Aside from all the basics, they allowed 3rd party applications on the device AND the desktop to talk to each other directly to sync custom data. I’ve bitched about this before.

As an avid Mac, iPhone, MobileMe and soon to be iPad user, I have to wonder what is happening with this at Apple. My real-world gripe is this:

I was just about to open OmniFocus on my iPhone specifically so that it would sync with the latest data on my MacBook Pro, which is set to sync via MobileMe (using a pretty ugly file based solution). Why am I even doing this? Why isn’t this data synced (a) when I dock my iPhone to sync all the other iTunes stuff, and (b) why can’t it automatically sync wirelessly

Well part (b) is easier to answer, although it is a three-fold answer. First, there’s no background app support to allow automatic sync of the OmniFocus app on the phone. That should be addressed by the Push API functionality except that OmniFocus doesn’t support Push API (server cost to them to do so) and even if they did support Push, iPhone SDK Push is not able to automatically pass the data to the application to force it to sync – the user must acknowledge the event and run the app on the phone manually. It’s a pile of suck, surprisingly, with a real feel of “disconnected device”.

Part (a) is more tricky to answer. It must be trivial for Apple to add this kind of support for direct-to-app syncing. They already have/had Sync APIs for OS X for a long time. Lack of support for this apparently makes no sense.

In conclusion I am very surprised that Apple has not updated the OS X Sync APIs so that:

• Third party apps can sync any data they like to/from the iPhone/iPod/iPad with iTunes as the conduit (that was the concept’s name in PalmOS if I recall)
• The transport for sync is completely hidden from the applications such that sync will happen transparently via Dock, Wifi (direct between devices on local Wifi network), and via MobileMe cloud if the device is not on the same Wifi network.

This is not rocket science after all. And yet we still have to know / think about what networks our devices are connected to, manually make sure we run them frequently etc. It is pretty lame, Mr. Jobs.

Related posts:

1. Apple iTunes and NAS usage – please fix it Steve!
2. My first few days with iPad
3. My Apple tablet predictions – for what its worth

Many many people have this desire: a single place for all their media: music, videos etc.

A NAS device is the place, backed up suitable. The problem is iTunes / Front Row just does not fit with this strategy if you have more than one computer / Apple device.

The core issue is that iTunes does not automatically rescan the media folder for new files. So you can point as many iTunes as you like to a shared location but they will only see the files they add to their libraries via purchasing or importing media.

So you end up with many devices in your house, all with a different view onto your shared media.

To access that new album you imported, you have to manually Add To Library on that iTunes instance, hoping it isn’t set to duplicate the files on the server. When this happens, as it certainly used to for me, you then end up with many copies of albums as you re-create your library from scratch on the various computers over the years.

Add to this the fact that if for some reason your connection to the NAS goes down when you run iTunes, it will revert the media path to the local folder, and you end up with a total mess – a bunch of machines with files only on some of them, their media split between local and NAS, and only the ones that added those files having them listed in their library.

The experience is so un-Apple it is shocking, and it causes daily pain.

It is important to remember that the iTunes Media folder is where it PUTS files that you buy/import – and that is ALL. (There is a new “Automatically add to iTunes” folder there, which seems half-assed to me). The iTunes Library is specific to each computer and is the list of media and the file path to each. This, unlike the media folder that actually stores the media files, is something you do not want to share between computers in many cases.

Now, iTunes 9 added Home Sharing. But guess what, this sucks and blows! Why? It (a) only shares iTunes-purchased content and (b) it duplicates the files to your local HD. Home sharing, I believe has a lot more to do with the iSlate/tablet and their new datacenter – music in the cloud crap – so you can sync your iSlate content without a cable.

