Web-based training comes in synchronous, asynchronous, and combined forms. The three basic modes I’ve experienced are these:

• The Webinar
  A synchronous, instructor-led session which attempts to capture the flavor and functionality of face-to-face instruction. The instructor presents through live demonstration, creating a screencast on the fly. Students generally interact with each other and the instructor through some sort of embedded chat application, though conducting Q & A via telephone or VoIP is also popular.
• The Screencast
  Students are given access to a screencast, previously created (a.k.a. “canned”), which they work through at their own pace. Follow-up with the instructor/designer is generally made available via phone, email, IM, Twitter, etc.
• The Thing That Should Not Be
  Students meet online at a specific time to view a canned screencast, with live followup after the show via chat, VoIP, or whatever.

The first two of these have their strengths and weaknesses. The third is simply an abomination.

The chief advantage of webinars is that, being instructor-led, they offer at least the possibility of immediate feedback. The chief drawbacks are logistical (you have to be in front of your computer, ready to go, at a certain time) and technical (your connection can always crap out on you).

The chief advantage of screencasts is the lack of logistical drawbacks. You’re booked solid this week? Fine, watch it next week. Work doesn’t allow you long blocks of time? No worries. Pause it and finish it when you can. Slow learner? Watch the tricky bits over and over until you understand them. The chief drawback is the lack of immediate feedback.

Webinars are quite a bit like traditional classrooms, where the instructor leads and you follow. Screencasts are much more like books: you work through at your own pace. Unlike books, though, you generally have the author’s contact information.

The last of these, The Thing That Should Not Be, combines the worst of both worlds. And I wasn’t even aware of its existence until recently. “The Thing” entails all of the logistical difficulties of webinars without the advantage of their immediate feedback and also without the level of control that screencasts provide. I can’t think of a compelling reason to conduct any training this way, and I hope it doesn’t become a trend.

Creating audio with Processing takes me way back to the Atari 400 I got for Christmas one year, with which I managed to, hamfistedly but joyfully, create various noise makers including a decent drum machine.

I decided to give Processing a try and quickly found Dan Ellis‘s page of sample code, which included a nice little sine-wave piano, which I stole, extended a bit, and share with you here:

• SinePiano v2 (requires Java, sorry).

I do this with permission from Dan, who has moved on from Processing and now likes to do things in Pd, which is an open source branch of Max.

There is, these days, a version of Processing reimplemented in JavaScript called Processing.js, but, in my tests at least, there’s no support for audio. So, sorry that I had to resort to the Processing IDEs Java Applet export. Performance is better if you run it straight from the Processing IDE, of course. And there’s a link to the source if you want to give that a whirl (or hack your own).

Keyboard layout is simple. The Z-row is a C major scale. The A-row is the same, one octave up. The Q-row plays the major or minor third, relative to the root note below it in the A- and Z- rows. The number row plays the perfect fifth for the same. So, as a diatonic instrument, there are no wrong notes. And you can easily play major and minor chords or intervals by pressing keys in any vertical column. Make sure you don’t have caps lock on, as I’ve only defined the lower case characters so far.

If you choose Data –> Subtotal, you get this dialog, which you use to define which column Excel should check for changes, and which column(s) should get a subtotal, and what sort of subtotal it should be (e.g. SUM, COUNT, AVERAGE, MAX, MIN, etc.):

After pressing the OK button and adjusting column widths, your sheet looks like this:

You can see that Excel adds subtotal lines for each change in date, as well as a grand total line at the bottom. If you single-click B5, B9, B13, and B14, you’ll see that Excel has written SUBTOTAL() functions for each of the three ranges created by the changes in the Date column, and has applied the same function to the entire range to get the grand total.

You can turn them back off by opening the Subtotal dialog a second time and clicking the Remove All button instead.

