Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.programming > #5640 > unrolled thread

What do they teach in school today?

Back to article view | Back to comp.programming

Contents

  What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-08-20 18:17 -0400
    Re: What do they teach in school today? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2015-08-20 23:56 +0100
      Re: What do they teach in school today? fir <profesor.fir@gmail.com> - 2015-08-20 16:10 -0700
    Re: What do they teach in school today? Kaz Kylheku <kaz@kylheku.com> - 2015-08-20 23:55 +0000
      Re: What do they teach in school today? Julio Di Egidio <julio@diegidio.name> - 2015-08-24 04:28 -0700
    Re: What do they teach in school today? Leo <chris.w.l.macleod@sympatico.ca> - 2015-08-20 21:21 -0400
    Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-08-23 09:33 -0500
      Re: What do they teach in school today? fir <profesor.fir@gmail.com> - 2015-08-23 09:47 -0700
        Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-08-23 14:45 -0500
      Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-08-23 23:27 -0400
        Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-08-24 09:48 -0500
          Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-08-25 00:49 -0400
            Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-08-25 09:58 -0500
              Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-08-28 19:30 -0400
              Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-05 20:49 +0100
                Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-11 00:26 -0500
                  Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-12 10:58 +0100
                    Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-12 08:06 -0500
                      Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-12 15:20 +0100
                        Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-13 00:53 -0500
                          Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-13 12:19 -0400
                            Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-14 00:32 -0500
                              Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-14 07:17 -0400
                                Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-14 09:29 -0500
                    Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-12 13:11 -0400
            Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-05 20:55 +0100
              Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-07 20:20 -0400
                Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-11 00:44 -0500
                  Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-11 08:29 -0500
                    Re: What do they teach in school today? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2015-09-11 17:05 +0100
                      Re: What do they teach in school today? Mark Carroll <mtbc@bcs.org> - 2015-09-11 17:07 +0100
                        Re: What do they teach in school today? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2015-09-11 17:27 +0100
                          Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-12 12:51 -0400
                      Re: What do they teach in school today? "Charles Richmond" <numerist@aquaporin4.com> - 2015-09-11 14:55 -0500
                        Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-12 12:54 -0400
                      Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-11 17:52 -0500
                        Re: What do they teach in school today? Ben Bacarisse <ben.usenet@bsb.me.uk> - 2015-09-12 00:23 +0100
                          Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-11 22:45 -0500
                          Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-12 12:56 -0400
                        Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-12 11:04 +0100
                          Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-12 08:10 -0500
                            Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-12 15:24 +0100
                              Re: What do they teach in school today? Richard Heathfield <rjh@cpax.org.uk> - 2015-09-12 15:36 +0100
                              Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-12 23:24 -0500
                            Re: What do they teach in school today? Robert Wessel <robertwessel2@yahoo.com> - 2015-09-16 13:08 -0500
                              Re: What do they teach in school today? Kaz Kylheku <kaz@kylheku.com> - 2015-09-16 18:54 +0000
                                Re: What do they teach in school today? Robert Wessel <robertwessel2@yahoo.com> - 2015-09-17 00:15 -0500
                                  Re: What do they teach in school today? Chad <cdalten@gmail.com> - 2015-09-25 21:08 -0700
                    Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-12 12:49 -0400
                      Re: What do they teach in school today? BGB <cr88192@hotmail.com> - 2015-09-12 19:56 -0500
    Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-05 21:21 +0100
      Re: What do they teach in school today? Mark Carroll <mtbc@bcs.org> - 2015-09-05 22:54 +0100
        Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-05 23:17 +0100
          Re: What do they teach in school today? Mark Carroll <mtbc@bcs.org> - 2015-09-06 09:59 +0100
            Re: What do they teach in school today? "James Harris" <james.harris.1@gmail.com> - 2015-09-06 11:11 +0100
      Re: What do they teach in school today? "J. Clarke" <j.clarke.873638@gmail.com> - 2015-09-07 19:59 -0400
      Re: What do they teach in school today? "Charles Richmond" <numerist@aquaporin4.com> - 2015-09-11 14:59 -0500

Page 3 of 3 — ← Prev page 1 2 [3]

#5847

On 9/12/2015 5:04 AM, James Harris wrote:
> "BGB" <cr88192@hotmail.com> wrote in message
> news:msvm9f$3gm$1@news.albasani.net...
>
> ...
>
>> there was an class for algorithms, but it was fairly rudimentary, like
>> implementing linked lists and bubble sort.
>
> Anything that teaches bubble sort has to be rudimentary! I remember one
> person saying that it should never be taught to anyone. The course below
> teaches eight (yes, eight) sorting methods and not one of them is bubble
> sort.
>

it is sometimes useful, either if one doesn't care (because N is always 
small), or (along with selection or insertion sort) as a fallback case 
for quicksort.

with the people there though, if they taught more advanced sorting, 
probably too few people would have understood it or been able to 
implement it. a lot of the people were having enough trouble getting the 
linked lists to work.


