NOTE: iPhones Firmware update 2.1, seems to have gotten rid of the problem. If now only Apple would fix the damn calendar bug

Have you recently sent an email from your beloved iPhone and have it delivered - empty? Then when you look at the sent on your phone you get the lovely: “This message has no content”.

For me - it was simple - the problem was my signature. Which constituted of a simple ‘•’. An elevated full-stop, also known as bullet-point.

There appears to be a bug in the iPhone that is as follows:

• If you send an email with a • as the LAST character, your sending bar at the bottom of Mail.app will continually say ‘Sending’, and the email will not even show up in your sent
• If you send an email with words, then • and then some more words - not a problem - it will be delivered

If I didn’t know better, I’d say that upon Unicode conversion upon sending, the email becomes corrupt through a rogue parser.

Quite an oversight on Apple’s behalf, and has only appeared since the 2.0.2 update.

Moral of the story - don’t have a bullet point as the last character in your emails, let alone your signature.

Have you noticed any strangeness in your iPhones behaviour?

•

Think of all the bills, invoices, warranty cards, tax refunds and bank statements that you file away, day after day. The problem here is that paper is bulky, and takes up a lot of physical space, especially if you’re an owner of a filing cabinet.

My hope for this task was to achieve a state where I can be light on my feet, so should I require to move places or send off a tax returns, I can now do so within a few minutes and not hours, and perhaps even to save myself a bit of sanity

Goals

• quick - no one wants to spend 5 minutes processing each document that comes in the mail
• searchable - putting in an account number of any kind, should bring up ALL the relevant document. So OCR here is a must
• manageable and easy - I don’t want to wade through hundred of folders to get the document that I want. Plus should I wish to group a set of files together for a submission, that should also be easily achieved

What will you need:

• document scanner
• OCR software
• some time to plan-out your storage strategy

Quick

Speed is an important factor when deciding on picking the right document scanner. I looked at several brands, but being an Apple user, am unfortunately limited to a certain domain of equipment with drivers that will work with OSX. I went with a Fujitsu ScanSnap S510M. The ‘M’ n the model denoting ‘Mac’, so naturally the only difference between that and the non-mac version is that its white and comes with apple drivers. Fortunately for me and you, it competes very well against the competition such as Canon DR 2050C

Searchable

The ability to find a scanned document by varying characteristics such as tags and creation dates and not merely by the hierarchy of parent folders, I find quite a necessity. Several applications that I looked at, had each their good points and down-right annoying.

DEVONthink Pro Office apart from the dated interface, lack of tagging ability {think folders}, $150 dollar price tag - is quite a lot like Lightroom for your documents. It has a lot of nice features, such as word occurrence calculators, which helps you pick out the most unique descriptions for each document. Devon keeps a database that you must also backup, of all the structure and data that you enter about each of the documents, so not the most agile of concepts.

Yep by Ironic Software is one of the more popular solutions. Thanks in-part to their clean design and ability to tag each document. It too relies on an database {XML-based plist} for the storage of its own meta-data. After spending quite a bit of time with Yep, I very much like the speed and almost ‘freshness’ of the app. Its drawbacks are also its strong points, it doesn’t have an OCR engine, which is great - since it leaves it up to you the user to figure out which one you wish to use. I tried both ABBY, and Acrobat; prefer the latter due to speed, and ability to do some compression with help of AppleScript before spitting out the resulting PDF.

I did notice a few bugs on first use

• by setting the ‘original document created’ field, the date doesn’t get saved to the file itself, and thus isn’t reflected in Finder’s ‘Info’.
• ability to create multiple smart collections of documents with the same name {screenshot below}
• when searching for tag occurrence within documents and you aggregate several tags together, an empty result-set is generated

NOTE: I sent the above-mentioned issues to the developers, and I have yet to hear back from them. Will update this section when I get a reply.

Despite the technical flaws, am a big fan of the workflow that Yep presents, yet their testing practices leave much to be desired.

Last but not least, Papers, an application that specializes in the storage and research of scientific papers. I would leave it at that, I didn’t find it useful for the task at hand, that’s not to say it isn’t good at what its designed for.

Manageable

Although tightly related with the previous section, manageability of your documents is something that will underpin how well you’re able to search for items once you get all those documents into some sort of a repository.

Unfortunately all the applications I’ve looked at, keep their own database of the meta-data. This is unfortunate for two reasons:

1. Switching cost - between applications is increased, as you’re more likely to stay with an app once you’ve imported a copious number of documents and succumbed to the apps way of thinking.
2. Backup - of not only the documents but also the database has to be thought of. This is where Devon makes it easy, by allowing you to place its database file into any location you desire, such as with your documents. What’s worse is that Yep doesn’t give you that freedom, and keeps its library within your Application Support folder. I received an email from ‘Ironic Support’ informs me that the only way to keep documents and data-store in one location, is if you keep all documents in your Application Support/Yep folder.

