VB or C?

[RobinS]

I'd love to know more about how these figures break down. I suspect
there are plenty of people who develop software as *part* of their jobs
who use VB, but I suspect that if you restricted the survey to those
who are full-time developers, the story would be significantly
different.

Do you mean you think the numbers would be much lower?

I'd like to know where they *got* those figures, because I certainly
haven't seen it reflected in the job market in the SF Bay Area.

I suspect there are many people like my father, who put
together VB programs as part of a job which would never normally be
called a development role, but who would count himself as a VB
developer. I suspect things would also change if you looked at people
developing "shrink-wrap" applications vs those developing "in-house"
business applications.

Do you suspect "shink-wrap" are done in C#, and in-house are in VB? That
would be my expectation.

Now, my paragraph above was mostly thinking of VB6, rather than VB.NET.
My guess (and all of these suggestions are just guesses and suspicions)
is that VB.NET has made more of an impact on the full-time developer
than VB6 did - although it would be interesting to see how many people
went from, say, Java to VB.NET compared with those going from VB6 to
C#. (I've heard of various people doing the latter, but none doing the
former. I'm sure they exist, I just haven't heard about them.)

I think there was a lot of VB6 usage by people writing in-house business
apps for small divisions. That's what my last two jobs were. The apps are
still running, and there is no interest in converting them to .Net because
they run just fine. It was easier for people to take up VB6 than VB.Net, so
I think there are a lot more "fly by the seat of your pants" applications
written in VB6 than anybody knows of.

None of this prevents VB.NET from being a perfectly viable language, of
course. I happen to prefer C# in various ways, but that's a different
matter.

I use whichever one has the most interesting job, and there seem to be more
interesting jobs with C#, so that's the way I'm going these days. But I
like VB, too.

Robin S.

 

[Jon Skeet [C# MVP]]

Do you mean you think the numbers would be much lower?
Yes.

I'd like to know where they *got* those figures, because I certainly
haven't seen it reflected in the job market in the SF Bay Area.

Well, total number of employed people isn't necessarily reflected by
number of open jobs, of course - but I guess a lot of these jobs
wouldn't be advertised as VB jobs in themselves.

Do you suspect "shink-wrap" are done in C#, and in-house are in VB? That
would be my expectation.

Yes - although I suspect that actually native code is still just about
ahead of both. Complete guess though.

I think there was a lot of VB6 usage by people writing in-house business
apps for small divisions. That's what my last two jobs were. The apps are
still running, and there is no interest in converting them to .Net because
they run just fine. It was easier for people to take up VB6 than VB.Net, so
I think there are a lot more "fly by the seat of your pants" applications
written in VB6 than anybody knows of.

That's certainly the impression that's around in the industry - and the
one that serious VB developers have always been trying to get rid of,
understandably.

I use whichever one has the most interesting job, and there seem to be more
interesting jobs with C#, so that's the way I'm going these days. But I
like VB, too.

I don't like the fact that there's so much library within the language,
as it were. I like a reasonably small language which is then used with
a large library.

 

[Michael D. Ober]

dim A
dim B

A = "5"
B = 1

What is A + B? What is B + A? With Option Strict Off, A+B and B+A may or
may not return the same value. Sometimes you'll get the number 6,
othertimes you'll throw an exception. With Option Strict On, the compiler
won't allow the addition, or, even the initial variable declarations since
they don't specify the data type. I have actually had programs fail in VB 6
because of this particular coding error. Option Strict On simply won't
allow it, which is why I feel this should have been the out of the box
default.

Mike Ober.

 

[Scott M.]

With Option Strict On, the compiler won't allow the addition, or, even the

initial variable declarations since they don't specify the data type.

Actually, I think this particular issue is not allowed with Option Explicit
being Off. It's default value is On, however. MS finially figured out that
it was better to enforce Option Explicit from the start (after many years of
it being Off by default in VB <= 6. I just wish they used the same logic
with Option Strict.

 

[Scott M.]

1)

#Ok, one jobs search engine down, 435,000 to go

I really think we need to see the list of 435000 job sites.

