Friday, September 21, 2007

iPhone for small data problems

Ok, so I broke down and bought an iPhone. The Mac evangelist in our office let me know within a couple of hours when the price dropped and that evening I went to the AT&T store and liberated one of the 8 GB models. I've had it since then and have used it around town and during travel.

There are plenty of reviews and opinions out on the phone, so I'll go over mine fairly quickly and then get to discussing it from a small data problem perspective. My previous phone was a Cingular 8125, so I'll be comparing it to that.

Overall Impression
Not perfect, but quite nice. The strengths are that Apple decided to make the device do a small number of things and it does them very well. Easy to use and intuitive.

Pros:
  • It serves as a great iPod.
  • There are a lot of subtle things in the UI that work well. For example, when I'm talking on the phone, it detects when I move the phone so that it turns on or turns off the display appropriately. When I have bluetooth available, I can switch it off for individual phone calls. There was obviosuly a lot more user centered work that went into the UI design than went into my 8125.
  • Finally someone got the cell phone browser paradigm right. Rather than trying to interpret a web page for a small screen, it just displays the whole page like it would be on a desktop, but makes it too small to see. Then you zoom in and out as appropriate.
  • Using WiFi was surprisingly easy. The last two phones I've had WiFi cards in them, but I could never get them to work properly with my home network or with hotels' wireless when I travel. But the iPhone would discover publicly advertised networks, ask me if I want to connect, and ask for the password if one was required.
  • The Edge network. I know that it is annoying slow, but it is really good enough for most browsing and if, as they say, the faster network cards are going to consume more power, I'd rather have something that is a little slower, but usable.

Cons:

  • The battery life is good enough for reasonably heavy use (web surfing, email, watching videos) for one day, but for no longer than that. I could go for about two days of heavy use with my 8125. On the other hand, I couldn't watch videos on the phone and the resolution of the screen wasn't near as nice.
  • Yes, the keyboard is usable after you've practiced for a while, but no where near as good as my blackberry or my 8125.
  • No next/previous song switch on the side. If I'm listening to the iPod and want to skip or repeat the current song, I have to take it out of its case, unlock it, and hit next. I ought to be able to do that more easily.

SmallDataProblem Perspective

The iPhone helps with the following small data problem issues:

  • Email -
    Pros:
    The iPhone does a good job of synchronizing with Yahoo and as an imap client through an exchange server (I didn't really try the google integration). I was particularly impressed with how it kept the folders synchronized.
    Cons: The downside is that you can't choose different "from" addresses in the Yahoo email client like you can on the web browser version. That's a significant problem if you are having a bunch of forwarded email addresses going to a single yahoo account.
  • Address Book -
    Pros - Although it does not synchronize directly with an exchange server (like the 8125 does), you can synchronize it with outlook, so you can effectively accomplish something similar.
    Cons - Synchronizing directly with exchange was really pretty nice on the 8125 and I miss it.
  • Voicemail - There has been a lot written about this and it's all true. Dealing with voicemail like you would deal with email messages is the way it should be.
  • Managing Music and Podcasts -
    Pros:
    My 8125 would, in principle, synchronize with Windows Media player. But the user interface on the Windows side is so convoluted, that it was not clear what was going on or how to get things synchronized properly. iTunes makes all this very easy. It's probably a little less flexible than what I could do with the 8125, but because it is so much easier, who cares?
    Cons: Managing podcasts is a bit odd. iTunes has it's own settings for deciding what to keep and when to delete. The iPhone also has its own slightly different paradigm. Also, when you manually download podcast episodes, it seems to override the general policies you set. As a result, I'm never exactly sure what is going to happen to the podcasts when I sync. Also, you can only establish one set of download/delete policies for all podcasts. I'd rather set policies on a per-podcast basis. So it would keep the onion videos indefinitely (or only until I delete them manually), but only keep the most recent evening news video download whether I had watched it or not.

One other note - there is a tradeoff between how powerful and flexible a device is and how easy it is to use. Good UI designers know how to pack a lot of both into a single system. Even though the iPhone is really a small computer (you can even set up an SSH server on it and log into it), Apple decided to hide some of that flexibility in exchange for ease of use. The 8125 did not do that. It was a little harder to use, but 3rd party applications and an open API made it more of a computer in practice.

