Wow, it has been a long time since I blogged.

I don’t have anything right not terribly compelling to post about philosophy or technical stuff, but I figured at the very least it might be nice to post about what I’ve been up to the last couple years.

Work
I switched jobs in 2012 to be a software developer (yay!), and I’m mostly loving it. The technical aspects and workflow fit me better then my controls/EE work did I think.

We released an app I helped finish up, it is in the play store and app store – it is for Trane Commercial building and automation systems, and is called BAS Suite.

Then I got to help architect and build, and for the last few years improve and add on to, another sorta-mobile product called Tracer Concierge.

Family
It has been a crazy and rough few years for me on the family front. Having a kid opened up and exposed existing holes in my relationship with my wife. The good news is that with the help of some really really great friends and a decent counselor we both feel like we’re over a lot of those hurdles, and filled in a lot of the relational holes that had been discovered.

If you follow me on social media you know that none of that stuff with my wife was helped by my Dad almost dying in 2014. He went through a pretty rough surgery where he was basically looking like he wouldn’t make it – but the hard work of the surgeon and staff, and I believe the will of God, pulled him through. You can read more about that on his Caring Bridge site

The Kid is 3 1/2, and a full blow threenager, and she is hilarious most of the time. We feel pretty lucky to have her.

Church
I’ve been laying a little low at church in a lot of respects. Trying to just navigate life partially, and partially because I picked up one pretty big leadership role – I am on the lay-leader preaching team. And it is awesome!

The nuts-and-bolts of it is that we meet once a month, discuss how we are improving our craft of public speaking, and then someone from the lay team preaches about once a month. That means I preach 2 or 3 times a year.

So far, I’ve gotten to speak twice, and I’ll be jumping into the pulpit for my 3rd time on November 8th. If you want to listen or check it out, click on over to the podcast, or here are links to the specific talks I did:
– August 17, 2014 – HEROES OF THE FAITH – WEEK 11: JESUS, BATMAN, AND BILL GATES
– June 14, 2015: VISION & APOCALYPSE: ZOMBIES AND REVELATION

And that is a broad overview of what’s new.

Well, I went an entire year without writing a post. But believe you me, I made up some great ones in my head.

The year has been an interesting one. Challenging in a great many ways, particularly at home. But really great and exciting in some other ways, particularly at work.

I did work some on the next version of my Modbus app that I talked about late last year. Never got super far into it – one part analysis-paralysis, two parts raising a toddler while trying to make sure my wife still knows I exist.

Other miscellaneous 2013 trivia and happenings from the Catlin home:

• Pearl Turned 1, and started talking. She now says entire sentences – mostly “Daddy I need this.” when she wants something from me that I have, or that she can’t reach.
• My first major project at work officially released to the Android and iOS stores. My second big one is I think internally released, and will roll out to our national account sales teams shortly.
• I learned how to cross-compile the linux kernel for our hardware, and get it to boot.
• I met my niece (my sister-in-laws daughter) for the first time.
• Went to Cleveland for my aunts wedding – saw a bunch of family that I haven’t seen in a long time.
• I built a new workstation/server PC, after a decade of faithful service, I felt like it was getting irresponsible to rely solely on that machine to keep my families photos and videos safe (and yes, I have cloud backup too, but still).
• Also at the same time I was building that new server my HTPC’s motherboard died, so I had to replace it somehow. On a very tight budget I squeezed both together, and managed to accomplish a new HTPC by buying a Raspberry Pi and installing OpenElec on it (which is a linux distro tuned to boot straight into XBMC).
• My best friend moved back to the US from Germany – Iowa to be exact. And my Sister-in-law moved to Iowa too. So instead of having no place to stay on the way to my parents house, I now have to pick between two really great options. And my best friend isn’t 7 hours offset from me, and in another country.
• Went to the boundary waters for the first time in 20 years with my dad. (Photos are on Google+). Found out I wasn’t as young as I used to be, and neither is my dad (though to be fair – when he was 32 he could have probably kicked my current selfs ass). But we came back in one piece, and had a really amazing time together.
• Braincell reunion at Rob’s wedding – which was a classy affair. Especially for the smelly kid. We drank, ate, drank, sang some songs, drank, danced a jig or 5, drank, and had wonderfulness well into the wee hours of the morning. And then paid the parents price when we had to get up to drive to the burbs to pick up our kid (from the always lovely Greg and Julie who graciously sat for us) in the morning.
• Ate a lot of great food, drank a lot of great drinks, and had general merriment with the greatest friends a guy could ask for. There are too many of you to list everyone, hopefully this blanket statement will suffice. Thank you every one of my dear friends, neighbors, and relatives, for another great year in which I didn’t die.

