Not A Review: The Erland And Crawford Ratings Explained
Al Crawford
I've had a number of enquiries about the two rating schemes I use in my
reviews. To hopefully sate the curiosity of those who've been reading my
reviews and been puzzled by the ratings I give, I'm posting this
explanation of both schemes. It'll also be made available for FTP from Dave
Datta's music archives at ftp.uwp.edu (131.210.1.4) so that in future if
anyone asks I can point them there :-)
Now I'm not about to expend valuable energy actually writing something
*new* here, so I'll use the two articles I have hanging around that I
usually fob off people who ask with. The first is a response to some mail.
I hope Jay Bromley doesn't mind me including two lines of private e-mail in
this posting :-) The second article originally appeared tacked onto the end
of what must have been, judging by the number of people who actually read
enough to find it, a very dull review I posted back in April.
(r.m.i. = rec.music.industrial, BTW)
----
The Erland Rating Explained
---------------------------
And lo, ja...@juliet.caltech.edu (Bromley, Jay W.) spake unto the masses
saying:
>
> Hi, just curious, what exactly is the Erland rating system?
Confusing :-)
> or is it in the FAQ?
Nope, it's not an r.m.i. exclusive thing, I use it in *all* my reviews, be
they r.m.i. relevant or not.
I've had a few pieces of mail about the rating system so it's time for an
explanation. Those to whom I've mailed explanations previously might notice
that this is different from the one I mailed them. This is to be expected,
since the Erland Rating has a certain undefinable quality :-) Well, that's
my excuse anyway.
The Erland rating system was first used in rec.music.misc a year or three
ago. It was invented by Erland Sommarskog, hence the name. At the time, I
thought it would be a good idea to try and arrive at some sort of ratings
standard system, and Erland's system seemed as good as any, so I started
using it. Needless to say, nobody else on the planet seems to use it
anymore, not even Erland, and it's probably the least standard standard
ever. Nonetheless...
The Erland Rating is on a -5 to +5 scale. The bottom end is -5, the top end
is +5. The scale *isn't* linear though. Nor is it exponential. It can be
roughly summarised as...
+5 - Probably the all-time greatest album in the history of the universe.
+4 - truly an all-time classic, one to buy and cherish forever.
+3 - a very good album indeed, and highly recommended.
+2 - a good album but with limited appeal or an album that has wider
appeal but is slightly less good.
+1 - a good album. Pleasant to listen to and certainly worth a listen.
+0 - competent but unoutstanding. It's not a *bad* record as such, it just
isn't particularly *good* either
-1 - an album you should have reservations about. It's not *really* bad, in
fact you'll probably decide to keep it, but you probably won't play it
that much.
-2 - the sort of album which, unless you are completely fanatical about the
artist concerned you probably shouldn't buy.
-3 - the sort of album which, even if you are completely fanatical about
the artist concerned you probably shouldn't buy.
-4 - really a very bad album indeed. Connosieurs of really bad albums
should buy this.
-5 - An album so bad you are unlikely ever to hear its like again.
Over the sum total of my reviewing I've dealt with about 85 releases. The
lowest I've ever given is -2, the highest +3. The extremes of the scale
will probably never be reached. My bias towards higher ratings I put down
to the fact that I *buy* my albums with my own money rather than getting
them as free promos, so I'm not likely to put out money on something I'm
going to dislike. Actually, the stuff at the very bottom of the scale is
probably worth keeping for the entertainment value.
The same scale applies for singles, although I don't generally pay the same
amount of attention to them - if I review an album, chances are I'll have
listened to it at least 4 times and probably more than 10 times before
writing a review, whereas I'll typically only listen to a single 4 times at
most before posting a review.
----
So, that's the Erland Rating explained. Now for the Crawford Rating. This
rating scheme is a great embarrassment to me. After you've read this
explanation, which was initially written in response to criticism of my use
of the Erland Rating earlier this year, I'll point out why...
-----
The Crawford Rating
-------------------
It has recently been brought to my attention in rec.music.industrial that
many people are unhappy with the Erland Scale, and consider that an eleven
point -5 to +5 ever-so-slightly-non-linear scale is too complex, and would
prefer something involving only a few words, preferably ones without too
many letters in them and which can easily be written in coloured crayon.
With this in mind, I spent the weekend working away on the Assurdo
Softechnologies AI and with its help have now devised a new scheme which I
feel is a definite improvement.
The new Crawford Rating scheme attempts to be rather more comprehensive
than the Erland Rating and recognises the fact that splitting all albums
everywhere into just eleven groups is rather subjective. In fact, for the
vast majority of albums, the Crawford Rating scheme gives a *completely
objective rating*, independent of the reviewers biases and surely a first
in music reviewing.
To arrive at a Crawford rating, we start out by rating the album in thirty
separate areas, ranging from originality and content through to BPM, cover
art, hairstyles, whether the artist has ever appeared on MTV and so on. To
give a more finely graded scale than the Erland rating, each of these
ratings is completely open-ended. Once this has been done, these thirty
numbers are used as coefficients in a system of second order differential
equations (the nature of which is closely guarded), which must then be
solved.
Solving this system of equations yields three numbers. These three numbers
are then used to lookup tables in The Big Book Of Mysterious Ancient Wisdom
which convert each number into a word, according to its numerological
value. If an album has been objectively reviewed, the first word should
always be a colour, the second a fruit or vegetable and the third a breed
of dog. Unfortunately, the size of these tables and the occult nature of
The Big Book Of Mysterious Ancient Wisdom mean that it isn't actually
possible to tell you what the rating *means*, but it's usually possible to
guess by looking at the Erland Rating.
Crawford Rating: Vermilion/Avocado/Schnauzer
----
So there you have it. The Crawford Rating is, well, it's a rather weak
joke. I've continued putting Crawford Ratings on reviews purely through
momentum well after the joke's sell-by date and will continue to do so just
to be awkward but the most important thing about the Crawford Rating is
that it has *no meaning whatsoever* and is intended by its very uselessness
to throw the Erland Rating into stark contrast. Unless, of course, you have
a copy of The Big Book Of Mysterious Ancient Wisdom, in which case...
--
Al Crawford - aw...@dcs.ed.ac.uk
Department Of Computer Science, The University of Edinburgh
Rm 1410, JCMB, Kings Buildings, Mayfield Rd, EDINBURGH, EH9 3JZ, Scotland
Tel: +44 (0) 31 650 5165 Fax: +44 (0) 31 667 7209