rating is important to gauge a player's skill, but this becomes secondary when you are playing to enjoy the beauty of chess!
Chess rating system

I think when the game is unrated one, there could be no loss of rating. Is it correct? Any experts here may comment!!

People on this site just amaze me. A rating is a number next to your name. It is basically so people can feel good about themselves when they inflate it higher and higher. Your rating is meaningless--especially on a chess website. Most people on this site manage to achieve a rating they feel is "good" and basically maintain it through endless unrated games, beating up on easy opponents, or just not playing at all. Your chess game on the board is what matters. Its laughable how easy it is to achieve a decent rating on this site. I sit around 17-1800 on this site, and i get toyed with on chess sites like ICC by people who are rated in the 1300s. The point is, a number doesn't make you a good chess player, and you can be a great chess player with an average number. What absolutely matters is your play. Stay away from the 1 minute games and actually try to appreciate chess.
You can't achieve a good rating through unrated games. Unrated don't count towards your rating. And while you COULD play weak opponents to get that good rating, it would take a lot more of them since they'd only give a point or 2 per win. Plus, if you ever ran into someone who was better than their rating suggests, which does happen with bot users or some weird occasions, you'd take a HUGE hit in rating that would take many, many games to make up.
As for 1 minute games, they don't show as live chess standard games. They are separated into their own section. Bullet is for the quickest games, Blitz the next quickest, and Standard is for games with higher time levels (I think 5 minute minimum). So you can't boost your rating with those quick matches, and accidentally playing one of those games won't harm your standard score. Same thing with Online Chess. It's separate from your Live Chess ratings, so if you win because others eventually stop playing, it still won't affect your Live Chess rating.
Therefore, a Live Chess Standard score does very much indicate a level of talent in chess. Furthermore, it shows your average opponent level too, to indicate how good the opponents you're beating are rated. My average opponent rating is over 1370, for example.
So in summary, while it might be possible to inflate a Blitz, Bullet, or Online Chess rating, the Live Chess Standard ratings I was talking about are very difficult to game.

So while I have an 1885 Online Chess rating, it means nothing to me. It only got that high because I played 2 online matches recently for a tournament my group invited me to, or I wouldn't have even tried. Online Chess, people might drop out and you lose by forfeits, so what's the point? Because of that, the average Online Chess rating is 1339 according to the link I gave before, whereas the average Live Standard rating is 1166.
http://www.chess.com/livechess/players.html?type=Standard
So while I don't see why a high Online Chess or Blitz/Bullet rating would be anything to brag about, a Live Chess standard score seems reputable. Personally, I consider it a mark of achievement, a sign I am getting better and growing in skill as my rating increases. I try to not only boost the rating itself, but the average opponent rating, now that I'm getting better, by playing better and better opponents, if only to show myself I am doing it the right way and improving.
For me it stands to reason that a player's rating, regardless of how it is calculated, would be more 'accurate' if each player played just one game at a time against an opponent with a similar rating. If everyone did this I feel ratings might mean something.

yes.. playing many games at once gets confusing for me sometimes. Kind of forget the original strategy or play to fast just to make a move.

Interesting. why would you play more than one game at a time?
Because games can last a very long time. Even a 1 day/move game can last a couple of months, easily, if moves are made close to the time limits. 3 day, 7 day, and 14 days per move last significantly lognger. I'm in a 14 days/move tournament that started in December 2009 and it is still in the 1st round on one of the games (vacations come into play here).
If you want to get a lot of games played you either play a number of online games (whatever you can handle) or play a lot of Live chess.

As Martin wrote, in correspondence (online) chess if you're only playing one game, you don't get to play much chess. Keeping a few games going at the same time still only requires a few moves per day and allows a greater variety of openings and playing styles to be experienced. All of the tournaments except some knockouts start out with multiple simultaneous games for each participant.

everyone starts at 1200. then as you play you get a new rating. it is all based on the Glicko ratings system :) check it out - it's a fun read!
the link is only for students that have finished calculus

True, but your opponents' ratings are affected by playing with someone with a randomly chosen rating-which in turn makes their ratings less accruate and eventually inflates the whole pool.
This rating system bears more flaws than benefits. Who reckons we should abolish it for a more substantial and valid method of judgement?

This rating system bears more flaws than benefits. Who reckons we should abolish it for a more substantial and valid method of judgement?
Put me in the "who gives a s**t" camp. What difference does it make? And why would someone who's been a member for less than a day and doesn't have a single game recorded yet be saying "we" should abolish the rating system?

It's interesting that many on this board consider the ratings here to be inflated.
I recently joined my local chess club and was surprised to be winning many of the games I played against players that are rated 1700-1900 through tournament play. Yet my chess.com rating has never been above 1520 (IIRC).

Its difficult to achieve a good rating if you lose alot too. Especially to lower rated players.
It only goes to show that ratings really is not a measure of how your opponent perform on a game... means Rule 1: Never under estimate your opponent.

It's interesting that many on this board consider the ratings here to be inflated.
I recently joined my local chess club and was surprised to be winning many of the games I played against players that are rated 1700-1900 through tournament play. Yet my chess.com rating has never been above 1520 (IIRC).
Unless your club play is under tournament conditions I wouldn't find that surprising. I have very good runs at my club (with casual games) where I can beat a couple players higher rated than me, sometimes consistently. Yet, they play differently, and usually more strongly, in tournament conditions.

Unless your club play is under tournament conditions I wouldn't find that surprising. I have very good runs at my club (with casual games) where I can beat a couple players higher rated than me, sometimes consistently. Yet, they play differently, and usually more strongly, in tournament conditions.
Yep, that's probably a large part of it: I did introduce myself as a beginner, and I got the impression that my opponents were being a little experimental.
Also, the time controls suited me. I like to play 10-minute games on chess.com even though it's possible to lose to anyone by just not concentrating for a moment.
At the chess club, the quickest games were 30-minutes each, so I didn't make any instant-resign mistakes.
(Actually come to think of it, my opponents' attitude to their 30-minute games against me was probably the same as my attitude to the 10-minute games on chess.com)
That chess.com players page is really useful, for those wanting info on where they stand in the rankings. I can't figure out where they hid the link to it on the site though, since I wouldn't have found it had not someone posted the link on here earlier. I think they should call it ratings and give it a link in the dropdown menu for this site somewhere.
Another recommendation, it would be nice if they'd show on it what the 50-60% rating cutoffs are, 60-70%, etc. It would help people know how they're doing and what the ratings represent.