[FoRK] Portable hotspots (or, why the iPad doesn't need 3G / future of Bluetooth questionable?)

Jeff Bone jbone at place.org
Tue Feb 2 10:29:26 PST 2010

Ken writes (in 1024-character column width, as usual... reformatted  
here for your consumption.  ;-)

> I'm a little bit confused (yeah, I know that's more or less my  
> permanent condition but let's say a little more confused than  
> usual). You seem to use WAN to refer to at least two communication  
> technologies.

Not clear where I did that, but I'm specifically referring to  
broadband cellular.  The point, really, is that you ideally don't need  
more than *1* such device / connection per personal "constellation of  
devices" assuming that 1 device can act as a router between that  
broadband cellular connection and wi-fi.  I.e., let's settle on a  
personal networking architecture where we've all got one portable,  
always-with-us broadband cellular interface and everything else talks  
wi-fi exclusively (and is routed via the broadband interface when  
appropriate.)  That interface would, of course, only be used in cases  
where other (i.e., not out-through-broadband) wi-fi is unavailable.

> Later in the post you seemed to use the term WAN as synonymous with  
> a WAN based on wi-fi wide-area networking. That's what confused me.

Not sure how you drew that conclusion, but that certainly wasn't what  
I was suggesting.  You can assume throughout the discussion that when  
I say "WAN" I mean broadband cellular Internet connectivity.  Internet  
connectivity via e.g. home wi-fi, some public wi-fi w/ a better pipe,  
etc. would generally be the preferred mode for Internet connectivity  
from any device in the personal constellation of devices, and routed- 
through-broadband the fallback.

> I agree with a preference for wi-fi for all wireless connections to  
> a user device versus Bluetooth for some and proprietary for rodents,  
> keyboards, etc. It would require that the user device be able to do  
> peer-to-peer wi-fi connections, which most cannot do right now  
> without a lot of screwing around (e.g. I'm sure I could figure out  
> some way to make my netbook and laptop talk directly to each other  
> over wi-fi without a router or switch between them but it wouldn't  
> be simple). Bluetooth provides that capability. Is it likely that  
> Bluetooth and propietary wireless connections to the user device are  
> the path of least resistance? E.g. easier to just create more  
> wireless tehcnologies for distance-limited uses than to fight with  
> whoever to get wi-fi (IEEE 802.11*) to do it? ...ken...

Well, if the personal architecture assumptions universally included a  
mobile router / switch / WAP with you all the time serving as the  
gateway between broadband and wi-fi (when necessary), then you don't  
necessarily need the pure-p2p aspect of things (though of course then  
none of your gizmos can talk to each other absent that device.)  OTOH,  
some of the p2p stuff is fairly trivial conceptually but I'm not sure  
how much fussing around you'd have to do to make it work w/ stock  
hardware and stacks.  Some folks --- Eye-Fi for example --- are hot on  
the very-limited-application uses of Wi-Fi, so at least the meme is  
out there.

Generally, I always bet on commodity / community standards vs.  
consortia standards in the long run...  The whole Infiniband / SAN  
thing, for example, has always amused me to no end.  An arms race that  
can't be won permanently, a pipe dream of VCs with more capital that  
needs to be "absorbed" / put to use than they have time to vet and do  
smaller deals.  (In particular, this was a huge issue in Austin in the  
Early 'Oughts...)

That help?


More information about the FoRK mailing list