Saturday, April 7, 2007

Toba 0.8.2 released

I just released version 0.8.2 of Toba on the web site. There were a number of changes, but the biggest change was with JASE (Just A Search Engine). JASE is a plugin for Toba that lets you take a directory of text files or XML files and create a search engine, available from within Toba, to search through all the files. Since the search engine uses Lucene, you get the full set of sophisticated Lucene search syntax options.


There were a couple of bugs that got fixed and features added. First of all, the known engines setting that you edit in the options window was clobbering the list of known plugins. That's fixed. Also, the feature for creating an index now goes recursively through the directory of source files so it's easier to go through large data sets. That, along with a couple of other little features listed in the change log made the release useful


Both the generic and the Mac OS X release packages are available.


Enjoy!

Saturday, February 10, 2007

Toba For the Mac!

Being written in Java, Toba is already platform independent. But there was a glitch on the mac where it was not recognizing right clicks. That's fixed. I've also created a Mac OS X "installer" so that you just have to unzip the zip download and double click on the icon to run the application. Just easier to use. The download is on the SourceForge download site.

Wednesday, February 7, 2007

Toba is Released!

Toba has been released. Version 0_8_0 was released on 2/7 (or 2/8 depending on your time zone). Check out Toba's main homepage or go straight to the download site.

Sunday, February 4, 2007

Toba's History

A screenshot of the graphics interface to Toba

I'd like to talk about how Toba got started. Years ago, when I was in graduate school, I wrote a program for taking class notes and keeping bibliography references.

I didn't really like using a word processor for taking notes. Either you ended up with one big file for the whole class or you ended up with multiple, disjoint which couldn't be searched (no desktop search tools back then). I couldn't afford the nice bibliography tools that were available at the time and figured that they were a bit of overkill anyway.

So the problem was storing, listing, and searching through bits of information that needed to be stored. I wrote the application in such a way that there was a single application framework and the editors for the bits of information were supplied through dynamically loaded DLLs. A couple of people who looked at it liked it and suggested other things that I could add to it - new bits of information.

Speed ahead a couple of years . I found myself at work with the same situation. I needed to keep track of a lot of bits of information - contacts, notes and to-do items - along with a suite of really specialized pieces of information - project opportunities, issues to be tracked.

I knew the types of information I'm going to need to track are going to evolve and change over time so the idea of a single, open application with an extensible framework would really do the trick. And now with open source products like Lucene and HSQL available, you can really pack a lot of power in an application framework without a huge amount of work.

I decided to add a couple of extra things - a command line interface for those of us who like that flexibility, saved searches so you can have lists of things that automatically update according to specified criteria and an internal scripting language so you can use the application itself as a development environment.

I got started writing it. The command line version of the application was usable within a couple of months. Eventually the graphical interface became usable. What started as a personal project to get a glorified, extensible, personal note application turned into a full fledged application.

So now I'm in the process of releasing it as open source. You can find the web site at http://www.smalldataproblem.org. I'm hoping that you'll find it useful as well and contribute new modules so the application can grow in new ways.

Tuesday, November 28, 2006

Welcome to the Small Data Problem Blog

Welcome to Small Data Problem - blog edition!

The "small data problem" is the need that most individuals have to manage a large number of small amounts of digital information. This includes digit information of all types, including contacts, misc notes about issues, digital music, logins and passwords, and any other small pieces of information that most of us have lying around.

There are a lot of different software packages out there, some of them free, some of them commercial, to help one manage their small data problem. In this blog, I'll be reviewing a lot of them over time as well as talking about the small data problem in general.

Also, I am releasing my own extensible solution to parts of the small data problem in a piece of open source software called Toba. You can visit http://www.smalldataproblem.org to learn more about this piece of software and download it for free. This blog will also comment on the development and use of that piece of software.

If you have any ideas for products you'd like to see reviewed here, please drop me a note at travis.bauer@smalldataproblem.org.