>> actual "serious" CS topics didn't really come up...
>
> For anyone who is interested there is a brilliant series of lectures on
> data structures at
>
>   https://www.youtube.com/playlist?list=PL4BBB74C7D2A1049C
>
> The first part of the course is on Java. The data structures part starts
> around lecture 21.
>

yeah.

[toc] | [prev] | [next] | [standalone]

#5849

"BGB" <cr88192@hotmail.com> wrote in message 
news:mt18h7$411$1@news.albasani.net...
> On 9/12/2015 5:04 AM, James Harris wrote:
>> "BGB" <cr88192@hotmail.com> wrote in message
>> news:msvm9f$3gm$1@news.albasani.net...
>>
>> ...
>>
>>> there was an class for algorithms, but it was fairly rudimentary, 
>>> like
>>> implementing linked lists and bubble sort.
>>
>> Anything that teaches bubble sort has to be rudimentary! I remember 
>> one
>> person saying that it should never be taught to anyone. The course 
>> below
>> teaches eight (yes, eight) sorting methods and not one of them is 
>> bubble
>> sort.
>>
>
> it is sometimes useful, either if one doesn't care (because N is 
> always small), or (along with selection or insertion sort) as a 
> fallback case for quicksort.

If I needed a sort I could write *and debug* quickly for even a 
fair-sized N I think I would go for insertion sort. Insertion sort is 
fairly easy to get right. What most algorithm analyses miss out is how 
efficiently modern CPUs optimize sequential memory traversals. Insertion 
sort basically depends on such sequential accesses so should be more 
efficient than a simple O() notation would suggest.

James

[toc] | [prev] | [next] | [standalone]

#5850

On 12/09/2015 15:24, James Harris wrote:

> If I needed a sort I could write *and debug* quickly for even a
> fair-sized N I think I would go for insertion sort. Insertion sort is
> fairly easy to get right. What most algorithm analyses miss out is how
> efficiently modern CPUs optimize sequential memory traversals. Insertion
> sort basically depends on such sequential accesses so should be more
> efficient than a simple O() notation would suggest.

If we're talking about writing from scratch without access to libraries 
(including standard libraries), I'd probably go for Shell sort, just 
because it's the one I remember best.

In Real Life, I'd pour everything into a self-balancing BST.

-- 
Richard Heathfield
Email: rjh at cpax dot org dot uk
"Usenet is a strange place" - dmr 29 July 1999
Sig line 4 vacant - apply within

[toc] | [prev] | [next] | [standalone]

#5857

On 9/12/2015 9:24 AM, James Harris wrote:
> "BGB" <cr88192@hotmail.com> wrote in message
> news:mt18h7$411$1@news.albasani.net...
>> On 9/12/2015 5:04 AM, James Harris wrote:
>>> "BGB" <cr88192@hotmail.com> wrote in message
>>> news:msvm9f$3gm$1@news.albasani.net...
>>>
>>> ...
>>>
>>>> there was an class for algorithms, but it was fairly rudimentary, like
>>>> implementing linked lists and bubble sort.
>>>
>>> Anything that teaches bubble sort has to be rudimentary! I remember one
>>> person saying that it should never be taught to anyone. The course below
>>> teaches eight (yes, eight) sorting methods and not one of them is bubble
>>> sort.
>>>
>>
>> it is sometimes useful, either if one doesn't care (because N is
>> always small), or (along with selection or insertion sort) as a
>> fallback case for quicksort.
>
> If I needed a sort I could write *and debug* quickly for even a
> fair-sized N I think I would go for insertion sort. Insertion sort is
> fairly easy to get right. What most algorithm analyses miss out is how
> efficiently modern CPUs optimize sequential memory traversals. Insertion
> sort basically depends on such sequential accesses so should be more
> efficient than a simple O() notation would suggest.
>

I have often gone for selection sort because for many tasks this can be 
implemented with a fairly minimal amount of code (or use insertion sort 
if items need to be added incrementally).

or quicksort for a generally quick sort...
or a balanced tree structure for fast incremental inserts.
...

it is probably more of an "it depends" situation.

[toc] | [prev] | [next] | [standalone]

#5865

On Sat, 12 Sep 2015 08:10:09 -0500, BGB <cr88192@hotmail.com> wrote:

>On 9/12/2015 5:04 AM, James Harris wrote:
>> "BGB" <cr88192@hotmail.com> wrote in message
>> news:msvm9f$3gm$1@news.albasani.net...
>>
>> ...
>>
>>> there was an class for algorithms, but it was fairly rudimentary, like
>>> implementing linked lists and bubble sort.
>>
>> Anything that teaches bubble sort has to be rudimentary! I remember one
>> person saying that it should never be taught to anyone. The course below
>> teaches eight (yes, eight) sorting methods and not one of them is bubble
>> sort.
>>
>
>it is sometimes useful, either if one doesn't care (because N is always 
>small), or (along with selection or insertion sort) as a fallback case 
>for quicksort.


