OU drops Smalltalk
John Lester
University (OU) newspaper says:
"And in response to requests from students, undergraduates will be taught
to write programs in the commercial sector's favoured Java language, rather
than Smalltalk, to give them a better chance of finding a job."
This speaks for itself. Still, it would be interesting to know what
contribution M206 makes to Smalltalk, and consequently what Smalltalk stands
to lose by the OU's decision to scrap it.
John (doing M206)
Thomas Gagné
students. Perhaps that's how I'm able to get my degree through the
mail--enough students requested it!
Java should be an elective. It would likely be a popular elective and even a
strongly recommended one, but the university is impairing their students'
understanding of objects through substituting a language with a (much) weaker
object orientation.
--
.tom
remove dashes in email for replies
http://isectd.sourceforge.net
Donald Raab
performs technical interviews to screen candidates for both junior and
senior level Java positions. I would rate 3-6 months of Smalltalk
experience equivalent to 2-3 years of Java experience (depending on the
competence of the individual). Of course having actual Java experience is
nice too. But it is much more difficult to teach somone to think OO than it
is to teach them the Java syntax, IMO. I am pursuing teaching Smalltalk to
developers that I work with me to help them take a leap outside of the box
that Java has so nicely closed around them.
Bottom line - Smalltalk developers are better Java developers
"John Lester" <us...@example.net> wrote in message
news:beutkh$hvh$1...@news6.svr.pol.co.uk...
Runar Jordahl
http://www.pvv.ntnu.no/~eddie/experiences.html
-Runar Jordahl
Lance Parkington
university have long gone. It is necessary to consider the
implications of spending time learning computer languages since some
return on investment (eg a job) required from use of this time.
Learning Java is easy but mastering it is quite difficult. Smalltalk
has advantages for the developer but does not provide what employers
are willing to pay for any more. I suspect the non-typing resulted in
too much hasty code construction that becomes just too
difficult/expensive to maintain as costs are being driven more rapidly
downwards by demand.
Strongly typed languages require just a little bit more thought before
coding and this can make a big difference to cost. Time spent in
testing badly written code can spiral out of control. A recent BCS
article suggested typically 30-40% time can be spent testing!
Hence learning something that may waste money is now perceived a waste
of time both by students and the educational institutions.
"John Lester" <us...@example.net> wrote in message news:<beutkh$hvh$1...@news6.svr.pol.co.uk>...
Nils Kassube
> are willing to pay for any more. I suspect the non-typing resulted in
You are confusing strong and weak typing with the use of explicit type
declarations. Smalltalk _is_ strongly typed. For example C is weakly
typed but uses static type information. These are two different
concepts.
> Strongly typed languages require just a little bit more thought
> before coding and this can make a big difference to cost. Time spent
Using dynamically typed languages like Smalltalk or Python it's much
easier to adapt to a quickly changing environment. In these modern
times that's very important and can make a big difference in cost.
> testing badly written code can spiral out of control. A recent BCS
> article suggested typically 30-40% time can be spent testing!
That's what unit tests are for.
Steve Taylor
> Strongly typed languages require just a little bit more thought before
> coding and this can make a big difference to cost. Time spent in
> testing badly written code can spiral out of control. A recent BCS
> article suggested typically 30-40% time can be spent testing!
Having written Smalltalk and Java in different jobs, I would say that
I'm aproximately twice as productive in Smalltalk, can debug more
easily, and tend to produce cleaner designs.
Steve
panu
>....
>Learning Java is easy but mastering it is quite difficult. Smalltalk
>has advantages for the developer but does not provide what employers
>are willing to pay for any more.
>
I think the should give a course in Smalltalk
the first semester, and Java on the second.
It's always very educational to look at a thing
from two different perspectives. I don't think
you can grasp the essence of Java without comparing
it to something else (something better).
Just my personal opinion. But I believe there's
already an overabundance of unemployed Java
programmers as well. So why not study something
that gives you a specialist niche, and is fun
at the same time.
It's easy to learn Java later on your own, from
countless Java books. What an educational institution
should provide is a through exposition of the basic
concpets of Object-Orientation. Java is a poor choice
for that.
-Panu Viljamaa
Lance Parkington
for Java was actually produced from the VisualAge Smalltalk
environment.
However the result of financial squeeze on educational instititions
seems the same as that on private companies. The number of Smalltalk
employers is now decreasing rapidly and hence no money in it for
trainers or academics. I have looked for another Smalltalk job in UK
for over 5 years and there is nothing! However no shortage of Java
opportunities!
panu <pa...@fcc.net_zerospam> wrote in message news:<Qrmdndmyzqx...@fcc.net>...
Vladimir Los
> opportunities!
Say "Thank you very much" to Bjarn. Due to C compatibility solution we have
ugly tradiotion: C -> C++ -> Java...
--
Non sibi!
Wlad [UR3LOS]
"Lance Parkington" <lvpark...@taz.qinetiq.com> ???????/???????? ?
???????? ?????????: news:de71524.03072...@posting.google.com...
M. Roberts
> contribution M206 makes to Smalltalk, and consequently what Smalltalk stands
> to lose by the OU's decision to scrap it.
The instructors of M206 might have just gone with the students'
requests, but one has to suspect that they had their own reasons as
well. After all, the instructor generally has the final say in the
course design.
Thus, I'm inclined to think that the decisive question about the OU
decision was more or less the opposite: I.e., What contribution was
Smalltalk making to M206? What was the benefit as seen by the
instructors? It would be interesting to hear their experiences and
thoughts about this decision. Knowing more could help us in future
efforts to get Smalltalk into universities.
Also: how extensively was the course design tied to LearningWorks?
I ask the latter because support for LearningWorks was discontinued
some time ago. Apparently, the source code for LearningWorks has been
completely lost, and so I imagine that no significant forward
development was possible.
This might have been a significant issue for the people running the
course.
M