Nik's Notes

Day to Day with Google ChromeOS and the CR-48 Netbook

2011-02-15T20:48:00.000-08:00

When evaluating the Chrome Netbook you have to take two things into consideration - the OS is an integral part of the user experience but if the hardware isn't good, it reflects on the overall satisfaction.

The Netbook hardware could be better.

When I open the lid, the OS boots up in seconds as advertised. That said, the network takes a lot longer to connect which renders the Netbook useless. The real boot time should be measured in time to usability, not time to seeing flashy lights.

The screen also flickers precariously when just opened - it's not a big deal but it doesn't instill confidence either.

Recenty I had to write up a meeting report and decided to use the Chrome Netbook. I love Google Docs and have no problem using it as a word processor.

When writing my report, on numerous occasions the mouse would jump to what seemed a random place in the page. I probably touched the touch-pad with the palm of my hand or something. That's not necessarily an excuse though - it never happens with my MacBook. I don't have to pay attention to where I am writing text on my MacBook - it writes where I expect it to write. With the Netbook I have to pay attention.

The search button started to drive me nuts. Google decided to replace the CAPS lock with a search button that opens a new tab. I didn't realize how deeply ingrained the urge was to hit the CAPS lock button... I have changed my settings so that it acts like CAPS lock again.

On the upside - I figured out some neat tricks to save web sites to pdf and then to Google docs and to capture screen images and copy and paste them into an email. The battery life is really great - I took the Netbook to work one day and deliberately left the power supply at home - not a problem.

Could I use it as a work computer? I think the answer would have to be not yet. Why? Because I am confident that Google docs or Zoho will evolve to provide more advanced features that will wean me from Excel, that Avery will add features that allow me to edit or create images without the need of an app, etc. etc. In fact it's more than that - I believe that SaaS apps will provide features that boxed software will never be able to provide.

Wow - I have a Chrome OS Netbook... Yay ..?

2011-02-09T09:37:00.000-08:00

To my utter surprise I came home yesterday and found a big box in my front hall. I had totally forgotten that I had volunteered to test the Chrome OS Netbook and didn't really believe that I would actually get one. And there it was.

Sleek matte black design, 12" screen, full size keyboard. Setting it up took minutes - all I needed was my GMail account and Google pulled all my info (bookmarks, etc.) from... hey, how did they do that? I didn't think I had sharing or sync enabled... hmmm

Anyway. So now I have a device that is basically an Internet terminal. Being limited to just a browser spurs creativity - you have to think out of the box to get simple things done.

First of all, you need to make sure you are always on a network - the Netbook includes Wifi and 100MB of free data/month or unlimited data @ $9.99/mo with Verizon - not that bad really (100MB doesn't get you very far).

I like to keep receipts so after going through the very simple Verizon set up process I thought I would keep a copy of the confirmation page. So how do I do that on Chrome OS? You can't really print - unless there's a "Windows PC available" - seriously??? How about just allowing me to print to pdf? Couldn't find that option.

This is going to be interesting. My setup is a little convoluted in the Google sense - aside from my GMail account I have several Google Apps accounts and keeping them separate is difficult on a normal PC (I use different browsers).

After the initial excitement of getting a shiny (or rather matte) new tech toy I just have to take a step back and ask myself some more existential questions.. such as: Why is Google pushing Chrome OS if they have Android?

Update: My buddy German pointed me to this great plugin: Send to Google Docs. Promises to do exactly what I was looking for before.

Faxes ain't goin' nowhere... But it is time to upgrade

2009-05-25T17:36:00.000-07:00

Even technophiles and geeks will have to admit that faxing is not dead - in fact it looks as if it will never die. As strange as it may seem to generation X (or is it Y or Z now?), there is still an established hard core user base that will just not let go.

So one day even you, the technologically super-advanced will be faced with the question: "So what's your fax number?"

Luckily, there are solutions for that, and they are pretty much all web based. eFax was probably one of the first out there. But when they cancelled their free service YEARS ago, they lost me as a customer. Not because I didn't like their service but because I didn't like their business model.

A lot of the software-as-a-service companies forget that there are users like me, who see value in their offering but probably use it very rarely. I'm not going to pay $20 a month just so that I can send or receive a fax 5 times a year.

Enter Trustfax - I now pay $30 per year which covers more faxes than I will ever need (sending and receiving) and if I ever have to send more, I can top up my account.

Here's what I love about it: I can send faxes directly from the web interface or I can send an email with an attachment, like a pdf, and Trustfax takes care of well.. the faxing. I can check the log on my web dashboard and I get an email receipt.