The complaint about bubble sort is that not only is insertion sort
faster (typically by a factor of two), insertion sort is a *simpler*
algorithm as well, usually requiring fewer lines of code to implement.
So what does bubble sort have going for it, except (as Knuth said), a
catchy name?

FWIW, bubble sort appears to be reasonably applicable to the one tape
sorting problem, but that's not really of much relevance, and even in
the past trying to sort with one tape was more desperation than
anything else.

[toc] | [prev] | [next] | [standalone]

#5867

On 2015-09-16, Robert Wessel <robertwessel2@yahoo.com> wrote:
> On Sat, 12 Sep 2015 08:10:09 -0500, BGB <cr88192@hotmail.com> wrote:
>
>>On 9/12/2015 5:04 AM, James Harris wrote:
>>> "BGB" <cr88192@hotmail.com> wrote in message
>>> news:msvm9f$3gm$1@news.albasani.net...
>>>
>>> ...
>>>
>>>> there was an class for algorithms, but it was fairly rudimentary, like
>>>> implementing linked lists and bubble sort.
>>>
>>> Anything that teaches bubble sort has to be rudimentary! I remember one
>>> person saying that it should never be taught to anyone. The course below
>>> teaches eight (yes, eight) sorting methods and not one of them is bubble
>>> sort.
>>>
>>
>>it is sometimes useful, either if one doesn't care (because N is always 
>>small), or (along with selection or insertion sort) as a fallback case 
>>for quicksort.
>
>
> The complaint about bubble sort is that not only is insertion sort
> faster (typically by a factor of two), insertion sort is a *simpler*
> algorithm as well, usually requiring fewer lines of code to implement.

Yes; for instance insertion sort doesn't have to calculate the silly
"are we done?" flag on every pass, unlike bubble sort.

The condition of being done proceeds from the way insertion sort
partitions the array between unprocessed and processed elements.



> So what does bubble sort have going for it, except (as Knuth said), a
> catchy name?

Bubble sort has perhaps this "virtue". Suppose that it is not required
that some items be sorted immediately, only that they eventually settle
into a sorted order, little by little.

Once in a while, when we have spare cycles, we perform a single bubble sort
pass, and the items are all closer to where they should be after that pass.
That could be a desirable property, in some imaginable situation.

Also, it seems as if bubble sort can be nicely parallelized. Combinatorial
circuits can compare adjacent elements and produce the values for the next
array. All that is left to do is supply a clock signal to latch the output back
into the array.  The implementation is easy because the elementary operations
of bubble sort are localized. By contrast, insertion sort requires the
treatment of a single element in one step of the algorithm to have a global
effect on the entire array.

> FWIW, bubble sort appears to be reasonably applicable to the one tape
> sorting problem, but that's not really of much relevance, and even in
> the past trying to sort with one tape was more desperation than
> anything else.

Bubble sort is useful in the education of newbie computer scientists.  It's a
nice example of an algorithm where some narrowly confined local effect, when
repeated over the data, brings about a global property.  It's not obvious from
the structure of bubble sort that it's getting all the items in order.  For
people who have no concept of an algorithm or computer programming, bubble sort
is eye opening.

Because bubble sort is "bad", that also has educational value; it can be used
as a springboard to into other algorithms, introducing the topic of how there
exist multiple algorithms to solve the same problem, which have different trade
offs, and that sometimes algorithms have so few virtues that they get
abandoned.

If you don't teach bubble sort, then you can't explain why insertion sort
is better than bubble sort. All you can do is tell the students is that there
exists something called bubble sort (which you will not teach) and that it's to
be taken upon your word that insertion sort is superior.

[toc] | [prev] | [next] | [standalone]

#5869

On Wed, 16 Sep 2015 18:54:49 +0000 (UTC), Kaz Kylheku
<kaz@kylheku.com> wrote:

>On 2015-09-16, Robert Wessel <robertwessel2@yahoo.com> wrote:
>> On Sat, 12 Sep 2015 08:10:09 -0500, BGB <cr88192@hotmail.com> wrote:
>>
>>>On 9/12/2015 5:04 AM, James Harris wrote:
>>>> "BGB" <cr88192@hotmail.com> wrote in message
>>>> news:msvm9f$3gm$1@news.albasani.net...
>>>>
>>>> ...
>>>>
>>>>> there was an class for algorithms, but it was fairly rudimentary, like
>>>>> implementing linked lists and bubble sort.
>>>>
>>>> Anything that teaches bubble sort has to be rudimentary! I remember one
>>>> person saying that it should never be taught to anyone. The course below
>>>> teaches eight (yes, eight) sorting methods and not one of them is bubble
>>>> sort.
>>>>
>>>
>>>it is sometimes useful, either if one doesn't care (because N is always 
>>>small), or (along with selection or insertion sort) as a fallback case 
>>>for quicksort.
>>
>>
>> The complaint about bubble sort is that not only is insertion sort
>> faster (typically by a factor of two), insertion sort is a *simpler*
>> algorithm as well, usually requiring fewer lines of code to implement.
>
>Yes; for instance insertion sort doesn't have to calculate the silly
>"are we done?" flag on every pass, unlike bubble sort.
>
>The condition of being done proceeds from the way insertion sort
>partitions the array between unprocessed and processed elements.
>
>
>
>> So what does bubble sort have going for it, except (as Knuth said), a
>> catchy name?
>
>Bubble sort has perhaps this "virtue". Suppose that it is not required
>that some items be sorted immediately, only that they eventually settle
>into a sorted order, little by little.
>
>Once in a while, when we have spare cycles, we perform a single bubble sort
>pass, and the items are all closer to where they should be after that pass.
>That could be a desirable property, in some imaginable situation.


