Rediscovering Google App Inventor

It’s been some time since I last had a play with Google Ap Inventor, the browser-based, drag-and-drop block style graphical interface for creating Android apps; well over a year, in fact, going by the date stamps on Getting Started With Google App Inventor – A Twitter Search Client and Getting Web-based Images into a Mobile App – App Inventor Formula Plotting App Using the Google Charts API. So when I was asked if I’d like a review copy of Ralph Roberts’ Google App Inventor [table of contents], I thought it would be a good excuse to have a play again (that and because I mistakenly thought it was another book entirely!(App Inventor for Android book and [Liam’s] first time as a Technical Editor). Oops..;-)

The review copy was a PDF version of the book, duly personalised…

Personalised review copy..

…so I’m afraid I won’t be lending it to anyone;-)

As anyone who’s tried writing material about web technology that is expected to persist for more than a few weeks will know, the fast pace of change of online applications can make it difficult to find a tone that will keep the content feeling relevant even when things have changed. With an October release date on the book, there was just time to include a prequel caveat and final Appendix addressing the change in ownership of App Inventor, as revealed in an August announcement that revealed that App Inventor would go open source, and MIT would act as stewards of it. An update to the software whilst the book was in preparation is also reflected in the conversational presentation of the book, and what I guess we might refer to as a dynamic remapping of some of the activities. (For example, “I tried TinyWebDB, but that was getting convoluted. Fusion Tables were slow getting data to the phone, and you had to make the tables public. Then, just a little over a week ago as I write, AI was updated and the Web feature added, which solved my problems.”)

As someone who buys a lot of physical/print copy computer books, I must admit to struggling with the ebook-as-pdf format. We do have an iPad in the house, but it’s not synched with my personal laptop, so I read book on a laptop, using a zoom level in a landscape screen that fit the whole of each pdf page onto the screen at any one time (CMD-right arrow then takes me on to the next page). I could have tried reading the book on my Kindle, but again my preferred way of reading is to try to fit a whole page on the screen, and this typically doesn’t work very well on the Kindle form factor. Splitting a page across a screen I find jarring whenever I get to a split page break, particularly where pages have an intrusive footer. And I didn’t want to print the PDF out particularly… One of the things I fail to understand about the continued desire of publishers (and that includes the OU) to reproduce A4 landscape pages in paged PDF documents when there is a likelihood that the reader will be using a device that responds far better to infinite paged scrolling, dynamically reflowing interfaces, I don’t know…

Another problematic issue faced by many computer books is how to balance text and screenshots; the text provides a narrative to follow, but the screenshots (particularly for GUI driven environments) provide the checkpointing. I have to admit to finding some of the screenshots hard to read, whether because the reosultion was too low, or because of poor contrast (yes, I know terminals often provide white/grey text on a black background, but that’s no reason to feel you have to use the same format in e-print…)

But the book – I hear you say – what about the content of the book? I’m coming to that…

The book is a beginners’ book, typical in format to many you’ll find on all manner of computing related topics. Part getting started (the first part of the book deals with getting the installation faff done and dusted), part overview (the second part of the book is an annotated Overview of Google App Inventor Blocks), part recipe book, part tutorial guide, the reader is led through a series staged exercises loosely relating to particular application types – communication apps, for example, database/storage apps or games/animation apps. The blocks overview give some interesting illustrative get-you-started use cases, and there are a couple of tangential nuggets (http://www.google.com/gwt/n as a path to a mobile friendly view of a web page for example) that can be incorporated into App Inventor hacks.

It’s little insights like this that I would have liked to see more of, things that make the book more than a disposable introduction. The same is true of the exercises; the stepping stones didn’t seem quite so plastic (useful/reusable/modifiable) to me as I’d have liked. Don’t get me wrong – I’m a great one for buying technical books I regard as one-shot, or disposable; the sort of thing I can read on a train to get a good idea of what the capabilities of a particular environment, application, maths framework or programming language are good for, but I never refer to again. Notwtihstanding the gripe above about PDF vs (implicitly) HTML, the PDF copy of the book was the wrong form factor – for me – for this sort of work; I couldn’t review it whilst soaking in the bath, for example, which is where I soak up the possibilities contained in all sorts of technical works…

The form factor can also affect the usability of the book as an instructional manual. A physical book takes on what you might think of as a second screen style affordance; I can refer to a physical book at the same time as working through an exercise on screen (really handy in the case of App Inventor, which has at least two main windows – a UI design window, and a blocks-manipulating programing window – or three windows if you’re also running an emulator to test your programmes.)

But the content? What about the content? As I said, fine… good coverage of the chunkier blocks, elements of programming (though at times confusing – concepts such as functions, procedures and subroutines all getting introduced at various stages and never quite being clarified or fully worked through). I didn’t get chance to work through the examples, but the spirit level cum pitch-yaw-roll app looked like it might be worth trying out.

So would I buy a copy of the book? If I saw it in hard copy, remaindered somewhere, probably. As a hard copy, at full retail price? Possibly, if I hadn’t already got up and running with App Inventor myself. As a book to have laying around at a kids computer club or in a maker space, it might inspire someone to get started with App Inventor and give them a few exercises to work through. As a PDF? No chance… I’d go Googling around “app inventor” tutorial instead…

PS I guess I’m not getting to get many invitations from Packt or their PR folk to review their books again, am I?!