It turns out that the setup of a private App Inventor server is easier than I thought. If you are used to Amazon EC2, it’ll be a doddle for you. The Google App Engine bit is absolutely trivial. Full MIT documentation is here.
Go to appengine.google.com and sign-up with a GMail/GApps address.
Create an App. This only means giving the App an name.You’ll have to confirm your identity by SMS from Google. We used a variation of the name of our school as the app name so that our private App Inventor will be accessible at http://nameofourschool.appspot.com
Now you can run the build server by going to the for-BuildServer directory you created in Step J above and just running ./launch-buildserver
In the AWS Console, Open up port 9990 in the security groups
Copy the public DNS name of your server from the AWS console
Now back on your Windows machine, go to c:\apps\appinventor\war\WEB-INF\appengine-web.xml
Change localhost in <property name=”build.server.host” value=”localhost:9990″ /> to the public DNS name of your build server
Change <property name=”use.whitelist” value=”false”/> to be true
Save the file
Now edit c:\apps\appinventor\war\whitelist and add the GMail/GApps addresses of every account that you are allowing to use this App Inventor instance. In our case that it me, 2 teachers and 30 pairs of kids, all using the school Parent’s Association GApps email addresses that I created.
In a Windows Command Prompt, cd to the c:\apps\appengine (or whatever) directory and then enter c:\apps\appengine-java-sdk-1.6.1\bin\appcfg.cmd -A myapplicationID update war (where myapplicationID is obviously the name you picked in Step 2)
The last step builds and uploads App Inventor to App Engine and runs it. If, after a few minutes, you get a Success message then you are in business! Just go to http://nameofyourapp.appspot.com and you can login with your GMail/GApps account and start using App Inventor just like on the original Google site.
As I said in the last post, the only major concern is what number of simultaneous users will kick us out of the free tier on App Engine. Also, if lots of people are packaging Apps for phones, the Micro EC2 instance will take an age to build them. This isn’t a huge issue for us as we spend 99% of our time either using the emulator or a live non-packaged debug session on the phones.
Shout if you have any questions or comments.
My next App Inventor post will be a status report on how the pilot in our school is going.
The concern that many people have over the upcoming legislation around the Charleton judgement went into high gear when it was announced recently that The Data Protection Commission had instructed Eircom to halt its ‘three strikes’ policy against music piracy. This fear was compounded by the relevant Junior Minister being shockingly dismissive and flippant towards anyone who expressed concern on Twitter. Finally, in the same week, initial ACTA legislation was weaseled through the EU Council by hiding it in an Agricultural & Fisheries session!
If you haven’t read the Charleton judgement, you should do so. There is some good stuff in there but the numbers quoted and accepted by the court are hysterically funny. They claim 675,000 of the 1,571,000 broadband subscribers in Ireland are involved in illegally downloading material. The extrapolations from a few basic numbers are equally risible. Seeing mentions of ancient networks like Limewire, eDonkey and Gnutella is just cringeworthy. Charleton’s favourable view towards a global file registry shows just how dangerous the courts can be when they lack the knowledge and expertise in an area. I assume UPC didn’t challenge any of this nonsense as it wasn’t the legal angle they were working and the just wanted a Safe Harbour judgement like the US.
Sidenote: IANAL, but doesn’t the DPC instruction fly directly in the face of the Charleton statement that “I am of the view that there are no privacy or data protection implications to detecting unauthorised downloads of copyright material using peer-to-peer technology;”?
When Minister Sherlock then tells us that there has already been a call “for submissions from the public” that none of us heard about, we have a recipe for disaster with the legislation effectively being written by EMI and their chums due to lack of a balancing lobby. One can only hope the Department is aware of the recent ECJ judgement on IP rights.
@psneeze Jim. I invited submissions from public. I spoke to all stakeholders.
The podcast I linked to recently drove home this idea very strongly to me. The reason bullshit like SOPA gets anywhere in the US is because the media lobby is incredibly strong and sophisticated. The technology lobby, which should be there to fight off disgraceful internet-breaking approaches like SOPA is barely in existence, is un-coordinated and is immature. In Ireland I don’t think we have any sort of lobby or if there is, it is so ineffective, I haven’t heard of it.
The basic mistake Charleton makes over and over in his judgement, including the final section is his belief that “Solutions are available to the problem of internet copyright piracy”. As Leo Laporte re-iterates throughout that podcast, no solution can ever work in the longterm when everything is “bits”. The only outcome that a technological approach to piracy in Ireland will cause is a giant leap in the revenues of international VPN providers.
We need a lobby group that is there to fight for the interests and rights of internet users and technology businesses that operate on the internet. It should have everyone from Irish ISPs and giant corporations like Google and Amazon to those who represent the fundamental rights of all individuals to full internet access. If left unchallenged, the media lobby will destroy the usefulness of the internet for everyone, in order to protect their dying business models.
In a similar vein, I noted a recent newspaper article where a member of the media lobby purposely tried to confuse US-style fair-use laws (which we lack) with making all copying free. This is the kind of FUD that needs to be smacked back hard by a balancing lobby.
The most successful grass-roots lobby we’ve had to date was Ireland Offline. Is it time to put something similar together, with tons of clout, to ensure we don’t end up with 19th Century mindsets being applied to a 21st century communications medium?
I’m not in a position to kick something like that off, but I’d damn well support it, if those with the skills and contacts did so.
I helped my Dad setup printing from his iPad to his HP wireless printer over the Christmas. The easiest way is just to email the private address of the printer using the HP ePrint service. But I spotted that they have an iPad App too and I installed it.