Please Apple, it needs to work like this:

How it should work

This is relatively simple:

1. First, never change the media path in iTunes if the previous path is not reachable. Tell the user what is happening so they can fix it
2. When new files are added, make a bonjour announcement to any other iTunes running (perhaps even wide-area bonjour to make iPhones/Slates pick it up) so that they can instantly add the file to their iTunes library
3. When a file is not located on the local disks, have a local cache for stuff that the user ACTUALLY PLAYS. My wife doesn’t like all that heavy metal I listen to, so let’s not fill her hard disk with a clone of it eh? iSlate and Apple TV / mac mini media hubs etc can pick up just the files that are needed.
4. For the occasion when not all the machines / iTunes are running, have iTunes do daily rescan in the BACKGROUND for any new files in the media folder and AUTOMATICALLY add these to the Library. This is not rocket science.
5. Maintain an “excludes” list on each iTunes library so the user can remove items from their local itunes Library (without deleting it from the NAS) and they will not be offered the file again in a future background sync.

Don’t give me stuff about non-purchased media not having ISRC codes to identify them and de-dupe. You can dedupe on SHA hashes of the media (calculate once and embed in the metadata of the file) and failing that trackdata, and then failing that – USER INTERVENTION eg “There are some new media files added and we don’t know if they are duplicates or not – help me”.

You can even put all these newly discovered files into a special “Newly discovered” place in iTunes where the user can yay or nay them – or have it set to auto-accept (default).

The more and more macs and related devices are sold to households the more shitty this problem becomes and you REALLY REALLY need to fix it Apple. Without the cloud. The cloud is not a solution for terabytes of media being instantly accessible in your living room.

Please. Just do it. iTunes Home Sharing was nearly it, but sadly failed completely to address this.

Related posts:

1. The future of computing UIs – TV and touch tablets and OS X
2. My Apple tablet predictions – for what its worth
3. Data sync on iPhone, iPod, iPad – the missing link?

It is noticeable that Mac OS X and related apps are moving to user interface paradigms that will work much better than conventional UIs on (touch controlled?) TVs and multi-touch tablets.

Some examples of this growing trend:

• Front row started it really
• Several Apple apps with full screen modes. Especially the new quicktime X and its large transport buttons, iTunes/Front row with Coverflow, iPhoto with multitouch trackpad support
• The new exposé screen layout, touchpad gesture and per-app exposé
• Spaces – something that will prove MUCH more useful if every application is full screen, and effectively each Space is a single Application. Then Exposé sort of becomes the current spaces “zoom out”
• Safari “Top sites” is, in my opinion, clearly targetted at touch (tablet) displays.
• iTunes LP media format
• This is the big one – iTunes “Home Sharing”. This is required to solve the problem of your TV/tablet as a media purchase platform. You will typically buy content at home on a TV/tablet and then expect it to be on your laptop and hence iPod.

So I expect, within a few years, for us to see a new paradigm at least for TV and tablets where apps run full screen and you swipe to switch application… and wonderfully simple UIs as a result.

If we can only do away with the keyboard, we might finally be rid of these rather old fashioned apps with lots of menus and shortcuts and yada yada.

Related posts:

1. Apple iTunes and NAS usage – please fix it Steve!
2. My Apple tablet predictions – for what its worth
3. My first few days with iPad

What’s happened is that I’ve found a couple of “minor” bugs, which have hit me in different ways and in different combinations in a client projects including the Weceem CMS 0.2 which is coming along nicely although we have had to battle with some of these problems. It seems like once you find something you can’t get away from it.

Anyway these issues have been raised in Jira and are scheduled for fixing soon. Here’s a breakdown of them and any workarounds:

1. Grails will no longer (?) let you bind null “noSelection” values to relationships. If you have noSelection=”['':'None']” or similar on a select box, if you select this on a nullable relationship field you will get a plethora of binding/validation errors. It seems the empty value is enough to have grails believe you are trying to create a new “inline” instance of the referenced class, which typically fails validation but may not depending on your constraints. Eg this could be creating extra instances of classes you don’t know about if you have lax constraints on the related class. Jira issue here.
   Current workaround: for a relationship to a “Template” object field named “template”, your action needs to do this:
   if (params.‘template’) params.remove(‘template’)
   if (params.‘template.id’ == ”) params.remove(‘template.id’)
2. It is currently impossible to query for null associations on 1:1 relationships using any mechanism – criteria, dynamic finders, HQL all return empty results. This ties into the previous issue – say you want a UI to select the “Author” of a book but you only want to allow Author to be used once in any Book(s). To present a list of Authors that have no “book” associated requires this kind of query. This is actually a Hibernate bug that needs votes + comments to get it fixed. Grails Jira is here.
   Current workaround: the insanely inefficient list and collect (beware of blowing heap):
   Author.list().collect { it -> it.book == null }
3. Invoking tags as methods and passing in bodies is a bit broken. Jira issue here – and a workaround for FCK editor is jira’d here. Basically if you invoke a tag as a method and need to pass in a body… your mileage may vary. The body is output immediately when executed, instead of returned as a String. So if you try to say have your own tag that then calls the FCK editor tag with some canned data it won’t work.
   Current workaround: change the tag you call to not evaluate the body() until it is needed in the output. Another option (untested) is to embed the invocation of the tag into a GSP fragment and render that using g:render instead.
4. The “transients” settings on GORM classes is not currently inherited if you have descendent classes. This might be causing you some embarrassment if you aren’t aware of it. Eg. If you have a domain class with descendents, check your database schema and your data to see if values that should be transient are being persisted – in particular passwords. Jira is here.
   Current workaround: Always declare a transients propert in your descendents even if you don’t need it, assigning the inherited value. Eg something like:
   class HTMLContent extends Content {
       static transients = Content.transients + [ ’summary’ ]
       //…
   }

Thanks to Graeme for his help confirming and nailing down these problems. Looking forward to fixes in 1.2

It seems that we really do need Grails users to hassle the maintainers of the components Grails depends on eg Hibernate, to get issues that break functionality in Grails fixed. All users of Grails are indirect users of these underlying technologies. This gives us rapid productivity gains, but can also cause delays getting issues resolved when they can’t be worked around in Grails.

So get the vote out – if you find a Grails issue dependent on another project’s bug, please vote for / comment on issues in those other issue trackers to up the pressure!

Related posts:

1. Weceem 0.8 Released – Highlights and background
2. Weceem 0.1 to 0.2 – changes under the hood
3. I’ve released another Grails plugin

Now this is particularly interesting to me as I have a fairly well developed idea for such a service, and had begun implementing this using the iPhone with a custom iPhone application (and of course a Grails application for the back end).

My immediate thought was “phew! Glad I didn’t spend any more time on that. Note to self: check own ideas for ‘behemoth trouncing risk’ in future”. Not to mention some relief that I wouldn’t have to implement the service myself now Google has “done it”.

However, then I started to think a little more and looked at their offering a bit closer – as much as I can with nobody using it and no iPhone support yet.

This made me realise a few things:

1. The behemoth does not always get it right, or rather tends to cater for the very high volume use-case which is not necessarily where the financial gain is to be mad
2. I have not seen their phone app yet, but I am wondering if it will have the right “drop dead simple” UI it requires
3. The behemoth when trying to handle the generic mass-market use cases, can not always create the seamless and simple UI required for users to love (and continue to use) a service
4. Most importantly – this kind of usage is not, in my view, what this technology is best for. I think the money to be made here is on smaller groups of users, and users in specific organisations.

Sure Google will have some plans further down the pipe – I’m sure an API will come (“Show where you are on your blog”, “Get location of friend X” etc). Its also surprising they don’t have an iPhone app (rather than an iphone customized web site) for this already.

However I think there’s a fair bit of money to be made with such a service that focusses on making sure parents know where their kids were when they said they’d check in, and for small-scale logistics for companies. The application UI -has- to be top-notch, and the functions it provides have to relate to the market place. There’s possibly a couple of different client apps that could be made to front the same back end.

So if you’ve got a some spare cash and want to fill what I believe remains a gap in the market, drop me a line and I can flesh out the ideas for you

Related posts:

1. What are all the things a startup web service needs?