I was being facesous, since 3 is clearly an insignificant number compared to
the many many job search sites out there, but here's enough to prove my
piont:

http://www.google.com/search?q="job+search"&sourceid=ie7&rls=com.microsoft:en-US&ie=utf8&oe=utf8
(Results 1 - 10 of about 21,800,000 for "job search". (0.14 seconds) )

2)

#Or, how about that Microsoft paid an "indepenent" consultant to alter
Wiki
#posted data with new *facts* disputing a recent report of the overall
#benefits of .NET vs. Java?

I would also really like to see a source for that.

Here you go:

http://www.google.com/search?q=micr...d=ie7&rls=com.microsoft:en-US&ie=utf8&oe=utf8
http://www.techcrunch.com/2007/01/25/microsofts-v-wikipedia-round-2

No it is not an opinion.

It is extremely strong statistical evidence.

No, it's anecdotal evidence.

Ancecdotal Evidence:

http://en.wikipedia.org/wiki/Anecdotal_evidence

Statistical Evidence:

http://www.stat.brown.edu/~jblume/slides/slides_ucinn.pdf


Now (for the last time for me), I will repeat that you your sample data is
limited (because of the redundancy in the data) and your sample frame is
skewed because it does not contain enough of the job search sources.

And, if you don't believe that:

http://en.wikipedia.org/wiki/Sampling_method

 

[Scott M.]

You don't want answers to your questions? Then why ask them?

 

[Scott M.]

Made up (your probability scenario above) or insufficient ones (flawed

I find it hard to take someone not capable of distinguishing between
no facts and insufficient facts serious.

The numbers are the facts.

The math is standard statistics.

Nothing manufactured.

Well, that's very interesting, since the link I posted earlier
(http://www.stat.brown.edu/~jblume/slides/slides_ucinn.pdf), written by
those uninformed folks at the little known Brown University seem to agree
with me and not with you about how to look at a sample and determine its
reliability.

Bye

 

[Tom Leylan]

I was being facesous, since 3 is clearly an insignificant number compared
to the many many job search sites out there, but here's enough to prove my
piont:

http://www.google.com/search?q="job+search"&sourceid=ie7&rls=com.microsoft:en-US&ie=utf8&oe=utf8
(Results 1 - 10 of about 21,800,000 for "job search". (0.14 seconds) )

You're doing yourself a huge disservice (I believe) when you post such
things. You have done a keyword search (look it up) and are trying to claim
the number returned is more than how many times it found that key word. It
is returning sites such as the Wall Street Journal and other newspapers,
bloggers and even newsgroup postings where people have used the words "job
search". They aren't job placement agencies.

http://www.google.com/search?q=micr...d=ie7&rls=com.microsoft:en-US&ie=utf8&oe=utf8
http://www.techcrunch.com/2007/01/25/microsofts-v-wikipedia-round-2

A quick check of the story reveals you have distorted (and I'll guess
purposefully) what occurred. There are claims being made and being
investigated which label a fact. The data in question was being challenged
and apparently nobody was addressing valid points about a bias in certain
articles favoring open source.

You wrote "here's enough to prove my piont (sic)" a point afterall that
suggests facts should be backed up with evidence and proceed to use the
keyword search result (of a single search engine) to prove it. How ironic.

 

[Tom Leylan]

I think we all want answers... do you think Dice can't produce accurate
reports on job opportunities? You seem to think that somebody else started
the thread. Do you think Science magazine didn't do their research on the
accuracy of polls and surveys?

 

[Tom Leylan]

Well, that's very interesting, since the link I posted earlier
(http://www.stat.brown.edu/~jblume/slides/slides_ucinn.pdf), written by
those uninformed folks at the little known Brown University seem to agree
with me and not with you about how to look at a sample and determine its
reliability.

As with so many of your tongue-in-cheek responses we can't tell when you are
being serious. The butter one, the attack on Wikipedia, change your mind
cite it but then site it as a source, etc. The paper you have referenced
was not written by "those uniformed folks... Brown University". Again
rather than stick to simple facts you distort them in some lame attempt to
seem more right. You are doing the equivalent of what you claim some
consultant did on Wikipedia, altering the facts to suit you.

