Things I've Learned Working From Home

In 2007, Maija and I moved to Finland, and I've been working from home ever since. What started as a 3-month trial that we assumed wouldn't work, has become a litmus for my organization; both our Executive Director and our CFO now work remotely, commuting back to the office only every few weeks. I'm lucky enough to get to stay home most of the year :) On my last few visits to the office I've been struck by the differences in what can be accomplished in the office, and at home.  I recently came across a post from Justin Wise, asking for product suggestions that could help remote workers. It got me thinking about the things I have to do intentionally, as well as the things my office counterparts struggle with. I'll share briefly some of the things I've learned, and then respond to Justin's request for products that would help.

Read More

Using OpenVPN in Windows 8

When I first started using Windows 7 x64 I had some challenges getting OpenVPN to work. I blogged about that here, and it's been the most popular post on this site ever since, by far. This morning I installed the Windows 8 Developer Preview, and thought I'd follow things up. The good news is that Windows 8 is remarkably stable/useable already. Getting OpenVPN to work was a cinch, but I'll describe the steps anyway as a couple of details weren't immediately obvious.

  1. Visit to download the client. Head over to the Community Downloads page to find it. For my installation I used the Windows Installer of version 2.2.1. Note that this installer includes the GUI, so you don't need to go anywhere else for that.
  2. Download and run the installation package. I didn't change any of the options during install. You'll be prompted to elevate to administrator, as you typically are when you install applications.
  3. Once installed, you'll need to copy your VPN settings as you would in any OpenVPN installation.


Now that you're ready to go, click on the Windows Start button (or hit the Windows key) to bring up the new metro-style start pane. Scroll/swipe your way to the furthest right pane, where your list of installed applications now sits. You should see OpenVPN GUI as the newest installed app. Before you click that bad-boy, you need to do one final step.

Right click the OpenVPN GUI app (or do something magical if you're on a touch interface) and the Metro options bar will slide up from the bottom. Click Advanced, and then "Run as administrator." If you'd not yet figured out how to remove applications from the Metro app list, you should be able to figure that out now from this screenshot ;)

OpenVPN GUI should now open, with elevated rights. It's normal behaviour is to hide in the system-tray, so you'll still need to double-click on it to launch your VPN.

Note that the above steps will only launch the app with elevated settings that one time. If you want to configure the app to run with admin rights automatically every time, which I assume you do... then you'll need to go back to the Metro start pane, right-click the app again, go Advanced and now "Open file location".  This opens the directory containing the shortcut. Right-click the "OpenVPN GUI" shortcut, hit Properties, and on the Shortcut tab hit Advanced. Check off "Run as administrator", click OK, Apply (you'll be prompted to give permission for this app to run as Administrator), OK, and you're done. Running OpenVPN GUI from the Metro start pane will now prompt you to click OK as you've been used to in the past.


Choosing a ChMS Vendor

Earlier in the year we began the search for a church management software vendor. I've been meaning to share the process we went through, as well as our results, but haven't had the chance to till now. We're in the midst of a large software project, replacing all our core business software with more suitable applications. One of the needs we recognised early on was for our pastors and ministry staff to have a tool that was truly well suited to their particular needs. The process we went through is equally applicable to most kinds of software, not just ChMS. We're using The Raiser's Edge for all of our larger constituent management, but we felt we owed it to the pastors to get them a best-in-breed tool, which we'd then integrate as necessary with Raiser's.

Update: A sanitized version of our Needs Assessment is included here. It will NOT match your needs exactly, but feel free to use it as a template. Please don't distribute it online yourself.

I spent some time looking online, and a number of blogs were hugely helpful in our early research phase, especially that of Joel Lingenfelter. After a few weeks, we settled on the following products to examine:

  • Fellowship1
  • Shelby Arena
  • Church Community Builder
  • ConnectionPower
  • ThinkMinistry MinistryPlatform