And now I’m raising a glass (no really – just picture it in your head if you have to) to another great year in 2014 – may it be even more awesomer than last year.

Despite the ranting title, I actually have a technical description of my problem, and then what went on. This is half for posterity so when I go to upgrade again later I won’t have to curse at my computer and the ubuntu forums, I will just be able to read my own blog.

Upgrading to 11.04

Ubuntu Linux upgrades every 6 months. A few months ago I read that 11.04 was buggy, and I shouldn’t bother upgrading until that settled out. So I waited until last month. Then I upgraded my server in the basement. My server holds media files and backups (now with super TimeMachine powers for backing up Macs), acts as my side-project web development server/code repository host, and records TV for me using MythTV.

Then I have a smaller, lower-powered, but able to play video at 1080p computer hooked to the TV upstairs, that streams TV from the basement server, or from hulu or whatever (except Netflix, but don’t get me started on how stupid it is that they don’t support linux).

So the server upgrade went flawelessly despite running on pretty outdated hardware – I built that machine in 2002, and haven’t done much to it since then.

Then another month goes by, and I decide to upgrade the front-end computer connected to the TV. This is a DISASTER.

As soon as I install the latest NVIDIA driver to be able to watch the video at 1080p and reboot, I get a blank screen. I try tons of stuff – tutorials, troubleshooting guides (this is apparently a pretty common problem), I even upgrade to the beta release of the upcoming version in October – Nothing.

So after days and days of spending time on this, and not being able to watch SYTCD or Dr. Who for a couple weeks, I give up and re-install ubuntu 10.10, which works flawlessly.

What do I think is going on from a technical standpoint: In the 2.6.38 and later kernels the bootloader can pass a parameter to the kernel to set the graphics mode (i.e. color depth and resolution), and that way the kernel can throw up a nice boot animation or graphic that is in the resolution you ultimately are going to run on the desktop. This avoids the black flash that happens on older windows and linux when the driver finally gets loaded and switches to the resolution – also called flicker-free booting. The program that does this nice hand-off or whatever is called Plymouth, which when I read about it over on http://www.phoronix.com and about how it would do all this wonderful stuff I thought it was great. But, it appears that it is pretty broken – especially with the proprietary NVidia linux drivers – which are the drivers that do all the neat stuff that I bought the NVidia motherboard to do (i.e. decode video super awesome w/ almost no CPU usage).

So, as I understand things it works like this when you boot Ubuntu:
Grub 2 loads and lets you select an operating system if you have multiples installed (I don’t), or lets you select failsafe modes or older kernels.
In the entry for the Kernel / OS it has some stuff about passing mode info – also there is some default that tells grub what mode to be in during this (so even your bootloader can looks less ugly).
When you start booting the kernel it loads your video driver and passes it off to plymouth to draw the pretty screen so it doesn’t flicker.
Then once that is all fine and dandy Xorg server (which is previously what set your video mode and caused the flicker at this point in older linux booting) gets a hand-off of control of the video driver and shows your shiny desktop. Except on my system it takes a giant crap, doesn’t leave any log messages or errors that can be deciphered, and shows you a blank screen while it hangs – silently crashed waiting in the background doing nothing but driving you insane – like a super zombie ninja with psychic powers of doom.

Anyway, I digress. I spent like a million hours trying to get this to work before downgrading back to 10.10 (and then having to re-fix my NVidia motherboards lack of audio out over HDMI – could we get this kind of crap fixed in linux please – audio and video are a giant fat mess – like 100 times worse than the switch to windows Vista, only it works way shittier and never gets fixed even after 5 years of running linux – Lame!). And, after all that, all I have to show for it is frustration, and a fear of upgrading.

