hello everyone,
I am writing this email only because i have 2 very
simple questions. After asking my questions i will explain them a
little bit so that your answer can be personalised according to my
purpose, understanding-level and choices.
Q: should i start SCHEME, LISP (& if LISP then which LISP) or RUBY?
Q; which books (for RUBY i already have the answer)?
1. presently i am learning through HTDP. i am working at PART-2 for now
and within next 3 months i will finish this book. In the "preface" of
the book it is very clearley written :
-- HTDP does not teach "how to programme in scheme" but rather it
teaches students "how to design programmes" hence for this purpose it
uses only a small number of scheme constructs and a dozen or so basic
scheme functions. someone who wishes to use scheme a sa tool will need
to read additional material--
that's ok because i did not pick that book for learning scheme. anyway,
from experience with this book i know that by the time i will finish
this book i will have a good knowledge of scheme language (around 40%).
2. now since i will have a good amount of knowledge of how scheme
language works i will want to know whether i should start scheme or
LISP or RUBY, if LISP then which LISP? ( i love LISP & RUBY more than
SCHEME ) i prefer a language which has greater prectical orientation &
that is the one reason i am posting this question since i do not know.
2nd reason is I am not so good at Mathematics. i am Bachelor of
Science, so i know more Maths than other people but still i am not so
good. On the contray i love to programme & enjoying my time through
HTDP.
3. 3rd reason is i want to start working on real-life software-projects
as soon as possible because my father is going to be retired by the end
of 2007 hence i need to start earning before time is over. also
"professional life" is exactly opposite to "student life" . from
experience as a "salesman" i know that student life is 'dream-world',
it is not real, it is not intended to be.
4. i have a friend in England he is coming to INDIA in march month, by
that time i will be on the last parts of the HTDP. he asked me if i
want to have any more books (that is how i had HTDP). it saves my
'shipment expenses'. that way i can save my money. pretty much for a
jobless person. that is why i am posting this question now, rather than
wainting to finish HTDP.
so i sum-up
1.will go through HTDP hence 40% of scheme languge is done there.
(so will not be a newbie like at present)
2. a language which is used in real-life software projects.
3. not so good at Maths but love to programme.
4. myself has less time to build mylife. must stand independent by
the end of 2007.
& in the end what books to follow?
i will really appreciate your time if you may spend some.
"arnuld"
...snip
>
> 3. 3rd reason is i want to start working on real-life software-projects
> as soon as possible because my father is going to be retired by the end
> of 2007 hence i need to start earning before time is over. also
> "professional life" is exactly opposite to "student life" . from
> experience as a "salesman" i know that student life is 'dream-world',
> it is not real, it is not intended to be.
>
If money is really needed by that time, perhaps you should aim for a
ordinary job first? I don't know how much experience you have with
coding like a pro. But you are not alone if you start with a
non-computer-programming job first.
My argument stands on the fact that it might not be easy to go from 0
to pro in 2 years. Being able to solve stuff in programming as a hobby
is different than real-life programming tasks. But it is a start!
Leif Dyvik wrote:
> ...snip
>
>> 3. 3rd reason is i want to start working on real-life software-projects
>> as soon as possible because my father is going to be retired by the end
>> of 2007 hence i need to start earning before time is over. also
>> "professional life" is exactly opposite to "student life" . from
>> experience as a "salesman" i know that student life is 'dream-world',
>> it is not real, it is not intended to be.
>>
>
> If money is really needed by that time, perhaps you should aim for a
> ordinary job first? I don't know how much experience you have with
> coding like a pro. But you are not alone if you start with a
> non-computer-programming job first.
>
Oh! what awesome advice. Good thing Bill Gates and Steve Jobs took
you up on it.
Wade
> myself has less time to build mylife. must stand independent by
> the end of 2007.
Two years is a long time. I wouldn't feel like that was a lot of
pressure, but maybe it's different in India.
> 2. now since i will have a good amount of knowledge of how scheme
> language works i will want to know whether i should start scheme or
> LISP or RUBY, if LISP then which LISP?
> ( i love LISP & RUBY more than
> SCHEME ) i prefer a language which has greater prectical orientation &
> that is the one reason i am posting this question since i do not know.
Practical in what sense...
If you have a difficult project you are trying to finish, then you have
come to the right place. Time to learn Lisp. Maybe you can convince
someone to pay you for that project, and maybe they won't care that you
solved the problem in Lisp.