As we are all riding the green wave: aside from saving paper because I now receive faxes in my email inbox, I can also ditch my fax machine, thereby saving one more device that needs ink cartridges that are (obviously) not the same as my other printers...

There is one enhancement I would like to see: they should add the sound of a dialing modem.. just for old times sake!

In case you wondered: this is NOT a sponsored post. I just like writing about services or companies that I find useful. Sometimes it is not product features that convince me but the business model. Listen up all you SaaS companies out there.

A Fresh Look at Personal Finance Software: Mint.com

2008-12-04T09:41:00.000-08:00

In March I blogged about my quest to find an alternative to Quicken in the post "Can I finally kick out Quicken?". At the time I compared Quicken to Moneydance - a cheaper and yet great alternative for personal finance management.

And then it struck me - I am SO OLD SCHOOL! What am I doing using traditional software if there is a beautiful Software-as-a-Service solution out there called Mint.com?
If you can get over the fear of having all your financial information floating in cyberspace, this is absolutely the way to go.Why? Because you don' t have to remember to log in to every account you own every time - Mint does it for you. I set up my Mint with 7 accounts - savings, investment, credit cards, you name it, Mint can take it.

What I love about it is that I can just log in at any moment and get a snapshot of my financial situation. No fussing around required - it just works.

What I don't like that much is the limited report capabilities. I can see a nice pie chart of what I spent in any given month but I miss my column graphs that show me how little cash flows in opposed to the huge amounts that flow out on a month by month basis.
That said, who am I to complain about a product that is completely free? How do they survive you may ask? According to a forum post:
"By showing our customers special offers that can save them money. When users accept an offer from the "Offers" page, Mint makes money from the referral."

Perhaps this is an opportunity for Mint to make some extra cash: a "pro" version that speaks to graph and report fanatics like me. I would definitely spend $20/year on a service like that. I just read that Mint may have another business model in mind - detecting spending trends. Read more in the Techcrunch post titled "Mint Joins The World Economic Forum, Knows That You've Cut Back On Starbucks"

It is no surprise that Quicken followed suit rapidly with an also free offeringnamed Quicken Online. I decided to compare both by tracking my small business financials. Turns out that Quicken shows incredible consistency in providing buggy software. This time it was a simple, two account setup - Bank of America and Paypal. Transactions that Mint showed (and that actually existed) did not appear in Quicken Online. One advantage that Quicken could have leveraged - integration with other Intuit products - was sorely ignored. One can only download transactions in csv format. I was hoping that I could export all my transactions and import them into my quicken app to do some reporting. No such luck.

Oh well... Nice try Intuit but Mint is the clear winner.

A Virtualization Convert - Parallels vs VMware Fusion

2008-11-27T16:00:00.000-08:00

I met a guy at a party not long ago who works for VMware. We had a lively discussion as I have been a Parallels user for over a year now and just upgraded to their 4.0 version.

This guy did not let loose and convinced me to give VMware's competitive product "Fusion" another try. I have to add that I actually tried the two about a year ago. Although Fusion seemed faster (which my ultrageek CTO at the time confirmed through in-depth tests) it lacked the Mac-ish user friendliness, so I stuck with Parallels.

Inspired by the party-guy's passion I downloaded VMware Fusion 2.0.1 and contemplated the consequences of messing with my tried and trusted setup.

After making sure I had enough hard drive space and convincing myself I couldn't really do too much damage as I was protected by fantastic Time Machine back up technology, I installed Fusion.
Here's where I hit my first snag: the new Parallels format is not compatible with VMware's import tool. Not a problem of course: all I had to do was retrieve the old format from Time Machine. With hard disk space getting thin I managed to retrieve the image and convert it to the Fusion format with the free tools provided by VMware.

The transition was remarkably uneventful... really, really smooth. When I launched Fusion I found a clean and easy to use interface - a big improvement to when I first tested it. After configuring all the options I started the VM for the first time. It went through automatic configuration routines, installed the VMware tools, similar to what happens with Parallels and after another reboot I was up and running.

Virtual Machines tend to hog a lot of processor and memory resources. Every time I launched Parallels, working on Mac apps turned into a pain (although watching a colorful ball turn can be so much fun...). I won't say it is blazing fast with Fusion but it does seem that a) Windows is running faster and b) Mac apps are still performant enough to not drive me nuts.

All in all, I must say: I'm a convert. I'm sticking to VMware Fusion.

Keeping my Mac and Google in Sync - Spanning Sync

2008-08-11T10:00:00.000-07:00

Undoubtedly synchronization is hard. That's probably why so many companies get it wrong. I love Macs but iSync is not one of the apps that deserves praise.