Doing the same to Quicksort would get the file sorted in much less
time.  To get roughly the equivalent work-per-pass, you'd process as
many partitions as you could while keeping the total number of
elements processed under the number of elements in the total input.
Admittedly this would not have Bubblesort's property that every moved
element will be closer to its final position after each pass.

If the dataset is small, it's probably sortable in a single chunk, if
it's large, this is going to converge with "sorted" very slowly.


>Also, it seems as if bubble sort can be nicely parallelized. Combinatorial
>circuits can compare adjacent elements and produce the values for the next
>array. All that is left to do is supply a clock signal to latch the output back
>into the array.  The implementation is easy because the elementary operations
>of bubble sort are localized. By contrast, insertion sort requires the
>treatment of a single element in one step of the algorithm to have a global
>effect on the entire array.


If I can actually afford order N custom hardware, I will probably
spring for one of the vastly faster parallel sort algorithms.
Admittedly, a simple comparison and exchange network might be the best
solution for very small input datasets.


>> FWIW, bubble sort appears to be reasonably applicable to the one tape
>> sorting problem, but that's not really of much relevance, and even in
>> the past trying to sort with one tape was more desperation than
>> anything else.
>
>Bubble sort is useful in the education of newbie computer scientists.  It's a
>nice example of an algorithm where some narrowly confined local effect, when
>repeated over the data, brings about a global property.  It's not obvious from
>the structure of bubble sort that it's getting all the items in order.  For
>people who have no concept of an algorithm or computer programming, bubble sort
>is eye opening.
>
>Because bubble sort is "bad", that also has educational value; it can be used
>as a springboard to into other algorithms, introducing the topic of how there
>exist multiple algorithms to solve the same problem, which have different trade
>offs, and that sometimes algorithms have so few virtues that they get
>abandoned.
>
>If you don't teach bubble sort, then you can't explain why insertion sort
>is better than bubble sort. All you can do is tell the students is that there
>exists something called bubble sort (which you will not teach) and that it's to
>be taken upon your word that insertion sort is superior.


I'll grant the pedagogical value for CS types, but Bubble sort is
taught to many non-CS majors (likely far more than there are CS
majors), and without the justification that it's an example of either
algorithm analysis or comparison to other sorts.

Still, there might be a few actual uses, but they're rare, and the
existence of (serious) outliers is not justification for teaching it
to anyone except CS types.

[toc] | [prev] | [next] | [standalone]

#5901

I spent four years learning the theory behind programming. Ie, why things work the way that they do. As a result, the university expected us to be able to learn something like C and Java on our own. The reasoning was that we should have been able to learn and master those languages on our own since we were given a solid theoretical background in Computer Science.

[toc] | [prev] | [next] | [standalone]

#5851

In article <msul9v$81j$1@news.albasani.net>, cr88192@hotmail.com says...
> 
> On 9/11/2015 12:44 AM, BGB wrote:
> > On 9/7/2015 7:20 PM, J. Clarke wrote:
> >> In article <msfh8e$cg4$1@dont-email.me>, james.harris.1@gmail.com
> >> says...
> >>>
> ...
> >>
> >> Well, to give an example, my current employer has an opening for a
> >> Quantitative Consultant (the same title I hold--don't let the
> >> "consultant" part throw you--it's a full time salaried position with
> >> full benefits), with the stated requirements being a bachelor's degree
> >> in computer science and two years programming experience.  The candidate
> >> also has to pass an aptitude test (a timed test which is half "word
> >> problems" in mathematics and half "fill in the missing piece" on
> >> flowcharts), and pass the interview.
> >>
> >
> > don't have a degree...
> >
> 
> went for a CS major, spent years in various colleges, fell one math 
> class short of a degree (while in the middle of a move). not bothered 
> here because few enough credits would transfer that it would be 
> more-or-less starting over. was like, "screw it", so no degree...
> 
> me and Calculus aren't really friends.
> 
> 
> hence, going for something simpler, now in a trade-school for industrial 
> maintenance...

Me and calculus werent' freinds either, not at first anyway--first two 
semesters were with a mathematician and I got Ds.  Third semester the 
professor was a retired United States Marine Corps General and he 
managed to explain things in terms that even I could understand.