There are maybe 70 vendors in this space, but the above 5 felt the most robust and headed in the direction we were going. We ruled out ACS as we found them pretty much irrelevant for a modern, web-powered ministry such as ourselves. We also ruled out TheCity, as it didn't appear to have the right focus for us.  Defining Our Needs The most important factor to choosing the right product, is knowing what your needs are. We'd made this mistake in the past, so I was determined not to do it again. Over a period of a couple of months, my team and I spent time with many of our pastors, and one particular pastor in particular. We noted down all of the things they felt that software could help them accomplish, and together we fleshed these out into detailed use cases and requirement lists. We passed these back to the pastors for review, and eventually ended up with 30 pages of "Needs Assessment", clearly defining all of the requirements that we had from a ChMS product. Each of the requirement sections and subsections had priorities, and lists of actors that required this functionality. For anyone else considering this kind of project, I cannot stress the importance of taking your time during this phase, and getting it done right. Resist the temptation to look at the cool features of the pretty software. Unless your pastoral people really need their brains jogged to understand the possibilities, then any time you spend looking at software will be time you don't spend defining your needs. Trust me. Acquisitions Galore! We started this process in January 2010, at which time the products I listed above existed as the list portrays. Fast forward to today, and there's been considerable change to that list, which I think is important to convey before we go any further. Fellowship1 was acquired by Active Network. Active had been trying to buy F1 for 3 years we're told, and eventually the management at F1 felt that Active were heading in a complimentary direction, and could help them reach their goals faster. We generally didn't take this into consideration too much in our evaluation, as we've seen acquisitions go both ways. A few months later, we got word that TheCity had been picked up by ACS, which was a fascinating turn-around, given TheCity's very web-ie nature, and ACS's distinct lack thereof. ACS had also picked up NSpire a few months prior, which is one of the product we're migrating away from. I can't think that ACS will be having an easy time managing all of their new customers and codebases... but I wish them luck. The next acquisition was a major surprise to us. Given the number of vendors making ChMS software, and the fact that there were at least 5, robust, viable products on the market with very similar features, acquisitions and mergers were likely to take place. This is why we weren't particularly surprised about the F1 acquisition. I also spent a few days with Blackbaud in Charleston, and they indicated they were quite interested in the ChMS space as well, whether by development or acquisition. What took us by surprise, was when Active Network went ahead and bought out a second, top-tier ChMS product: ConnectionPower. For our particular needs, we found ConnectionPower to be the weakest of the products that we looked at, but it was a rich product nonetheless, with a strong customer base. In a period of 6 months, Active Network grabbed themselves around 3300 customers: major market share in this space. They've announced that they'll discontinue the ConnectionPower product, and roll it's unique features into Fellowship1 (which is being re-branded at some point). I'm skipping over anything else about ConnectionPower in this post, as the product is irrelevant now. The Evaluation Process I spent a few weeks getting to know the vendors, discussing their ideals and goals with their sales folks, and reading as much as I could find about each solution online (from the vendors, and more importantly NOT from vendors). We also reviewed pricing, technical requirements, corporate profile, customer feedback and various other things not relating to the actual useability or features of the solution. These last ones had significant bearing on our choice of solution, especially the presence of Canadian customers. We've learned to make this an important point for us, as have most Canadian entities looking at the US market for software; and it severely hurt the chances of a few products, specifically Shelby Arena and ThinkMinistry MinistryPlatform. At the time we evaluated, Church Community Builder also couldn't handle financial transactions outside of the United States, but more on that later. After that initial process, I used the Needs Assessment that I explained above, to build a scoresheet of functionality that I thought we could cover in a demonstration. We then scheduled demos with each of the vendors, supplied them with our scoresheet and full Needs Assessment, and assembled a team to evaluate the solutions. That team consisted of me, two of my technical staff (my DB guy and my training guy) as well as 2-4 pastors, depending on the day of the demo. The demos were all at least 3 hours long, some of them closer to 4 hours. During the demonstrations, I had each of the pastors with their own copy of the scorecard, marking down grades on the functionality as it was shown to them. Once the demos were done, then we compared the grades and discussed the solution at length. We settled on grades for each piece of functionality, and then reviewed the overall score and compared it to how we felt about the product in general. We awarded generous bonus points for things that smashed it out of the park. Once we'd finished the demos (which took the better part of 3 weeks) we started comparing the solutions to one another to try and normalize our scores. We did have to go back to a couple of products and see them again, as we inevitably missed things, or didn't realise something we should have asked beforehand. Once we'd massaged each solution's numbers to a point we felt was fair, then I built some cunning formulas. Mathematical Love After we'd spent 15+ hours looking at software, it was clear that ANY of them could get the job done for us. What wasn't clear was how each product performed overall against our specific priorities; there were so many trees we couldn't hope to see (or even remember) the forest. I decided to take a fairly mathematical approach, the description of which you can skip over if you want. I'll be including some PDFs and Excel sheets shortly, so you can reproduce some of this yourself, in case my describing this to you makes your eyes glaze. Since the pastors had given us priorities for each of their requirements, we could extrapolate these out to point pools, which when combined with the scores, would produce weighted averages. Put it this way: Priority 1 = 100 points Priority 2 = 75 points Priority 3 = 50 points If "Functionality A" received a 7/10, that's 70%, and if "Functionality A" was Priority 2 to us, then it got itself 52.5 points. Line up the points awarded beside the max points possible, run a weighted average calculation on those columns (technically a sumproduct divided by a sum), and you'll come up with a score referencing your priorities. We scored each individual piece of functionality this way, and then rolled those values up to subsections of functionality, which were globally prioritized and scored again. The ultimate result was a score out of a 100, for how well this product performed on the things that were most important to us. We also totalled the non-prioritized raw scores, to provide another evaluative point. Human Readability We charted all of that, so you could see the numbers of the competing products clearly. We then built out additional charts with other working-sets of functionality, such as: Young Adult Campus tasks vs Connecting a new visitor. We didn't prioritise these ones, just looked at the vendors' scores for each set. We also examined consistently highest scores, as well as consistently lowest scores, to see what further trends might emerge.  As you'll see in a moment, the scores were so close we really felt we had to try to push them harder to find weak spots. Fellowship1 vs Church Community Builder vs Shelby Arena vs MinistryPlatform The results that we came up with for each product are totally specific to our needs, a part of me is hesitant to post our grades and results online because of this. I will say this once again: the most important thing you can do, if you're going through this process yourself, is to define your needs. Exhaustively. Because you're going to spend good money, and a good amount of time implementing this software, and then you're going to be married to it. We evaluated these products againts OUR needs, and so should you. With that in mind, here's a brief overview of each one. Fellowship1