My objective was simple - ensure that all the relevant data on my Mac is backed up on the web and with relevant data I mean contacts, calendar, email and if possible docs as well but that's a 2nd in priority. I went through multiple hand-held devices (Nokia, Sony-Ericsson, Palm, Blackberry, etc.) that synchronized OK with iSync but I caught myself holding my breath every time I would initiate a sync.

I also tried .mac (before it turned into me.com) to see if Apple could hold their promise of a happy synchronized world. They couldn't - not back then. I gave up on .mac but decided to give it another try when it morphed into MobileMe. The promise of "Exchange for the common person" was too much of a draw. I must say I am in awe.

It works... Apple certainly got their act together on this one. I can actually sync contacts and calendar items over the air with my iPhone, MacBook and me.com account. I LOVE it.

That said - I'm still locked into my pretty Apple world. I still need integration with Google. I use GMail a lot, privately and at work, so integration with Google is important. Specifically, I need my Google work calendar to sync with my Mac - I'm not too worried about contacts right now and email integration works fine.

That's where Spanning Sync comes in. They really know their stuff. Setting up multiple calendars is a breeze (I have 4 for some reason...) and I have had none of the common synchronization problems such as duplicate entries, entries disappearing and things like that.

I have not installed the beta of 2.0 that promises synchronization of contacts as well - mainly because I wanted to sort out my MobileMe setup first. I think I will wait until it is officially released for a change. I certainly will upgrade though - I like not having to manage multiple contact databases across disparate systems.
My initial focus was on the calendar though so now I have a seamless sync with Google via Spanning Sync and as long as one of my Macs is switched on (laptop or desktop) also a seamless sync across my many Apple gadgets with MobileMe. It would be great to have a server version of Spanning Sync that takes care of synchronizing Google and MobileMe in the cloud - maybe I should suggest it to them...

If you want to try it out click on the link below - that way you can save some money if you decide to buy it.

Stop Motion Video on a Mac - Take II

2008-08-07T09:25:00.000-07:00

In my last post I described how I set up a stop-motion video "surveillance" system - the objective is to document the construction of new school buildings.

Overall it has gone pretty well so far. You can see the results on the Chabot Elementary contstruction update page.
Here's a quick summary of experiences:

Imagecaster is doing a great job managing the cameras. It would be great if I could not only determine the time of day it is active but also the day of the week. Right now we have to manually delete all the pictures taken over the week-end where there is no activity. There also seems to be a minor but - I am managing two cameras on one computer (two instances of ImageCaster) and sometimes ImageCaster crosses over images from one camera into the other camera's folder. Again - something a manual edit can take care of quickly.
Logmein has been a god sent as I had to log in remotely many times to fix or adjust things. I even got technical support despite using the free version.
I needed to adjust the parameters for Quicktime a few times to adjust the resolution of the movies. After a month, the movies were approaching 100MB in size. After altering some parameters I managed to reduce the size to under 30MB with acceptable quality.
Automator rocks and so does AppleScript - this is coming from someone who last programmed in 1989 (in Assembler!). I'll include some code samples at the end of the post.

There are two other technologies I haven't mentioned that I needed to make this happen:

Ethernet over power lines
FTP client

Ethernet over Power Lines
One of the rooms that hosts a computer and two cameras did not have Internet access. I tried Wifi but the room was out of reach. The solution was Ethernet over power. After a bit of research I found the Panasonic HD-PLC Ethernet Adapter Starter Pack. I had read about Ethernet over power lines before but never tried it. I don't generally read manuals so I just plugged the two devices in and it worked! I had Intenet access instantly.
This was a really simple solution and cost around $80.

FTP Client
After I had managed to pull together all the technology to create the images and the movies, I still needed software to upload the movies to the web server. A pre-requisite here was the ability to automate the process with scripts. For this task I used Transmit3 by Panic.com.
Transmit has great scripting capabilities and also prewritten scripts - I used one to back up all the individual photos onto the server. I asked for free licenses for my non-profit project and was granted one - thanks to the people at Panic!

One FTP client that I couldn't get to start on 10.4 but works fine on 10.5 is Cyerduck. It is also scriptable and seems very full featured but as I couldn't get it to run on OS 10.4 I used Transmit.