The slides are the output of an Assistant Professor at Brown (he is also
co-founder of Analytical Edge) but not by "Brown University". It is
significant in it's own right but no more accurate to say "Harvard doesn't
believe in the Holocaust" if one of it's professors writes a paper disputing
it. Try to stick to facts when trying to persuade people to stick to facts.

When all is said and done you haven't answered one question. If the reports
of poisoned dog food have no statistical evidence of being factual are you
still buying those brands of dogfood? Or (because you will avoid the
questions and simply state you don't have a dog) would you buy the dogfood
if you did have a dog?

Would you have any hesitation to visit a city in foreign country where 120
tourists were killed in the last 12 months? In lieu of a study would you
say there is no evidence to prove you are in any danger? Yes of course you
would say that... now would you actually visit the place? How about 240
tourists killed in 6 months?

Wouldn't the available facts suggest you should personally eat the possibly
tainted dogfood while visiting that town?

 

[=?ISO-8859-1?Q?Arne_Vajh=F8j?=]

I was being facesous, since 3 is clearly an insignificant number compared to
the many many job search sites out there, but here's enough to prove my
piont:

http://www.google.com/search?q="job+search"&sourceid=ie7&rls=com.microsoft:en-US&ie=utf8&oe=utf8
(Results 1 - 10 of about 21,800,000 for "job search". (0.14 seconds) )

And you mean that google search say that there are 435000 job sites ?

That does not seem obvious to me.

The number 435000 does not appear when I search.

And I can not see any way of getting from 21800000 (actually
I get 23900000 but ...) to 435000.

Not in any way substantiating facts of what you said.

Here you go:

http://www.google.com/search?q=micr...d=ie7&rls=com.microsoft:en-US&ie=utf8&oe=utf8
http://www.techcrunch.com/2007/01/25/microsofts-v-wikipedia-round-2

If you bother reading the article then you would have read
that it was about office file format not about Java versus .NET !

Not in any way substantiating facts of what you said.

Arne

 

[Scott M.]

You're doing yourself a huge disservice (I believe) when you post such
things. You have done a keyword search (look it up) and are trying to
claim the number returned is more than how many times it found that key
word. It is returning sites such as the Wall Street Journal and other
newspapers, bloggers and even newsgroup postings where people have used
the words "job search". They aren't job placement agencies.



A quick check of the story reveals you have distorted (and I'll guess
purposefully) what occurred. There are claims being made and being
investigated which label a fact. The data in question was being
challenged and apparently nobody was addressing valid points about a bias
in certain articles favoring open source.

You wrote "here's enough to prove my piont (sic)" a point afterall that
suggests facts should be backed up with evidence and proceed to use the
keyword search result (of a single search engine) to prove it. How
ironic.

So, you'd say my sample frame was flawed then? How ironic. Not too mention
that I never presented this as undisputable fact, did I? And, based on your
reply, I guess you just proved my point!

 

[Scott M.]

And you mean that google search say that there are 435000 job sites ?

That does not seem obvious to me.

The number 435000 does not appear when I search.

And I can not see any way of getting from 21800000 (actually
I get 23900000 but ...) to 435000.

Not in any way substantiating facts of what you said.

See my reply to Tom on this.

If you bother reading the article then you would have read
that it was about office file format not about Java versus .NET !

Not in any way substantiating facts of what you said.

And this relates to the point that Wiki can be changed by anyone and
therefore I don't like it? You have now changed your tactics away from the
point at hand, did MS pay (or attempt to pay) to have Wiki altered?

And how does my being incorrect about the topic of what was changed (but not
my main point in bringing Wiki up) relate to your *statistics*?

I've noticed that those who change the subject in mid-thread away from thier
assertions to something that is not the topic being discussed and dwell on
insignificant things that do not advance their case have done so because
they don't have anything credible to offer on their original point.

 

[Scott M.]

I also noticed that you didn't comment on my other reply to you about what
statistical evidence actually is. As I just pointed out, it's probably
because you haven't found a way to dispute it, so you change the topic.

 

[Scott M.]

What does this have to do with you chastising me for giving you an answer to
a question you asked?

I think we all want answers... do you think Dice can't produce accurate
reports on job opportunities?

No, I don't.

You seem to think that somebody else started the thread.

Yes, Arne did.

