Paul Golding's Radio
These are just my thoughts on wireless etc.
They do not represent the view of my employer as I am not employed
I'm the boss :)



Subscribe to "Paul Golding's Radio" in Radio UserLand.

Click to see the XML version of this web page.

Click here to send an email to the editor of this weblog.

Join Paul's mail list
 

 

Monday, October 14, 2002
 

It works - Vodafone finally finds me!

After months of trying "find + seek" on the otherwise poorly designed Vizzavi WAP portal, finally - today - it found me! Well, almost.

Previous attempts had infuriated me by insisting that it could not locate me and I had to manually enter an address, which did not seem to work unless I entered a postal code! Come on guys - how am I going to know that off the top of my head? Talk about poor usability. They might like to read Jakob Nielsen's report on WAP usability and think again.

Anyway, today it worked. Well, it found me within 2 miles, which is not bad I guess. But that just goes to show that cell-based ID (which I'm assuming they use) is never going to be as good as GPS except in micro-cell areas (which I'm hardly in on the edge of town here). I'm still in shock from Orange's LBS project manager's claim that Three (Hutchison3G) are not really ahead of the game by staking their interest in assisted GPS for their users. Hello? Well, which would you prefer? Knowing where you are in the nearest 50 metres or the nearest 2 miles? But then that's because Orange's accuracy graph was clearly focussed on micro-cell areas. Fine if you live in the city centre.

The strange thing about the find + seek is that it gave me a precise address where it thought I was located - "6 blabla street" - which seems really odd when you first get the information on the WAP page. It almost feels like a fault. By being so precise in the address, it gives the impression that it can locate you to the nearest house, but has just got it wrong. Whereas what it is probably trying to do is give you a fixed address as a real reference point, rather than simply saying "you're in blabla street" which could be rather long.

Whilst on the topic, I just noticed that CPS announced that Texas Instruments is going to include their enhanced observed time difference (E-OTD) technology into the TI baseband chipset for GSM.


6:50:32 PM    

Get paid for doing something wireless (Genius!)

Check out http://www.pinpoint.com/developers/benefits.html - these guys (inc. O2) are finally doing what I wanted in 1997/8 when I developed the Zingo wireless portal concept for Lucent - a portal/platform/API to enable any developer to deliver his app and get paid for it. Hey! That's a bright idea isn't it?


6:28:58 PM    

I must be a dummy then....Enterprise JavaBeans for Dummies recommended

I constantly review books that I recommend to my course students, especially on the whole J2EE/EJB area where it can be quite hard to grasp various concepts (like skeletons/stubs). I actually found that for people at the support level who need to know what's happening on the app servers they run, then Enterprise JavaBeans for Dummies is actually quite a good read.


3:33:42 PM    

Give me a break <BR>

My brother just posted me the news from ZDNet about Opera's claim to have "solved" the problem of browsing normal web pages on small devices. Really? Give me a break! I thought we had buried this myth a while ago, as discussed on my course, so it looks like I'll have to give it a mention.

Anyone with any experience of web authoring knows that unguided transcoding from ordinary web pages into small-device browsers does not work (will never work) except in a few sterile cases. Apparently, what Opera does is reformat pages into one column so that the user only has to vertically scroll. Granted, this may increase usability slightly in the mobile context, but it's hardly going to solve the problem that in my opinion is simply unsolvable and so such claims should come with a fools gold warning label.

What Opera does do - and I think this is what they are trying to promote here with their new browser version - is provide embedded browsers with full HTML support. This is interesting as it addresses the issue of not having to code in another ML for small devices (e.g. WML or cHTML). The new desktop version of Opera is able to display pages in small device format too. Why would you want to do that? Clearly so that you can see what it looks like if you're authoring pages. So in effect, the real intent of Opera is to modify the authoring process going forward, so the original claim is a loaded one really.

There are too many issues here to be dealt with in a small posting, so I may write all this stuff up in a whitepaper and post it (I needed a topic to write about for my first weblog whitepaper - you may have noticed I'm a newbie blogger - is that a noggler or something? don't know much of the jargon I'm afraid). WAP no longer uses WML, so someone should tell that to a few of these analysts who seem woefully out of date in their "analysis". It uses XHTML-WP which is a profile of XHTML-Basic. What that gains is another story (see forthcoming whitepaper - i.e. subscribe to my log).

One thing to look at for those interested in authoring for all possible end-user devices is Volantis. This is a very useful piece of software that you use in the delivery of pages and works especially well in conjunction with BEA's Weblogic Server, both as used by Three (Hutchison3G). You can think of Volantis as a kind of Cyscape's BrowserHawk in reverse. BrowserHawk is purely a browser-capability detection solution, but the main value add is that Cyscape are tracking browser capabilities for you. Similarly Volantis are tracking device browser capabilities, even down to the variations that occur from one firmware release to another (recall the dreadful experience of coding WAP for the 7110 only to find that "all" the phones were different).

Volantis actually takes care of the page formatting for you based on the assumption that you have coded the pages either using their JSP tags or their extensions to HTML (and now they are supporting inbound XML). The re-formatting can be device or device-family tailored and is all under the control of a graphical layout tool included with Volantis. Even URL re-writing (as URL length is also a variable capability on devices) and page-length "chunking" are taken care of by Volantis.

Ultimately what we need is a networked version of Volantis and their design tools, or something similar (hey! there's a cool product opportunity here!!) that authors can use, the idea being that the mobile operators should host the solution and provide it free-of-charge as part of their mobile network API offerings. WAKEUP OPERATORS!!!! It should be your job guys to make sure we can edit pages for any of your customers!!! Don't you want wireless data to succeed?