The Code
For those who are interested: here are the scripts I used within Automator. The first one is to convert images into a stop-motion movie:

-- AppleScript to get Quicktime to create the stop motion movie

-- This is to avoid that the script times out:
with timeout of 10000 seconds

 -- Quicktime actions:
 tell application "QuickTime Player"
     activate

     -- point quicktime to the first file and define the frames per second (fps)
     open image sequence "Macintosh HD:Users:UserName:Pictures:picture1.jpg" frames per second 6

     -- export movie in a format more suitable for the web. I played around with Quicktime and saved the configuration settings in a file
     export front document to "Macintosh HD:Users:UserName:Movies:movie1.mov" as QuickTime movie using settings "Macintosh HD:Users:UserName:Movies:ExportSettings" replacing yes
     -- instead of a file you can also type e.g.: using settings preset "Lan/Intranet"

     close front document saving no
 end tell
end timeout
tell application "QuickTime Player" to quit

Here's the script that uploads the video to the web server:

-- Using Transmit from Panic.com to upload movie files to the web server

-- This is to avoid that the script times out:
with timeout of 2000 seconds
  
   -- Transmit actions:
    tell application "Transmit"

       -- Prevent interactive alerts from popping up during script execution
       set SuppressAppleScriptAlerts to true
      
       -- Create a new session window for the script
       set theDoc to (make new document at end)
      
       -- send commands to the frontmost document window
       tell current session of theDoc
          
           -- Connect to password protected server
           connect to "mydomain.org" as user "login" with password "password" with connection type FTP using port 21 with initial path "/media"
          
           -- Set your stuff listing to "~/Movies" folder and upload .mov file
           set your stuff to "~/Movies"
           upload item "movie1.mov" with resume mode replace
       end tell
   end tell
end timeout
tell application "Transmit" to quit

Yes, it's that simple!

A WebCam, an iBook and Stop Motion Video

2008-06-03T20:48:00.001-07:00

I volunteered for a project at my kids elementary school: record the historic remodeling of the school with web cams.

The idea is to end up with a stop motion video that compresses the construction work of many months into a few minutes of video.

It goes without saying that the budget needs to be kept at a minimum so I decided to use as much equipment as possible from the school inventory and beg for the rest.

The school owns many iBooks so I started with that as a basis. Given their age, they still run on Mac OS 10.4 which restricts the other parameters: the software and the cameras to use.

After a bit of research I learned that there are some versions of Logitech webcams that plug and play with Mac - in fact it was astonishing to see that there are hardly any 3rd party cameras that are good quality iSight replacements.

The camera I decided on was the QuickCam® Pro for Notebooks, listed at USD99.99. I checked with Logitech and after a little trouble getting hold of the right person, found a very generous soul that sponsored all the cameras I needed!

On the software front I tried multiple applications. The one I liked most was ImageCaster by Econ Technologies. It works great on the Mac and has all the features I needed to record images at pre-set intervals. It can even create a web page for you.
Econ Technologies also were very generous and provided the licenses I need for the school project for free.

ImageCaster manages taking all the individual images and dumping them in a folder, I was still stuck with how to convert them all into a movie. I tried iLife applications - loading the pictures into iPhoto and then using iMovie to grab them and make them into a movie. It worked fine but I wanted to automate the process and it just didn't seem elegant enough.

Then I learned that QuickTime Pro had a very simple feature called "Open an Image Sequence" - Just what I needed. Not only is it simple - all you do is point to the first picture of a series - you can also set the frame rate (how many frames, or in this case pictures, per second). I also learned that Quicktime is scriptable! Not that I knew how to script but I sensed it would come in handy.

Putting it all together
So I had all the pieces, now I needed them all to work in harmony. Setting up ImageCaster was a breeze. I could define what folder I wanted the images to be loaded into. I decided after some tests that I would go with one picture every 5 minutes.
Then the hard part started - automating the process of converting them to a movie and loading the movie to a web server.

I decided to play around with Automator and Script Editor. Automator is basically a graphically oriented way of programming and very well integrated with Mac applications. It also allows you to include custom scripts - something I needed to control Quicktime Pro. After some initial problems due to my lack of programing skills it actually turned out to be really easy (thanks to some big help from capitalj on the macosxhints forum).

The script literally only consists of opening QuickTime, pointing it to the right start file and then closing it... it is that easy.
Automator made the rest really simple - copying files from one location to the other for example. Best of all, it allows you to save the script as an iCal plugin which in turn provides an easy way to schedule the events in the Apple calendar.

So my test setup was complete:
- ImageCaster records images from the Logitech camera every 5 minutes into a folder on the local hard drive of an iBook
- An iCal event triggers a workflow (created with Automator and Script Editor) twice a day that
has QuickTime create a 12 fps (frames per second) movie out of all the files in the folder, then grabs the movie file and overwrites the moive file on the web server.

That's how it works in my test setup. I'm keeping my fingers crossed that it works as smoothly in the real life implementation.

