Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

My experience becoming an MCAD

1 view
Skip to first unread message

David Kavanagh

unread,
Aug 31, 2005, 1:45:04 PM8/31/05
to
I discovered this newsgroup a week ago and found it contained some helpful
information on preparing for MCAD/MCSD certification. I'd like to repay the
favour by posting my experience of passing the first 3 exams, on the way to
being an MCSD.
First a bit of background on myself: I've been developing software
commercially for just over 10 years. Mostly VB, some Java, C++ and C#. Most
of my work has been for the .Net platform for the past couple of years.
Increasingly clients were asking if I had any MS certifications so I decided
I would go ahead and attempt to get some.
Since Microsoft were doing a Free Second Shot deal, I decided that I might
as well sit the exams now (before studying) with the expectation that I
would fail. That would give me a good idea of what I would be tested on, and
I could then brush up on these areas and retake the exam for free a couple
of weeks later. Since this offer finished at the end of August, and my
nearest test center was almost fully booked, I was forced to squeeze the
three exams into 2 days - something I would definitely not recommend.
My results were:
305 VB.Net and Web Apps 867
306 VB.Net and Windows Apps 1000
310 VB.Net and XML Apps 946
Obviously I am very pleased with these results but also somewhat concerned.
As a typically "real world" developer, I know certain areas of .Net very
well, but there are other areas, such as Cryptography, that I have never
even looked at. However, my scores indicate that even in those unknown
areas, I must have generally got the answers correct. Surely, this should
not be possible?
This is the approach I took to all of the exams:
1. I made a fairly quick pass through every question in the exam, lasting
around 30 minutes of so. If the question was on a subject I knew, and was
not too complex, I immediately chose what I believed to be the correct
answer. If I was not familiar with the topic, or just not confident in my
answer, then I quickly made a best guess based mostly on instinct, and
marked the question for review.
2. I now knew I had answered all of the questions that I definitely knew the
answers to, and had at least made a guess at all of the others. I could now
relax. I had plenty of time left so I could now concentrate on those harder
question that I had marked for review.
3. I took each marked question in turn and followed the question answering
method that I've detailed later in this message to make my best guess.
4. I took one final look at each of the questions. Here I was not attempting
to work out if my answers were correct or not. Instead I focused on the text
of the question and the answer. Does the question really say what I think it
says or have I misread it under the pressure of the exam. Does my selected
answer say what I think it says? Pay particular attention to syntax and
variable names - sometimes it hard to notice that a variable such as
daGeneral has been used instead of the correct dsGeneral, for example.
5. Click finish and hope for the best!
My method for answering the more difficult questions was as follows:
1. Before taking in any of the detail of the question, just get a rough idea
of what it is asking. I found that some of the long detailed questions were
just there to confuse you with unnecessary detail. A couple of times I read
a long description of a particular problem, and the solution to be used,
which would then be followed by a one line question such as "What is
authentication?".
2. Take a detailed look at the question, looking out for clues. For example,
any mention of firewalls, IE5, SQL Server 6.5 are really great clues as to
the correct answer. These factors have probably been mentioned for a reason
and so you should consider how they may effect your choice of answer.
3. If the correct answer is still not obvious, then consider which of the
answers are not correct. Often you will be able to immediately dismiss at
least half of the possible answers as being totally irrelevent, or obviously
incorrect syntax. If you have to choose between example code, pay particular
attention to where each answer differs from the others as this can often
help you spot the error in a particular answer even if you are not
particularly familiar with the syntax.
4. Look out for clues in other questions. There was one question for which I
needed to choose the exact syntax for something I had only used a few times.
Fortunately, a later question actually contained a similar piece of code.
5. If you still can't tell which of the remaining answers is correct, then
go with your gut feeling. You may have seen something similar before, but
don't remember it. Alternatively, one may look more correct than the other
because it uses a more ".Net way of doing things".
I'd just like to finish this off with my opinion on those study guide books
that you can buy. I've noticed much of the discussion on this group is on
whether one is better than the other. My personal recommendation is that you
avoid them. I had a quick look though a couple of these books (both highly
recommended on this newsgroup) and found that although well-written, they
didn't cover the material in anything like the detail required for the exam.
To be fair, if you read these books from cover to cover, you probably would
pass, but I suspect you wouldn't get a good pass, and you certainly wouldn't
be particularly effective as a developer. A few people on here advocate
using MSDN. This is an excellent resource when developing, but for the most
part I'd say it is too detailed for study. It's main purpose is to explain
the intricate details of a particular technology or function call. Generally
the bigger picture is harder to find, and this is where the exams are
focussed. Most of what I know about VB.Net (but not application design)
comes from "Progamming Microsoft Visual Basic.Net" by Francesco Baleno. Over
the past couple of years I've probably read most of this book and I'm
convinced that any VB.Net certification candidate who reads this from cover
to cover will not only pass the exams with an excellent score, but also
become a much better programmer.
Sorry this message has grown to be so long, but I hope somebody may find it
helpful.
Regards,
David

