from across multiple peoples working progress areas.
2. When working with a team across multiple companies, I prefer to use Groove,  the fact that it works offline and syncs in the background well makesit much more likely to be used.  I also like the fact that it provides a multiple tools including a discussion area where people can refine Ideas and post status information.  We sometimes use it to provide a sort of team blog.  Everyone I know who uses Groove wishes it included RSS notifications so that we only had one place to look for change (our RSS readers).
3. When working with a team within the company my preference is to work locally on my desktop and Tablet and publish to a collaborative space at key points, if I am working with a small team I will also write a highlight report with links to documents, blog posts, useful web sites etc.
4. I think RSS will have a huge impact — making it much easier to coordinate activities within and between teams as well as many other scenarios.

In summary I am essentially looking for a more integrated and seamless experience, hesistate to propose solutions because there are people better at figuring these thing out than me but here are some things I have seen and liked. 

1. I really liked the idea of accessing web based document libraries through the windows shell, especially if I have right click menu support for all key actions like check-in/out.
2. I really like the idea of being able to perform actions offline and then have these actions execute transparently when my PC detects that it’s online again.
3. I like the idea of making RSS subscriptions to document libraries and associating them with folders in the filesystem and seeing file enclosures appear as if by magic.
4. I want all change reported to me through RSS and sometimes I want enclosures to be delivered as well. 
5. I want intelligent handling of URLs, ie if I am offline I want them to resolve to local replicas or caches on-line I want the option (Lotus Notes handles this pretty well)
6. I want tags to work across applications and not just at the file system level. 
7. I like the Vista idea that allows applications to register preview, search and property handlers to create a more unified experience
8. I really like the “blog this” concept and would like as many applications as possible to have one click integration into Live Writer or simillar.  That way I have a single place to manage blogs and persist drafts and previous posts
9. I like the idea of Groove but I would like a central Groove server as well so that a star topology was supported for replication as well as a mesh.  Sometimes with small teams individuals working across time zones are not online enough at the sametimes.

A few customers have recently pointed out to me that they are going to have a significant number of early adopters of Office 2007 in advance of their migration to SharePoint 2007, this will be particularly true where SharePoint is used at an enterprise level.  Previously I had assumed that SharePoint (being server side) would migrate in advance of Office.

This got me thinking about how little I had seen from Microsoft about the issues of compatibility between mismatched versions of SharePoint and Office.  As my area of interest right now is Office 2007 used with SharePoint 2003 here are a few of the issues I can guess (possibly incorrectly) at and there will probably be a great many more:

1. Groove 2007 only provides support for offlining document libraries whereas the previous version also supported discussions and some other tools.  So users of Groove 2007 are going to see reduced functionality - assuming Groove 2007 works with Office 2003 at all. 
2. Although some of the missing functionality in Groove 2007 migrates to Outlook 2007 this migration will only be relevant when the SharePoint 2003 environment is migrated to 2007, in addition it’s only likely to be suitable for users of Groove within the enterprise.
3. Excel 2007 doesn’t support bi-directional replication of lists and publishing of lists to SharePoint 2007,  although existing lists created with SharePoint 2003 will continue to work.  I am assuming that Excel 2007 when used against SharePoint 2003 will have the same limitation. So users will see a loss of functionality, similar functionality will be available in Access 2007, ie the ability to publish tables to SharePoint and take bi-directional replicas offline, however this functionality in Access 2007 will only be relevant when the SharePoint 2003 environment is migrated to 2007, and of course you need to have a SKU that includes Access.
4. I don’t expect that SharePoint 2003 will support the new Office XML file formats for searching, and document property promotion and demotion.
5. In SharePoint 2003 documents were checked out to a hidden area on the server and so could only be worked on online.  SharePoint 2007 and Office 2007 now check documents out to a hidden directory under [My] Documents, I am not sure which behaviour if any will work with mixed versions
6. Word 2007 includes a rich interaction model with SharePoint - which features of this will work?

If anyone knows of a good information source on this topic, please leave a comment, or if you know of any issues I have missed or issues I have highlighted incorrectly please leave a comment as well.

I have been reviewing a friends presentation for a few days, its content is complex and very interesting, the slides have a lot of information and he has included extensive speaker notes.  However I have struggled to follow the story he is trying to tell and come up with a memorable takeaway from the presentation.

So today I started to try and capture in 1 line the key point from each slide, several of the slides had the same point and some seemed to have no point that was relevant to the story being told.  I got about half way through rationalising and sequencing when looked back at my work and realised that it was actually much easier to understand the story as a set of key points that proceeded in a logical progression than to look at the presentation. 

Tonight I found myself unable to sleep and came across an old post by the great Peter Cochrane.  In it he condenses a keynote he had recently given down to a sequence of key points In a couple of minutes it was pretty easy to understand the story he was trying to tell and also to critique it, exactly as I was trying to do, so I am inspired to continue.

Now I have no doubt that a well crafted presentation can have considerable impact especially if delivered by a skilled speaker however I think a list of key points presented in textual format as a story is much easier to distribute, digest, critique and remember.