That’s handy, but wouldn’t it be handier to turn this on and off via VBA? I thought so, and it’s pretty easy to do. Fire up the VBA Editor (Developer –> Code –> Visual Basic), double-click the sheet which contains your data (e.g. Sheet1) from the Project Explorer on the left, and create two subroutines. The first, AddSubs() will turn on the subtotals. A second, RemoveSubs() will turn turn them off. Here is the code for AddSubs():

Sub AddSubs()
    Worksheets("Sheet1").Activate
    Selection.Subtotal GroupBy:=1, Function:=xlSum, TotalList:=Array(2)
End Sub

It’s pretty simple. You can see that the worksheet is specified by name in the second line. In the third line, the GroupBy value (i.e. 1), is used to specify the column where Excel should look for changes. In VBA, the columns are numbered from left to right. So 1 is Column A, our Date column. The TotalList array specifies the columns which should get a subtotal. So 2 is Column B, our Visits column. If you have more than one of them, just separate each number with a comma (e.g. Array(2, 3)). The Function parameter tells Excel what sort of subtotal to create. You feed it one of several contants (e.g. xlSum, xlMin, xlMax, etc.).

Turning subtotals off is equally easy. So long as you get the worksheet name right, you should be in business. Here’s the code for RemoveSubs():

Sub RemoveSubs()
    Worksheets("Sheet1").Activate
    Selection.RemoveSubtotal
End Sub 

To use either one, click your mouse inside any of the cells that contain your data (here, any cell in the A1:B10 range) choose Developer –> Code –> Macros, select the macro from the list, and press the Run button.

Following Microsoft’s recent announcement that Internet Explorer 10 will not support plugins (including Flash) when running in its new Metro style, opting, as Apple has done, to embrace HTML5 as the future, the question becomes, where does Adobe fit into all of this?

I’ve long thought that the way forward for Adobe is to retool its apps to support HTML5 output. With respect to Adobe Captivate–the tool I use most in my work–I’m happy to say they’ve made substantial strides in this direction. For the past few weeks, I’ve been beta-testing Adobe’s Captivate-to-HTML5 converter. And the initial testing has been so promising that Adobe is now offering it on Adobe Labs so it can reach a wider testing audience. Is it perfect? Of course not. But I do like it and I’m very happy that Adobe is heading this direction, because I like building things in Adobe Captivate and Flash Professional.

[Updates:]

The Adobe Captivate Blog has a post about the HTML 5 converter and some samples of converted projects.

On Windows 7, it lives here: ¹

C:\Users\AppData\Roaming\Microsoft\Excel\XLSTART\PERSONAL.XLSB

On Windows XP, it lives here:

C:\Documents and Settings\Application Data\Microsoft\Excel\XLSTART\PERSONAL.XLSB

In either case, the easy way to create PERSONAL.XLSB is not to muck around with the file system directly. Instead, just record a macro in Excel and, when you’re prompted to save it, choose Personal Macro Workbook from the Store Macro In drop-down menu on the Record Macro dialog box. Record yourself typing a few numbers and adding some formatting to them or something similar. Afterwards, you can open up the Personal Macro Workbook via the Visual Basic button on the Developer tab later and delete whatever you record. Once you create it, your Personal Macro Workbook will be listed in the VBA Editor as “VBAProject (PERSONAL.XLSB)”.

Some Additional Details for the Truly Nerdy

The location of the Personal Macro Workbook is a little confusing, because older versions of Excel stored it in an XLSTART folder buried inside the Program Files folder. Even though you can’t write any files to it, Excel 2007 and Excel 2010 also have XLSTART folders inside the Program Files directory. Here are the paths:

For Excel 2010:

C:\Program Files\Microsoft Office\Office14\XLSTART\

For Excel 2007:

C:\Program Files\Microsoft Office\Office12\XLSTART\

“Office 14″ is the internal name for Office 2010. It’s actually the 13th version of Office, but Microsoft saw fit to skip naming it Office 13, due, one suspects to superstition. So “Office 12″ is Office 2007, “Office 11″ is Office 2003, and so on. If you happen to be stuck running Excel 2003 on a Windows XP box, you’ll find PERSONAL.XLSB at this path:

C:\Program Files\Microsoft Office\Office11\XLSTART\

.

defaults write com.apple.Finder AppleShowAllFiles YES

You have to relaunch the finder to make it take effect. There are few ways to do that. One is to choose Force Quit from the Apple menu (or press Command-Option-Esc), select Finder from the list of running applications, and click the Relaunch button (you’ll be asked if you’re sure, requiring you to press a second Relaunch button). Another way is to hold the Option key, then click and hold the Finder icon in the Dock, then choose Relaunch from the context menu. ¹

To reverse it, so that you (or your wife, or kid, or pet) don’t accidentally delete something, you repeat the same command, but substitute NO at the end, like this:

defaults write com.apple.Finder AppleShowAllFiles NO

As before, you’ll have to relaunch the Finder for the changes to take effect.