Geoff

unread,
Aug 31, 2005, 8:16:12 PM8/31/05
to

Thanks for that explanation it was very helpful...just began studying
myself :o)

--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.344 / Virus Database: 267.10.17/85 - Release Date:
8/30/2005


serge calderara

unread,
Sep 1, 2005, 1:36:02 AM9/1/05
to
HI,

I am actually peparing to my 2 lasat exam for beeing MCAD and the one I am
studying now is the Web application part. I already got my Windows
application cert quite easily because I am working all days on that type of
application for customer so it helps me a lot.

For the exam that I preparing now, the Web application with ASP.net, its a
bit hard for me o start because I have never been involved yet in such real
ASP project.

So I was wondering if you could give me a tip in order to start studying
easily and in a correct way. I give myself one month delay to succeed for
this exam

How should I approach this study ? should I build a real ASP application of
my choice as a sample project ?

THnaks for your experience recommandation
regards
serge

David Kavanagh

unread,
Sep 1, 2005, 4:00:24 AM9/1/05
to
Hi Serge,

I found the Web Applications exam to be the most difficult. I think this was
probably because I come from a similar background to you. I mostly do
Windows apps, together with a few backend services such as Windows Services
and Remoting. Although I have done Web front-ends from time to time, I
always have a book open next to me while I'm doing them.

I would begin by ensuring you are very familiar with all the web controls -
the methods, properties and events of each one, and exactly how they are
included in an ASP.Net page. There were lots of questions on this, and I
often struggled on the details of these. There were also a lot of question
on accessing remote Web Services, so make sure you know exactly how to do
this. Lastly, there was a few giveaway questions on Deployment. These are
great because the topic is pretty simple and once you know it, you can
answer the questions easily. Of course there were also plenty of ADO.Net
questions, mostly related to retrieving a dataset from a web service,
updating it and then sending back the changes.

One thing I did notice on this exam was that many of the questions were
extremely long, much longer than any of the other exams. I found it
difficult not to be intimidated by the amount of detail given in a question
but the truth is that often this was just 'scene setting' and did not
actually impact the answer.

A good approach would probably be to build a simple web service that returns
a dataset in response to a database query. Then write a shared assembly
which you should strongly name and install in the GAC. This shared assembly
should include all the logic to access the web service and retrieve the
dataset. Then write a web front end with lots of data bound controls to
display the contents of the dataset (preferably hand-coding all of the html
rather than dragging the controls onto the page). Add some validation to
individual controls and a facility to send back the database changes. Build
a deployment project for the web front-end and the shared assembly.

Regards,
David.

David Kavanagh

unread,
Sep 1, 2005, 4:05:48 AM9/1/05
to
One last suggestion: Make sure you know when to use to use ViewState and
when not to - I think I saw 2 questions on this. Also know the differences
between the various caching techniques and when to use application caching
and when to use session caching - this often arises during lengthy "case
study" type questions.

serge calderara

unread,
Sep 1, 2005, 9:51:10 AM9/1/05
to
sorry its me again,

I have post one topic here about a small report analysis project on which I
do not really know how to approach it fro my final user.
Would it be possible for you to check my post and may be give me an idea of
solution ?