I'd suggest if you haven't passed calculus with transferable credits 
yet, get the Schaums Outline and work through it--remember that some of 
it is just grunt-work memorization.  If you can do all the exercises in 
the Schaums, you're not going to have any real trouble with coursework, 
at least not unless you've got one of those idiots who spends three 
semesters teaching proofs and never asks you to actually solve a 
problem.

[toc] | [prev] | [next] | [standalone]

#5856

On 9/12/2015 11:49 AM, J. Clarke wrote:
> In article <msul9v$81j$1@news.albasani.net>, cr88192@hotmail.com says...
>>
>> On 9/11/2015 12:44 AM, BGB wrote:
>>> On 9/7/2015 7:20 PM, J. Clarke wrote:
>>>> In article <msfh8e$cg4$1@dont-email.me>, james.harris.1@gmail.com
>>>> says...
>>>>>
>> ...
>>>>
>>>> Well, to give an example, my current employer has an opening for a
>>>> Quantitative Consultant (the same title I hold--don't let the
>>>> "consultant" part throw you--it's a full time salaried position with
>>>> full benefits), with the stated requirements being a bachelor's degree
>>>> in computer science and two years programming experience.  The candidate
>>>> also has to pass an aptitude test (a timed test which is half "word
>>>> problems" in mathematics and half "fill in the missing piece" on
>>>> flowcharts), and pass the interview.
>>>>
>>>
>>> don't have a degree...
>>>
>>
>> went for a CS major, spent years in various colleges, fell one math
>> class short of a degree (while in the middle of a move). not bothered
>> here because few enough credits would transfer that it would be
>> more-or-less starting over. was like, "screw it", so no degree...
>>
>> me and Calculus aren't really friends.
>>
>>
>> hence, going for something simpler, now in a trade-school for industrial
>> maintenance...
>
> Me and calculus werent' freinds either, not at first anyway--first two
> semesters were with a mathematician and I got Ds.  Third semester the
> professor was a retired United States Marine Corps General and he
> managed to explain things in terms that even I could understand.
>
> I'd suggest if you haven't passed calculus with transferable credits
> yet, get the Schaums Outline and work through it--remember that some of
> it is just grunt-work memorization.  If you can do all the exercises in
> the Schaums, you're not going to have any real trouble with coursework,
> at least not unless you've got one of those idiots who spends three
> semesters teaching proofs and never asks you to actually solve a
> problem.
>

I had that once, didn't understand the proofs either, or what exactly I 
was going on or what I was supposed to be doing in the class, and the 
teacher I suspect was one of the biggest pricks I have probably ever 
seen IRL (he basically came off like Q from Star Trek just without the 
special powers...).


this was enough to make me actively annoyed with "math people" for 
several years, until I later noticed that most of them were more like 
"Mr. Data", which is mostly just mildly annoying (them being really big 
into pointless esoterica, rather than things which have some semblance 
of being at least potentially useful).

then again, most of my stuff isn't really so useful either, but at 
least, I make things that exist and do stuff...

but, I have a limitation: I have found that my ability to understand 
mathy stuff doesn't go all that far beyond algebra.


like, in some ways, I am basically more like Zim...

[toc] | [prev] | [next] | [standalone]

#5820

"J. Clarke" <j.clarke.873638@gmail.com> wrote in message 
news:MPG.30401c0c36885213989ac3@news.eternal-september.org...
>I was talking to a recent computer science graduate today.  He had 
>never
> seen C, never seen Fortran, had no notion of a shell script or batch
> file, had never heard of ActiveX or .net or gcc.

What *had* they taught him? There must have been something.

> So what does somebody spend four years doing in a modern university
> computer science program?

AFAIK from online university courses I have perused universities do 
teach programming properly (though maybe not as diversely as they could) 
but schools (i.e. for up to 16 or 18 years of age) seem to have 
abandoned proper computer education completely. ISTM that schools teach 
children how to *use* computers, especially computer applications, 
rather than how to control a computer; they focus on things like 
spreadsheets and word processors. They are probably more keen to ensure 
that kids are not frightened by computers, or perhaps it is the 
teachers' own fear that they are trying to overcome.

From things I have seen on TV recently it seems that when educators do 
teach kids to program it is using things like Scratch which allows 
programs to be composed by dragging constructs and dropping them in a 
gui. There's an example of it at

  https://scratch.mit.edu/

I can see some value in it for tiny tots perhaps but, otherwise, aargh!

Perhaps children don't have, or aren't expected to have, the same 
attention span and educators think they will get bored if they have to 
type something that won't give them a GUI or make sounds. Printing the 
numbers from 1 to 10 is no longer exciting, eh!

I thought to reply to this partly as I have seen a similar amazing (not 
good-amazing) change in the books on computing available in local 
libraries. When I was young libraries had books on programming languages 
and how to write things like compilers.