But your practical issue is just to get a job. I think it would be
easier with Ruby. Recently, I hear that there is a lot of demand for
Ruby on Rails developers who can produce "Web v2.0" online
applications.
If I were desperate to get a programming job next month, I would spend
a few days learning as many Rails tricks as I could (ignore the actual
Ruby language - you mostly don't need that) then start marketting
myself as a Rails expert. I know someone who has been doing this for a
few months, so I would ask him for some tips. Fortunately, I am not so
desperate. That is why you see me on comp.lang.lisp instead :-)
If you go the Ruby/Rails route, don't forget about Lisp. Lisp is the
better language, and you should plan on coming back after Web 2.0 goes
through the same boom and bust that 1.0 did. Of course, you have two
years. Maybe you are not so desperate and can just focus on becoming a
good programmer for a while. While you are picking up basic programming
skills, any language can help you on that path. Lisp in particular is a
particularly good choice because it supports a high level of
abstraction without unnecessary complexity.
>If I were desperate to get a programming job next month, I would spend
>a few days learning as many Rails tricks as I could (ignore the actual
>Ruby language - you mostly don't need that) then start marketting
>myself as a Rails expert. I know someone who has been doing this for a
>few months, so I would ask him for some tips. Fortunately, I am not so
>desperate. That is why you see me on comp.lang.lisp instead :-)
On the topic of Rails... can anyone recommend a Lisp library that would
make a good rival for Rails?
There was a Rails presentation at a local Linux Users Group meeting,
and I am going to give a Lisp presentation at the same LUG. I would
like to be able to demonstrate that Lisp can match the rapid
prototyping speed of Rails, preferably without the autogeneration of
1000s of lines of code (as is common in Rails) which would need to be
maintained later. I am imagining spending ten minutes on an
introduction to Lisp, then do twenty minutes of web programming. While
there will be a bit of teaching, I am mostly looking for "Wow," with
the possibility of following up later. I have no web programming
experience, but I am confident that I can pick up the basics in good
time. The options I've seen so far are AllegroServe, KPAX, TBNL, and
CL-HTTP, but I don't know much about any of them so far.
Another issue is database access. There is probably a bit of that built
in to some of the above web programming libraries, but what is a good
way to access a database such as MySQL? The emphasis is on quick and
easy (for setup and programming), not on performance. If it's simple
enough, it would be great to show database setup on stage. The ease of
database setup is one of the big selling points of Rails (less code
than most frameworks spend doing XML sit-ups...), but the Rails
presenter had to skip this step entirely due to the time limitation
(here's how we create a Rails application "rails myapp" ... and here's
the finished product). I can have the libraries installed in advance,
but aside from that I'd really like to demonstrate incremental
development.
As ealier posters said, if you want to earn money then learn more
commercially widespread languages like C# .net, or php and mysql. That
kind of stuff is vastly more in demand than lisp and scheme. Also
consider C/C++.
But once you're earning money and have the security to investigate lisp
I would start with Practical Common Lisp and also look at the Structure
and Interpretation of Computer Programs, which teaches scheme (and
programming) and is free online if you google for it. Same goes for "On
Lisp" which is also good but gets advanced quickly.
Justin.
Eric Lavigne <············@gmail.com> wrote:
+---------------
| Another issue is database access. There is probably a bit of that built
| in to some of the above web programming libraries, but what is a good
| way to access a database such as MySQL? The emphasis is on quick and
| easy (for setup and programming), not on performance.
+---------------
Well, you should definitely start by looking at
<http://www.cliki.net/database> and especially
at <http://www.cliki.net/CLSQL>, but since I
personally greatly prefer PostgreSQL to MySQL,
I find <http://www.cliki.net/Pg> [a direct socket
interface to PostgreSQL] to be all I need, e.g.:
> (defun pq (query)
"PQ -- Does quick & dirty PostgreSQL query to my private
database, and then releases the SQL connection.