Next time I will do two things differently:
1 – Just create a new partition, and install the newer one to that, and dual-boot – so I can watch Dr. Who or recorded TV on mythTV (new season of Glee! – AWESOME-Sauce!) – and still try out the new stuff to see if it is broken.
2 – back up my config files I change to my server. In general I’m starting to work up a real backup plan – and one of those things needs to be getting config files that I’ve changed or setup identified, and then backed up to a seperate folder which is backed up to the cloud. I’ve got a lot of hours into configing my server (in particular) and it is just built up over the last couple of years of just screwing around and making stuff work. Does anyone know a good way to go about that? Particularly, I don’t want to back up binaries, or version-specific files – but I do want to be able to upgrade and not have my stuff break (I’m looking at you ejabberd). And I want a good way to automate that backup (and I’m really getting annoyed w/ rsync, so unless you have a magic way to set it up easily, don’t suggest it – plus, in the world of dropbox, why can’t we get more seamless backup solutions for linux that monitor file system changes instead of backing up on a schedule?). And then a way to easily know where to put those files when I want to restore them on a new install or something.

Anyway, that is rambling now – but point is – on number 2, suggestions for things that have worked well for you are much appreciated.

To recap – Ubuntu – horribly broken on my Nvidia system (zotac micro-atx motherboard w/ 9300 graphics/chipset), need to be smarter about how I approach upgrades on critical and appliance devices (i.e. servers and single-function computers like my mythtv frontend) – Suggestions about backup strategies welcome.

One of the major players in the world of industrial control automation is Rockwell Automation. They make a vast array of products (or private-label from other companies to fill in their gaps), which include but are not limited to: Industrial servo controller/drives, variable frequency drives, PLCs (Programmable Logic Controllers – think hardened PCs that are made to run one piece of software in a loop, very quickly, and never crash… tho I have crashed them before, but that is another story), temperature controllers, circuit breakers, servo motors, HMI’s, etc.

They also make their own software to program their products. The most popular packages being: RSLogix5000, and RSView Studio – for programming their PLC’s and HMI’s respectively.

I use their stuff every other week or so when I’m not in the middle of a major development project (in which case, if the project is based around their hardware, I use their software every day).

Just a couple weeks ago, my PC was feeling pretty slow, and starting to crash occasionally in obnoxious ways, so I asked our IT guy if it was OK if I formatted my computer and reinstalled. At this time I casually mentioned that if we wanted another person to test-pilot Windows 7 I’d be up for it (a hand-full of our people have tried it with much success, though corporate isn’t quite sure they are ready to support it yet on every new PC coming in the door). He thought that was a good idea, so I did a little homework and found that while most of my software that is special for my job (like the Rockwell software) worked in Windows 7, almost all of it did NOT work in the 64bit version.

So I installed the 32bit version for now, and went on my merry way. But, now I’ve started to unravel some of the little annoyances, and things I’ve had to do to make them work right, and I thought I would share for anyone who follows in my footsteps. But before I dig into that too hard – I should note, I hosed up my computer for a day or so, trying to move my profile location to my second partition (not recommended, ask me if you want to do this and I’ll tell you what I messed up). That seemed to have caused some headaches, but re-installing most my software fixed the vast majority of my problems.

K, so there is the big list of issues:

Ultraware 1.64:
Windows XP SP3 – all our PC’s where I work, including mine prior to formatting for Windows 7, can’t compile projects. It gives me a Win32 Heap error. I fixed this by downloading the latest package of cygwin or mingwin (can’t remember which) and extracting the cygwin1.dll and replacing the one in the windows system folder.
Windows 7 – has to be run as aministrator (find the file and right click on it, go to compatibility and check “run as administrator”). If I recall correctly I couldn’t get it to compile otherwise, but I think I could open the program.

Panelbuilder32 3.82.01
Windows 7 – Couldn’t get it to open unless it was run as administrator. If I double-click a file to try and open it from the file explorer, I get an error and it won’t open, even though the application opens. If I use the File -> Open dialog it loads correctly.

But the big one here, is that when I go to download the file from the software to the device I get a blank error box, with just one of the exclamation point icons, and no text. I can click OK and that is about it. I had to grab an XP computer with the software on it to load it. I’m going to try downloading the “Windows XP” virtual machine from microsoft and putting it in that, and see if I can make it work somehow. But until then, I can edit and save file, but not load them onto hardware – which is a pretty major part of any of these software packages.

RSLinx 2.56
Installs fine, seems to load, at some point it stopped giving me an icon in the system tray (Maybe it never did and I just didn’t notice), and I can’t bring up the UI at all, even though it says the process, and subsequently the service is running.

The fix I found on Rockwell’s user forums. You can load the ‘control panel’ for RSLinx, and there you can stop the service, uncheck the box that says “Always run this as a service” and then restart the service. That gave me the icon in the system tray and let me configure communication drivers from within it. I also tried running it as administrator and turning off UAC. Dunno if that made a difference in the overall picture, because they didn’t fix the problem, but that was something that was done in case my control panel trick doesn’t work for you.