My local library, now, has books like a beginner's guide to spreadsheets 
but when I looked for a book on programming, at least in Python, they 
had absolutely nothing. Even more incredibly, expanding the search to 
all the libraries in the same council-controlled area also resulted in 
*no* books whatever on programming in Python. And yet a search for 
programming jobs locally brings up just now 75 jobs mentioning Python 
within 10 miles of London advertised in the last week.

I don't know what the world's coming to. Am I sounding old? ;-)

James

[toc] | [prev] | [next] | [standalone]

#5821

"James Harris" <james.harris.1@gmail.com> writes:
(snip)
> From things I have seen on TV recently it seems that when educators do 
> teach kids to program it is using things like Scratch which allows 
> programs to be composed by dragging constructs and dropping them in a 
> gui. There's an example of it at
>
>   https://scratch.mit.edu/
>
> I can see some value in it for tiny tots perhaps but, otherwise,
> aargh!

My son's in secondary school in Perth, Scotland. Everybody did Scratch
and Build Your Own Blocks. He chose to continue with computing, and is
thus now on VisualBasic and will be doing Java at some point. How
typical that is I've no idea.

(snip)
> I thought to reply to this partly as I have seen a similar amazing (not 
> good-amazing) change in the books on computing available in local 
> libraries. When I was young libraries had books on programming languages 
> and how to write things like compilers.
>
> My local library, now, has books like a beginner's guide to spreadsheets 
> but when I looked for a book on programming, at least in Python, they 
> had absolutely nothing. Even more incredibly, expanding the search to 
> all the libraries in the same council-controlled area also resulted in 
> *no* books whatever on programming in Python.
(snip)
> I don't know what the world's coming to. Am I sounding old? ;-)

(You seem to be posting from a British perspective; I hope readers
elsewhere will forgive my continuing that below.)

Mine isn't great either, Perth certainly not: I instead drive over to
Dundee and their central library's "Opportunities Room" /does/ have a
few books. (Admittedly, it's not obvious the room exists, it's around a
corner at the end of the top floor and I don't see a sign!) My wife
liked one of their Java books, and I have about their only decent, if a
bit old, XML book on hold now, somebody else has it out. (I have some
XSLT joy coming up at work.) It's certainly not a wide or up-to-date
selection, but it's better than nothing.

Of course, there's a lack of academic books too. Those with access to
good university libraries are very lucky. I suppose they're expensive.

There's a general lack of study resources, though: in many other
subjects in these modern times there's a lack of "apprenticeships" in
industry and a lack of textbooks that are decent for self-study instead
of being designed as an adjunct to some taught course. They don't even
put those nice Open University lectures on television in the early
mornings any more and if one looks at local night school courses it's
all pottery and whatnot. In computing we're lucky that at least decent
self-study books for practical programming can be bought and one
typically doesn't need expensive resources to be able to work on them.
Indeed, outside the Microsoft world a lot of industry standard stuff is
available for free download, like the whole stack from PostgreSQL up
through Hibernate, Lucene and Spring to Tomcat and JavaScript with
jQuery and Backbone.

There /is/ a British Computer Society sign up in Dundee's central
library but what actually happens in connection with it I've no idea,
which is slightly embarrassing with my being a member myself. Maybe
someday I should pop my head around that door and actually ask.

No Python books does sound pretty sorry. Even Perth has "Beginning
Programming with Python for Dummies" (albeit only one copy that is
"awaiting shelving", not yet reservable) though no obvious O'Reilly.
(I've not studied Python myself, mostly I bluff my way through it based
on what I know of OOP and Perl and Haskell.)

-- Mark

[toc] | [prev] | [next] | [standalone]

#5822

"Mark Carroll" <mtbc@bcs.org> wrote in message 
news:87io7obkht.fsf@ixod.org...
> "James Harris" <james.harris.1@gmail.com> writes:
> (snip)
>> From things I have seen on TV recently it seems that when educators 
>> do
>> teach kids to program it is using things like Scratch which allows
>> programs to be composed by dragging constructs and dropping them in a
>> gui. There's an example of it at
>>
>>   https://scratch.mit.edu/
>>
>> I can see some value in it for tiny tots perhaps but, otherwise,
>> aargh!
>
> My son's in secondary school in Perth, Scotland. Everybody did Scratch
> and Build Your Own Blocks. He chose to continue with computing, and is
> thus now on VisualBasic and will be doing Java at some point. How
> typical that is I've no idea.

Do you mind my asking what age range they teach Scratch to? No need to 
reply if you'd rather not. I just wondered if it was at the start of 
Secondary School - age about 11 or 12, perhaps, or do they continue to 
get pupils to use it up to age 16? By "chose to" I presume you mean that 
he went on to VB at around 14 or 15 but perhaps not.

...


> They don't even
> put those nice Open University lectures on television in the early
> mornings any more

True, and before they stopped I remember the type of OU programme 
changing from arguably dull but interesting because of being detailed 
and informative to jazzy and slick but bland and lacking info. It seemed 
to me that the programmes became more to motivate than to educate. Just 
my view.

