A couple of comments from a former SSD firmware engineer:
SMART is an industry standard in the protocol that is used to retrieve the statistics from the drive. There are some SMART attributes that are fairly standardized, and then each vendor may have vendor specific attributes that report whatever information those vendors think is useful. In general, if you have SMART attributes that are indicating bad things, then yes, you should get the data off that drive immediately and then stop using it. Most failing drives will start failing SMART attributes somewhat before the drive catastrophically fails. Probably the key one to look at would be the number of grown bad blocks on the drive - I would be worried if it were more than a few.
Overall I believe that SSDs are on average more reliable than mechanical spin drives. I have seen data that claims they are about twice as reliable as HDDs, but that is from my former employer, so maybe take that with a grain of salt. On the other hand, there are no moving parts and no issues with altitude, dust, shock, vibration, humidity, or head alignment, so it seems plausible.
The SSDs in the Macs that @braumeister mentioned in post #18 almost certainly support SMART. What Apple probably means is that they don't want to encourage their users to rely on SMART data. My guess, having worked with Apple, is that they don't like any standard that they don't fully control. I bet if you pulled the SSD from one of those machines, stuck it in a Windows box with a SMART utility, you could read the SMART data just fine. Remember, SSDs are OS and hardware agnostic.
I'm going to disagree a little with what njhowie said about SSDs getting full in post #27. First, because of how SSDs write data, they don't need to move files around the way traditional HDDs do as they get close to full. Second, even when an SSD reports that it is 100% full, there is still spare space (often called overprovisioning in the literature) for it to be able to operate at full speed. Finally, when SSDs are life tested, they are written until they are 100% full, so they are engineered and spec'd and manufactured to last the full waranteed lifetime of the drive (usually expressed in terabytes written, or TBW) when they are full.
@gauss, you should have no worries with the situation you describe in post #28. First, for all of the reasons in the previous paragraph. Second, the way SSDs work under the hood is that there is little to no correlation between the host logical block address and the physical disk address where your data is stored. To maximize longevity, the data is written in different places, so when the host writes to LBA #137 the first time, it may be stored on flash chip #3 block #6 page #487, but the second time the host writes to LBA #137, it may be stored on flash chip #8 block #9 page #2. IOW, there is absolutely zero correlation between your host partition and where the data is stored on the drive.
Further, the SSD keeps track of how many times it writes to each block and writes in such a way as to use them up equally. That being said, unless you're a ridiculous power user, if you do the math you'll find that you'll probably wear out your SSD after decades of average use.