Here’s the example I am referring to:

1. Throughout the 70s, 80s and 90s people who knew deep technical stuff (nerds) were derided and discounted. The management attitude was that these people were irrelevant and a pain. Deep tech understanding was not seen as necessary to manage anything. How the world has changed - today some of the richest people in the world are ex-nerds!
2. This retrograde management attitude had a lot to do with the greater than 85 per cent failure rate of IT programmes through that era, that continues today in industry, defence, education and healthcare. Know-nothing managers are a menace to any industry and profession.
3. Not including the end user, not understanding the technology and not understanding the difference between data, information and knowledge is not only dangerous - it turns out to be very expensive!
4. The biggest universal mistake has been to take the old paper processes and transplant them to the screen, and then create even more paper! IT presents a much bigger opportunity to change organisations and operations but, unfortunately, people seem unable to adapt and change in more than one dimension at a time. Contrast the old (50- to 100-years-old) companies to the new (10- to 20-years-old) and it is stark in the way they use IT to create, run and advance the business.
5. Increasing numbers of mobile workers means the notion of centralised databases are going to be more difficult to sustain. In a lot of companies the transition will be from filing data away in a predetermined structure to finding what a worker needs when they need it, a far more Google-like existence.
6. We have to think about data, information and knowledge being collected, collated, created and stored by a wide variety of sources and not just think in terms of centralised operations. Mobile workers and young people are a new source of everything. For example, the young jump straight to Google and Wikipedia as their sources, whilst a mobile workforce is hunter/gatherer-like, collecting and creating on the move to meet their immediate need. This is a far cry from the deskbound cultures of old.
7. And then there is modelling, any kind of modelling, from crowd behaviour and flow in a retail store, to market modelling and ecological economics. We can no longer afford the crude management decision-making tools of the past as it is all getting far too complex for the knee-jerk reaction!
8. CIOs and their teams have a prime responsibility to keep companies and boards ahead of the game. They have to be the IT and potential threat radar, the thinkers, the modellers, the guiders of the corporate hand. Keeping on top of the latest Office patches isn’t where the action or responsibility lies.
9. Young people will help transform everything. They think and act differently and come with new expectations and skill sets that often outclass and outflank the established order of the IT department. Most likely they will not work for a centralised and controlling regime, and will certainly usurp the old ways of doing things. This new attitude and skill set needs to be embraced as an opportunity for change rather than being a target for punitive action.
10. Everything is moving to the edge of networks and organisations - computing power, communication, skills, information and knowledge.
11. Increasingly the future will be about taking risks - not blind risks - but calculated, modelled, tried and tested risks. And the CIO has a new and key role in the process. IT isn’t an adjunct function of the company; it is central to success and as such needs to be recognised as an asset by boards and managers in general. Unfortunately, IT is one of today’s least-loved corporate functions and seen as some form of creative chastity belt. This has to change fast if organisations are to grow and prosper - the clock is ticking!

I love this article by Joel Spolsky which talks about the tendency of some architects to over abstract.  I have spent many years as an architect, just as many as a systems integrator and quite a few as a developer, manager and project manager but I know what he’s talking about.  I sometimes find myself doing the same thing.  So every few months or so I make a point of reading his article just to remind myself not to fly too high!

Here are a few words from the introduction to tempt you into reading the whole thing:

When great thinkers think about problems, they start to see patterns. They look at the problem of people sending each other word-processor files, and then they look at the problem of people sending each other spreadsheets, and they realize that there’s a general pattern: sending files. That’s one level of abstraction already. Then they go up one more level: people send files, but web browsers also “send” requests for web pages. And when you think about it, calling a method on an object is like sending a message to an object! It’s the same thing again! Those are all sending operations, so our clever thinker invents a new, higher, broader abstraction called messaging, but now it’s getting really vague and nobody really knows what they’re talking about any more. Blah.

Much as I like full text search tools like Windows Desktop Search and X1 they are not really optimized for searching ALL of the files on my PC just by filename, and file meta-data.  So I was pleased to come across Locate, on my PC it builds an index of 365,000 files and 30,457 directories in under an hour and is lightning fast at finding them again - even when searching every file on your PC.  I also like the fact that the results list is like a normal explorer window and supports all the normal right click options etc.

In fact as I have used Locate over the last week or so I realized that most of the time I don’t need full text indexing at all.

Stu has tagged me and since he’s a friend of mine I will oblige, however not being a fan of chains this branch will end with me.

Thinking of 5 things you don’t know about me is a bit if a challenge as my blog contains a lot of information already including an overly long 100 things about me list (apologies for duplicates).

However here we go 5 things you might not know about me:

1. By the time I was 16 I was making a good income - for a school kid - with a large window cleaning round. By 17 I was also employed by my school as a handyman, and had the keys to the whole school, I spent most of my early evenings repairing desks, doors etc.  By 18 I was working in the holidays as a painter as well.  It’s not surprising therefore that although I didn’t get pocket money I was the richest kid in school!
2. My first real hobby whilst I was in sixth form was scuba diving and I payed for all the gear, holidays and training myself using the income from my jobs.  I really enjoyed it but eventually gave up on it because it was too time consuming and diving off the coast of England lost its appeal.  I was also a competitive swimmer and long distance runner both of which helped me progress rapidly as a diver.
3. I have lived in 17 different places, excluding visits and holidays, but including 2 long stays in hospital.  Several of these “homes” have been pretty grotty, but I was happy enough and sometimes it’s good to remember that.
4. I love kids, especially babies - which I am very good with - but after giving birth to 4 great girls my wife is not too keen on any more.
5. I have devoted most of my working life (20 years) to improving the day to day productivity and enjoyment of the people I deliver IT solutions too.  I almost never think of IT in terms of the technology, it’s all about the people who’s lives I am going to impact.  Several times I have delivered systems that might have resulted in people losing their jobs but in every case I found ways to keep them gainfully employed.  Despite having mild Asbergers syndrome I get very concerned over the welfare of my peers, team members and customers and although lots of people tell me this is a weakness I consider it a blessing.