Thursday, August 25, 2005

IM Gets a Boost

This week saw some interesting announcements from the Instant Messaging space. My fascination with Zipit Messenger has awakened my interest in IM, especially protocols and interoperability issues.

Google launched their own IM client called Google Talk, which is very thin on features and just ho-hum for someone like me that uses this stuff every day. Skype on the other hand introduced an SDK for IM developers that will help them address interoperability problems and leverage the Skype platform. Intel announced collaboration with Skype to ensure that its upcoming Centrino mobile line takes full advantage of Skype's technology (especially VOIP).

"It’s important not to build a walled garden."
- Janus Friis, Skype

That's an interesting statement from a company in the Internet telephony business - shows they're paying attention to applications of IM and VOIP, as you'd expect. Most people I know think Skype is a company to watch. And most people I know are Googling less as they rely more heavily on RSS. If this was Wired (and it's not), then Skype would be Wired and Google would be Tired.

Friday, August 19, 2005

A Brief Intro to Podcasting

Worked with my boss Mahboob on a podcast, providing a brief intro on the evolution of podcasting. There is a small error in my remarks. I said "RSS stands for 'Really Simple Subscription'". Of course RSS stands for 'Really Simple Syndication' - I was trying to explain the subscription process and got carried away.

Mahboob's end of the conversation suffers from a slight technical flaw, an echo effect apparently due to my not wearing headphones while recording. Apologies.

[ammended 20050820]

Wednesday, August 17, 2005

Lack of Data for Proper Assessment

Can schools predict student success based only on 'exit exams'?

Consider the case of California, where passing scores in exit exams for basic math and English proficiency are now mandatory before a high school diploma can be issued. Exit exam requirements are intended to give employers, students, parents and the public assurance that a diploma signifies that students have the knowledge and skills to succeed after high school. The tests now in use measure proficiency at 8th grade level, which most students achieve even before entering high school.

Crunch time is coming for the first batch subject to this requirement. So far, about 90 percent of California's incoming seniors (the last year before graduation) have already passed at least one test. While the state knows that the remaining 10 percent still need to pass at least one of the critical exams, that don't have precise data. Says one article: "Because the state doesn't have individual student data, it can't precisely gauge how many students have passed both tests. Conversely, it can't tell how many of the seniors who haven't passed the English test are the same students who have failed the math test."

Thus the 10% figure understates the number of students who still must pass at least one portion of the exam to get a diploma - and they only have one year in which to achieve it. This lack of data is appalling, and highlights the need for continuous assessment in schools. Continuous assessment will provide better data to school administrators and will help teachers intervene earlier to help struggling students.

Statewide there is a wide disparity in pass rates by race and ethnicity, a phenomenon known in education circles as the 'achievement gap'. The biggest challenge for districts which have large immigrant populations, such as those in Southern California, will be helping English learners pass the test. Only 58 percent of incoming seniors who don't speak English fluently have passed the English test and only 70 percent have passed the math test.

Exit exams can identify whether students are minimally proficient, but by the time results are out - it's already too late. Also, the level of proficiency tested is that which students should have attained before entering high school, so the four years spent in high school classes are not optimally utilized.

California schools need to implement continuous assessment, not snapshots.

Monday, August 15, 2005

Is Radio Dead?

A lot of folks in the podcasting community suggest the death of radio is imminent. Among them is Steve Gillmor in his recent podcast "Broadband on the Run". Must be an allusion to McCartney's Band on the Run. WKRP tombstone (courtesy

Certainly there is an emergent model for podsafe music. Information wants to be free. But I believe that (Steve's co-host) Doc Searls is right, namely that the traditional model for music distribution is not dead and there will be some level of co-existence for a long time. And I think Gillmor is wrong to play Hey Jude in the background of his podcast (unless perhaps he's engaged in review and commentary). And in the same vein, Adam Curry is wrong to play mash-ups as if they were podsafe. Neither is 'fair use', and neither escapes current copyright restrictions, since I and others are being invited to download the programmes.

Moreover, I think that Gillmor's programme suffers from the 'fog of war' he seems to be declaring on copyright owners and publishers. Namely, the programme falls short of clarifying why radio is in the throes of transformation, what drives this transformation, and how radio is being transformed.

Internet radio is nothing more or less than the delivery of radio programmes via TCP/IP. That delivery can be either live or on-demand. I listen to Virgin Music via Shoutcast live on my Treo phone and that's Internet radio. I listen to downloaded BBC programmes via iPodder or iTunes and that's Internet radio. I listen to BBC live via my Windows Media player (incidentally bypassing their custom HTML player) and that's Internet radio.

In copyright terms, what's different about how I listen is not the Internet versus free-to-air delivery, or the live versus on-demand aspect. It's the packetized streaming versus download aspect. Once 'radio' content is downloaded, any notion of being able to collect royalties on the basis of number of 'listeners' is lost. With this loss of control, copyright owners are naturally reticent to release music for download. And in many cases, they have confused packetized streaming with genuine downloads.

