Miles Nordin on Mon, 1 Jul 2002 20:17:02 +0200 (CEST)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

[Nettime-bold] Re: <nettime> mobile: the next killer app (some comments)


>>>>> "rs" == Rich Stillman writes:

    rs> But take a close look at instant messaging.

There are two big problems with IM.  While wireless IM might be nice,
I think we should want wireless email first.

1. email is inherently a one-to-many discussion medium.  One can
   subscribe to mailing lists, or just put lots of addresses in the
   To: header.  IM is, like voice, a person-to-person medium, so it
   preserves the political limitations of the telephone network as
   compared to the Internet.

2. IM is a closed standard under AOL's control.  Open clients
   reverse-engineer AOL's protocol.  Outside AOL's hedgemony, Instant
   Messaging is a balkanized mess of competing and
   increasingly-ridiculous standards.  My favorite right now is
   gale.org, but I haven't actually used any of them extensively.

   Even if you could make IM one-to-many, which you can't easily do
   equivalently to email given the system's pervasive assumption of a
   synchronous conversation, the community could not grow past the
   borders of the system integrator's brand name and the
   wireless-handset medium.

email, OTOH, is ubiquitous and flexible.  It's also gatewayed to stuff
like FAX, web-to-mail, ftp-to-mail.  And if you limit the discussion
to fundamental functionality rather than color wallpaper and bitmap
emoticons, email is a superset of IM.

Granted, email is missing some of the sweet IM window-dressings like
ad-hoc conversation threading and idle-time-stalking.  The
single-window conversation-threading is kind of pretty, but the
stalking features are basically irrelevant with wireless: what's the
point of keeping track of who is ``logged in'' when you are carrying
the handset and therefore always logged in?  In any case, whatever IM
extras we can achieve should come second, after providing email's
essential and open, non-exclusive communicative ability.

These pleas for wireless IM disturbingly echo email's
pre-mainstream-Internet past where the Internet was one of many types
of email---there was also Compuserve mail, MCI mail, Delphi mail,
Prodigy mail, Bitnet, Fidonet, WWIVnet.  You tried to be on the same
mail system as your friends, and most people were on several systems.
Business cards would list a grab bag of addresses along with fax and
``telex'' numbers.  From Compuserve one could communicate with most of
these, but only through ``gateways'' using weird gothic addresses,
introducing formatting garbage, requiring unpredictable delays, and
often charging per-message or per-line.  No doubt this appeals to
telcos with their history of ``roaming agreements'' and closed ``text
messaging'' systems (with most US carriers you can receive truncated
60 - 160 char emails sent to a magic ``text message'' address, but you
can't send this kind of text message from your handset to an email
address.  Now, there are many subtly-different kinds of text messages
even on a single carrier's network.  Some of them ring and others
sneak into your remotely-held ``inbox'' like voicemail, all with
different length limits, address namespaces, costs, and archiving
rules.).

Anyway, I don't think IM is useless.  Considering stuff like finger,
ytalk, irc, and Zephyr, the IM concept is almost as old as Internet
email based on SMTP.  I just think simple, properly-integrated
wireless email needs to come BEFORE wireless IM, and that the current
US trend of AOL-branded Blackberrys and AT&T keitai is far more
suspicious than Sadie Plant.

    rs> Current screens are too small for Web browsing.

    rs> Connection speeds are too slow for data transfer.

I disagree with this almost unilaterally.  Existing systems have
totally appropriate hardware and network speeds for doing things like
reading the nettime list.

The screen size does defy the current crop of web-oriented
flashturbation marketing graphic designers, but I think ``new media''
types will embrace the keitai screen successfully.  Remember, existing
handsets are 65536-color 120x160 screens.  In the reasonably readable
font that I'm using right now, 'xterm -font 6x10', here's how big such
a screen is:

 120x160 keitai                 newspaper, USA Today
+--------------------+     ___
|   Months before    |     /|\    Months before WorldCom says it
|WorldCom says it    |      |  began to boost earnings through
|began to boost earn-|      |  alleged accounting fraud, the com-
|ings through alleged|      |  pany faced intense pressure to
|accounting fraud,   |     3.3 please Wall Street that worsened
|the company faced   |     cm  as its problems grew.
|intense pressure to |      |     Last week's disclosure that the
|please Wall Street  |      |  telecom giant hid $3.9 billion in ex-
|that worsened as its|     \|/ penses during 2001 and the first
|problems grew.      |     ___ quarter of
|   Last week's dis- |
|closure that the    |
|telecom giant hid   |
|$3.9 billion in exp-|
|ses during 2001 and |
|the first quarter of|
+--------------------+

Think of it as an extreme version of the polite way you are supposed
to fold your newspaper on the train, except that the keitai should do
smooth scrolling with an analog IBM-gerbil-dick or Kyocera-scrollwheel
instead of requiring you to awkwardly refold and elbow your neighbor.

The screen size problem is that on current poorly-integrated American
systems, there is no smooth scrolling, only jerky per-line or per-page
scrolling.  

What's worse, you have to press ``next page'' after every 1 - 5
screenfulls and wait for a new page to download, which takes about 10
seconds with Nextel Online (compared to only 1 second with DoCoMo
i-mode, a better-integrated wireless data network of equivalent
bandwidth to Nextel's).  You spend 3/4 of your time pressing [next
page] and 1/4 of your time reading.  You shouldn't spend ANY time
pressing [next page], because keitai should download pieces of article
before you scroll to them by using software read-ahead.  NONE of this
has anything to do with screen sizes or network capacity!

The network speed problem is that email isn't sent to the handset
until you're reading it---another software/integration issue.  If it
were, then the existing networks' speed becomes both adequate and
irrelevant: you never wait for the network.

Devices like the Blackberry and the i-mode/EZweb/J-Sky celfones in
Japan send the entire email into the keitai before it rings.  This
way, you can scroll through mails quickly, solving both problems.

    rs> Memory and processing power are too limited for significant
    rs> local applications.

I've seen zero evidence of this.  I see that software in US celfones
is often of very low quality (especially that OpenWave ``phone.com''
WAP browser that mangles all the European sites), but I see no
evidence that this has anything to do with memory or processing power.
Nor have you actually named any ``local applications'' that this
nebulous level of memory and processing power supposedly precludes.
This is nothing but FUD.

    rs> Keypad design makes data entry too difficult

Again, this argument is rarely presented as an obstacle to Graffiti,
yet I believe skilled 10-button keitai/handset typists easily beat
skilled Graffiti(r)(tm) users, and unlike the plastic-pencil-pushers,
the keitai users can do it one-handed if they want to.

For some reason, certain people seem very resistant to the evolution
of devices in telephone handset shape to other purposes, in spite of
contrary evidence of its practicality in the Japanese (and Korean?)
markets.

I have a much longer rant about this:

 http://sakima.Ivy.NET/~carton/gotwap.html

--
There will be plenty of room, I'm convinced, for any wireless service
that doesn't frustrate its customers---something many American
wireless providers do routinely---and i-mode emphatically does not
frustrate its current customers.	-- Dr. Tachikawa, NTT DoCoMo

_______________________________________________
Nettime-bold mailing list
Nettime-bold@nettime.org
http://amsterdam.nettime.org/cgi-bin/mailman/listinfo/nettime-bold