RSLogix5000 16.03
Just worked, and seemed to go great and do everything it was supposed to after I go RSLinx working. No need to run as administrator or any of that.

But, to any Rockwell people that stumble across this, I will say in general, asking me if I want to install adobe flash player EVERY SINGLE TIME I open your software, is a huge ‘no-no.’ If I want flash player installed, I will install it. And what the hell do I need it for anyway? Your start page is worthless to me if I’m not a total newbie. There are better ways to give me access to sample programs and help files. Flash is not the answer, or the future – or at least that is what Apple tells me – at any rate, it is annoying. Cut it out, get rid of that. (Note: I haven’t run v17 in a long time, and haven’t ever run v18, so if you got rid of this nuisance, congratulations – any chance you would patch v16 for me to get rid of it too?)

Latest Activation tools work fine, latest control flash for v16 CPU’s worked fine, so did the BootP server (though for whatever reason it didn’t seem to stick on a power cycle earlier today, I might track that down this afternoon).

So my big suggestion here, is that 1) if you are going to rely on GCC and cygwin for your compilation environment, make sure you’re updating your software with the latest bugfixes to the compiler toolchain. (cudo’s to you for sticking with ANSI C as your language of choice in the Ultra5000 BTW – cuz your normal structured text in RSLogix5000 blows, and I’m disliking IEC61131-3 less and less by the day – though it would be nice if you supported it natively instead of your hodge-podge structured text…OK, I’m getting off topic now). And 2) Please, PLEASE, PLEASE! support windows 7 with your older products (panelbuilder32, ultraware, etc), and get native x64 support into some of the older versions of the software like V16 of RSLogix. (For those of you who don’t know, if you have a controller that has its code written in V16 of the RSLogix software, you have to have V16-based firmware installed on the controller, and if you try to install V17 or V18 and load up the file you saved in V16, it will give you an error, ask if you want to convert. And if you do convert it, you have to change the firmware on the controller to match the rev. You cannot edit a file in V18, and save it with a target of V16 devices – which is super annoying because to work on multiple projects spanning features from multiple generations you have to have V16, V17, and V18 all installed on your machine, and when you load a file it pulls up the correct kernel of the software to support the format the file is saved in).

Finally, a word to controls people in general. Why do you prefer a single file for everything? I hate working with binary files when I’m writing code in structured text. There is no good standard way to do version control. Bzr, Git, Subversion, etc – 90% of their features are tossed out the window when you do this. Rockwell, I applaud the use of C for the Ultra5000, but seriously, packaging those text files into a binary blob, ensuring that I can’t get any benefit from DVCS or diff/comparison tools – not cool.

This goes for GE, Wago, and all you others out there. Our world is increasingly about writing code. Let’s not re-invent the wheel and make walled gardens everywhere, and instead adopt some standards text, or ASCII based file format that we can do simple version control and file compares on. Hell, rockwells proprietary diff tool exports ladder logic into XML, let’s just do that, every function or function block could be an XML file in a directory tree, with a single project XML file at the top connecting them all together. Or anything like that, something simple could be done.

OK, off my soap-box. Anyway, hopefully some of this is helpful for those who follow in my windows 7 path.

It is not saying anything new to point out that the iPhone changed the way we thought about and interacted with smart-phones. In the years following the Android phones brought us a vision of what a cloud-centric phone OS on many vendors hardware could provide if it was well supported. And the Palm webOS phones showed us what intuitive multi-tasking and a great notification system could provide.

With the improvements to the hardware and software over the two following years of iPhones and their respective software we got most the features we could want. I assume some sort of multi-tasking is coming in the 4th iteration. So we have reached a place where the revolution has evolved to a mature place of mobile-specialized OS’s and great hardware.

But where is the next revolution? I don’t think it is in the tablet space, which is being heralded as the next big consumer computing cash-cow. I think at least the next big evolution, or progression would be context-aware computing and inter-computer interaction.

What I mean by that is that I have massive headaches in some of the artifical boundaries, and poorly thought-out aspects of how our phones/mp3 players, tablets/netbooks, and regular laptops and desktops interact with eachother.

In order to explain this, let’s focus just on phones and go through the situation we have today.

I have drawn this little diagram that highlights what is considered basically the best way to ‘sync’ data on your phone.