That describes why music copyright holders resist download services, and why most streaming services have only been permitted to offer talk radio and not music programmes. But with the advent of podcasting, a new force is unleashed, and that is the ability of the podcatchers to aggregate listeners in sufficient quantity that some musicians will prefer to go direct in order to build a following. Thus they release music outside the traditional distribution system, ie- podsafe music.

Podcasting creates a commons for the exchange of music between producers and listeners. But it is not a marketplace because there is no economic model - yet.

In the early days of open source software, it was said that open source had no economic model, but that is no longer true. Today, open source integrators charge customers for maintenance and support. In my view, podsafe is going to evolve in a similar manner to open source and other commons-based activities. Podsafe distribution will become the new 'airplay' and while not generating cash directly, will stimulate revenue for artists via subscription services, licensed compilation downloads and even CD sales. Adam Curry can fairly be described as a radio DJ for podsafe music.

Radio is not dead, it is being reborn in a new distribution environment. And this distribution environment is vastly richer than the staid traditional one, with tremendous potential for personal, educational and even corporate broadcasting.

Saturday, August 06, 2005

Zipit Messenger Review

My Zipit Messenger arrived via a Target Store in Eugene Oregon (where my brother lives). It was well packed in a cardboard box with a bit of bubble pack, and there was no shipping damage. The unit itself was contained in a blister pack which was heat-sealed. Zipit Messenger in blister pack

The ZM power button was easy enough to find and operate. The battery compartment is sealed with a screw, but the rechargeable battery is already installed and pre-charged. The unit has a status light to show that power is being switched on. After a few seconds the ZM logo appears on the screen and the word "Zipit" is spoken from a tiny speaker at the back of the unit.

I tested the unit in my office, which has a Wifi network. To say that ZM works right out of the box is almost an understatement. The unit immediately found two networks, mine and another nearby, and prompted me to log onto my own network. I did and was prompted to configure my IM accounts. I have an MSN and Yahoo account and configured both. The unit reported that the MSN account worked but failed for some reason to connect to the Yahoo account. Since most of my contacts are on MSN, and I don't use AOL, I left it at that.

The last step in this 'setup' was to customize my emoticons - I just left the default settings. But I did note that the emoticons are linked to six special and easily identified 'function' keys at the top of the keypad. This could be useful to create macro keys for multiple choice questions in a quiz application.

Then I was presented with my MSN buddy list and selected a colleague for my first chat using Zipit. It worked flawlessly and I was typing away just minutes after removing the ZM from its blister pack. The screen is a graphical grayscale display which is very legible. It is not backlit, but it's brightness is adjustable, and it would be fine in an office or classroom.

My complaints are that the keys are a bit klunky and to edit mistakes you need to backspace over earlier typing - there is no cursor. This of course is just a function of the software, not a hardware limitation. The keyboard might be a bit easier for big thumbs if the left and right portions were separated by a middle gap. But the display is very legible, the prompts are clearly stated, and the QWERTY keyboard layout is familiar enough for instant use. I can type at about the same speed as on my Treo 650, which has a marvelous keypad.

The keypad also includes a big ZM key in the center, which is used to bring up the buddy list while communicating. There are some other special keys, including a menu key (for configuration changes) and a pair of 'next' and 'previous' keys, which can be used to toggle two or more chat sessions. There is a directional cursor key (one single key which functions like a joystick), which is used to scroll messages in a chat session. Numeric keys, punctuation and closing a chat session required prefixing my key press with an 'Alt' keystroke - which may be conceptually challenging for small children, but is natural enough for anyone who's used a mobile phone or computer.

The clamshell cover nicely latches closed, and the unit goes into a low-power sleep mode. In this state the device is rugged enough to toss around and small enough to fit in your pocket, backpack or even a small purse. The ZM is lighter than my phone, and I eventually carried it around the whole first day without noticing the weight. While the battery appears to last for hours, I didn't test that. The unit is powered down by depressing the power button and holding it for a second or two.

The physical form factor is excellent in several respects. The unit sits firmly on a tabletop while you peck at the keys. You can easily type single-handed while doing other tasks, such as talking on the phone, or perhaps while taking written notes in a meeting or in a classroom session. The power button is nicely recessed so that the unit cannot accidentally be turned on or off. The clamshell lid latches securely closed.

I was particularly pleased with the power setup. The ZM comes with a miniature charger that supports both 120 and 240 volts - it's fully internationalized. There is a headphone jack on the back of the ZM, but there were no headphones supplied with my unit. Other than the chirpy "Zipit" welcome message, I did not really encounter any audio features.

The internal Wifi antenna seems to be very powerful. I was able to sync to more than a dozen networks while driving around in a taxi - sometimes getting 3 signals at one time. Of course, I could not log into most networks.

In summary, Zipit Messenger is rugged, contains a powerful Wifi antenna, has a well-designed and internationalized power supply, a very legible graphical screen display, and a workable keyboard including user-definable macro keys.