Success ==> (values results nil) ; list of rows+1 lists of strings
Fail ==> (values nil error) ; SQL-ERROR condition object"
(handler-case
(with-pg-connection (conn "rpw3" "rpw3")
;; single query needs no transaction
(let* ((res (pg-exec conn query))
(cols (mapcar #'car (pg-result res :attributes)))
(rows (pg-result res :tuples)))
(values (cons cols rows) nil)))
(error (cc)
(values nil cc))))
> (my-pq "SELECT * FROM toy WHERE c2 = 'video'")
(("season" "media" "title" "seq")
("xmas" "video" "The Grinch who Stole Christmas" 4)
("xmas" "video" "Home Alone" 6))
NIL
> (my-pq "SELECT season, count(season) FROM toy GROUP BY season")
(("c1" "count") ("easter" 2) ("fall" 1) ("spring" 1)
("summer" 1) ("winter" 1) ("xmas" 3))
NIL
> (my-pq "SELECT foo FROM toy")
NIL
#<POSTGRESQL::BACKEND-ERROR {58932CC5}>
>
-Rob
-----
Rob Warnock <····@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
Eric Lavigne wrote:
> I would
> like to be able to demonstrate that Lisp can match the rapid
> prototyping speed of Rails, preferably without the autogeneration of
> 1000s of lines of code (as is common in Rails) which would need to be
> maintained later.
You would think that lispers would be slightly more sensitive to common
misconceptions about a language. What you just wrote is the equivalent
of saying "preferably without the need to represent every data
structure as a list (as is common in lisp)." Rails is no more about
autogenerated code than lisp is about lists; most non-trivial rails
code doesnt rely on scaffolding.
As for your demonstration, I wish you luck, but for "hook a database to
the web" type apps, you aren't going to find anything in lisp that's
comparable to rails. Better to pick a different domain.
From: Bill Atkins
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <87vewbimqu.fsf@rpi.edu>
··············@gmail.com" <·············@gmail.com> writes:
> Eric Lavigne wrote:
>> I would
>> like to be able to demonstrate that Lisp can match the rapid
>> prototyping speed of Rails, preferably without the autogeneration of
>> 1000s of lines of code (as is common in Rails) which would need to be
>> maintained later.
>
> You would think that lispers would be slightly more sensitive to common
> misconceptions about a language. What you just wrote is the equivalent
> of saying "preferably without the need to represent every data
> structure as a list (as is common in lisp)." Rails is no more about
> autogenerated code than lisp is about lists; most non-trivial rails
> code doesnt rely on scaffolding.
I don't know what you mean. Rails is very much about code generation.
Code generation is what makes ActiveRecord go; it's how ERB templates
are run. Generating code is at the heart of what makes Rails
convenient - the programmer doesn't have to do tedious things like
specify mappings between relational tables and classes or write custom
validation functions for common patterns. Everything in the script/
folder in a Rails app is a straight-up code generator. Maybe I'm
looking at this incorrectly.
> As for your demonstration, I wish you luck, but for "hook a database to
> the web" type apps, you aren't going to find anything in lisp that's
> comparable to rails. Better to pick a different domain.
Haha. You sound pretty sure about that. Wherefore the confidence,
codykoeninger? What makes you think Rails is the final word on web
frameworks?
--
Bill Atkins
Bill Atkins wrote:
> I don't know what you mean. Rails is very much about code generation.
> Code generation is what makes ActiveRecord go; it's how ERB templates
> are run. Generating code is at the heart of what makes Rails
> convenient - the programmer doesn't have to do tedious things like
> specify mappings between relational tables and classes or write custom
> validation functions for common patterns. Everything in the script/
> folder in a Rails app is a straight-up code generator. Maybe I'm
> looking at this incorrectly.
Yes, I would say you are looking at this incorrectly. The examples of
'code generation' you cite do not need to be maintained by hand, any
more than the expansion of macros need to be maintained by hand. The
OP (well, offtopic OP anyway) specifically mentioned maintaining 1000s
of lines of autogenerated code; this seemed to be a swipe at rail's
abilty to generate scaffolding code (which could then be tweaked and
maintained by hand), not the ORM layer.
> Haha. You sound pretty sure about that. Wherefore the confidence,
> codykoeninger? What makes you think Rails is the final word on web
> frameworks?
You're putting words in my mouth. I didn't say rails is the final word
on web frameworks. If I actually thought that, I would be hanging out
with all the web2.0 folks rather than on c.l.l. I said if you're
trying to demonstrate the ability to whip out a 'hook a database to the
web' type of app, you wont find anything lisp-based that will beat
rails. Clearly there are more complex application areas; this is why I
said pick a different domain.
As for why I said you wont find anything lisp based that will beat
rails in that narrow domain . . . as an experiment, pick a neutral
observer - say a friend who doesnt know much about programming - and
have them look at the rails book as compared to, say, the UCW (lack of)
documentation. Or show them the following two movies:
http://versions.tech.coop/movies/lol.mov
http://media.rubyonrails.org/video/rails_take2_with_sound.mov
Then ask them which one they'd rather throw a web app together in.