• Publishing Captivate to SQL using ASP, Part I
• Continuing with Captivate
• Publishing Captivate to SQL using ASP, Part III
• Publishing Captivate to SQL using ASP, Part III (Version 2)

I’ve gotten several requests in the past to create an ASP version of Cp2DB, so this should be of interest to many.

I’m originally from Texas, but I lived long enough in Arkansas that I think of it as home, even though I’ve been in Charleston ¹ for more than seven years. Arkansas has always had something of an identity complex. To anyone in the north who has even heard of it, it is a southern state. ² To quite a few people in the deep south, it is too far west to really count as southern.

I have beautifully simple litmus test for southern states: any state that fought on the wrong side of the Civil War qualifies by that fact alone.

With the exception of a brief stint in Philadelphia, I’ve lived in the south my entire life. And that’s not entirely a bad thing. There are a lot of things I like about the south. ³ But one thing I’ve never liked about the south is what I think of as Civil War Denialism–a condition whose primary symptom is the inability to admit, whatever the extenuating circumstances, that the south was, in fact, on the wrong side in the Civil War and deserved to lose.

Admitting that the south was wrong to secede, and, even more importantly, to secede over slavery, isn’t to say that the north was some Utopia of racial equality, or that the Union’s motives were entirely pure. But about the core issue–the existence of slavery as a legitimate economic institution–there should be no confusion. Anyone who claims that the Civil War “wasn’t really about slavery,” a claim I’ve heard my whole life, hasn’t bothered to read the “Declaration of Immediate Causes which Induce and Justify the Secession of South Carolina from the Federal Union.” Was “states rights” and issue? Certainly it was. But so was human rights. Arguing over jurisdictions doesn’t change that, it just shifts the argument from the ethics of slavery the semantics of law, while failing to address the former. ⁴

The thing that struck me most about the NPR story was that the various defenders of the “good name” of their Confederate ancestors have such trouble admitting that those same ancestors were simply wrong about some things. The absurdity here is they were likely wrong about many things, from the most trivial of things to the most important. I’m willing to bet that these various sons and daughters of Civil War veterans would be unwilling to defend their ancestors’ point of view on most any other topic. So why feel the need when when it comes to this one?

I feel no need to agree with or, in those cases where I disagree, “respect” my ancestors’ opinions on any topic. They were their own people, trying to make sense of the times in which they lived. I’d no more ask them for advice on human rights than I would on dental hygiene. We’ve learned a lot about both since April 12, 1861.

In both cases, what you see here is raw output from a simple capture in demonstration mode, using Captivate’s default settings. The point here isn’t to show off my screencast production chops. The point is to compare raw output when capturing a virtualized app compared to a native one. I’ve kept the default slide duration (3 seconds), so you may have to pause here and there to really see what’s happening. I’ve also kept the default slide quality (low). Feel free to avail yourself of the audio on/off button on the playbar, as that typing sound effect can get a little tedious.

First Test: Microsoft Word 2010 running on a Windows XP (SP3) virtual machine in VMware Fusion 3 on Mac OS X (10.5.8/Leopard):

Word 2010 via VMware Fusion 3

Second Test: iWork Pages ’09 running on Mac OS X (10.5.8/Leopard):

iWork Pages ’09, natively

The comparison shows some useful things. First, Adobe Captivate 5 was able to capture actions in Word 2010 even though it had to navigate through VMware Fusion’s virtualization layer to do it. That’s impressive. The mouse was captured as a separate object, just like it always is, which means you can reposition or hide it in post-production (which has always been one of my favorite features of Adobe Captivate).

On the down side, none of the highlight boxes were properly positioned, and quite a few extraneous text captions (a.k.a. “callouts”) were added, saying “click the Windows XP Professional window.” So, clearly, there were times when Captivate couldn’t determine what to name an interface element. If I were going to do a real project this way, I’d probably disable automatic highlights and text captions and just add my own manually, as that would, on the whole, be less trouble than repositioning and/or deleting all of these.

So, the answer to the question that prompted this post is a qualified yes. Adobe Captivate 5 can, in fact, capture apps running in guest operating systems in VMware Fusion 3, but the raw results of such captures will require more post-production cleanup than you might like.