Bias in Eathquake reporting?

So in seismology news, an earthquake struck off the coast of Vancouver Island; the story is still developing. When I first read the story, it was reported that a 6.1 quake had struck the area; now, the report is 5.8. The last earthquake reported-- I believe the one in Southern California not too long ago-- was similarly reported as initially being larger than later measures. Why the discrepancies? And from what I remember-- and this itself may be skewed-- I can't remember the last time an earthquake were reported as being smaller than what it was eventually adjusted to.

I remember being in the Loma Prieta earthquake in 1989 and having it initially reported at 7.1, whereby it's understood now to have been a 6.9 quake. The Wikipedia entry may clarify a bit of the mismatch: "...the earthquake lasted approximately 15 seconds and measured 6.9 on the moment magnitude scale (surface-wave magnitude 7.1)."

Are the surface readings what we get first, only to have the moment magnitude scale provide greater accuracy later? I was under the impression that you head right on over to the seismometer and see what it has to say, then catch some lunch.

Do reports of bigger earthquakes lead to bigger stories? Perhaps, but news sources report what the USGS tells them...if they do lead to bigger stories (and bigger audiences), it's not from their doing. Can they consistently give too large of initial readings? Why not adjust the process?

(By the way, on the Richter scale page on Wikipedia, they list a magnitude 10.0+ earthquake as "epic." I found that humerous for some reason. Probably too much dissertation work. Though also noted on the page was the fact that the China earthquake in May was initially estimated at 7.8 and later upgraded to 8.0...the only upward adjustment noted on the page.)