Archive for April, 2008

If you like Twitter, you’re going to LOVE Jott!

Written by Cornelius J. van Dyk on . Posted in Blog

OK, so I’m testing the waters with all this new micro-blogging technology. The question is this… just what benefit, other than being able to spout off something random that comes to mind for your Twitter minions to see, does something like a Twitter have for a professional user. This is what I’m trying to figure out. There are many ways people suggest using it and probably the most notable would be to have private networks that function conversationally. But then again I’m left wondering… what about IM? To really be useful to the mobile workforce, a technology must be on the mobile phone. With the use of tools such as TinyTwitter, you can get your tweets on your mobile phone, but in the private network context, how is that any different from email? Mmm…

Anyway, so as I’m left pondering the value of Twitter, I started working on a pet project for the mobile phone. My endeavor led me to search for a speech recognition SDK for Windows Mobile. Alas, there is no such thing, at least not one that includes recognition components, at the present time. Anyway, my search for the SDK led me to the Windows Mobile Speech newsgroups where I found a reference to a new service called Jott. I checked it out and was immediately intrigued by the possibilities.

What does Jott do?

In short, they provide a service for taking voice and converting it into text as well as the ability to pass said text on to any other service out there.

OK, so that’s very nebulous right? So consider this. Almost anything that is done in text today, can be handled through voice with Jott. Having just found it, I can already identify several time saving options:

  1. Twitter plugin. Jott already has a Twitter plugin so you can tweet to your heart’s content simply by leaving a voicemail. No more typing while driving. Talking on the phone, while still not recommended, it much, much safer and can at least be done hands free and more importantly, heads up and looking outside.
  2. Google calendar plugin. I know, I know. I’m a Microsoft guy and I live in the SharePoint world. Unfortunately, I have not found a good way to keep my personal Outlook calendar and my wife’s personal Outlook calendar in synch. We are rarely if ever on the same network and I don’t have an Exchange server. So I used the Google Calendar Synch plugin to bi-directionally synch her calendar with my Google Calendar and then using the same plugin on my machine, can get updates made between my personal Outlook calendar and the Google Calendar. I never put anything personally into my Google Calendar but always via Outlook which makes it a disconnected solution. So now anytime I make an Outlook calendar entry, it pushes (when connected) to the Google Calendar and then down to my wife’s calendar. The reverse also applies to her making appointments. No more trying to synch calendars manually. Well Jott just took that to a whole new level because its Google calendar plugin allows me to call and setup an appointment via the phone and have it push to both Outlook calendars. I hate typing on my mobile phone, even with the QWERTY keyboard of the Tilt… and now I don’t have to.
  3. Notification groups. OK, so let’s say you have a large group of people that need to be notified of a certain event, such as the birth of a baby or something. In the past, you’d have to spend quite a bit of time notifying everyone individually. With Jott, you can import all your contact and then simply create a group. Then when you call in, you just send the Jott to the group and Jott will automatically notify everyone through either email or text messaging. Pretty sweet and a real time saver.
  4. Replying to text messages or email. OK, all mobile professionals read email on the go right? But if you’re running to make your connecting flight, you don’t have time to stop and send an email or text message. WithJott, simply call and leave a voicemail to take care of it.

Like I said, these are just the first thoughts I have for my own personal use. The service’s voice recognition is pretty darn good too! Even though the text messages could be better, the email even allows you to playback the voice recording and they don’t have any adds cluttering up and slowing down their site or email.

This seems like a good service to me. I’m certainly going to play with it in more depth. I would put them in the same space as Grand Central who, as you know, got bought out by Google recently. The technologies are not competing, but complimentary so I would not be surprised if either Google or Microsoft targets Jott in the near future for a buyout.