> and if one looks at local night school courses it's
> all pottery and whatnot. In computing we're lucky that at least decent
> self-study books for practical programming can be bought and one
> typically doesn't need expensive resources to be able to work on them.

Books are available. Over the years I have bought scores of them. With 
some cheap second-hand gems (and without meaning to boast but just to 
relate) I have a personal library of computing books which is vastly 
better than any modern library I have been in.

...

> No Python books does sound pretty sorry. Even Perth has "Beginning
> Programming with Python for Dummies"

Well, that title speaks volumes (no pun intended) about the level of 
information library users are exposed to today.

> (albeit only one copy that is
> "awaiting shelving", not yet reservable) though no obvious O'Reilly.
> (I've not studied Python myself, mostly I bluff my way through it 
> based
> on what I know of OOP and Perl and Haskell.)

From memory, Python in a Nutshell by Alex Martelli is a good source of 
quick info if you already know how to program.

James

[toc] | [prev] | [next] | [standalone]

#5823

"James Harris" <james.harris.1@gmail.com> writes:

> "Mark Carroll" <mtbc@bcs.org> wrote in message 
> news:87io7obkht.fsf@ixod.org...
(snip)
>> My son's in secondary school in Perth, Scotland. Everybody did Scratch
>> and Build Your Own Blocks. He chose to continue with computing, and is
>> thus now on VisualBasic and will be doing Java at some point. How
>> typical that is I've no idea.
>
> Do you mind my asking what age range they teach Scratch to? No need to 
> reply if you'd rather not. I just wondered if it was at the start of 
> Secondary School - age about 11 or 12, perhaps, or do they continue to 
> get pupils to use it up to age 16? By "chose to" I presume you mean that 
> he went on to VB at around 14 or 15 but perhaps not.

Yes, he started Scratch at the start of secondary school. After the
first two years of school, at least in Perthshire, they get to make some
choices about what they study; those who chose computing got to continue
programming in their third year with VisualBasic instead. He doesn't
know what happens to the children who don't choose computing, but he
mentions some computing part of the "enterprise" track (there's more
emphasis on business than when I were a lad) that is mysterious to us.

There are some "code clubs" that teach Scratch toward the end of some
primary schools around here, which I think might largely be a volunteer
effort; a colleague at work disappears some afternoons to run one at a
primary school.

>> They don't even
>> put those nice Open University lectures on television in the early
>> mornings any more
>
> True, and before they stopped I remember the type of OU programme 
> changing from arguably dull but interesting because of being detailed 
> and informative to jazzy and slick but bland and lacking info. It seemed 
> to me that the programmes became more to motivate than to educate. Just 
> my view.

Many documentaries got weird too. Just recently I watched some BBC thing
about waves. We touched the tip of some interesting nuggets, but half of
it veered on to the presenter saying how dumb he'd been about waves, how
it all connected personally with his late mother, and chatting to a
philosopher about how waves are like people. Some years ago there was
some series about ancient cities or somesuch that was positively
unwatchable given its focus instead on the personal journey of those who
had made the discoveries (what those discoveries were, I didn't have the
patience to find out).

(snip)
> From memory, Python in a Nutshell by Alex Martelli is a good source of 
> quick info if you already know how to program.

Thanks. It's been interesting to glimpse how it changes over time, e.g.,
Unicode handling, also how there's enough Python 2.x investment that it
looks like it will be supported for a long time yet despite Python 3.

-- Mark

[toc] | [prev] | [next] | [standalone]

#5824

"Mark Carroll" <mtbc@bcs.org> wrote in message 
news:87io7nhqjh.fsf@ixod.org...
> "James Harris" <james.harris.1@gmail.com> writes:
>
>> "Mark Carroll" <mtbc@bcs.org> wrote in message
>> news:87io7obkht.fsf@ixod.org...

...

>>> They don't even
>>> put those nice Open University lectures on television in the early
>>> mornings any more
>>
>> True, and before they stopped I remember the type of OU programme
>> changing from arguably dull but interesting because of being detailed
>> and informative to jazzy and slick but bland and lacking info. It 
>> seemed
>> to me that the programmes became more to motivate than to educate. 
>> Just
>> my view.
>
> Many documentaries got weird too. Just recently I watched some BBC 
> thing
> about waves. We touched the tip of some interesting nuggets, but half 
> of
> it veered on to the presenter saying how dumb he'd been about waves, 
> how
> it all connected personally with his late mother, and chatting to a
> philosopher about how waves are like people. Some years ago there was
> some series about ancient cities or somesuch that was positively
> unwatchable given its focus instead on the personal journey of those 
> who
> had made the discoveries (what those discoveries were, I didn't have 
> the
> patience to find out).

Your reaction sounds very familiar to what mine has been when seeing 
similar programmes. I don't really know if it's just me getting older - 
all generations tend to lose touch to a certain extent with the world as 
it changes - or if programme makers really do miss opportunities to 
produce something informative that would still appeal to many.