Beautiful, robust & powerful, F1 is the Cadillac Escalade of ChMS software. It's customer base is the largest, and it's well deserved. Excels in member management and reporting, especially. Unweighted, raw totals score:  84.52% Weighted, prioritized (smart) score: 87.72% Church Community Builder

Friendly, powerful & organic, CCB is the Lincoln Navigator of ChMS software. It's been around a bit longer, isn't quiteas pretty as F1, though it makes up for this in every way (and is by no means ugly, in it's own right). Their multi-site/multi-campus capabilities are especially well thought out, and their communications tools are feature-rich. Unweighted, raw totals score:  87.58% Weighted, prioritized (smart) score: 87.81% Shelby Arena

Shelby purchased the Arena software from a church who'd developed it themselves, and are now focussing all their development on it. It's a robust, well featured product, that was held back for us by an unclear interface and a strong "Microsoft feel". Their communications and reporting tools were very strong. Unweighted, raw totals score:  81.35% Weighted, prioritized (smart) score: 76.45% Think Ministry MinistryPlatform

The new kid on the block, these guys have built out comparable features to all the others, in a fraction of the time. We liked a lot about it, but ultimately it's aimed more at administrative staff than pastors, and for us that was the wrong focus. It's got enterprise constituent management written all over it, and has the most flexible family/relationships & multi-congregation concepts we've seen yet (make's multi-site look one-dimensional).

Unweighted, raw totals score:  81.39% Weighted, prioritized (smart) score: 79.70%

Files NeedsAssessment - Catch the Fire - April 2011 (Word Doc, generally sanitized) I'm requesting that you don't distribute this document online yourself. It was for OUR needs, and will need significant changes to match YOUR needs, but it may be a suitable template for you. ChMS Solutions Rating - Catch the Fire - September 2011 (Excel Workbook, somewhat sanitized) ChMS Solutions Rating - Catch the Fire - September 2011(PDF printout of the above) Down to the Wire