I like lisp. Drew seems like a nice guy, as does bill clementson. But
that's some bad press . . .
··············@gmail.com" <·············@gmail.com> writes:
> As for why I said you wont find anything lisp based that will beat
> rails in that narrow domain . . . as an experiment, pick a neutral
> observer - say a friend who doesnt know much about programming - and
> have them look at the rails book as compared to, say, the UCW (lack of)
> documentation. Or show them the following two movies:
I went to a good presentation of Ruby on Rails a few weeks ago. The
presenter was very good and showed some demos and code. However, I
found the whole thing to be a quite a kludge. Personally I find the
combination of Webactions and CLSQL much more elegant and simple. Here
I can define a database in Lisp, and have the HTML document use Common
Lisp Server Pages to update forms from instances of database objects
and the action functions update the objects and the database from
entries given by the user.
Petter
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?
From: Bill Atkins
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <87oe228tu3.fsf@rpi.edu>
··············@gmail.com" <·············@gmail.com> writes:
> Bill Atkins wrote:
>> I don't know what you mean. Rails is very much about code generation.
>> Code generation is what makes ActiveRecord go; it's how ERB templates
>> are run. Generating code is at the heart of what makes Rails
>> convenient - the programmer doesn't have to do tedious things like
>> specify mappings between relational tables and classes or write custom
>> validation functions for common patterns. Everything in the script/
>> folder in a Rails app is a straight-up code generator. Maybe I'm
>> looking at this incorrectly.
>
> Yes, I would say you are looking at this incorrectly. The examples of
> 'code generation' you cite do not need to be maintained by hand, any
> more than the expansion of macros need to be maintained by hand. The
> OP (well, offtopic OP anyway) specifically mentioned maintaining 1000s
> of lines of autogenerated code; this seemed to be a swipe at rail's
> abilty to generate scaffolding code (which could then be tweaked and
> maintained by hand), not the ORM layer.
The code generated by the script/ files certainly does have to be
maintained by hand.
>> Haha. You sound pretty sure about that. Wherefore the confidence,
>> codykoeninger? What makes you think Rails is the final word on web
>> frameworks?
>
> You're putting words in my mouth. I didn't say rails is the final word
> on web frameworks. If I actually thought that, I would be hanging out
> with all the web2.0 folks rather than on c.l.l. I said if you're
> trying to demonstrate the ability to whip out a 'hook a database to the
> web' type of app, you wont find anything lisp-based that will beat
> rails. Clearly there are more complex application areas; this is why I
> said pick a different domain.
I think I owe you an apology. I had never seen you post here before,
so I assumed you were just a Rails fanboy who'd happened to drift into
c.l.l. I have had comically frustrating experiences with Rails users
who decide to make judgments on Lisp (e.g. "Lisp only provides a
subset of Ruby's power" or "Lisp people ignore 90% of the Ruby
features," both in the infamous "Ruby is an acceptable LISP [sic]"
discussion). I reacted hastily, and probably sounded a lot harsher
than I meant to.
In any case, I don't think there's a problem with Common Lisp having
good databse bindings, even if there are "more complex application
areas."
--
Bill Atkins
> As for why I said you wont find anything lisp based that will beat
> rails in that narrow domain . . . as an experiment, pick a neutral
> observer - say a friend who doesnt know much about programming - and
> have them look at the rails book as compared to, say, the UCW (lack of)
> documentation. Or show them the following two movies:
>
> http://versions.tech.coop/movies/lol.mov
> http://media.rubyonrails.org/video/rails_take2_with_sound.mov
>
> Then ask them which one they'd rather throw a web app together in.
I'm not really a neutral observer, but if I had to deliver something
tomorrow it would be written in Rails. It's well documented and easy to
install, and the generate scripts make it easy to hit the ground
running.
> I said if you're
> trying to demonstrate the ability to whip out a 'hook a database to the
> web' type of app, you wont find anything lisp-based that will beat
> rails. Clearly there are more complex application areas; this is why I
> said pick a different domain.
Any suggestions for an alternative domain? 30 minute time limit for the
presentation. I'd prefer to do it on something web related, mostly
because I'd like to learn about that application area myself, but
completely different ideas are also welcome.
