As I've been updating my sites, I've been updating my Recommended statements in reviews. You'll see them with a date or dates, as in:
- Recommended (2022) — The current year, so only one date appears in my current iteratation. I'll probably be updating this to say "2022 to present" to get things down to just the following two categories:
- Recommended (2014 to 2018) — A range of years during which I recommended the product.
- Recommended (2012 to present) — Another range of years during which I've recommended the product, but open ended.
The ranges are important to note. If the date range extends through to the present, that means I'm still recommending the product here in 2022. But if the date range ends, it means that I decided that something happened that no longer makes it something I'd currently recommend. With cameras, that tends to be that a newer model that came out or that competitive products just got far better. With lenses, it can be that cameras that came out with higher resolution that start to show issues with the lens, or that better lens choices in that focal length came out, or perhaps even that better choices at a lower price came out.
I update my recommendations once a year, generally during the summer months, but it takes me awhile to go through every site and page to make sure that what I'm recommending is still accurate to my present thinking. I've done that already this year for zsystemuser.com, will be completing that for sansmirror.com probably later this week, and will then work on doing the same for dslrbodies.com.
Likewise, I check my star ratings and update them if needed. However, I want to remind everyone that my starts are not a linear scale; the stars are assigned based upon expectations:
- 1 star = don't even consider the product, as it's far below what your expectations of it might be for a product of its definition and price.
- 2 stars = the product comes in below what most people would expect for its specifications and price. For optics, for instance, that means that a lens won't perform at the level most people would expect at that price.
- 3 stars = the product basically meets the expectations you'd have for it given its specifications and price. A lot of folk seem to think that 3 stars means I wouldn't recommend the product, but it's just the opposite: I think the product will probably perform in this aspect as you'd expect.
- 4 stars = the product is clearly "above average" and more than you'd expect in this category. I'm not sure I've ever rated something at 4 stars and not recommended it.
- 5 stars = is reserved for exceptional products that are far beyond what you'd expect in the category.
I see people all the time say "but Thom only rated [product] at three stars out of five." Yeah, but that means that the product is going to meet your expectations ;~). Clearly, very few products will ever get a five star rating from me for any category I rate. I'm a tough grader and I grade on an expectation curve.
_______________________
I should point out that others who claim to use linear ratings probably aren't. In statistics classes you'll learn that the ability to properly assign a value on a linear scale for something is something that people have a difficult time doing consistently, let alone well. I'm amused by dpreview's apparent 100-point scale ("91% overall score"). Statisticians would say they'd be hard-pressed to make consistent evaluations on a 7-point scale. But we all like big numbers, don't we? ;~)