Oh dear god. An ususable, counter-intuitive steaming pile of garbage. It doesn’t even have a Wizard to help you setup. Just a blank screen and some buttons.
I finally sorted it out after 15 minutes of messing. And then reverted to using the email method.
I’d like to congratulate HP on killing the TouchPad. Even they realised that they haven’t the first clue how tablets work.
The nice people in Packt Publishing sent me a free review copy of Google App Inventor Beginner's Guide by Ralph Roberts recently. As someone who adores App Inventor, I was thrilled to see the book being published. I'll do a full review later as I haven't finished it yet but I wanted to get some initial thoughts out there as I think the book is worth buying.
There are two big problems with App Inventor at the moment. One is obviously the shutdown by Google in three days time, before MIT is ready to take it over. I think this is a bad mistake by Google and means a lot of the goodwill towards the project may be lost. Luckily MIT is releasing early versions of the code so you can setup your own server. This will allow some of us to keep going in the gap period.
The second problem is the quality of the initial tutorial material. Whilst the examples are of kid-friendly Apps, the descriptions are far too technical and concise. I've had to re-write some of them for a non-technical Primary School audience and even the re-writes have needed simplifying.
So a book like this is really badly needed by educators and overall I'd recommend purchasing it if you want to get to grips with App Inventor. I only have two real criticisms. It's very obvious they have quickly hacked in some mentions of the Google shutdown. A 2nd Edition will probably be required in March/April once the MIT move is complete. But that's hardly the author's/publisher's fault.
The second criticism is that I think the first few chapters need a re-do. They flip between extremely simplistic descriptions and details like setting up Linux device drivers. This makes it quite confusing for beginners. They should pick one platform (Windows really) and base everything on that. All the nitty gritty of Linux/Mac etc should move to the Appendices. Also the Emulator and connecting to phones shouldn't be mentioned until it is time to run the first tutorial App. In fact, rather than going through each feature in each screen one-by-one whether basic or advanced, Chapters 1/2 should consist of a very basic tutorial from going to the App Inventor site to running on the emulator or phone.
Once they get into actual tutorials, the book flows quite well. I'd love to see a stripped down version of it for kids K-12 that could accompany the main book for maybe $2-$3 per child. I know 60 who'd love to have something they could bring home after school!
I've always loved the easiness of Posterous but it still has enormous problems with comments and Social Logins and I'm getting really tired of it. The lack of an official Export feature (yes I know about the API) always worries me too.
My default would be to switch to good old self-hosted WordPress plus something like Amplify for sharing interesting pages as posts.
But I just had the scary thought that Google+ could be "good enough". The Data Liberation Front means I don't have to worry about lock-in and the sharing features are pretty darned good. But of course all my old posts would remain stuck on Posterous and I can't have a custom domain.
Boxee finally told us what we all knew was coming, they are dropping support for PC/Mac/Linux. And with that, we now know they are done. I predict they will either shut up shop before this time next year or be sold for chump change by their VCs. Rumours that they are already hawking themselves around to CE companies sound reasonable.
Boxee got all of its buzz and online promotion from early adopters like me who loved their fork of the XBMC code-base. It worked brilliantly on both Windows and Linux for me. Then the updates stopped as they tried to release the code for the D-Link Boxee Box. This was an unmitigated disaster. The box that was released was, by all accounts, unusable. Even now, I see constant problem reports.
This was followed by the partnership with Iomega who finally managed to stagger out with a box recently, probably 6 months late?
And whilst all of this has been going on, what has happened? The rise of the Second Screen. The sad fact is that I don’t need Boxee any more, I just need something (anything) that can play a wide range of media formats. i.e. pretty much any media app or box out there. All of the Social stuff has moved to Tablets and Smartphones. It never really worked on the big screen.
As for super-smart connected TVs? Oh please. I spent the 90s working on digital TV software for the likes of Philips and Toshiba. Despite all their attempts to be more, these companies have never succeeded at doing any more than shipping LCD displays. The future belongs to Android and iOS with Social EPGs in our hands controlling everything. And those Apps won’t be, can’t be, Boxee.
In the olden days (as my kids call the 90s), we turned on the radio in the car and played music or stuck a tape in a tape deck. In recent years we'd plug a phone, iPod or USB stick into a car stereo. Now we are messing with streaming more. It's still as patchy as hell in Ireland with all the mobile blackspots on our motorways but we had a reasonable amount of success yesterday doing the M8-M7-M3 run.
Sidenote: And what a marvellous road that M3 is. Anyone not using it on principle? Nah, didn't think so.
As we lashed along the M3, with each of us calling out songs to play next, I realised we were using a huge amount of technology to get Led Zeppelin, Rock and Roll, playing out through the car speakers. Here's a diagram of it.
The two key bits are Bluetooth A2DP to the car stereo and my HTC Sensation acting as a Wifi portable hotspot so the youngfella wouldn't incur any mobile data charges. Apart from the odd bit of buffering, we reached Cavan with Queens of The Stone Age, No One Knows, at top volume.
I think it's good to stop everyone once in a while and marvel at the magic of all of this. My first job was writing code for a GSM basestation and there were thousands of people involved in just that. Think about every piece of the media delivery chain and how many people were involved in creating it. We humans can be pretty awesome sometimes.
Oh and this was the top tune in our car, played over and over on the trip. A soundtrack you should definitely buy. A bajillion times better than the dire movie.
If you ever did anything on Google App Inventor, you have 9 days to download your code.
I had to log in 30 times to download all the projects of the kids in our school (30 teams of 2). Next step is to setup a private instance of App Inventor on App Engine and EC2 in lieu of MIT being ready.