The Scanner

An important consideration for this whole task is the scanner itself. So what are you looking for in a document scanner? It doesn’t really matter what I say here, as price inevitably dictates peoples choice. Personally I saw the time it takes me to collate, organise, and file my .. err files extremely wasteful. It ended up being about how much my time is worth to me in the long run.

After doing some research the decision was between two models, the Fujitsu S310M and the S510M. The baby {given its diminutive size} of the two, S310M has one of the most attractive features in a document scanner, it can be powered merely off the USB ports on your laptop. So if you travel a lot, and its important to scan research papers {if not available in PDF-form already} or contacts, then you can’t go past the S310M. I chose to go with the S510M hoping that the extra $100 will result in a better quality scanning ‘head’ if I may call it that; so your choice should also depend on how many documents you’re going to be scanning. If you’re merely going to be doing a couple of pages a day, then grab the S310M alternatively if you are swamped with papers on a daily basis, or are going to be re-scanning your complete paper filing cabinet it might be an idea to go for the big brother S510M.

The shots below are to allow you to see comparatively how big or small this document scanner really is. I didn’t want to use the metric system, nor the imperial system - to prevent confusion. So I picked a unit everyone is familiar with. So the scanner is about 1.3 iPhones tall, with a footprint of about 1.1 iPhones, with the length of just under 2.3 iPhones.

Both scanners do duplex scanning, thus double sided documents are not a problem. What impressed me most is how well the paper feeder actually works. I’ve had the experience of working with a Canon DR 2050C and if you mis-align a page, or put something of obscure size, such as a shopping docket through - you were bound to get it jammed, not so with the S510M. I threw about a hundred sheets of A4, A5, dockets, scrunched receipts from the corner PC store, throughout the duration of the day and without a single twitch it generated a double-sided PDF for me; with the help of Acrobat 8.0 Professional and some AppleScript it OCR’ed and knocked out backgrounds for me too. In instances where I was a little eager and mis-aligned the paper, the software automatically corrected any skewing and rotated the page upright - nice!

Conclusion

When I first embarked on setting myself free from dependence on the myriad of documents, I was very hesitant to lay down the $’s for a scanner. Albeit since it arrived, it has been a blessing. As soon as I scan a document, back it up {on that in the next post}, I shred it. Hence to anyone that values their time, and ability to streamline some of the mundane tasks of paper management I couldn’t recommend the idea of a paperless office more.

Have you made your office paperless?

Having recently purchased the unit, especially for the purpose tagging all the photos for an upcoming trip, I flew away extremely happy, witnessing the outstanding performance of this little device. You can see a screenshot of Google Earth displaying the track below; its the takeoff from Sydney towards Tokyo.

What sets this unit apart from other loggers and GPS units on the market, is its ability to not only acquire a signal within the boundaries of an airplanes fuselage, but keep that sync as the airplane is banking. This latter achievement is the result of the -158dBm sensitivity rating of the unit.

Geotagging

In order to couple the data off the unit with your images, first of all you must make sure that the date+time is the same on both the unit, and your camera. Then based on the time of each point recorded and the time a photo was taken, the closest data point becomes the location of where the photo was taken.

There’s nothing you can do about the GPS, since its sync’ed with the satellites and stored in UTC. What is important though is to adjust the cameras time to the Time Zone of your location, this will save you much heartache later on when you have to adjust the times.

For my needs, I thought I’d try out HoudahGeo, and GPSPhotoLinker. I very quickly dismissed HoudahGeo, because I was unable to try it ‘properly’. By which I mean use it unrestricted for a period of my journey. I don’t believe its possible to evaluate certain software unless you can actually try your complete workflow with the software, and see if it fits, HoudahGeo unfortunately provides you with a handicapped version.

In comes GPSPhotoLinker, which is actually a free app, and gets the job done surprisingly well. What you’ll notice is that once you combine your photos with the exported data from the GPS, it will actually look up the City and Suburb data as well and embed that into the EXIF along with the GPS location. This is extremely useful for my Lightroom catalog where I can pick things based on the City as looked up by GPSPhotoLinker - brilliant! This is extremely helpful when you’re island-hopping, as is inevitable in Greece, or are merely in a new place.

As it stands, I think its a wonderful little unit that has served its purpose extremely well. I would highly recommend it, and will not be leaving the house without. It now lives at the top of my camera bag.

