I read the whole article; I did NOT read the whole thread.
When I was an insurance adjuster I was trained in arson investigation (several courses including fire department controlled burns). From my perspective I can say without reservation that trying to determine the cause of fires is almost futile in every case.
They train you to look for evidence of accelerants, which show up as specific kinds of stains on floors, and which are a clear indicator of arson. Couple problems with that though; most people keep flammable liquids in their homes, under the kitchen sink stuff, nail polish remover, rubbing alcohol, and basements full of thinners and paints. Many items of home furniture contain polyester fabric and polyester foam, which when melted and then burned leave the same traces as splashes of gasoline or kerosene. This goes for almost anything made out of plastic as well. Over half of my class voted a large stain underneath a TV tube was accelerant, when it had in fact been the TV itself, and the fire started in another portion of the room altogether via electrical means.
They also train you to look for the path of the fire. They make it all seem so easy and common sense while you are there, but I happened to be there when the chief arson investigator for the Anderson IN fire department could not correctly identify the origin of two fires, nor the mechanism that started them. Even our class came up with many alternatives to how the fires started and how they spread. The video tapes proved almost everyone wrong, which was part of the point, that fire investigation is NO SCIENCE.
We were then told stories about how some people were sentenced for murders they did not commit based on faulty fire department reports. One I remember was how a man was put to death over an alleged arson that killed his wife and children. It seems the fire report stated that accelerants were used. Years after he was executed it was discovered that the flooring in the home was glued together using a very turn of the century method involving gasoline as solvent for glue, completely invalidating the accelerant theory.
These fire investigators in the article that say shit like "Fires don't destroy evidence, they create it.", "You learn that fire talks to you.", and "The fire tells a story.", are all self-serving nincompoops not paying any attention to their own trade.
Arson investigation evidence, which is unfortunately passed off as infallible (along with DNA, fingerprint, fiber, and pressure mark evidence), should be highly scrutinized and treated the same way as hearsay. It is anything but infallible.
Cliffs
Things I learned about fire investigation
1. Almost impossible to tell how a fire started or whether it was started intentionally.
2. Accelerant sniffing dogs can find a drop of gasoline on someone hours after it was dripped onto the person, and from several blocks away.
3. Swimming pool powdered chlorine and brake fluid are fun.
4. Poor guy got the shaft due to overzealous retards in positions of authority.