The post is named : Help me too choose
The 3rd topic below yours

Thnaks agin

David Kavanagh

unread,
Sep 1, 2005, 10:27:52 AM9/1/05
to
There is a lot of overlap with the Windows Apps exam, particularly with
regard to ADO.Net and security, so once you have successfully passed that
one, I think it is fairly easy to pass the Web Apps exam. One month of study
of a couple of hours per day should be plenty to pass but, depending upon
your experience, may not be enough to get a good score. I don't know whether
exployers look at the score or not, so I don't know how important that is.
I'd suggest you plan for one month's study and if you do not feel confident
by the time the month is up, just postpone the exam for a couple of weeks.

Regards,
David.

Wor Tony

unread,
Sep 1, 2005, 1:53:56 PM9/1/05
to

"David Kavanagh" <david.k...@contension.com> wrote in message
news:2I6dnSGaRr_...@giganews.com...

> your experience, may not be enough to get a good score. I don't know
> whether
> exployers look at the score or not, so I don't know how important that is.
> I'd suggest you plan for one month's study and if you do not feel
> confident
> by the time the month is up, just postpone the exam for a couple of weeks.
>
David
I think there is only a concept of passing/failing. A higher score for one
candidate
does not indicate a "better" pass. I've read this on the MS web site
somewhere
but I can't find it at the moment. Also, your on-line transcript shows only
the exams
you have passed and not scores.

AP


David Kavanagh

unread,
Sep 1, 2005, 3:28:47 PM9/1/05
to
I think that is true but some employers now believe that the MS Certs are
too easy to pass and so are looking at your score. I know when I did mine at
a Pearson Vue test center, I was given a printed transcript which included
my actual score (broken down into different categories). This transcript
also included an authentication code that an employer can type in to the
Pearson Vue website to confirm that it is genuine. The Microsoft transcripts
though only include a pass/failure, presumably because Microsoft think that
a pass should be sufficient.

In an earlier discussion on this newsgroup, a number of people were debating
whether certification was good for your career. I noticed that a couple of
people mentioned that, as recruiters, they would regard it as a negative
point if someone just scraped a pass and they would have been better off not
mentioning it at all.

Regards,
David.

Victor Guillen

unread,
Sep 1, 2005, 3:44:54 PM9/1/05
to
I don't agree at all, because a Certification exam at first glance validates
the basic skills and habilities of a determined subject.
If a job candidate is a real developer (either windows or web) the employer
may require demos. In my case, two prospective employers required demos from
me, and the best way to show them, is give the URLS of the websites I have
developed in other companies, and also, a demo of my coding style.
If a job candidate can give those demos, it would be more than sufficient to
demonstrate and validate what the resume, interview and the papers say.

I took recently the 70-315 exam, and I didn't pass it with shining scores,
however, I have a live website running under ASP.NET for one of my
customers, and at the same time, I'm migrating (aka rewriting) the ASP
project into ASP.NET, and I can defend my point, that if a developer can
demonstrate what he claims to know with live websites, or software projects
developed, who the hell would care about your score?

"David Kavanagh" <david.k...@contension.com> wrote in message

news:Y6GdnR2cKrN...@giganews.com...

David Kavanagh

unread,
Sep 1, 2005, 3:59:13 PM9/1/05
to
Good point. No matter how good your score is, it is never going to be a
replacement for a proven track record. It may however, help you to get the
interview in the first place. At that point the certs become pretty
irrelevent as the interviewer will access your technical (and non-technical)
skills for themselves - often by looking at your past work.

I guess the point that I was making is that if there is a trend for at least
some employers to look at your score, you should attempt to make life easier
for yourself by aiming for the highest possible score rather than just a
pass in the shortest time possible. Personally I believe that once you have
a track record, the certification should not matter too much, except perhaps
to show that you have a good allround technical understanding, and that you
have kept your skills current.

Regards,
David.

Victor Guillen

unread,
Sep 1, 2005, 4:02:43 PM9/1/05
to
Yes, your point is valid also, but in the end (for the people who passes
with braindumps) if they don't have experience to show once they're on the
computer, the employer will realize about that this candidate is not a true
developer

"David Kavanagh" <david.k...@contension.com> wrote in message

news:osKdnQqKd_6...@giganews.com...