Do you think Science magazine didn't do their research on the accuracy of
polls and surveys?

Did I say that?

Since you and Arne continue to respond with everything you can think of
EXCEPT for addressing what I have stated about Arne's post; it's clear to me
that the reason is because you have no credible way to prove that Arne's
sample frame is not a basis to conduct and come to any kind of accurate
statistical analysis. Forget the toast and Ginger Ale analogies, they were
just attempts to try to make you see how flawed Arne's assumptions were.

Bottom line (and if you are going to reply, it would be nice if you replied
about the whole point of this, rather than something that does not go to the
point), The sample frame provided and the outcome decried as a statistical
analysis and therefore credible are just not correct. I think the link YOU
posted proves that quite nicely when coupled with the Brown University .pdf
link I provided. I really would like to hear you try to dispute this,
simple (and relevant) set of facts.

 

[Scott M.]

I also noticed that you didn't comment on my other reply to you about what
statistical evidence actually is. As I just pointed out, it's probably
because you haven't found a way to dispute it, so you change the topic.

 

[Andy]

Please read, with your eyes open, what you are commenting on and I'd offer
some friendly advice that it's not really a good way to start a conversation
with an insult to someone you have never communicated with, who has never
said anythiing insulting about you.

I have been reading, thanks, but apperently you have not been. We've
posted back and forth a few times before my last reply.

With regards to being rude or insulting, it seems you started that
thrend in this thread (which I was following before I ever posted into
it).

At any rate, the problem you are having is with the 3:1 C# to vb.net
job availabity ratio. A quick survey of the three largest online job
posting sites IS a good sample, but for some reason you refuse to
acknowledge that.

Then when I point that out very clearly in my last post, you do
nothing to refute it, instead acting as if I have been rude in some
way. I do not feel anything in my previous post was rude, yet your
other replies to other people I do think have been. I guess if you
can't make your case, you say that the people who disagree with you
are 'bad people.'

 

[Scott M.]

LOL Tom! When a study or paper is presented by an employee of a company or
institution, and that study was done "on the clock", the study or paper
becomes a study "of" the company or institution. Your whole rant otherwise
gave me a good chuckle though.

The paper was written by an expert in the field from a pretty prestigous
university. If you can't tell when I'm being sarcastic and when I'm not
with such a simple thing like how qualified a professor at Brown is, you
have my sympathy about how hard it must be for you in daily conversations
you have with others.

BUT, more importantly, what in the paper agrees with my assertion that the
sample Arne used is flawed and his *probability problem* is just that a
problem for his case and what supports Arne's supposition that his case is,
in fact, based on a good sample frame and his math is viable?

You also want to critize me for citing a source YOU provided to help make my
point? Did I or did I not already acknowlege I was wrong about Wiki? Have
you ever been wrong about something and change your mind about it? If not,
you don't get a chance to learn and grow.

You and Arne are now just attacking anything you can (my sense of
sarcasm/facetiousness, my analogies). Anything OTHER than my simple point:

Arne's sample frame is not complete and the 1 (ONE) basis for his sample
frame (online sources) is too small. His math produces results that are
easily misinterpreted and therefore what he has claimed to be a correct
*statistical analysis* is anything but.

The Brown paper proves what Arne did what NOT a "statistical analysis" and
your Wiki link proves he started with a flawed sample frame.

 

[Tom Leylan]

So, you'd say my sample frame was flawed then? How ironic. Not too
mention that I never presented this as undisputable fact, did I? And,
based on your reply, I guess you just proved my point!

Not actually. I'd say you chose a word commonly found in a web page and
claimed the number of times that word occurred is close to the number of job
agencies. Why not just claim it is one-quarter of the total job agencies in
the US?

The numbers you have a problem with was the result of a query for jobs on a
job site. The statement (and I didn't make it) was that there is a 3 to 1
ratio and a couple of queries returned those numbers. Poor sample or not
they held up your number didn't.

Frankly if you believe that nobody here can understand the results of an ad
hoc query you have larger issues to deal with. There is no "point" to
prove.

 

[Tom Leylan]

And how does my being incorrect about the topic of what was changed (but
not my main point in bringing Wiki up) relate to your *statistics*?

We can conclude that you didn't read it.