I took the
test early in the week. Passed, thank goodness. I had big plans for a really fancy and information-filled post. Instead you will get this (ha!). I do have about 15 minutes before dinner, and I type pretty fast, so you'll have a big thing to read (if you care to sort through the random thoughts).
Reading HF was of course the starting point. Front to back, doing everything (exersizes - no cheating - really thinking). Then again I went through the book, but this time skipping the concepts, which I of course 'got' the first time. That second pass was an attempt at the memorization. Unfortunately, this reading was dragged-out over a long time, and my brain tends to loose data after time. Soooo, I had to figure out a way to jump back in after a month of not looking at it. I photocopied just the chapter mock tests and randomized by cutting up all of the questions (I wanted a 'final mock test' but didn't want to waste the real one). I got 77% to 80% on every 'batch' of questions I took. Again, no cheating at all. I took the free tests on jdiscuss and got 67% to 77%. After each of those tests (and the cut-up chapter tests) I analyzed every question, and got to the bottom of it (usually by looking it up in the specs, occasionally running a few lines of code - but that only proves the way my container implemented it). Anyway, I thought I was ready to take the final mock.
The HF final mock answers had been typed into a spreadsheet a long while back, before I knew anything about the topic. See, I was planning ahead to get two runs at the final mock test. It goes like this... take the mock test, get the grade, but you don't see which ones you got right or wrong, just your score. That way, you can get another shot at it without being as 'tainted'. Well, I got like 51% the first time and 54% the next time. I got a lot of the same ones wrong, but there were about 8 questions that I swapped (got right the first time, wrong the next and the other way around). So I concluded I had taken the test too early.
Although I didn't like the idea of spending MORE money with Whizlabs than I had spent on the frickin' book (not 'cuz the book was too expensive, it's just that I KNEW I got SO much more out of the book than I would ever get out of these tests, it just sort of pissed me off that they got more cash than Kathy and Bert). Anyway, I bought the tests from Whizlabs. But let me tell you, I don't think it really helped. Those tests seem engineered to give you a 70. They ask you some questions that when you get right down to it (eliminate the chaff), it's asking you a plain-jane question, but there are other questions that are about impossible to get right (like having A, B, C, D, E, F, G, each being random true and false questions that don't relate to each other at all... just straight quotes out of the spec). By the way, you will be glad to know that NOTHING EVEN CLOSE to that kind of crap is on the real exam, which, by the way, I am not planning to disclose until the very end of this rant; I'm not very proud of my score, but it's going to be an HONEST answer.
About memorization, I made some 'flash
cards' for myself using powerpoint. I've shared them here
http://sengsation.freeprohost.com/SCWCD_Memorization.html One of my testing strategies was to use the time during the initial "survey" portion of the test to write down some of this memorization stuff. I did that, and I only looked at it once. I didn't even need to look at it, I just looked at it because I had written it down, and knew it was over there. So I guess what I'm saying is that the wrote memorization of the stuff that's on that link above wasn't valuable, but the PROCESS of making the flashcards WAS valuable. For instance when the inevitable question about listeners comes-up, you don't just have to know what the API is, you have to know when and why you'd use one over the other.
Okay, you've been through enough of my rant to hear that I passed with 18 points to spare. You have to know what a passing grade is to know what I got. The test seemed pretty 'easy' on the "IQ-TEST" scale; there wasn't a lot of "not", "except", double negatives to worry about. I got dinged too many times to count on those Whizlabs test with "select all such and such that are not true". The test had a few "NOT's", but only when it MADE SENSE to ask the question that way, not put in there just to screw you up.
My idea going into the real test was that whatever honest score you got on the HF mock test, you could add 20 points, and have yourself a pretty good estimate of what you'd end-up with. But the problem with my calculation is that only the people that get good scores seem to post them (imagine that). So when I came out with a 51 on the mock, 20 points would take me to 71... that's safe. But I started second guessing my formula... maybe it wasn't valid for low scores. I recalibrated to a 30% increase. That works out to about 20 if you get 70 on a mock test, but only 15 if you end-up around 50, like I did. So I was looking at a 65... that's too close for comfort (thus the purchase of the Whizlabs tests that I ended-up not really learning much from). Let me take that back. I think I 'might' have learned a few tricks that could have been sprung on me during the real test, but the test didn't spring that many tricks (that I could detect). What buying the tests did do for me was to give me things to look-up... to keep me going. I'd sort of run out of stuff to do with the book, and the lack of quality of free tests was starting to get on my nerves (many weren't up to 1.4, and so many crappy, unrealistic questions). The Whizlabs had an okay set of questions, but much higher on the "IQ test" side than the real test. And always a set that were of the "come-on!" variety (7 unrelated true-false in a single question "pick the 4 true ones").
A ringtail hawk landed in the tree right outside the window while I was taking the test. She couldn't have been more than about 15 feet from me! I stared at her for a long time, oh, TIME! I've got to concentrate on the test! It took me almost two hours to get through the test - first pass. There wasn't a lot of code to step through or anything, but I always try to think through everything, eliminate the wrongs, and check the rights, and see if they agree. If not, do some more thinking. But doing that, I had about 7 or 9 that I had 'big question marks' on (meaning I figured I had my 'final answer' on 60 of them). There were obviously five or so more that I didn't have a big question mark on that I should have! But the point here is that I didn't have much extra time. I flipped through those 7 or 9 and honestly I could have used more time thinking about them. 'Everyone' says time shouldn't be a problem, and maybe I would have taken extra time to move correct answers to wrong ones. Right, for the majority of the questions, I knew the answer and I moved on. A few of them, ... oops FAJITAS! I gotta go. Good luck to all that follow.
--Dale--
Edit: Took the period off of the link
[ March 01, 2005: Message edited by: Dale Seng ]
