1) Provide constructive positive feedback when possible
If you like something, then you should try to tell the author what exactly it is you liked about it. Most questmakers/tile artists/music artists enjoy receiving constructive positive feedback in particular, and pointing out things you like about a submission is likely to encourage them to continue in that direction.
Example: If you like someone's palette usage, tell them what exactly stands out about the palette. For example, I tend to really like subdued palettes... of course, peoples' tastes will vary and be subjective, but that's to be expected.
2) Provide constructive negative feedback when possible.
If you dislike something, then you should try to tell the author exactly what it is you disliked about it. As long as you're constructive and carefully exercise tact when giving your opinion, then most artists will take into consideration what you are saying.
Example: If you think someone's dungeons are too linear, tell them that. Give advice on what they can do to make it less linear -- e.g. using more backtracking, putting locked doors a bit further away from keys.
A good review will contain both a balance of positive and negative feedback, if possible. Try to carefully think about how you felt when playing a quest and analyze both the good and bad parts of it. That way, an artist will have a good idea of what direction to take with future projects of theirs.
3) If you don't have anything helpful to say about something you don't like, you should avoid saying it at all.
Nobody likes being told that their work sucks without a good reason why. Most people will take it personally if you just say something along the lines of "This was terrible!", because this is essentially meaningless and unhelpful and doesn't talk about anything objective. In these cases, unless you can come up with a good reason for saying you don't like something, it's better to not say anything at all... and it's certainly never a good idea to outright tell someone that they suck and should quit.
4) Accept feedback gracefully, even if it's negative and even if it's harsh.
If someone brings up something they don't like about something you've made, even if it's negative and even if it's just blatantly saying your quest is awful without a good reason why, you should always avoid attacking them in retaliation. These actions are unjustifiable, even as self-defense.
If someone has a sincere, well-thought out reason for not liking something you've made, even if you strongly disagree with it and refuse to implement their suggestions, it doesn't necessarily mean they have a vendetta against you... perhaps the thing you've made isn't for everybody. :-/ You don't have to change your creative ways for anybody, but you don't have to take the criticism personally just because you don't agree with it, either. Just state that you're not going to change your work and move on.
5) If someone doesn't accept your criticism, you shouldn't force it down their throats.
... I'm not sure what else to say about this one, admittedly. :P If someone's decided that they don't agree with your criticism for whatever reason and has decided they're not going to change their quest, you can't really do anything about that. Repeatedly giving someone the same criticism or telling them "oh, you just can't accept criticism" for sticking with a design decision isn't going to help anyone -- if anything, it's going to irritate people and make them feel like they're being antagonized.
6) Read descriptions/readmes before reviewing.
If the description says something about the quest/tile sheet/song and you know you won't like it, then that probably shouldn't be a factor for your review. It's possible it was just intended for a different audience than you. As an example, if you're reviewing an enemy gauntlet quest and it explains that clearly, you probably shouldn't rate it 1 star just because there are no puzzles. If you're reviewing something that's meant to be a piano-only music piece, you probably shouldn't say "hey! There should be other instruments!". ... You get the idea. :P
7) Regarding specific item types...
When rating specific things in the database, it's generally a good idea to take a few things into account for that specific type of item:
When rating quests, it's generally a good idea to play a fair chunk of the quest before actually reviewing it. A quarter might be okay, but at least half of it is probably a safer bet. Of course, the best reviews are probably from those that have played the entire thing start-to-finish. It is generally not a good idea to review quests based on screenshots alone. (With that said, it's still generally a good idea include screenshots when submitting quests, if you can, as good screenshots will increase peoples' interest in your quest and get them to play it... but they shouldn't be the sole determinant of a quest's quality.)
When rating loose tile sheets, two things need to be considered: Both how good they look and how well they work. The latter one is important, as I think it's something that's often overlooked. Here's some things to keep in mind when rating tile sheets... these are straight from Espilan's old topic, but if you guys can think of any more things that should be kept in mind when rating tiles, feel free to add them in a reply.
- Are they really usable, as long as they're not too hard? Are any included instructions easy to understand?
- Do the ones that animate animate well?
- Are they recolors or straight rips from a ROM or other source? These seem to be frowned upon due to how easy they seem to be. Be careful when submitting them.
- Does everything sound right to you?
- Is it too loud? Is it too quiet?
- If it is a Tracker submission and it loops, does the loop happen at the right time?
- If it is a MIDI submission, is the extention 'MID'? Zelda Classic can't use the 'RMI' format.
8) The "Golden Rule"
Peoples' tastes vary and are subjective. It's definitely okay to have an opinion and to express that opinion, but you can't expect everyone to think the same way as you.