• log data to internal memory
• make sure there’s LOTS of memory
• and lots of battery

After looking at the Sony GPS-CS1, Wintec WBT-100 and GlobalSat DG-100, I concluded that the BT-1000P is by far the best unit for my needs.

• First of all, the GPS chip supports up to 51-channels, with sensitivity down to -158dBm, which allows me to get a satellite lock even at the university concrete monstrosity of a car-park.
• It has integrated bluetooth, so I can use it as a replacement for my ageing Holux-236.
• Most importantly, it can log around 200,000 points, with my current configuration of strings, only 122,398 (@5 second intervals, that’s 7 days of 24 hour logging before I have to clear it!)

What pushed me to purchasing after looking at the specifications though is the battery.

• The same as you find in most Nokia phones [BL-5C], which can be picked up cheaply off eBay
• Delivers a claimed 32 hrs of logging, although I was able to get 26 hours without any problems, before the little red battery starts to flash. Mind you - I didn’t fully charge it, and just started running it soon after receiving it in the post.

Now I needed the unit to work on a mac, which the BT-1000P doesn’t do out of the box. The good news is that there’s a project called bt747 that lets you interact with the BT-1000P.

Upon connecting the unit via USB the rest of the steps are quite simple:

1. Download and install the USB->Serial driver
2. Download bt747
3. Switch the BT-1000P to ‘LOG’ mode
4. Copy RXTXcomm.jar and librxtxSerial.jnilib into /Library/Java/Extensions
5. Run run_rxtx.sh

You should get an extremely ugly GUI

Just hit ‘Connect Port Nbr’ and you’re ready to go. You should get something similar to: After this, all you have to do, is run around the block a few times with your unit in LOG mode. Come back home, run the app again then: Under the ‘File’ tab, set the Output directory to a location that has world write access, then hit ‘Apply & Set the above values’ Alternatively create a directory in your /var and sudo chmod 777 <directory>.

1. Now -> ‘Log’ tab
2. Hit ‘Get Log’ button {this might take a while, depending on how long your jog was}
3. After its complete, feel free to select the output format, either CSV, GPX or KML.

If you chose to export a KML, and you have Google Earth installed, you should get a nice looking track. Unfortunately it is evident that the BT-1000P doesn’t filter its points prior to writing them out, resulting in data that isn’t representative being recorded, hence you can observe the spike.

Now to get ready for my last exam tomorrow morning, and then to start packing for the trip. Once I’m back, will report on how the unit goes in the field, and I’ll include some geocoded images.

UPDATE

Upon the comment below by Benn, I remembered a feature that this unit has, that every time you take a photo on your camera - you can also hit the big red button on the tom of the unit. This marks the POI in the data stream. Merely allows you to come back for any manual intervention and find interesting parts easier.

Everyone agrees that WebKit is the faster browser engine. Plus its my favourite browser/engine in OSX

Problem is that it doesn’t update itself automatically - and every night there’s a beautiful new build that comes out - which I simply must have.

So in order to save you time - here’s a little script that I put together this afternoon that will download the latest build at the push of a button.

Initially I wrote it with wget, but realised not everyone will have it installed - hence here’s a version with curl


#!/bin/bash

# getWebKit.sh - Obtain the latest WebKit build
#
# Created by Roman Tarnavski on 2008-04-13.
# For an updated version head to www.romant.net/
#
#
# Distributed under a Creative Commons Attribution 3.0 License
# For more details on the license: http://creativecommons.org/licenses/by/3.0/

var_dir="/var/tmp/"

echo -n "Downloading WebPage ..."
location="http://nightly.webkit.org"
curl http://nightly.webkit.org/index.html -o ${var_dir}index.html > .get_webkit.log 2>&1
nightly=`cat ${var_dir}index.html | grep ".dmg" | cut -d '"' -f 2 | uniq`
echo "Done"

filename=`echo $nightly | cut -d '/' -f 5`

echo -n "Getting WebKit ..."
curl $nightly -o $var_dir$filename >> .get_webkit.log 2>&1
echo "Done"

if [ $? ]; then
 echo -n “Installing WebKit …”
 rm -rf /Applications/WebKit.app
 hdiutil attach -quiet “$var_dir$filename”
 cp -R /Volumes/WebKit/WebKit.app /Applications/
 hdiutil detach -quiet /Volumes/WebKit
 echo “Done”
 exit 0
fi

#cleanup
if (( $? )); then rm ${var_dir}index.html && rm $var_dir$filename ; fi


 Grab it and let me know how you go

MacBook Pro

Dell 2408WFP

You can evidently see that the Dell shows a purple instead of a blue. Very disappointed was I to think that a screen on which I just dropped some hard earned, can’t even display “BLUE”!

and then…

I just thought I would open it up the images in Firefox - just to check…. and voila

Unfortunately after a few more tests like the above - it seems across OSX is the same thing - ie. Finder when displaying the above photos - merely shows it in shades of purple as well.Now the only reason I even came across the fact that it could be an application specific error - is when I look at the photo in Lightroom 1.4, depending on which screen I have the image on - it changes the colours to suit the screen. So my question is - if Lightroom can do it - why can’t OSX?

