Embedded Systems Survey

Embedded Systems Survey

[Tucker Taft]

If you use Ada for embedded system development, you might
want to visit:

    http://www.embedded.com

to vote accordingly in their on-line survey of language use.

-- 
-Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
Technical Director, Distributed IT Solutions  (www.averstar.com/tools)
AverStar (formerly Intermetrics, Inc.)   Burlington, MA  USA

Re: Embedded Systems Survey

[Marin D. Condic]

Tucker Taft wrote:
> 
> If you use Ada for embedded system development, you might
> want to visit:
> 
>     http://www.embedded.com
> 
> to vote accordingly in their on-line survey of language use.
> 
While I did vote early and often, :-) the results of these sort of
Internet polls are basically meaningless. They are a self-selected
sample. Its like a radio talk show call-in poll. The callers are
generally the die-hard audience and their opinions reflect the views of
the show.

You think there may be some way of getting an honest answer to the
question of how many developers prefer Ada for a given problem domain?

MDC
-- 
======================================================================
Marin David Condic - Quadrus Corporation - http://www.quadruscorp.com/
Send Replies To: m c o n d i c @ q u a d r u s c o r p . c o m
Visit my web site at:  http://www.mcondic.com/

"I'd trade it all for just a little more"
    --  Charles Montgomery Burns, [4F10]
======================================================================

Re: Embedded Systems Survey

[Larry Kilgallen]

In article <3907DD01.4D0B9987@quadruscorp.com>, "Marin D. Condic" <mcondic-nospam@quadruscorp.com> writes:

> You think there may be some way of getting an honest answer to the
> question of how many developers prefer Ada for a given problem domain?

First you have to enumerate "all developers" for that problem domain.

Then you decide on a sampling algorithm and ask away.

The "no response" and the "none-of-your-business" samples will
affect your "margin of error", as does the "always lie on surveys"
mentality.

If it were not language advocacy, but something else, like
preferred toothpaste, would _you_ answer ?

Giving people money for answering skews the results.

> You think there may be some way of getting an honest answer to the
> question of how many developers prefer Ada for a given problem domain?

No.

Re: Embedded Systems Survey

[Marin D. Condic]

Larry Kilgallen wrote:
> First you have to enumerate "all developers" for that problem domain.
> 
> Then you decide on a sampling algorithm and ask away.
> 
> The "no response" and the "none-of-your-business" samples will
> affect your "margin of error", as does the "always lie on surveys"
> mentality.
> 
This might skew your results as well. What if there is a high
correlation between "Ada Programmer Mentality" and "Always Lie On
Surveys" mentality? :-)


> If it were not language advocacy, but something else, like
> preferred toothpaste, would _you_ answer ?
> 
Depends. Does it make me get on a mailing list or something? Is the
*government* behind it trying to get even *more* information about me so
they can better control every aspect of my life?

> Giving people money for answering skews the results.
> 

I've had a little statistics training, so I'm confident that an
experiment could be conducted which would get a reasonably fair answer.
The problem is more one of the mechanics of it. A survey on the web is
inexpensive but suspect from the start for obvious reasons. A magazine
or direct mail survey has problems of cost, sample set and responses.
(Do Ada programmers read the same magazines as C++ or Java programmers?
Is one group more likely to respond than the other?) I suppose if you
could identify a random set of embedded system programmers and called
them with a two question survey (Which language do you use on your
current project? Which language would you prefer to use?) you might get
a meaningful result. But where would you get your list? Or the money to
do all the calling?


> > You think there may be some way of getting an honest answer to the
> > question of how many developers prefer Ada for a given problem domain?
> 
> No.

Maybe. People tend to be pretty honest about what they prefer. The
problem is getting an honest survey.

MDC
-- 
======================================================================
Marin David Condic - Quadrus Corporation - http://www.quadruscorp.com/
Send Replies To: m c o n d i c @ q u a d r u s c o r p . c o m
Visit my web site at:  http://www.mcondic.com/

"I'd trade it all for just a little more"
    --  Charles Montgomery Burns, [4F10]
======================================================================

Re: Embedded Systems Survey

[tmoran]

> > You think there may be some way of getting an honest answer to the
> > question of how many developers prefer Ada for a given problem domain?
>
> No.
  There's a nice trick for getting statistics on something people won't
tell you.  Ask, for instance:  "Think of a number between 1 and 10.  If
your number is less than 5, or you prefer Ada, say Yes.  If it's greater
than 5 and you don't like Ada, say No."  Calibrate by asking another
group just the number part and not the Ada part, then the difference in
percentages tells you the percentage of people preferring Ada, without
telling about any particular person.

