Funnily, I usually do the opposite. I'll set my first weld wet, because it's typically fool-proof, and when you've got a large initial stack (my initial stacks are usually 3" tall) for many of my patterns, and the vast major of patterns I build are re-oriented end-grains or "mosaics". My experience has shown me a higher consistency of good welds on the far edges of billets with flux, and with the above type of patterns, it's pretty important, as I like to re-orient and draw them very thin for making W's and those types of patterns get as much action in the first weld as possible, while having to grind away the least amount of "lines" on the edge once it's re-oriented.
If you don't make a lot of these types of patterns, this probably makes little sense to you, as with side-bar and pressed in patterns, you almost never bother with the edges, and by the time you've restacked a few times, and the billets wide, the loss is pretty minimal, lots of people that sell damascus, myself included, usually don't bother even grinding the edges, and most people just know to stay away from the very edge, but it's a whole different game with complex mosaics and end grains, where if things go wrong, you can chase a bunch of the billet grinding past any bad edge welds once they're re-oriented to the top and bottom of your billet.
So I typically do my first weld wet, and then all subsequent welds dry. For me, it's the fastest, with the most consistent results. After a couple of restacks, there are no "flux lines" to be seen in the finished billet, and I can't remember the last time I've had a flux inclusion. I've heard some "myths" bandied about about this or that weld style being stronger than flux welds, but I've seen no evidence that this has been tested under any objective scenario.
To be clear what I'm referring to here are not bad welds, I'm referring to the depth on the sides where the weld isn't perfect. Which is something you only ever notice when you're reorienting billets 90 degrees, which of course, can put a minor flaw, right in the center of your billet.
Also, flux is a buffer, a shield, from oxidation and carbon burn off, that's what will keep welds from setting properly. It's not soap, it doesn't clean anything out. You can hydrocarbon or fluxless weld rust and mill scale covered stock easy as pie if you know how, and the rest of your process is right. However, flux has it's advantages even if you aren't using it for welding. For patterns with a large number of restacks or extensive soaking sessions, a light fluxing can mitigate a significant amount of over-all decarburization. Even if you have a perfectly reducing atmosphere, without anything protecting the billet, you'll notice significantly increased scale formation,as soon as you pull the billet out of the forge, an order of magnitude higher than if the tiniest bit of flux is sprinkled on the outside of the billet, and there hasn't been any real research to see how amplified the carbon burn off is inside the forge without flux.
I'm not saying it's perfect, and I use it only under particular circumstances and sparingly, but this recent trend of vilifying flux has gotten a bit out of hand in my opinion. It's a tool, that has advantages and disadvantages like all do. I don't think it does any of us any good to pontificate in absolutes, just because a new trend has come into favor.
I personally use all techniques, depending on the pattern, the size and end shape of the billet, the starting material, and whatever other caveats come into play for a particular job. It's important for all of us to find which techniques work best for us with our equipment, materials, and process.
Also, thousands of the best knives ever made by the best makers of our time, were made with damascus where every weld was done dripping flux, with nary a flux line to be seen, inclusion, or other imperfection, in the finished product. Flux lines go away after HT, because they're not trapped flux (they're weld zone discolorations), unless there's a significant problem with a weld otherwise. The only time it's really an issue, is for someone like me who sells damascus billets to other makers, they're unsightly, mostly, because of a large amount of mis-information being proliferated by people that aren't as knowledgeable about the subject as the purport themselves to be. I will say that if you do have a major flux inclusion, at least you will typically notice if before you're making a knife from it, as it'll usually cause a significant blister, which is plain as day. An isolated bad weld area without flux, will not make itself known until you grind into it.
It's easy to avoid them, because dry welding (after the initial) stack, is efficient, and also fool-proof, and has become my preferred technique regardless (dry welding the initial stack is on the other hand, excessively tedious, you have to either build a can, which is time consuming, or weld the entire sides up with whatever your preferred method is, even MIGing at light speed, it takes forever), so no big deal, but is there any objective evidence that one method or another yields a better result? Not that I've seen, unless I've missed something since Kevin did some pretty extensive testing of flux and fluxless welds, which didn't from my reading, yield any conclusive evidence one way or another, although many people chose to interpret it the way they already "felt" was better.
To be clear, when I say dry welding, I mean welded seams or canned welds (they're fundamentally the same), not hydrocarbon or just open stacks with zilch, although I've done many hundreds of those welds successfully also, for me though, every once in a while I'll get a "rogue" bad weld or inclusion, with the open stacks with no hydrocarbon or anything, that I typically can't explain, and trust me, I'm not welding by feel. I've got pyrometers, timers, and an extensive process, that usually, is foolproof. Although I'm sure I've just jinxed myself.
Also, there are kiln washes that can be essentially flux proof, and plenty of options to reinforce a forge that flux can't touch (super high alumina extreme duty firebrick for example). Unfortunately, in the knife world, not many people have ever delved deep enough to figure this out, other than offering a few limited, super expensive products, in a range of thousands, that the rest of industry understands.
Sorry, I know this ended up being a bit of a rant, and I'm not trying to call anybody out, I just hate seeing blanket "dogma" being passed out as objective truth. I know it's the nature of social groups to try and distill everything down to a specific consensus, but as craftsmen, I don't think we do ourselves any favors trying to assert absolutes in such complex areas of the art. Hell if we really cared about having the best scenario from the standpoint of the steel when forge welding, we'd probably be doing our welds in charcoal, since it's pretty much the optimal medium to assure no net carbon loss, at least, when it's perfectly setup and tended, but I think most of us agree, that's simply too tedious, and allows too much room for error, when things aren't perfect.