Drowning in Social Networks

2008-05-21T14:47:00.000-07:00

I would consider myself technically adept and on top of technology fads. One trend I was late to engage in was social networking. In fact, I didn't actually engage in it, it just took me hostage.

The barrage of email requests from friends to join Facebook basically wore me down to a point where I caved and signed up. From then on my social networking career was on autopilot. With no effort from my part whatsoever my network grew and grew.

Although I find the phenomenon very interesting I am puzzled about the myriad of useless applications available to users. Why would I want to send someone a virtual cocktail, let alone pay for it??? A wonderful summary of the uselessness of Facebook is portrayed in this YouTube video (I'm afraid it is in Spanish...) - that was posted on my FunWall

If you don't understand Spanish - believe me, it's hilarious.

I am not trying to indicate that social networking is useless. I have to admit that my initial statement was not quite correct as I was an early adopter of LinkedIn - only at the time, I didn't realize it was called a social network. LinkedIn solved my problem of not being able to keep my rolodex up to date - that's why I signed up.

Now I think it is an incredibly powerful tool - especially if you are searching for a job: How often have you applied for a job by submitting your resume to a web site? You might as well put the resume in a bottle and throw it in the ocean...

LinkedIn actually shows you how you can connect to the recruiter - the better connected you are, the better the chances you have to get an interview.

That's what social networking is all about - it is based on trusting your friend's advice and opinion.

Another one I got roped into is Plaxo. I knew it from the past as a glorified address book application. Now it has metamorphosed into a social network. Not only that, they just got acquired by Comcast... what the???
I don't understand this world any more. Admittedly, Plaxo seems to be the only application that has a decent tool to help clean out duplicates (it's still not perfect - there's an business opportunity for a good address book cleaning application!) but that can't be why Comcast bought them. This is just as bizarre as eBay buying Skype.

So now I am not only connected, I am in some cases connected twice or three times to the same person through multiple networks. LinkedIn has certainly carved out a great niche and made it useful. As for the others, if you build an application, the usefulness will come.... some day...

The Involuntary IT Guy - how LogMeIn made my life easier

2008-04-17T08:59:00.000-07:00

Are you blessed with computer knowledge above and beyond the average users'? Have you ever spent HOURS on the phone talking your father / mother / brother / aunt / neighbor through sending an email or installing the latest version of Firefox?

Well, I have. And being a bit of a geek, I have been on the lookout for a good remote administration solution - one that let's me log in to my relative's computer and take it over. I tried quite a few.

The basic criteria I was looking for was:
- free (or very cheap, one time fee)
- easy to use (so non savvy people can install it)
- has to work with both Mac and PC
- reliable

I tried VNC - but I am not that much of a geek that I went as far as creating my own VNC server. So I found services that did that for me. One early technology was Hamachi, in fact it is so good that it got acquired by LogMeIn - more about them later.

I tried Skype plug-ins, Java based apps, MSN has a feature to hand over control to an administrator, there are some great apps that wrap around VNC and make the connection process a snap - but the one I found only worked on windows. They all had their pros and cons - some didn't work across platforms, others to hard to explain how to install...

Then a real IT guy (someone who gets paid for it) told me about LogMeIn.

As mentioned before - they acquired my first favorite solution Hamachi - Hamachi only solved part of the problem though - the VNC server part. LogMeIn takes it a step further and adds a management console so that I have a list all the PCs that I have access to and see their state (available, off) and more importantly - they make the installation easy.

It is so much easier to SEE what is going on than to imagine what is happening - LogMeIn is saving me some gray hairs.

Apple will support Microsoft Activesync! hurray....?

2008-03-17T14:30:00.000-07:00

I watched the March 6 event keynote where Steve Jobs introduces the SDK and it was good... but not Apple-esque, mind-boggling great.

Yes, having Exchange integration is going to make life a lot easier for a lot of cool dudes in suits who can now get rid of their Crackberry and fully engage in the love affair with their iPhones. My feeling of disappointment stemmed more from the fact that I believe that Apple could do so much more.

What for example? Here's a suggestion: make collaboration easy, usable and fully transparent AND accessible to ANYONE (not just corporate users).

The fact that MS Exchange is currently the only complete solution out there should be an indication that there is a business opportunity, especially because it is only available to enterprise users. I believe more and more non-enterprise users are ready for calendar and contact sharing (to say the least).
Apple shines in pulling together 3rd party bits of innovation and making them into something out-of-this-world-fantastic (like the iPod, iPhone, Air, etc. etc.) and available to the average (but very cool) user.

