As it is now, the professions do not really exist in salem and I guess a lot of players that were initially interested in salem had the idea of getting some profession and being good at it and setup some sort of business based on it. Players would buy and sell items based on their respective proficiencies in various areas. At the moment, I see little reasons to specialize, a lot of things seems to be missing at the moment (I understand it is beta), but it is not clear to me by reading the forums where the devs are going with the skill system and how exacly they want it to be important in the game. I saw that at some point Loftar pointed out that he wanted the skills to be only minor aspect of the game when it comes to its impact on the salem experience (if I understood correctly. viewtopic.php?f=12&t=322&p=3387&hilit=profession#p3387). I think it would be rather disapointing if one cannot feel that the time invested into some crafting specialization did not give you the exact feeling of being superior to others in that particular area. Someone could for instance be one the best baker in the area and make money out of it. (It is not possible with the current system and economy.)
The professions have already been discussed on the forum (viewtopic.php?f=12&t=2505, viewtopic.php?f=12&t=960&p=9396), but the system I am proposing is mostly based on systems already available and is quite different from those already discussed.
If I understand Loftar's statement correctly, it is more that they don't want the skills to basically determine the whole usefullness of a given player. This would indeed penalize a lot the casual players who don't invest as much time as some other. The whole community would suffer needlessly from such things with reduced server population. I believe it is possible to make the skills and professions good enough so that some people will put a lot of effort into it, while not penalizing the others that are not specialists so much as to make them useless. Some balance must be reached.
Again I am suggesting modifications that seems to me very intuitive and thus players do not need to understand the whole underlying mechanics to get it. I suggest that some sort of crafting quality factor be added (It could consist of a purity modifier that is directly included in the purity of the final item, or a separate quality of manufacture variable. Ex: A buddies on a branch could have 50% purity while having 25% quality. OR if you don't want to add another factor, the purity of the final product could be affected by the skill of the crafter. It could only lower the quality and not increase it, meaning a 100% crafter would get the quality of the input products.). This would of course affect the items inspirational/healing/gluttony values. Ex : Crafter1 has skill 40% of skill Crafter2 has (100%). Crafter1 buddies on branch would (starting from 100% purity inputs) turn out to be lower than that of Crafter2. The later would get a 100% purity product. (The formula could be linear, but I would prefer something like, 50% of purity of inputs + 50% coming from crafter's skill. A crappy cooker can very well ruin a very good raw steak and reduce its purity but not totally ruin it.) Now this simple idea introduce the concept of professions, I can imagine players with good raw materials (high purity raw steak) and low cooking skills, want the food to be cooked by some specialists in exchange for some silvers, thus creating the cooking profession. You can think of similar reasoning for all the skills and craftable items in the game. An example is given later.
Factors that affect the quality value of the final product would include: quality of raw matrials, quality of item needed (fire, stove, smelter, etc. with weigths according to some logic.). Buddies on a branch could consist for example of 80% fish purity, 15% fire and 5% stick quality for final quality/purity values. It would also be affected by the cooking skill level of the player (Actually more the skill that unlocked the recipe).
Foods could only have a purity value, but manufactured stuff and equipment could very have some quality rating and/or a purity one. A good quality carpenter bench would yield higher quality planned boards for example. (Reaching the theoretical limit of 100% is not easy and should not be, hence it is theoretical.) Or a good quality smelter would see the metal poured from it with a better quality. (Ex: Iron ore determine 40% of output quality, lime 20%, 20% coal and smelter 20%). That would create an incentive to build high quality structures to further process goods (creating demand for good crafters -> economy). If structures would need repair after a couple of use, you would still need a good crafter to repair it to lessen the quality drop. This way items need to be replaced at some point and the economy can get going.
The whole idea of introducing professions is one of customization of characters and also one to make the economy system going. I discussed the economy in another thread (viewtopic.php?f=12&t=3045). Also, by having some purity to the forageable items will help create those jobs and more (viewtopic.php?f=12&t=3028).
The skill of a player would be determined simply by a weighted average of the proficiencies required to unlock the skill that is being used to do the action. For example, if someone is crafting buddies on a branch, the skill directly linked to that recipe is fishing (600 F&W + 600 H&G). Thus the higher the F&W and H&G are, the better his recipe will be. Since both require 600, they would contribute each 50% to the final quality score. If Fisherman1 has 2000 F&W and 1200 H&G, his crafting score for this recipe would be 0.5 * 2000/600 + 0.5 * 1200/600 = 2.666. Now the higher this score, the better the outcome. But in order to compare apples with apples, we have to compare his score against the best fisherman in salem to make sense of it. I would suggest at this point that the best fisherman (highest score) on the server have 100% quality of crafting (he is the best afterall). (This requires the system to regularly refresh its list of the best score achieved for a character on a given server.). Lets take the numbers 12000 F&W and 10000 H&G for Fisherman2. (If more than one skill unlocks the given recipe, an average of the two skills with some weight could be used. (the same principle would apply, these tables would need to be hardcoded))
That means that your score is not compared to some hardcoded value, but is actually compared to the best there is around which kind of make sense. It also creates a perpetual competition among specialists to be the very best, thus reducing the possibility for a player to be exceptional in a variety of professions (This creates demand and supply for the economy). The relation need not be linear, which means that a fisherman with score 10 is not necessarily twice as good as one with 5. I think it would be a good start if the relation was on a square root scale (All scores would be put in square root). Ex: Fisherman1 (0.5 * 2000/600 + 0.5 * 1200/600)^(1/2) = 1.633 and Fisherman2 (0.5 * 12000/600 + 0.5 * 10000/600)^(1/2) = 4.28. Now it is clear that on that scale, the two are not that much different thus making the extra grind somewhat questionable, but not useless either. In this particular example Fisherman2 is about 2.5 more skilled that Fisherman1. If Fisherman2 is the best on the server, his buddies on a branch would inherit 100% crafting quality (or purity if raw quality is 100%) and Fisherman1 would get ((1.633/4.28) * 100%) = 38% (It could be changed, just the numbers are taken for example purpose. A balance must be reached.). This little example illustrates an important point of my suggestion, that is, everyone can be useful in salem without much effort and yet it rewards the very best, grindy type of people that want to make it their profession. (A square root relationship basically means that for each time you have a 4 times higher score than someone else, you are in fact only twice as effective as they are.)
Please comment, and make sure to read the other threads since this idea is made to fit together with the other ideas I am proposing. Also, if you think you have a better idea, spit it out. Any questions you may have are welcomed and I will answer my best.