I agree with Petter Gustad. At the moment I am working with WebActions
which is really nice. In theory you won't ever have to write any html.
You can concentrate on the Lisp code and the webdesigner manages the
html side (well, you probably have both roles in one person, but
anyway, it is a clear and nice separation).
Give WebActions a try and install the cvs version of portable
AllegroServe.
André
--
André Thieme wrote:
> I agree with Petter Gustad. At the moment I am working with WebActions
> which is really nice. In theory you won't ever have to write any html.
> You can concentrate on the Lisp code and the webdesigner manages the
> html side (well, you probably have both roles in one person, but
> anyway, it is a clear and nice separation).
> Give WebActions a try and install the cvs version of portable
> AllegroServe.
I'll do that. Probably not until next week, though, as I have too many
school-related responsibilities right now.
So for now the lineup looks like WebActions/AllegroServe, CLSQL, and
PostgreSQL.
"Eric Lavigne" <············@gmail.com> writes:
> So for now the lineup looks like WebActions/AllegroServe, CLSQL, and
> PostgreSQL.
That's my choice as well. I'm running CMUCL under Linux. Here's a neat
trick on how to start the webserver under screen from a crontab at
boot time:
$crontab -l -e lisp
gives:
@reboot $HOME/src/web/rc.d/startpaserve >> $HOME/src/web/rc.d/startpaserve.log 2>&1
startpaserve contains:
echo reboot at `date`, starting portable allegroserve under screen
screen -D -m -S aserve /usr/bin/lisp -eval "(asdf:operate 'asdf:load-op :web)" -eval "(mp:make-process #'web:start-web-server)" -eval '(mp::startup-idle-and-top-level-loops)'
I have a web.asd which will build my server. A function called
start-web-server will call net.aserve:start. I start the process as a
user lisp and listen to port 8080 then use ipchains to forward port 80
to 8080. Any time after a reboot I can call "screen -r" to connect to
the REPL. Then do:
(asdf:operate 'asdf:load-op :swank) ;; only once, and
(swank:create-swank-server 4005) ;; to start a slime listener
Then launch emacs and do M-x slime-connect and interactively compile
new functions, inspect data left by users using the server, and all
other wonderful SLIME things...
Petter
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?
"Eric Lavigne" <············@gmail.com> writes:
> I'm not really a neutral observer, but if I had to deliver something
> tomorrow it would be written in Rails. It's well documented and easy to
Neither am I, but if I had to deliver something tomorrow and something
which has to be maintained and developed further the day after
tomorrow, it would be written in Lisp using AllegroServe/Webactions.
Petter
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?
On Mon, 23 Jan 2006 05:21:24 +0100, ·············@gmail.com
<·············@gmail.com> wrote:
>
> Eric Lavigne wrote:
>> I would
>> like to be able to demonstrate that Lisp can match the rapid
>> prototyping speed of Rails, preferably without the autogeneration of
>> 1000s of lines of code (as is common in Rails) which would need to be
>> maintained later.
>
> You would think that lispers would be slightly more sensitive to common
> misconceptions about a language. What you just wrote is the equivalent
> of saying "preferably without the need to represent every data
> structure as a list (as is common in lisp)." Rails is no more about
> autogenerated code than lisp is about lists; most non-trivial rails
> code doesnt rely on scaffolding.
>
> As for your demonstration, I wish you luck, but for "hook a database to
> the web" type apps, you aren't going to find anything in lisp that's
> comparable to rails. Better to pick a different domain.
>
Have you seen the KPAX movie?
It dosn't demonstrate a lot of features like lisp server pages
and database access with CLSQL but it gives a idea.
http://homepage.mac.com/svc/LispMovies/lisp-movie-2-reddit.mov
http://homepage.mac.com/svc/RebelWithACause/
--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
> Have you seen the KPAX movie?
> It dosn't demonstrate a lot of features like lisp server pages
> and database access with CLSQL but it gives a idea.
> http://homepage.mac.com/svc/LispMovies/lisp-movie-2-reddit.mov
> http://homepage.mac.com/svc/RebelWithACause/
>
I've seen it, and KPAX is most likely going to be in my toolset. I
haven't had a chance to install it or go through the documentation yet.
The autogenerated docs look mostly useless, except perhaps as a quick
reference for experienced users. Hopefully the examples will be enough
to get me started, though.