The basics are that the best way to sync small data – email, contacts, calendar, and carrier data comes from the ‘cloud.’ The cloud here being any internet-based or telephony-based data that lives on a server somewhere far away and is ubiquitously available via a link to the net.

That same data is accessible via web services or the same as mobile protocols to your PC.

But when it comes to larger data, files, music, video, pictures, and some larger data sets from your app, the best way to move it between the PC and other computing devices is to connect a cable, and utilize the computer program specific to the device you are syncing. Every phone as a different non-standard way to handle this.

What works with this?

• iTunes lacks features for power users, but it gets the syncing job done well enough (finally as of iTunes 8 which gives you the ability to change the ‘media type’ from music to audiobook/podcast/etc).
• Cloud syncing is GREAT for information that best lives in cloud (i.e. Email, contacts, calendars)
• Phones in general have gotten pretty good for playing music, making calls and passing text messages, and GPS.
• Computers are great at creating and storing files, documents, and media.

What Doesn’t Work

So if you think about use-cases my phone usage based on what does and doesn’t work, and a lot of other people I know too, works out about like this: If I’m out and about, I get my email, set appointments in my calendar, create new contacts, and check and update facebook and twitter. This is of course in addition to taking calls. But the minute I sit down at a PC, I do all that stuff on the PC – because it is just easier to edit everything (except maybe twitter) on a PC. Especially longer emails and things that benefit from more typing. But the only reason I pick up my phone is to take a call or answer a text. And it is kinda a pain in the ass once I’m settled at a PC if my phone is sitting somewhere a ways away.

So, here is my list of thoughts on how we might address some of those things that doesn’t work and maybe improve the user experience.

• Let’s connect our phones to our PCs more so we can pass information more easily between the two.
• Start with the phone calls – let’s pass the phone calls to the computer and use the computers mic and speakers.
• The simple answer is have the computer support a hands-free profile, and the phone would connect to it like an earpiece. Then some type of notification would have to come up when there was a call.
• The more complex answer would be to run a VOIP or some type of messaging client on the PC, and the phone would receive the call, and forward it as a VOIP call to the PC.
• Step 2 is to expose a common driver/API architecture to expose onboard features of the phone or devices on the computer to eachother. So when the phone connects to the PC via USB or Bluetooth, let the computer see a USB hub with a camera, GPS, mic/speakers/audio device, etc. (See table below for a running tally of what I’m thinking)

Computer sees Phones as:

Device	Via
USB Hub	Driver / virtualization layer?
Bluetooth adapter	Driver / BT profile (a pass-through/relay type profile)
External Audio Device	Driver
Touchpad Mouse / Keyboard	BT Profile / driver via USB
Camera	Driver
GPS	Driver
External Storage Device/SD Card reader	Driver and network sharing
3G Modem	Driver / Bluetooth PAN
External Display (windows-sideshow style)	Driver (or app maybe?)
Voice Gateway – Expose the GSM modem or VOIP gateway	Driver

Phone sees computer as:

Device	Via
External Storage	Network Share / special app
Net connection	Driver / Bluetooth PAN
USB/Bluetooth Keyboard	Driver/BT profile/Server App
External Audio device w/ mic and speakers	Driver
USB Hub – the current apple ability to have peripherals added to the phone is vastly under-utilized	Driver / Extention to the BT profiles to encompass Wireless USB (since wireless USB is on the way out)
Video / Webcam / still cam	Driver
Secondary CPU / Processor core and GPU	Virtualization layer

• The next level would be at the application and OS.
  • Seamlessly move apps from the phone to the PC so you can interact with them from the PC. This way you can pick-up where you left off.
  • Allow a framework or API to sync applications on connection of PC to computer.
  • An example – listening to music on the way to work, get to your desk, phone prompts “Do you want to transfer this to the PC app, move the App to the PC, or continue on the phone?”
  • Move to PC app – music pauses on phone, music app (iTunes, winamp, etc) opens, and starts playing the same file off the phone from where it was paused.
  • Move app to phone – see diagram
    