Re: Embedded Systems Survey

[Robert I. Eachus]

tmoran@bix.com wrote:

>   There's a nice trick for getting statistics on something people won't
> tell you.  Ask, for instance:  "Think of a number between 1 and 10.  If
> your number is less than 5, or you prefer Ada, say Yes.  If it's greater
> than 5 and you don't like Ada, say No."  Calibrate by asking another
> group just the number part and not the Ada part, then the difference in
> percentages tells you the percentage of people preferring Ada, without
> telling about any particular person.

  This proposed method has several serious flaws.  First, as stated,
anyone who answers "No" doesn't like Ada.  Second, having people think
of a number introduces a bias which you attempt to correct for by the
second survey.  But notice that clever readers of the survey will answer
"Yes" if they like Ada without guessing a number.  So there will be a
bias in the answers if those who don't like Ada tend to guess different
numbers than those who do.  If you really need to assure people that the
survey is anonymous, you have to use a method which people will trust. 
The easiest way I know of to do this is a (snail) mail survey.  Best I
know of is to bind a postcard (not a blow-in) into a magazine that you
are using to define the population of interest.  If the card is designed
to be folded and stapled, most people trust the survey to be anonymous.

   Of course, that still doesn't mean that their answers will be
honest.  On "Do you want to be a millionaire?" I am constantly amused by
the numbers in response to the "Ask the audience" lifelines.  Even if
two answers have already been eliminated, there will be votes for those
answers.

Re: Embedded Systems Survey

[tmoran]

>If you really need to assure people that the survey is anonymous,
>you have to use a method which people will trust.
  "Mail the postcard" surveys have their own problems.  Adding an
innocuous question with answer obviously unknown to the interviewer
is one way of increasing anonymity, even with a numbered postcard or
a face-to-face interviewer.  I think it's clever.  Since it adds
known noise to lessen error, I suppose it's analogous to dithering
for improving printing.

Re: Embedded Systems Survey

[Marin D. Condic]

tmoran@bix.com wrote:
>   "Mail the postcard" surveys have their own problems.  Adding an
> innocuous question with answer obviously unknown to the interviewer
> is one way of increasing anonymity, even with a numbered postcard or
> a face-to-face interviewer.  I think it's clever.  Since it adds
> known noise to lessen error, I suppose it's analogous to dithering
> for improving printing.

Yeah, but the biggest problem is that of a self selected group. If
you're interested in finding a reasonably accurate answer concerning the
number of software engineers that prefer/dislike Ada, you need to define
the group by some other means. Those who would voluntarily fill out a
card in a magazing may be biased heavily by being Ada Hawks or Anti-Ada
Bigots. The magazine itself is going to have a biased population of
readers in some ways and probably is not going to reflect a reasonable
cross section of all software professionals.

People are likely to be honest about whether or not they like/dislike a
given programming language. It's not like you're asking them something
terribly personal or something sensitive which might lead them to lie
about it. The problem is identifying a good sample set to ask.

MDC
-- 
======================================================================
Marin David Condic - Quadrus Corporation - http://www.quadruscorp.com/
Send Replies To: m c o n d i c @ q u a d r u s c o r p . c o m
Visit my web site at:  http://www.mcondic.com/

"Nothing in the world is more dangerous than sincere ignorance and
conscientious stupidity."

    -- Martin Luther King, Jr
======================================================================

Re: Embedded Systems Survey

[Georg Bauhaus]

Marin D. Condic (mcondic-nospam@acm.org) wrote:

: If
: you're interested in finding a reasonably accurate answer concerning the
: number of software engineers that prefer/dislike Ada, you need to define
: the group by some other means.

You could still ask just anyone and shrink the group only
afterwards.
One way is to ask about Ada and not mentioning the name. :-)
Q1: Which programming lanugages are you familiar with?
Q2: What do you think are the outstanding features of these
languages?

Then conclude what programming language this person "really"
likes.

Re: Embedded Systems Survey

[Gary Scott]

"Marin D. Condic" wrote:
> 
> Tucker Taft wrote:
> >
> > If you use Ada for embedded system development, you might
> > want to visit:
> >
> >     http://www.embedded.com
> >
> > to vote accordingly in their on-line survey of language use.
> >
> While I did vote early and often, :-) the results of these sort of
> Internet polls are basically meaningless. They are a self-selected
> sample. Its like a radio talk show call-in poll. The callers are
> generally the die-hard audience and their opinions reflect the views of
> the show.
> 
> You think there may be some way of getting an honest answer to the
> question of how many developers prefer Ada for a given problem domain?