David Kavanagh

unread,
Sep 1, 2005, 4:13:57 PM9/1/05
to
Well I certainly would never advocate using a braindump. This would be like
lying in your interview. It may get you the job but when it becomes clear
that you do not have the skills, you'll get fired.

On the subject of braindumps, I never even knew they existed, or at least
wouldn't have believed they contained genuine questions, if it wasn't for
the number of people complaining about them on this newsgroup. Every time
they are mentioned, this becomes a great advert for them!

David.

Victor Guillen

unread,
Sep 1, 2005, 4:18:01 PM9/1/05
to
Yes, that's true. But as I said before, if someone is not a true developer
or true IT Professional, the people around him/her will realize it

"David Kavanagh" <david.k...@contension.com> wrote in message

news:fMKdnZ2dnZ15IRDQnZ2dn...@giganews.com...

Sean

unread,
Sep 2, 2005, 12:13:02 AM9/2/05
to
I got extreemly lucky with my web apps exam. 3 questions were pure ADO.Net
and most of the others were questions that most .Net windows developers could
have answered without knowing much ASP. The real winner for me was an obsure
XML question which just happend to something I ran into by mistake the night
before the test while looking for something else not even related to my
studies for the exam. Obsure questions I understand they give more points for.

So as it turns out I only studied for Web apps for about a week (full time)
and passed the first time which has not been the case for any of the other
exams given I dont have any real life experience. The windows exam I studied
a TON (6 full times weeks I think) so much of what I learned there spilled
over to the web apps test. Pure luck on the questions I had.

Sean

unread,
Sep 2, 2005, 12:16:02 AM9/2/05
to
By braindumps I assume you mean litterally actual questions from the real
test bank. I have found legal mock questions extreemly helpful in pointing
out a deeper understanding of material.

Someone green (like me) can understand what something IS, but its also
helpful to understand what something IS NOT. Surprisingly that is not as
intutive as one might think.

Sean

unread,
Sep 2, 2005, 12:36:01 AM9/2/05
to
That last comment about recuiters doesnt really make much sense to me.
Sure a higher test score is better if they ask for it but taking and passing
the exams are clearly better then not doing it at all given same experience
level in question.

Furthermore, firms use their employees as their "resume" for clients and the
more certified people you have on your staff the better. So if I am looking
to fill a position that requires X years of experience I would perfer
canidates at that level to be certified rather then not regardless of the
score.

I am a HUGE advocate of cerifications but in no way do I think they replace
experience. They just give candidates of same given experience an advantage
(or at least should).

The question I have for naysayers of certs is this "give the same level of
experience what should a canidate of that level do that shows a competive
edge against others of the same level"?

Perhaps why there are so many naysayers of certs is becuase they are being
given positions higher up then what reflects their job experience and if that
is what has been going on I can understand the resentment. I do certs just to
compete among people in my same level of experience.

David Kavanagh

unread,
Sep 2, 2005, 3:51:02 AM9/2/05
to
My comments regarding the potential negative effects of just scraping a pass
are not based on any personal experience - I was simply repeating what had
been written in this group by others who claim to have been in a recruiting
position (use Google to look at the archive). Personally I think that anyone
who has taken the time to get certified should get full credit for taking
their career seriously and educating themselves. In the past, I have
recruited a number of programmers, although none have ever been certified.
If I had seen two candidates which had identical experience, but one had at
least one cert (no matter what the score), then they would have had the
advantage. You also make a good point regarding companies wanting as many
certified employees as possible.

Unfortunately, certifications are not viewed very highly in some
organizations. I suspect this might be because those doing the recruiting
often do not have them and so either do not know how difficult they are, or
feel insecure. I've noticed similar behaviour with those with college
degrees - if they have one, they insist that everyone has one, if they
don't, they think they are worthless.

David.

David Kavanagh

unread,
Sep 2, 2005, 4:00:25 AM9/2/05
to
Practice questions are a good idea. Not only are they a great way to learn
the subject, but they are a great way to get used to how the questions are
phrased. In our everyday work we are rarely given a quick overview of a
problem and asked to pick the best choice from four options, within a couple
of minutes. This is a skill that should be practiced before the exam. I'm
just started looking at the 70-300 exam, and I think this one is best
approached by doing as many practice questions as possible.