OK, so are you ready to check it out? You guessed it…




BEWARE! The age of TWAM (Twitter Spam) is upon us…

Written by Cornelius J. van Dyk on . Posted in Blog

OK, I’m trying to give Twitter ( a fair shake. So far there’s quite a bit of time consuming fluff chatter on there, but there’s some tweets that are actually pretty good. Unfortunately, when you’re tweeting the same d@mn message fifty times per day, your a$$ is going to get blocked. That puts you in the same league as certain buzz-ards that steal other people’s content. I’ve only blocked two people so far, but I think I need to be very diligent about who I follow. There is no way I could follow, or would want to follow, 20000+ people! Anyway, we’ll see how Twitter works out for me. So far so good…



VMWare Snapshot inspirations

Written by Cornelius J. van Dyk on . Posted in Blog

I received the following tips on the usage of VMWare Snapshots from fellow SharePoint MVP, Dan Holme. I thought it might be helpful to others…


      Firstly, keep in mind that SYSPREP does MORE than just “new sid”.  If you’re in a super-simple environment (i.e. 2-machine lab), newsid.exe is fine. In production you must also be taking care of other ‘unique identifiers’ such as WSUS client id, for example.  Be wary of using newsid.exe.  What I do on my VMs is bake a SYSPREP.INF script into the system, where all configuration is set except for anything I *want* to configure e.g. machine name.  I run sysprep, shut down, and snapshot.  The first time I return to THAT (sysprepped) snapshot, sysprep runs unattended and life is great. I then snapshot the “new” system (with name and uniqueness) as a ‘fresh’ capture of that machine.

Here’s some inspiration for newbies to snapshots:

This was my WS2008 RC0 system that I had set up for sessions at Windows Connections in Vegas last November.  You can see where I turn the server (sysprepped) into a DC then a separate snapshot with all the sample accounts (I use MS’s sample accounts for MS content and my own ‘fun’ names for Connections).

This is my standard Vista VM… you can see the two snapshots that “match up” with the above (to answer someone’s comment about working with multiple machines)

Note RC0/Office 2007 snapshot matches with “with clients” snapshot on WS2008 RC0 system

Note WSS START (paused/already started) snapshot matches with “WSS Baseline” snapshot on WS2008 RC0 system

And this WS2003 VM branch is used to spawn DCs for clients as well as dev/testing/demo for an application we built called MyGroups.  You can see how I’m leveraging the “Sysprepped” snapshot here.



Linked clones

Another great feature of VMware is LINKED CLONES.  The “Sysprepped” snapshot above has linked clones, which use the same “baseline” disk then create what VPC calls “differencing disks” (except VMware does it much more easily!).  The joy of linked clones is that you can be running them both / all at the *same time*, so you can have 2+ servers running, for example, off of the “Sysprepped” base, with only one instance of the ‘giant’ baseline disk for Windows.

Important tips

Running Server (or Vista) in VMware, always

  1. Set the mouse theme to None (turn off Aero)
  2. Turn off mouse shadow
  3. Install VMware tools for better drivers
  4. Turn hw acceleration in the guest OS display properties to full-on. 

Note that sysprep ‘resets’ the mouse and video drivers to ‘basic’, so you can either

  1. Reinstall VMware tools after sysprepping or
  2. go into device manager and update the driver (choose ‘pick manually’ and pick VMware’s drivers from the list).




What do I gain or lose by using SharePoint site collections rather than sub sites/webs?

Written by Cornelius J. van Dyk on . Posted in Blog

Terry, asked in the forums what you gain and/or lose if you architect your sites around site collections rather than sub sites/webs. Here’s my take on this issue: If you use site collections you lose:

  1. Aggregation/rollups. Components such as the Content Query Web Part that can perform rollups of lists across entire site collections and thus sub sites, cannot roll up across multiple site collections. This can generally be offset by the use of the Data View Web Part, but you need SharePoint Designer to use that web part since it is not available through the UI.
  2. Peer site references. You can define columns to source their data from peer sites within the site collection, but not across site collections.
If you use site collections, you gain:
  1. Strict security isolation. For example, the creation of site security groups for each sub site, makes those site security groups visible to any administrator of any peer sub web within the site collection. In strict security context cases, the names of these groups, which usually involve the site name, could reveal the existence of some site that you might not want everyone to be aware of. It just helps to isolate security to the site level.
  2. Better backup granularization. Doing site collection backups can be much faster and more efficient when sites are broken into many site collections rather than one.
  3. Better stale site management. SharePoint’s stale site management features are, as you might have guessed, tied to site collections. This means that if you have a site collection that contains 10 sites of which 9 are never used but one gets some usage, then that site collection, and the 90% stale content, never gets flagged for review.
  4. Better quota management. Quotas can be assigned at the site collection level. If the site collection contains multiple sub webs, it becomes harder to determine which site(s) are most active and consuming the majority of the allotted quota.