I hadn't thought about lisp server pages, since I was quite happy to
write the whole thing in Lisp. On the other hand, a non-Lisp audience
may find LSPs to be a bit less overwhelming. They also have the
advantage of letting people get involved without understanding Lisp. I
need to look into this...
On Tue, 24 Jan 2006 04:22:59 +0100, Eric Lavigne <············@gmail.com>
wrote:
>
>> Have you seen the KPAX movie?
>> It dosn't demonstrate a lot of features like lisp server pages
>> and database access with CLSQL but it gives a idea.
>> http://homepage.mac.com/svc/LispMovies/lisp-movie-2-reddit.mov
>> http://homepage.mac.com/svc/RebelWithACause/
>>
>
> I've seen it, and KPAX is most likely going to be in my toolset. I
> haven't had a chance to install it or go through the documentation yet.
> The autogenerated docs look mostly useless, except perhaps as a quick
> reference for experienced users. Hopefully the examples will be enough
> to get me started, though.
>
> I hadn't thought about lisp server pages, since I was quite happy to
> write the whole thing in Lisp. On the other hand, a non-Lisp audience
> may find LSPs to be a bit less overwhelming. They also have the
> advantage of letting people get involved without understanding Lisp. I
> need to look into this...
>
The real documentation is in the paper Nicky Peters wrote:
"Can a 40 year old Computer Language do Web Applications?
Using Common Lisp to build Web Applications."
http://homepage.mac.com/nickypeeters/index.old.html
'Chapter 3 KPAX application framework' and from there on..
Which describes the scope and to some extent interface of KPAX.
> I hadn't thought about lisp server pages, since I was quite happy to
> write the whole thing in Lisp. On the other hand, a non-Lisp audience
> may find LSPs to be a bit less overwhelming. They also have the
> advantage of letting people get involved without understanding Lisp. I
> need to look into this...
Not neccesairly a good idea. Many 'real' web cites are a collaboration
between
designers and programmers. Lisp Server Pages allow a designer to
design the page in (X)HTML and the programmer to then add behaviour to the
page.
I'd say if all you are doing is print a web page and add a few values
LSP might be for you. It is certainly a lot easier to test the CSS
against a webpage that to have to grind a page every time.
--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
> > I would
> > like to be able to demonstrate that Lisp can match the rapid
> > prototyping speed of Rails, preferably without the autogeneration of
> > 1000s of lines of code (as is common in Rails) which would need to be
> > maintained later.
>
> You would think that lispers would be slightly more sensitive to common
> misconceptions about a language. What you just wrote is the equivalent
> of saying "preferably without the need to represent every data
> structure as a list (as is common in lisp)."
I probably have lots of misconceptions, but I didn't just pull them off
someone's webpage :-)
I own a copy of "Agile Web Development with Rails." I've read the first
150 or so pages, followed along with all the exercises, then gone on to
build a couple of small web apps myself (just practice, nothing good
enough to publish). I have also seen several presentations by Rails
developers, including the rails_take2 on rubyonrails.org. So, if I have
any misconceptions regarding Rails, they are based experience with
Rails itself and with the recommended practices that I have found in
its documentation.
The pattern I keep seeing is "script/generate ..." followed by hundreds
of files getting autogenerated and dumped in my working directory.
Maybe there's a way to avoid maintaining these by hand? This is just a
feeling, but after the 4th or 5th time doing this I started imagining
that I was misusing a macro out of some library by macroexpanding it
right into my source code (then tweaking the result to better suit my
needs). Naturally, it was hard to stick with Rails when strange
thoughts like this were floating around in my head :-) so now I have to
find a way to do it in Lisp instead.
> Rails is no more about
> autogenerated code than lisp is about lists; most non-trivial rails
> code doesnt rely on scaffolding.
I certainly got the impression that scaffolding tends to get replaced
by hand-written code in a mature project (and to a lesser extent in new
projects). On the other hand, I am also under the impression that such
scaffolding is the big selling point of Rails. Maybe there's something
bigger that I haven't run across yet?
> As for your demonstration, I wish you luck, but for "hook a database to
> the web" type apps, you aren't going to find anything in lisp that's
> comparable to rails.
I was thinking that something like cl-prevalence might give me the edge
in this area. It's cheating a little bit, since it's not really a
database... My impression is that you give up the ability to interface
with non-lisp applications in return for having practically no database
setup issues. I haven't researched this thoroughly, though, so there's
a good chance I've got it completely backwards. Maybe it's better to
keep it simple with something like CLSQL, since database setup isn't so
hard anyway.