One of my main irritants in these surveys is the definition of
"developer".  They usually completely omit programmers that are "non-cs
scientists and engineers" (a sizable segment).

> 
> MDC
> --

Re: Embedded Systems Survey

[Robert A Duff]

"Marin D. Condic" <mcondic-nospam@quadruscorp.com> writes:

> While I did vote early and often, :-) the results of these sort of
> Internet polls are basically meaningless. They are a self-selected
> sample. Its like a radio talk show call-in poll. The callers are
> generally the die-hard audience and their opinions reflect the views of
> the show.
> 
> You think there may be some way of getting an honest answer to the
> question of how many developers prefer Ada for a given problem domain?

I don't know, but it seems to me more important to ask whether the
resulting software is cheaper, safer, or better in some other way,
rather than whether the developers prefer it.  ;-)

- Bob

Re: Embedded Systems Survey

[Ted Dennison]

In article <wccya5rjyfu.fsf@world.std.com>,
  Robert A Duff <bobduff@world.std.com> wrote:

> I don't know, but it seems to me more important to ask whether the
> resulting software is cheaper, safer, or better in some other way,
> rather than whether the developers prefer it.  ;-)

When asked in a poll of software developers, either question is liable
to produce the same answer. Better yet would be to test it somehow (or
at least poll customers, then check against the implementation
languages).

--
T.E.D.

http://www.telepath.com/~dennison/Ted/TED.html


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Embedded Systems Survey

[Robert A Duff]

Ted Dennison <dennison@telepath.com> writes:

> In article <wccya5rjyfu.fsf@world.std.com>,
>   Robert A Duff <bobduff@world.std.com> wrote:
> 
> > I don't know, but it seems to me more important to ask whether the
> > resulting software is cheaper, safer, or better in some other way,
> > rather than whether the developers prefer it.  ;-)
> 
> When asked in a poll of software developers, either question is liable
> to produce the same answer. ...

I don't believe you.  I think there are lots of programmers who like
languages for reasons other than the quality of the resulting programs.

>... Better yet would be to test it somehow (or
> at least poll customers, then check against the implementation
> languages).

Yeah.

- Bob

Re: Embedded Systems Survey

[Marin D. Condic]

Ted Dennison wrote:
> When asked in a poll of software developers, either question is liable
> to produce the same answer. Better yet would be to test it somehow (or
> at least poll customers, then check against the implementation
> languages).
> 
Cheaper, faster and better are things that are testable by experiment.
(Maybe and *expensive* experiment, but definitely something you can
measure.) Opinions about what is cheaper, faster, better won't tell you
much. I'd be willing to bet that if you polled 1000 Java programmers
about what is cheaper/faster/better they would say Java. Poll 1000 Cobol
programmers and the answer is likely to be similar. And all those
opinions may be wrong - the answer might be APL or Pascal or something
else.

As a promoter of the use of Ada, I'd like to know a) what do people
prefer, b) why do they prefer it, c) what they dislike and d) why do
they dislike it. Knowing that, we might better market the language to
those who are not currently using it.

MDC
-- 
======================================================================
Marin David Condic - Quadrus Corporation - http://www.quadruscorp.com/
Send Replies To: m c o n d i c @ q u a d r u s c o r p . c o m
Visit my web site at:  http://www.mcondic.com/

"I'd trade it all for just a little more"
    --  Charles Montgomery Burns, [4F10]
======================================================================

Re: Embedded Systems Survey

[Marin D. Condic]

Robert A Duff wrote:
> 
> "Marin D. Condic" <mcondic-nospam@quadruscorp.com> writes:
> > You think there may be some way of getting an honest answer to the
> > question of how many developers prefer Ada for a given problem domain?
> 
> I don't know, but it seems to me more important to ask whether the
> resulting software is cheaper, safer, or better in some other way,
> rather than whether the developers prefer it.  ;-)
> 
It may be more important to make a business case for Ada, but that's not
something you do by survey. We couldn't go out and conduct a poll that
asks "Which language do you believe to be cheaper, safer, etc." and have
the results tell us much that is meaningful. People's opinions about
what costs less to develop in are meaningless. Whereas, when conducting
a survey, asking people what they *prefer* generates something useful.