  • Keep on phone – nothing changes
  • The final step would be having the apps be aware of the computer connection. Similar to the ‘app syncing, but instead the phone app could be transferred to the PC, but when it got there it would have a specific UI for that situation. It would see the CPU on the PC as a resource available to it, and start drawing a more desktop/laptop-centric UI with some expanded feature set. That way the transition to the PC would produce an experience tailored to the PC and have a seamless transition. This would have to have some type of virtualization layer that exposes the CPU and possibly memory or GPU exposed to the phone OS in a way that it can use.
  • This would allow for more ‘in-between’ situations to be handled with a middle UI for an app. So an iPad could have a version of the UI, the computer another, and the phone another – but all do similar tasks.
  • Moving away from the PC-phone interaction and on to messaging. I’m envisioning a new messaging hub. A timeline where you can see all your calls, emails, texts, chat messages, and whatever else you want. All this information pushed to you (instead of polling – which is so 1999). While the implementation details don’t have to be particular on this protocol, it needs to be open, and extensible (that is why I keep thinking about XMPP for the task, like what Wave, Gchat, and facebook chat are based upon). This protocol should have a built-in way to keep track of a message ‘thread’ so that no matter what medium you use, you can see how the chain of messages connected. This protocol should also support calendar events / invites as part of its payload, and it should have some way of a person updating their contact information remotely and if you are ‘subscribed’ to it, you would get the update (if they allowed you to, it would have to be a two-way subscription – with fine-grained controls for this)
  • This would need to have enterprise management features – as in the ability to sync a global contact lists, and manage the security policies of who can see what or book resources.
  • Bonus points for the protocol being aware of when a service like Google Voice forwards you a text message as an email, so you don’t double-up on notifications.
  • This would be a great way to handle generic notifications – meaning if it is extensible it could work for facebook updates, twitter posts, blog/RSS notification if the user so chose to add those features.
  • The final point that I think we should address in the PC-to-Phone interaction is syncing. We need a better, and flexible way of syncing.
  • I believe that as a starting base-point or guideline is that the phone OS makers should pick a single folder in the file-system, and make that a ‘shared space.’ Meaning apps can install to a program location, but they should store their data in the shared space. They should enable file sharing over the network, and the ability for the phone to act as an external storage device (so the phone appears as if it is a USB hard drive). This folder should be appear as a shared folder over the network, and should appear as the root folder when the phone is connected.
    • Some basic common sub-folders within this folder should be agreed upon. I’m thinking:
    • /media
    • /media/video
    • /media/music
    • /media/podcast
    • /documents
    • /app data
    • A matching folder structure should be tied to the computer for backup and ease-of-sync purposes.
  • Secondly I think that having a ‘dropbox’ like tool running on any approved PC, and the phone, which watches for changes to the files in the shared folders. If the PC is in range or connected via USB/Bluetooth this tool would sync the file changes. I think this is particularly important for the documents that can be easily edited, but it gives you a way to back up pictures and videos easily also which is a plus.
  • Finally an agreed upon transaction model for this syncing for libraries. I just mean that some file types, or apps build up a collection of metadata on files, and call that a library. It is fancy way of saying that an application has an internal database of the files and the info about the files that are important to it. iTunes, the music apps on the phones, winamp, songbird, iPhoto, etc.
  • Let’s standardize the way changes are communicated to be bidirectional. This could be a part of the tool above, but if a change happens to a file, or a file is added to the library, when the phone and the PC connect either one can say “Hey, I’ve added files or modified files – do you want the update”.
  • I don’t care what the ‘database’ is that is tied to the app managing your media files, but the way in which the changes are communicated so that a transaction (file updates, etc) take place need to be standardized. By not necessitating a specific back-end to this process we can keep our current DRM schemes I believe for music and video.
  • Also, let’s standardize a smart-playlist-like functionality for syncing collections like podcasts, audiobooks, and series of talks or lectures. So that you can individually say this file, or this series, or this collection is to be synced. If it changes, sync it to wherever doesn’t have the changes, and if there is a conflict – modified in both places between syncs, ask me what to do.
  • The key is, that this has to be seamless for the users – which is why I like dropbox, it is always running and as soon as I save a file it is watching it uploads it to my cloud storage. This could be like that, only just between the PC and the phone and could handle libraries built out of links to files and metadata. (you could add the cloud component too, but a more standard way of handling this would be good)

  So, there you have it, my cell-phone OS wishlist. This is true of the new Windows Phone 7 Series (someday I believe Microsoft will get this naming thing down so it isn’t so silly all the time) which I just heard about 10 minutes ago. Nothing in there suggests to me that Microsoft is doing anything in this direction beyond their typical exchange server support.
  
  Now, my question to you smart-phone users out there; What is your beef with how your PC and phone interact? What would improve that? And, would any of this help you use a phone or a tablet in a more effective and satisfying way? Or is it just me who doesn’t like the way things are now?