Some things I don't get: Apple already uses IMAP in .Mac to solve the email problem, why not use the iCal Server for calendaring (is it just me or is anyone else still trying to find the calendar in .Mac?).
If Apple thinks that pulling the pieces together is too much work, there are some companies out there that have done it for them. One example is Funambol, an open source company focused on mobile messaging; another one is Sharpcast with their universal sync platform.

My point is: there is no ideal sync solution today - not even Exchange. Anyone who has had to deal with Exchange will agree with me that it has its challenges. Plus - it is limited to email, calendar and contacts.. why?

Apple could offer seamless, transparent, push sync of email, contacts and calendar PLUS transparent synchronization and backup of photos, music, applications and files. Yes, everything Phil Shiller, Apple's sVP of Worldwide Product Marketing showed that Activesync can do and A LOT MORE.

Phil showed how a contact is available immediately across the network no matter where it is edited. The same should be available for any kind of data. Imagine taking a picture with your iPhone and when you get home, the photo is already in iPhoto and in the .Mac Gallery. You download a tune at Starbucks and when you get home it is in your play list in iTunes.

What gets really cool is when you play around with sharing features, where I share one photo album and one group of contacts with Aunt Petunia and other photos and contacts with my colleagues. But I am getting carried away listing the endless opportunities.

My point is the following: Apple doesn't need Microsoft Exchange. Apple could pull together existing technologies and put something far superior on the market in under 6 months.

If they did, the average user (like me) would adopt it rapidly, start networking with family and friends and eventually push it into the enterprise.

Head on competition with Microsoft Exchange? Now that sounds like fun!

Can I finally kick out Quicken? A closer look at Moneydance

2008-03-04T06:56:00.000-08:00

I have wanted to kick Quicken out for such a long time... Every year when Quicken releases a new version my hopes are high that finally, this version will fix the bugs and undo the awkwardness of use. So I shell out the $50 or $60 or whatever it costs.. but no... no luck. Some things improve, some things get worse, some stay the same - but overall I remain disappointed.

Disclaimer: this is NOT a sponsored post, although I wish it were - I would love to make some money writing about my experience with Quicken and Moneydance. Hey, Moneydance - if you see this, why not start a campaign with Snapbomb so that you can get more reviews like mine from people that are suffering from the Quicken quasi monopoly! For that matter - Intuit should do a campaign too, to get some real feedback.

Quicken is undoubtedly one of the leaders in the personal finance software market. There is also Microsoft Money but that's not an option for me as I am a Mac user. So for us Mac fanatics, there hasn't been much of a choice in the past years. Another factor is that once everything is set up, it is very hard to move to another software. Sadly, there is no "export from Quicken, import to your new favorite application" function that works smoothly. So my experience has been that it is pretty much a clean cut: stop using Quicken and start the year afresh.

My strife with Quicken
So what is it that bugs me about Quicken? And I have to clarify: I don't ask for much! All I ask for is a little simplicity in life!

Finance is monotonous and in my book, a burden (no offense to all my CPA, CFO and accountant friends out there...). Personal finance software should relieve me of the burden. It starts with small things like auto-completion.

Quicken will claim to have it, but when you download transactions from your online bank (which is the only way I enter transactions), it forgets. A simple example: Checks.
In the process of downloading transactions, my bank shows a series of entries with the description "check" and the category "bank charges" (how ingenious). Thanks to the check number cross I can verify what I used that particular check for (looking at my check book or the online scan of the check). Based on that information I edit the description, e.g. change it to "baby-sitter", a process I have to do pretty much every month. Now you would think that it shouldn't be too hard for Quicken to figure out that the description "baby-sitter" should automatically invoke the category "child-care" given that I do this every month... but no, Quicken insists on defining that transaction as a bank charge.

You may think that's just a minor quirk (although after the 1,000th time it gets annoying...). Here's a bigger one:

Don't ever think that you can just do a quick "accept all" when downloading transactions form your online bank or credit card. You will soon realize that Quicken is not smart enough to match transactions on the same day of the same amount if the description is not the same - for example: credit card payments.
I can't count the times I have wasted hours trying to find why my accounts don't balance, only to find that I had two CC bill payments with different descriptions: one that I downloaded from my bank account and one that I downloaded from my CC account.
To make my life simple, Quicken should at least ask me if two transactions are the same if the date and the amount are the same. To make things worse - I can't actually manually tell Quicken that a matching transaction already exists.

Quicken has trained me to be suspicious of how my finance software treats my finances...