As I said earlier, any of these solutions could have worked for us in the end. The lowest score was 76%, which is hardly bad. From the scenario charts and the math, we could see that were really looking at Fellowship1 and Church Community Builder. The other solutions just weren't quite playing the way we wanted to play. Now you'll notice that the weighted scores of F1 and CCB are INSANELY close. When I showed these scores to both companies, I think they were each a little disturbed how closely they had scored to one another (though on different functionality, some of the time). Given how close their scores were, and how much our various staff were enamoured with either solution, I jumped on a plane and spent a day in Denver followed by a day in Colorado Springs. Culture All things being equal, which they very nearly were, we had to make a decision about which product would fit our culture the best. I spent 6 hours or so each with the respective staffs of F1 and CCB, met people in roughly equivalent roles, and chewed the fat as much as I could. I told some jokes, I asked hard questions, went for a drive with my account managers, and did whatever I could to find out who these guys (and girls) were. If I could have assigned a numerical score to each one, it would have looked very similar to the scores they both got above. At the end of the day, both companies are run by awesome teams, with great vision for helping the Body of Christ. For the record, I have no reason to believe the other 2 solutions are run by any-less awesome people, but I didn't meet them myself. Our Pick In the end we chose CCB. Their smaller size felt comfortable to us, and in the time we spent together we felt that we had a closer DNA match. The lack of financial support outside the US was not an issue for us, as we have to take all our payments through The Raiser's Edge, as it is the primary donations software that we're using. CCB has many Canadian customers, each of which have either found a way around this, or didn't need the functionality. CCB have a plan in place to address this, and while they didn't commit to time frames I imagine it won't be an issue this time next year. We signed contracts with them in August and are racing towards an October/November launch window. If any of this has been of value to you, I'd love to hear about it in the comments. We put hundreds of man-hours work into this, because we strive to be good stewards. It's all for God's glory, after all.

New MIDI mappings for Traktor Pro and Numark Stealth Control

The Load Track button for Deck B broke on my MIDI controller about 8 months ago. After a whole lot of hassle, I've finally got the controller back, fully repaired. This week I've remapped the layout to better reflect what I'm using Traktor for at present. It's fairly simple, but quite a bit more effective than my last one.  The Pitch Bend buttons have been re-purposed to jump to cue points, or to change the loop length when a loop is active. Cue Play resets the loop length to 4 beats. Download the TSI here (may need to right click and Save As).  

Numark Stealth - J Puddy Style

King David - Streaker, rapist, hero

King David - Streaker, rapist, hero

King David enjoys status higher than a lot of Christendom and Judaism's other heroes and figureheads. Even Jesus is called the Son of David. I've been thinking recently how much this exposes flaws in modern Western Christianity. Somehow, King David managed to be the only person in the Scriptures that God said was a man after his own heart. Countless books have been written about David, or following the model of David, or otherwise encouraging a Davidian element in our lives. Obviously he is venerated and held in hugely high regard. And yet, we know that in a moment of passionate love for his God, he ran naked and nude down the main street of town. We also know that in a moment of passionate lust, he raped a man's wife and then murdered her husband. Rock on!

Read More

Fixing a sliding door jam on a Toyota Lucida

Today is January 1st, 2011. It's the first day of a new year, and so I set about doing something I had never done before: car repair. James recently flattened the battery on my mum's van, and after recharging the battery the sliding door doesn't open anymore. A quick google search told me this was a common problem, so with a few helpful tips from others on the web, I got going (if you've ever posted anything about a Toyota Lucida, I probably read it... THANKS!). If you find yourself in a similar predicament with this van, here's how to fix it.

  1. Start by finding a thin cloth or piece of fabric, and run this around the door handle (on the inside). The door handle must be removed before the door panel can come off. There's a small ring that holds the door handle in place, and teasing around the door handle with some cloth will eventually catch it and pull it loose. It looks like this:

  2. Now look for a small screw inside the door lock switch, it's angled upwards so you may need to look from below. This can be unscrewed, the plastic enclosure around the lock can be gently pried off. Next, go to the back end of the door, and look for 2 snap clips (I don't know what they're really called), that hold the door panel in place. To remove these, you simply push in the center piece, and then the whole thing will slide out. There are two of them about a foot above one another.
  3. Now you can gently but firmly pull the door panel off, going slowely around and pulling at the edges. There are latches inside that will pop loose as you tug on them. The whole panel can be removed and put aside. Once that is off, you'll see a plastic sheet covering various electrical components. This sheet is for water-proofing, I'm told, and it's stuck on with a gooy adhesive that can be gently pulled off. It should remain sticky and you can just stick it back against the door after (or tape it if it looses its stick).
  4. The black tubular thing is a solenoid, a motor that powers the electrical door catch/release. This is the source of the problem, as with a power outage to the van it has gotten out of sync with the car's electrics. You need to disconnect the connector with the red + red/blue wires, this is the power source feed. Disconnecting that partially isolates the solenoid, and allows you to hit it with power, in the next step.
  5. This step requires you to have a car charger OR a spare battery OR long enough jumper cables to run from your existing car battery. Your goal here is to fire the solenoid, so you need to provide 12 volt power into the solenoid and get it to manually fire, which will then get it in sync with the car. I had a car charger, and found some wire from a broken bicycle light. I stripped back some wire, attached one wire to each of the red and black alligator clips from the charger (you can do this with any wire, and potentially with your jumper cables to the car battery), and then I taped off the exposed metal from the clips with electrical tape, just to be careful.
  6. Turn on the charger/connect the battery, and carefully connect the red cable from the battery/charger into the solenoid plug, on the red side (look at the solenoid end of the connector and you can see the cable colours). Touch the black to the other wire inside the connector, and the solenoid will activate (you'll hear the familiar mechanical noise). You'll need someone else to help you, as at the moment you activate the solenoid, the door needs to be pulled open. Don't leave the power activated on the solenoid for longer than necessary the pull the door open.The door should now be opened. Slide the door towards closing it just enough to reconnect the solenoid. Now try carefully closing the door again (carefully because you've got exposed electrical gear). The door should catch and you should hear the mechanical close noise, and then you should be able to open it again successfully. Congratulations, you've fixed the problem :) Now re-attach the plastic sheet, re-attach the door panel, the lock covering, and the door handle. The metal clip slides back around the door handle plastic and the two inward pointing parts of the clip slide into the holes in the handle, which will then clasp onto the handle pole. Well done.

It's 2011. What are you going to do this year that you've never done before? I have a goal of reading 26 books this year, which is roughly 20 pages a day. After this successful job I think I'd like to read a book on car repair, especially seeing as I bought my first car in October (a '96 Ford Escort).

Happy new year, folks.

Using OpenVPN in Windows 7 64-bit

Update: Using Windows 8? If so, click here for special instructions. Update: For Windows 7 RTM, all you need to do is download the latest beta version of OpenVPN (2.1 RC20 at time of writing), install like an:y other program and then run OpenVPN GUI as an Adminstrator. Pretty straightforward now, thanks to the good people at OpenVPN!

When I tried installing the latest stable OpenVPN (2.0.9) in Windows 7 RC1 x64, I received a message stating the TAP driver is unsigned, and this version of Windows ONLY supports signed drivers. That may prove to be an interesting gotcha... I'm trying to think of what other hardware/virtual hardware I have that may have unsigned drivers.

Anyway, here's a fix.

Basically you want to download from here and then set the compatibility (right click > Properties > Compatibility tab) to Windows Vista, and set it to run as an administrator. The install went fine for me, and the TAP driver is now signed in this RC. As mentioned in the brief note, but is quite important, you need to also run OpenVPN GUI as an administrator, otherwise any of the network level stuff (such as routing of networks) won't work.

Traktor midi mappings for Numark Stealth Control

The Numark Stealth Control doesn't ship with a TKS/TKI file for Traktor3/Traktor Pro. I'm sure an official one will be released at some point, but I've got mine working just fine as it is. The only tricky part was the LED mappings, for which you can't use the regular midi learn feature in Traktor. I just mapped them out manually one by one. Attached are the fruits of my labor:

Standard (without my own customizations) TKS file for the Stealth Control CSV file with the mapping matrix for the LEDs

Openness, candidness and honesty

For as long as I can remember, I have valued open and clear communication. I strive to communicate myself clearly and concisely, and I am frustrated when I fail to communicate myself, or when I witness mis-communication. I firmly believe that friendships, families, corporations, and governments function BEST with the minimal limits to their communication, [...]

Read More

With, not for – another thought

What if Jesus hanged out with prostitutes, corrupt tax collectors, and other sinners… not because he primarily wanted to convert them, or because he wanted to illustrate some lesson, allegory or principle…but because they were his friends, and he loved them. Thoughts inspired by nakedpastor. [...]

Read More

With, not for

With, not for

I look forward to the day when people, en masse, realise that they don’t need to (and can’t) do anything to earn God’s favour, and start to simply live with Christ (rather than for Christ). This is from Wayne Jacobson’s book – He Loves Me – via my good friend Dallas. The emphasis is mine. As you...

Read More

The Ten Commandments

It’s become my conviction that fundamentally we are not supposed to follow the Ten Commandments. I mentioned this in my post about The Shack… the perspective that I fundamentally follow Christ, and he may lead me where he wills. I want to dig into that a bit here, because I think choosing to follow the [...]

Read More