It's true that information is massless, and eliminates that obstacle to interstellar trade. But it's still subject to the laws of physics, which place limitations on trade in information.

The speed of light means that negotiations and exchanges would take a long time, effectively creating a sphere in which trade is feasible within the lifetime or attention span of the parties. The time lag would apply pressure to decide quickly whether to accept an offer; making a counteroffer and waiting for the reply introduces painful delays.

The inverse-square law limits bandwidth, and is only partially mitigated by using a focused beam. First you have to broadcast omnidirectionally to find your trading partners. There's a reason why our deep space probes using decreasing bandwidth the farther from home they get, because of the relationship between power and bandwidth. So there's an energy cost in trading information, and because different parties would have different energy costs depending on their technology, the information traded would be valued differently by the different parties.

The basis in physics creates challenges at higher levels. You need to establish a common modulation scheme (AM vs. FM vs. SSB etc.), data format, protocol, and language with trading partners, and that will take a long time due to the aforesaid limits. We can probably count on the universality of the bit (nothing's more fundamental than on/off), but not much else. The universality of the bit means that binary representation is probably also universal, but you have to agree on bit order and how to represent on and off. After that, though binary is directly translatable when it represents numbers, its use to represent other values, such as tokens in a language or pixels in an image, requires agreement by both parties.

If there's an established trading civilization out there already, some of the procedural points could be handled by a broadcast "Welcome to the galaxy" channel that loops instructions and specifications for participation on the other channels. Sort of a channel guide for the galactic communications infrastructure. Such a channel would be designed to be easy to pick up, and since we haven't yet, it's doubtful whether it exists.

In all this, I'm anthropomorphically assuming that trade is conducted according to Adam Smith's description, which requires all that handshaking and negotiation and back-and-forth. But today's open-source software concept suggests an alternative: Just fire off your information without prior negotiation for compensation, and reap the harvest indirectly by receiving everybody else's similar transmissions.

A first pass through game theory suggests that that might be nuts, because it games out that "betrayal"--taking without returning--is the most advantageous. And maybe that's why we don't hear any open-source information broadcasts from out there. But yet it seems to be working for a lot of enterprises and individuals here on Earth, through a mechanism that can only be described as "what goes around comes around". Economic karma, in other words. Open source software or information looks, on the surface, like an act of philanthropy, but like a lot of philanthropy, it's based on enlightened self-interest.

You're right that interstellar trade in information is a lot more likely than trade in physical objects. Still, it has a few obstacles of its own, and unfortunately brings up the question, if interstellar trade in information is so attractive, why aren't we already picking up the data streams from this trade?