Depends on several factors such as the number of bolts, the bolt pattern, length of the channel, stiffness of the tubing. To consider the joint fixed I would want to see multiple bolts in a relatively large pattern. My recommendation for most cases: be conservative and assume the joint is pinned.
Take a step back and look at it from this perspective: does the joint allow for rotation? Mechanics says that pinned joints allow rotation but not translation.
Without seeing the connection or having more detail one can't say for sure. For modeling purposes, I would also assume it is a pinned connection.
Remember also that there is no point in providing for a "joint" that is fixed yet the structural tubing that the channel will connect to is flexible enough under the loading to allow rotation. This would undermine the purpose of the fixed connection.
Here is my simple response: Bolted connection can be simple or fixed. It all depends on how you detail it and how you design the steel member for the end forces. As stated above by fellow engineers various connections allow rotation and or prevent it. I have a copy of an excellent connection illustration in PDF format that is published by the AISC. It uses real life pictures of connections with engineering details and it classifies these connections. I believe it is based on the connection tree that Dr. Duane Ellifritt built at the University of Florida. The file is about 4 MB. I think that you still can download it from the AISC web site
Coming back to the original question, many connections are intermediate between pinned and fixed. In this case, the simple answer is to be conservative. That may mean for example assuming pinned for the supported member but it may mean assuming fixed for the supporting member. If you are uncertain and you are stressing the members highly, then the safest approach is to assume each in turn and design for the worst case. If the members are only lightly stressed, then it does not matter.
Becareful in that it is not always conservative to assume a pinned joint. You really need to look at both extremes to be certain your design is safe.
For example, a pinned-pinned beam, point loaded in the middle, will have a maximum bending moment in the middle, zero moment at the end, and highest shear at the ends. A fixed-fixed beam has a moment at both the end and the middle, and a high shear at the ends.
Now if the beam is uniform along it's length, then it can be easily seen that the pinned-pinned condition is conservative for all sections. However, it the beam tappers towards the ends/attachment points, then you may have a situation where the ends are more critical under a fixed-fixed situation.
This situation of Degree of Restraint is very well explained by Flabel in "Practical Stress Analysis for Design Engineers", starting on page 39.
So a little engineering judgement is required here.
I have tried locating the file, on my own, at aisc.org, as you have specified. There are several excellent resources available for free downloads, but, I failed to locate the one which you refer to.
The determination whether the connection to be considered is fixed or pinned depends on the connection detail and relative stiffness between connected member. It is usually conservative approach that the connection is considered as pinned connection and most structural engineers do that.
If you consider the connection between channel and tube, it is proper that the connection is pinned when the channel is bolted to the wall of tube that is transverse to longitudinal axis of channel, because the wall of tube is small stiffness. If with additional plates, the channel is bolted to the side wall of tube which is pallel to longitudinal axis of channel, this connection would be considered as fixed connection.