You can't argue about what someone likes or dislikes. Asking them why
they like/dislike something also generates useful information. If we
have a clear picture of perceptions, preferences, etc., we might be able
to better address the *marketing* aspects of Ada, instead of just the
technical or business aspects of the language.

Just a thought. I knew all those marketing classes I took would pay off
one day. :-)

MDC
-- 
======================================================================
Marin David Condic - Quadrus Corporation - http://www.quadruscorp.com/
Send Replies To: m c o n d i c @ q u a d r u s c o r p . c o m
Visit my web site at:  http://www.mcondic.com/

"I'd trade it all for just a little more"
    --  Charles Montgomery Burns, [4F10]
======================================================================

Re: Embedded Systems Survey

[Ted Dennison]

Tucker Taft wrote:

> If you use Ada for embedded system development, you might
> want to visit:
>
>     http://www.embedded.com
>
> to vote accordingly in their on-line survey of language use.

I note that so far our ballot-box stuffing effort has resulted in a result
of 50.1% Ada utilization. :-)


--
T.E.D.

Home - mailto:dennison@telepath.com  Work - mailto:dennison@ssd.fsi.com
WWW  - http://www.telepath.com/dennison/Ted/TED.html  ICQ  - 10545591

Re: Embedded Systems Survey

[Ken Garlington]

The best part: you apparently can vote as many times as you want!

"Ted Dennison" <dennison@telepath.com> wrote in message
news:39075401.8FA0E054@telepath.com...
> Tucker Taft wrote:
>
> > If you use Ada for embedded system development, you might
> > want to visit:
> >
> >     http://www.embedded.com
> >
> > to vote accordingly in their on-line survey of language use.
>
> I note that so far our ballot-box stuffing effort has resulted in a result
> of 50.1% Ada utilization. :-

Re: Embedded Systems Survey

[Ted Dennison]

53% and rising now. :-)

Ken Garlington wrote:

> The best part: you apparently can vote as many times as you want!

Perhaps we could write a stuffer-bot, using AWS. That might get us up into the
90's. :-)

--
T.E.D.

Home - mailto:dennison@telepath.com  Work - mailto:dennison@ssd.fsi.com
WWW  - http://www.telepath.com/dennison/Ted/TED.html  ICQ  - 10545591

Re: Embedded Systems Survey

[tmoran]

>Perhaps we could write a stuffer-bot, using AWS. That might get us
>up into the 90's. :-)
  I could write one in Ada using Claw, but that wouldn't be right.
(Not all of us Chicago-born have the same attitude toward ballot
stuffing.)

Re: Embedded Systems Survey

[Robert I. Eachus]

> I note that so far our ballot-box stuffing effort has resulted in a
> result of 50.1% Ada utilization. :-)

   I was slightly surprised by the popularity of Forth and C++, is it
possible that someone else is ballot box stuffing?

Re: Embedded Systems Survey

[Richard D Riehle]

In article <39075E96.BC16B5D8@earthlink.net>,
	"Robert I. Eachus" <rieachus@earthlink.net> wrote:

>> I note that so far our ballot-box stuffing effort has resulted in a
>> result of 50.1% Ada utilization. :-)
>
>   I was slightly surprised by the popularity of Forth and C++, is it
>possible that someone else is ballot box stuffing?

Robert,

It is important to realize that a high percentage of ESP readers
are programming microcontrollers such as the 8051 series.  For this
processor two of the more popular programming languages are Assembler
and Forth.  The latter, Forth, is actually quite apprproriate when one
is concerned with both efficiency and close mapping to the architecture.

As an additional point of interest, at one time, the largest Forth 
Interest Group (FIG), was in Leningrad (now St. Petersburg).  Also,
Forth enthusiasts, after considerable debate for obvious reasons, 
decided not to call their organization the Forth Users Group.

Richard Riehle

Re: Embedded Systems Survey

[Gautier]

Ted:
> > If you use Ada for embedded system development, you might
> > want to visit:

> >     http://www.embedded.com

> > to vote accordingly in their on-line survey of language use.

> I note that so far our ballot-box stuffing effort has resulted in a result
> of 50.1% Ada utilization. :-)

Click! Back. Click! Back. Click! Back. Click! Back. Click! Back. Click! Back.
Click! Back. Click! Back. What a stress... Someone to relay ? 8-}))

______________________________________________________
Gautier  --  http://members.xoom.com/gdemont/gsoft.htm