Although I've never seen them, I believe that when people refer to
braindumps, they are not referring to practice questions, but to the actual
questions that have somehow been smuggled out of the exam and published,
together with supposedly correct answers. Whether they are real or not I
wouldn't know, but if they are then this is cheating.

David

DST

unread,
Sep 2, 2005, 4:01:57 AM9/2/05
to
David,

First of all: congrats!

I find it reassuring that an experienced developper can pass the exams
with the knowledge he acquired on the job.

I confirm the observations you made about the exams. You have to know
your stuff, but there is also a question-solving strategy you must
follow to be really succesful on your exam. Your post details this
strategy far better than I ever could.

70-300 is a case study exam. The case studies (3 in total) are several
screens long, and you have around 10 questions.
This is my strategy:
1) Read the case study twice to understand the requirements before you
start reading the questions.
2) Read and answer each question while refering back to the relevant
portions of the case study (you'll have a global understanding of it
after having it read twice and remember where to find the details
relevant to the question)
3) Some of the questions will have more than one technical correct
answer. You'll have to take the requirements into account to select the
best answer. For example, the case study could mention that you don't
have the budget to buy a new server. The you know that installing your
web application on a new IIS server is not the correct answer.
4) Mark questions that give you a problem for review and move on to the
next question
5) Reread the full case study and review your marked questions.

If found that a lot of questions could be answered by careful reading
of the case study, without the use of any other knowledge you might
have.

And you don't need to know MFS to pass this exam. I had a few
methodology questions, but they were so general that common sense and
experience with multi-programmer development projects was enough to
know what to answer.

I've also applied this strategy with success on exam 70-298.

DST

David Kavanagh

unread,
Sep 2, 2005, 4:20:02 AM9/2/05
to
Many thanks for your suggestions. I'm probably doing this exam next but due
to work commitments, I don't know when it's likely to be.

I do have a couple of questions:
Did you prepare specifically for this exam, or was your on-the-job
experience sufficient?
Did you find time to be a problem in this exam? I've seen this mentioned a
couple of times.

Thanks again,
David.

DST

unread,
Sep 2, 2005, 5:01:43 AM9/2/05
to
I've read the MS Press and Exam Cram books for this exam.
It was an interesting read, but I clearly remember being disappointed
after the exam because I was convinced I could have succesfully taken
the exam without any preparation. I'm not using MSF and it's very
unlikely I'll use in in the future, so the study time was kinda
"wasted" (it's never wasted when you learn something new in IT, but I
could have spend my time stuying something more useful for me).

Time was not a problem (I think it took me about one hour). But be
aware that I get a "Time extension for non-native English speaker
living in non-English-speaking country" (30 minutes).

I've been writing software for 25 years now (15 years professionaly),
I've worked on small one-man projects and large projects, I've been a
project manager of a team of 10 developpers, I've done projects were
I've only had to program and others were I had the full life cycle,
including the installation and maintenance of servers (IIS, SQL), ...
What I mean is: my experience is broad and diverse, and this helped me
a lot with exam 70-300.

Good luck on your exam!

DST

Victor Guillen

unread,
Sep 2, 2005, 9:19:06 AM9/2/05
to
I believe that now we're getting to the point:

If you have experience and can provide proof of your work, your
certification gets an individual value at your favor, even if the recruiter
is reluctant on certifications.

If a developer has experience, it doesn't have to be hard to write the code
required to solve a business problem and if the developer has certification,
this gives more credibility and respect for himself/herself and coworkers

"DST" <didier....@gmail.com> wrote in message
news:1125648117.7...@o13g2000cwo.googlegroups.com...

Sean

unread,
Sep 3, 2005, 9:16:01 PM9/3/05
to
I am curious if anyone has ever had a recruiter actually frown on someone who
has a certification. This all sounds a lot like myth to me and I would
personally like to know if anyone here has had that happen to them. I know
that some companies REQUIRE one to be certified to even be looked at and that
is why I find the reverse view from recruiters a bit hard to believe.
0 new messages