March 23, 2008

Is WiMAX a disaster?

Is this account of WiMax correct

In an astonishing tirade to an international WiMAX conference audience in Bangkok yesterday afternoon, CEO Garth Freeman slammed the technology, saying its non-line of sight performance was “non-existent” beyond just 2 kilometres from the base station, indoor performance decayed at just 400m and that latency rates reached as high as 1000 milliseconds. Poor latency and jitter made it unacceptable for many Internet applications and specifically VoIP, which Buzz has employed as the main selling point to induce people to shed their use of incumbent services.

Any Copenhagen or elsewhere WiMAX customers around who would care to comment?

Posted by Claus at 07:38 PM | Comments (0)

March 06, 2008

App-centric or Data-centric

Anyone remember the OpenDoc vs OLE document war of the 80s? It was a war over architecture that became a war over control. The OpenDoc model - advocated primarily på Apple but soon adopted by the ABM (Anything But Microsoft)-crowd. The idea - in opposition to OLE - was to have simple UI parts deal with specific types of data and then orchestrate the parts as needed for the complex stuff. The OLE model is what we know from e.g. Word. Humongous apps that do everything.
I've written about the battle before. OpenDoc has some pleasing characteristics experience-wise and architecturally, but commercially OLE won, because OLE affords a lock-in strategy. Which is exactly the same reason Microsoft is still fighting for this model by trying to convince the world that the arcane document format needed to support the lock-in is in fact an open format. Pure nonsense, and its clear why it is. If not, MS's own inability to reimplement their own format on another platform should be a good clue as to how good OLE is at providing lock-in.

Anyway, long introduction for something else: The "Silos or formats" debate we're seeing now is a replaying of the exact same arguments for the exact same reasons with the exact same public vs. private reasons for taking positions. Tim O'Reilly is fond of analogizing thee lock-in strategies to Intel Inside Data-loyalty, but the OpenDoc vs OLE debate is actually a lot better, since there's more directly relevant history there.

Posted by Claus at 05:56 PM | Comments (0)