It gets worse:
Quicken has bugs!
Ok, that's not surprising - it is software after all. But these bugs drive me nuts. It has happened multiple times with cross account transactions (e.g. paying a credit card). This is how it works:
I download a transaction from my checking account that includes a payment to my credit card. I have to double check whether Quicken got the credit card account right (as normally it doesn't) and then accept the transaction which is then recorded in both accounts. For some reason, sometimes (not always) the credit card account will show the transaction but there is no dollar amount. It drives me insane - financial software I can't trust...

In search of better software
Google revealed a number of financial software packages but mainly targeted at the enterprise. I did find one that I hadn't heard of before: Moneydance.

A software package that claims simplicity and works on Mac, Windows and Linux - what a concept! Plus, it only costs $29.99 - that's a no-brainer.

I tried it out a few times - first by trying to export EVERYTHING from Quicken and importing it to Moneydance. It did sort of work but it imports all the quirks and errors I take responsibility for due to my lack of accounting skills. Plus I didn't want to spend hours checking whether the import was correct or not. My addiction was still too great to Quicken that I never fully took the plunge.

So this year I decided to start afresh and forget about my financial history (I'll leave that in Quicken) and use Moneydance for all new transactions.

Why Moneydance is cool
I must say - after the first two months I am delighted. It seems to do just enough. Not too much and not too little.
Online connectivity is essential for me - I pick my bank based on whether I can download transactions. Moneydance works like a charm, in fact it works better than Quicken with my brokerage account that has a checking account attached - Quicken would keep telling me that it didn't recognize the bank but would import the transactions if I downloaded the QIF file... it would just get confused. Not Moneydance - it simply works.

There are still things I need to figure out i.e. I haven't tried entering a budget or managing a mortgage but I consider that higher finance - right now I am just checking out the basics.

There are other aspects that are better than Quicken, i.e. the ability to manage transactions in multiple currencies - I know most people won't need that but I do. In Quicken it is impossible. In Moneydance it is simple. You just specify the currency when you set up an account. Exchange rates are updated automatically through the Internet.

In conclusion
So am I finally ready to kick out Quicken? Not yet. I am actually running both software packages in parallel for now. I want to test Moneydance for a few more months before I let go of Quicken. I most certainly will not shell out the $60 for an upgrade this year.
Even if I still carry the burden of Quicken, I must say Moneydance is very promising - I think this may actually work out!

Why Embedded Software isn't a Billion $ Industry

2008-02-13T09:15:00.000-08:00

In my last post I described technical or developer related challenges the embedded software world faces. In this post I will look more at the business challenges to try to understand why embedded software is not a billion $ industry.

The technical challenges basically define whether a product is sellable. Let's just assume you can overcome those, then you need to face the challenges of pricing and business model.

So lets move on in the series of Problems with Embedded Software:

4. The Value Gap

Embedded software is mostly a "technology" or rather a "software component" as opposed to a complete application. It solves a specific problem, i.e. how to manage a device or how to communicate with other devices (through standard protocols) or how to secure communications, etc.

From the equipment manufacturers perspective - especially from their accountant's perspective - it's another item on the bill of materials (BOM). This is where the "Value Gap" comes in.
The embedded software vendor spends millions on developing a product and thinks it is the best thing since sliced bread. The OEM is developing a product and sees the embedded software component as a tiny part of the whole puzzle that constitutes a product. That tiny part should only get a tiny percentage of the BOM.

This is how I see the product value chain:
1. Complete products and full applications rock - that's where the $$$ is

2. Frameworks are useful and can help with product quality, deployment, management etc. - they can get some $ too

3. Software components are bottom feeders - they get ¢ if they are lucky. They are the resistors and capacitors of the hardware world

5. The runtime business model
The runtime business model aka royalty aka per-unit-fee aka many other things is fundamentally flawed. It is like a drug for embedded software vendors promising unlimited upside potential if that one product would just take off and sell in the quantities the OEM projected...

The flaw I see is in the association of value to the embedded technology. A runtime model assumes that the value is greater the more units are shipped. In fact the value of an embedded software component is to the developer who saves time creating the complete app. It is totally irrelevant how many units are shipped - the software stays the same.

So if a product actually DOES become successful and starts shipping in the millions, the OEM switches to a margin optimization mode which translates to squeezing suppliers and replacing parts that figure too prominently on the BOM. Embedded software becomes a victim of its own success.

6. Supplier dependency

Last but not least (but I think I have ranted enough now...) - embedded software companies tend to be small. OEMs don't like to depend on small companies that could go under. In the HW world it works because the component providers are huge and financially stable.
The OEM is always faced with the decision of taking the risk and betting on a tiny software company or building the software in-house where he has complete control.