Now I'm asking myself how do I run my weblog to be readable on certain mobile devices. I'm going to have to go look to see if someone's already done it. Hmmm - I can smell cooking, must go now...


1:34:09 PM    

Petrol Stations...Again...and remote printing

Already posted earlier about the cell masts hidden in petrol forecourt signs and I just had to follow-up with other thoughts on garages, as I seem to have them quite often. I mean, everyone uses them so they are places of potential interest to people on the move, so ideas are always popping in my head whenever I visit one (other than the idea to get rid of my v.thirsty Ford Exporer).

First off, I just had to point out (and I wanted to write to garages about this) that those guards they put on pump handles to indicate out-of-action are pathetic and annoying. I mean, you have to pull up near the pump usually before you notice. In a crowded station (e.g. M4 during the rush) it can be a right pain to have to back up and find another pump. Chaos usually ensues. So why don't they put a simple traffic light system on each pump so you can see green lights for go and red lights for stop (out of action)? Seems simple. Any garage forecourt designers reading this (unlikely), please take note.

One of my ideas a while back when working on a product to access Microsoft Exchange via SMS, was to enable forwarding of attachments to some kind of print vending machine in the garage. Simply forward the attachment to the garage network, get an SMS serial number back, pop to the nearest outlet (that would have been notified in the SMS using location sensing) and after inserting a credit-card and the serial number, get the print-out.

That just set me thinking about remote printing and my earlier post about remote file storage via mobile phones with a USB cable. Another variation is to enable the handset to dump a file straight to a printer, ideally with Bluetooth, but also via a printer-cable adaptor straight from the phone port again. That may necessitate having a printer driver on the phone, but not necessarily. The printer-driver could be anywhere on the network and spew out its data stream back to the phone over a 3G pipe and transparently pushed by the phone onto the adaptor. That way, you could plug into any printer you like, wherever you happened to be.

I know the idea of printing via fax needs also to be added to the thought process here. Obviously you could forward to a fax to get a print out, but that assumes you have the fax software on your PC or network and that just uglifies it for me. I mean what if you don't have a fax setup like that? I guess you could always forward the file to the remote storage area where your handset could access it via its packet data connection whilst simultaneously spewing it out in fax format to the nearest fax machine via its "switched" telephony connection.

Most go make some coffee.....


10:36:27 AM    

Going to the toilet

Hear the one this week about IBM dropping their patent 6,329,919  "System and method for providing reservations for restroom [toilet] use"?

Of course, sounded ridiculous at first, but turns out to be quite sensible when you read the patent that conveys the intent was to make it easy to visit the tiny toilet on the airplane without having to queue (hazardously) in the aisle.

Deceptively simple (and silly as a patent) - the passenger with a tense feeling in their bladder simply presses the button to "reserve" the next toilet slot (hey - "bog slotting"). Seems like a good idea to me.

It got me thinking about seats and remote control - via wireless of course (what else do you expect from me?). Maybe you could reserve toilets via SMS in a theatre and so forth. But more attractive is being able to order snacks via SMS from a cinema seat to avoid that huge tension that builds up whilst deciding whether desire for snacks warrants missing a part of the movie, or which part of the movie is sufficiently devoid of information that it can be missed for a snack-trip. Of course, you could have an interactive SMS menu designed using Xsonic DataNow (shameless plug).

But what about WAP? [Yes - I know that's something we all ask in wireless, but I don't mean "whatever happened to WAP", as I say in my course - "WAP is back, not WAP is cr*p" (WAP2.0 that is).] An interactive snack-menu from the seat via WAP? Maybe. This is where you would need a context-sensitive menu that kicks in when you enter the cinema. Well, that would be easy if there was a pico cellsite in the complex to detect accurately your location and push out a WAP-push message, but given that they go out of their way to stop phone usage in cinemas, perhaps that's unlikely.

Then again, maybe they haven't thought this through. Maybe they should promote phone usage in cinemas, especially during the trailers for new movies. How about being able to book tickets from the phone whilst watching the trailer? What about getting picture-messaging ads for the next films with a link for booking? What about getting discount coupons (free coke or whatever) sent to your phone?

It would probably be even better to sync up things happening on the phone in real-time to each trailer allowing ample opportunities to get bookings, merchandise sales and so on. But this would require something more elaborate than what is currently on offer via GPRS devices. Getting the hook between the user and the movie screen could be done manually (without location-sensing etc.) if the user has already subscribed to the service (incentives could be used of course) and their viewing schedule is known by the backend system. Timebase synchronisation should be fairly straightforward (though not sure what cinema projection kit encompasses). [Note: I am pondering about the use of WiFi in this and again thinking that WiFi may be inevitable in future consumer devices irrespective of their wide area 3G capabilities.] [NNote: What about live film reviews from the audience?] [NNNote - reviews could be left as "Air graffiti" of course.]

Well, too many ideas are buzzing through my head on this one, so I'll let you fill the gaps with your own imagination as I have to get on with drinking my tea (bonus cup from breakfast) and updating my course (something I was doing already for the next run, but I just thought to include this example). But this kind of interactive experience seems a possible way to capture interest in new generation wireless services.


9:56:21 AM    


Click here to visit the Radio UserLand website. © Copyright 2002 Paul Golding.
Last update: 11/4/2002; 7:57:59 PM.
This theme is based on the SoundWaves (blue) Manila theme.
October 2002
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Sep   Nov