>> From memory, Python in a Nutshell by Alex Martelli is a good source 
>> of
>> quick info if you already know how to program.
>
> Thanks. It's been interesting to glimpse how it changes over time, 
> e.g.,
> Unicode handling, also how there's enough Python 2.x investment that 
> it
> looks like it will be supported for a long time yet despite Python 3.

Yes, there is still a lot of resistance to going to Python 3. It is 
probably better overall but a little too different for many tastes.

BTW, ISTM that the code 'ninjas' like to play with the introspection 
facilities of Python and produce obfuscated code. *If* that too-clever 
complexity is avoided then Python can be a great language.

James

[toc] | [prev] | [next] | [standalone]

#5829

In article <msfioj$i8c$1@dont-email.me>, james.harris.1@gmail.com 
says...
> 
> "J. Clarke" <j.clarke.873638@gmail.com> wrote in message 
> news:MPG.30401c0c36885213989ac3@news.eternal-september.org...
> >I was talking to a recent computer science graduate today.  He had 
> >never
> > seen C, never seen Fortran, had no notion of a shell script or batch
> > file, had never heard of ActiveX or .net or gcc.
> 
> What *had* they taught him? There must have been something.
> 
> > So what does somebody spend four years doing in a modern university
> > computer science program?
> 
> AFAIK from online university courses I have perused universities do 
> teach programming properly (though maybe not as diversely as they could) 
> but schools (i.e. for up to 16 or 18 years of age) seem to have 
> abandoned proper computer education completely. ISTM that schools teach 
> children how to *use* computers, especially computer applications, 
> rather than how to control a computer; they focus on things like 
> spreadsheets and word processors. They are probably more keen to ensure 
> that kids are not frightened by computers, or perhaps it is the 
> teachers' own fear that they are trying to overcome.
> 
> From things I have seen on TV recently it seems that when educators do 
> teach kids to program it is using things like Scratch which allows 
> programs to be composed by dragging constructs and dropping them in a 
> gui. There's an example of it at
> 
>   https://scratch.mit.edu/
> 
> I can see some value in it for tiny tots perhaps but, otherwise, aargh!
> 
> Perhaps children don't have, or aren't expected to have, the same 
> attention span and educators think they will get bored if they have to 
> type something that won't give them a GUI or make sounds. Printing the 
> numbers from 1 to 10 is no longer exciting, eh!
> 
> I thought to reply to this partly as I have seen a similar amazing (not 
> good-amazing) change in the books on computing available in local 
> libraries. When I was young libraries had books on programming languages 
> and how to write things like compilers.
> 
> My local library, now, has books like a beginner's guide to spreadsheets 
> but when I looked for a book on programming, at least in Python, they 
> had absolutely nothing. Even more incredibly, expanding the search to 
> all the libraries in the same council-controlled area also resulted in 
> *no* books whatever on programming in Python. And yet a search for 
> programming jobs locally brings up just now 75 jobs mentioning Python 
> within 10 miles of London advertised in the last week.

Just a note on libraries but I spent a year working in one recently, the 
main city library for the state capital of a state in New England.  
There was a policy "no textbooks", with even a Schaums Outline 
classified as a "textbook".  Nobody would tell me the reason for the 
policy.

[toc] | [prev] | [next] | [standalone]

#5840

"James Harris" <james.harris.1@gmail.com> wrote in message 
news:msfioj$i8c$1@dont-email.me...
>
> "J. Clarke" <j.clarke.873638@gmail.com> wrote in message 
> news:MPG.30401c0c36885213989ac3@news.eternal-september.org...
>>I was talking to a recent computer science graduate today.  He had never
>> seen C, never seen Fortran, had no notion of a shell script or batch
>> file, had never heard of ActiveX or .net or gcc.
>
> What *had* they taught him? There must have been something.
>
>> So what does somebody spend four years doing in a modern university
>> computer science program?
>
> AFAIK from online university courses I have perused universities do teach 
> programming properly (though maybe not as diversely as they could) but 
> schools (i.e. for up to 16 or 18 years of age) seem to have abandoned 
> proper computer education completely. ISTM that schools teach children how 
> to *use* computers, especially computer applications, rather than how to 
> control a computer; they focus on things like spreadsheets and word 
> processors. They are probably more keen to ensure that kids are not 
> frightened by computers, or perhaps it is the teachers' own fear that they 
> are trying to overcome.
>
> From things I have seen on TV recently it seems that when educators do 
> teach kids to program it is using things like Scratch which allows 
> programs to be composed by dragging constructs and dropping them in a gui. 
> There's an example of it at
>
>  https://scratch.mit.edu/
>

There is another "drag and drop" language from the University of California 
at Berkeley called Snap:

http://snap.berkeley.edu/

-- 

numerist at aquaporin4 dot com

[toc] | [prev] | [standalone]

Page 3 of 3 — ← Prev page 1 2 [3]

Back to top | Article view | comp.programming

csiph-web