Update

After re-calibrating both monitors with the Gretag Macbeth i1 Display 2, I have composed some windows together to show the difference between different applications and their adherence to the needed colour management.

The applications I used for this test, and in this specific order are:

1. Safari3.1 (5525.13)
2. Lightroom 1.4
3. Firefox 3.0b4
4. Preview 4.1 (469.2.1)
5. Photoshop CS3

Dell 2408WFP

The peculiar thing about the Dell, is that Lightroom actually showed a very strong purple instead of blue, and then after about a minute - without me physically interacting with the computer - it simply changed to the correct colour. Which is depicted in the shot above.

MacBook Pro

Analysis

It seems that in the initial shot with the 2408WFP, Safari obviously does the worst, showing a blatantly purple where there should be blue. Other applications vary slightly in the shades of blue, but otherwise display a fairly closely matched colour.

When it comes to the MBP to display the same set of images, Safari, Lightroom, and Photoshop - all display a shade of purple instead of blue.

The annoying part is that whilst writing this - if leaving the  specific application on either screen - it adopts the correct colour profile, and displays the correct colour.

As such - I have no choice but to personally conclude that it seems that OSX isn’t informing the application that it has been moved to another screen in a timely manner.

There have been reports that the Dell uses a new backlighting system that the i1 is finding difficult to actually meter, yet to me it seems there’s more software problems at play here under the bonnet of OS more than anything else.

Can’t believe I stumbled onto this shortcut, as its exactly what I wanted to do. You see, I didn’t want to disable the multimedia keys as defined by default in OSX - as it would lead me to having to hold Fn to do one-key thing such as volume. So I went about pressing everything I could - and by mistake I came upon the following shortcut.


Fn + Tab

Which hides both - your left and right panes.

Quite useful for both development view [at least until Adobe decides to provide secondary monitor support], and grid view.

Before:

.. and After:

Late last week I purchased the MBP from Apple’s Online store. Thinking about the resale value, I decided to spec up the bottom model. I added:

• Glossy Screen
• 250GB HDD

Picking against matte was a hard decision, and only decided to go with it due to the everyone I’ve spoken to, for some “darker blacks” reason preferring glossy screens - so I thought I’d go with it for resale value sake.

Unfortunately upon the much anticipated and late arrival of the MBP, thank you TNT - I discovered that not only where the hard drive was replaced was the seam in widened between the aluminum and the gray bezel. To add to the exceptionally poor quality control, when the laptop is closed, the lid remains ‘UNEVENLY‘ elevated from the base, and thus producing a very ugly gap.

As much as I enjoy using OS X, and Apple products in general, I am far from impressed about the build quality of their laptops. Apple should spend more time on testing the usability of their products, as sometimes it seems form takes over function.

Brendan decided to take some photos. Enjoy. Sorry for the bad quality - its his crappyN95 to blame

vs.

Unfortunately until recently - you had to maintain two accounts in order to utilize the above-mentioned applications. One for your ‘Google Account’ [you@gmail.com] – and your domain account [you@yourdomain.com]. Zoli has written a post which in great detail explains the process. For those brave enough - a summary of the steps follows.

So how do you pseudo-integrate the two accounts together? Quite simple as it turns out [check the warning below]

1. Log into your Google account and go to Manage Account
2. Under ‘My Services’ - Click Edit Services
3. Delete Gmail
   • IMPORTANT: Specify your domain account email as your new login account. [you@yourdomain.com]

   You’ll have a few verification stages – follow the instructions.

That’s it - you’re done

The password for your Google Account remains the same, and chances that it is different to your domain password [unless you use a 1 password fits all policy].

You can now utilize your iGoogle page, Google Reader and any other pages you’re accustomed to that poor integration required two usernames to utilize.

I mentioned the fact that this is pseudo-integration due to the fact that as far as Google is concerned – they [your Google account and your Domain account] are two complete separate accounts; this is shown by the fact that it still requires you to log into your domain when checking email, and the mere fact that there are two password for seemingly the same account.

The expectation is that one day Google will truly provide a seamless, single-sign-on solution. Hopefully soon. At least now you can participate with Groups and Reader as a domain account.

Warning

· All your Gmail data will be deleted within 2 days. So make sure you migrate your data first if you wish to keep it.