THE SOLUTION TO THE PROBLEM
Ok - the title is a little pompous... but I do have some ideas.
Embedded software vendors need to realize that they are selling a commodity and treat the product as such. The mere fact that they are building a standard technology and make it available to any OEM would qualify it as a commodity in my opinion.

I believe that the pricing for software components should be very low, in fact I love the open source business model that makes it available for free.

So how can embedded software companies make money? -> Follow the value

1. Leverage the fact that you as the creator of this cheap/free software know it better than anyone and can offer paid services to adapt, change, enhance, integrate and support it.
2. Try to create a razor/razor-blade model: give away the software for free and sell tools that make using it and testing results really easy
3. Take the software component and based on the knowledge of your markets, customers and uses, pick an application and build it.

But if you want to aim high and make the big bucks - go for the full application or product.

The Problem with Embedded Software

2008-02-11T14:01:00.000-08:00

After 8 years in the embedded software market I have formed my opinion on why embedded software isn't ruling the world.

What is embedded software? Software that runs in "things" such as phones, cars, planes and refrigerators. PCs are excluded from the list - they are a "thing" category for themselves.
That is an oversimplified description but let's leave it at that for now.

Seen from 10,000 feet one could easily think that the embedded software industry should be booming. There are more and more connected intelligent devices that are run by more and more complex software. I would even bet that most equipment companies employ more software developers than hardware developers today.

So why isn't this a multi-billion $ market? There are multiple reasons why - here are a few:

1. The ego problem: Selling to developers
The equipment vendors employ many smart developers who are trying to get very complex devices to do what they are supposed to do. As an embedded software vendor you are essentially telling them that you know how to do one part of what they do better.
For some reason software development hasn't evolved the way hardware development has. In hardware, it is the standard to piece together a product from third party components. In software, most of the code is made in house.
My theory on why: Software development is not just a science but (and that bring us to reason #2):

2. Software development is an art
Software development has unlimited flexibility which is both a strength and a weakness. The strength of being flexible enough to adapt to any circumstance is a weakness when it comes to standardization and compatibility - a requirement when one wants to use third party "parts". There are many efforts to foster standardization. Why doesn't such a powerful concept catch on faster? Because coding is an art and coders share many characteristics with artists. Any standardization effort limits the creative freedom of a developer and is therefore met with resistance.

3. The scalability problem
One reason why the enterprise and PC software market are so huge is because they benefit from fairly standardized platforms. You write your software and you can sell it to an installed base of millions and millions of customers.

There are variations but they are manageable: the biggest chunk of the market runs on Wintel and if you really want to be nice you can port your application to Mac OS X and Linux and you will probably be able to serve 90% of people using a computer.

Not so in embedded.

The vast selection of processors and operating systems is what drives up the cost for embedded software companies: the broader the market you want to target, the more processors and operating systems you need to support. The result is that cost grows hand in hand with revenue.

There are efforts to standardize but generally limited to a niche. Google recently introduced Android, a Java based platform that has the potential to evolve into THE standard platform for cell phones. Android could also be used in other applications too, such as printers and teller machines. It will take a while before we see anyone using Android in robot control systems or other mission critical applications.

Google's Android is really the first serious effort of standardization, leveraging the Java paradigm of software portability that allows equipment vendors to tap into a vast pool of developers. That said, Java is in its infancy in the embedded world and still needs to prove it can grow up to meet all the stringent demands.

In this post I focused more on developer and technology related issues. In my next post I want to take a closer look at business issues.

Entering the world of blogging

2008-02-11T13:24:00.000-08:00

This is it, I have entered the world of public blogging.

Mind you, it is not my first blog. I toyed around with open source software (b2evolution) hosted on my own server for a bit. The objective was two-fold.

a) to see if I could get the software running and
b) to see if I liked blogging.

It worked - I used it like a diary more than anything else. I was aware that it was publicly available (if anyone could find the URL) but I was primarily writing for myself. That said - after a couple of weeks it actually got spammed... amazing how sophisticated spam engines are.

So why a public blog?

I think a blog should be more than just a diary. It's the new news outlet. Blogs should contain useful information that others can use. It should trigger discussion. That's how blogs become valuable.

Then there's another important reason: Blogs are part of the Internet economy. Blogs have links that point to stuff. Lots of links to that stuff ensure that the stuff is visible in search results. If it is visible it will get clicked on. That's how companies sell more stuff.

It is a way to filter through the terrabytes of data out there - the cumulative wisdom of millions of bloggers helps sift through the dirt and uncover some gems.

I have personally benefited from reading blogs. Now it is time to jump on the train and see if I can contribute too.