> Better to pick a different domain.
Almost certainly true :-) but I'll try anyway.
> As for your demonstration, I wish you luck
Much appreciated - I think I'm going to need it. Whether or not Lisp
turns out to be up to the task, my web programming skills are
practically non-existent. That makes it even more worthwhile, though,
as I expect to learn a lot.
"arnuld" <·······@gmail.com> writes:
> hello everyone,
>
> I am writing this email only because i have 2 very
> simple questions. After asking my questions i will explain them a
> little bit so that your answer can be personalised according to my
> purpose, understanding-level and choices.
>
>
> Q: should i start SCHEME, LISP (& if LISP then which LISP) or RUBY?
If Lisp, Common Lisp.
> Q; which books (for RUBY i already have the answer)?
For Common Lisp, I can't help but recommend my own, _Practical Common
Lisp_ which you can check out on the web at:
<http://www.gigamonkeys.com/book/>
Some other folks here seem to like it as well so maybe someone will
back me up on this. Or you can read some blurbs at:
<http://www.gigamonkeys.com/book/blurbs.html>
-Peter
--
Peter Seibel * ·····@gigamonkeys.com
Gigamonkeys Consulting * http://www.gigamonkeys.com/
Practical Common Lisp * http://www.gigamonkeys.com/book/
From: drewc
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <87hd7v5iru.fsf@rift.com>
Peter Seibel <·····@gigamonkeys.com> writes:
>
> For Common Lisp, I can't help but recommend my own, _Practical Common
> Lisp_ which you can check out on the web at:
>
> <http://www.gigamonkeys.com/book/>
>
> Some other folks here seem to like it as well so maybe someone will
> back me up on this.
Best ... Book ... Ever!
drewc
--
drewc at tech dot coop
arnuld wrote:
> Q: should i start SCHEME, LISP (& if LISP then which LISP) or RUBY?
Since you say you "love lisp" more than scheme, choose lisp. But the
difference between scheme and lisp is not as important.
> Q; which books (for RUBY i already have the answer)?
Finish HTDP. It will make you a better programmer no matter what the
language. Don't worry about real-world use of a particular language,
your understanding of problem solving will help you find employment.
After that, many of the books already mentioned are available free on
line -
"on lisp" http://paulgraham.com/onlisp.html ,
"practical common lisp" http://gigamonkeys.com/book/ ,
"structure and interpretation of computer programs"
http://mitpress.mit.edu/sicp/full-text/book/book.html
If you decide on Lisp and want to spend money on a book, I would
recommend:
"Paradigms of Artificial Intelligence Programming: Case Studies in
Common Lisp"
http://isbn.nu/1558601910
even if you have no interest in AI, it shows how to solve non-trivial
problems in a powerful way.
> so i sum-up
>
> 1.will go through HTDP hence 40% of scheme languge is done there.
> (so will not be a newbie like at present)
>
> 2. a language which is used in real-life software projects.
>
> 3. not so good at Maths but love to programme.
>
> 4. myself has less time to build mylife. must stand independent by
> the end of 2007.
>
> & in the end what books to follow?
>
> i will really appreciate your time if you may spend some.
>
> "arnuld"
well, thanks everyone here for such good advices and spending time on
my "stupid" mail. after reading all the 23 responses here i searched
entire "newsgroups" archives for choosing over a language and the best
advice i had is: "Common LISP", it took 5 days to ssearch last 18 year
archives and then read them all, taking print-outs and underlining
important points and then wrting out thge "extract" onto the back of
print-outs. i did a lot of work and this thing we call programming
boils down to "Common LISP".
anyway i am getting " a little bit frustrated with HtDP (@ htdp.org)
because i am doing it from last 8 months and still not able to get
beyond part-2 of the book, it has 8 parts in total. i think the problem
is that one needs to have an instructor if one wants to learn with
HtDP. i am not able to do that anymore, instead i want to do Common
LISP. trust me this is really great. i wonder how people wander
aimlessly in other languages thinking they are good programmers with
only C, JAVA and even SCHEME into their bags. (well, SICP is exception)
point is: newbie to programming wants to start in Common LISP.(i like
CLISP)
i am on my own, no teachers, no courses, no previous background of
programming, only mentor is ( are ) the people @ comp.lang.lisp, who,
from my experience are very talented and helpful.
where & how do i start in Common LISP?
( one more thing, i did not read them but heard about them, how about
"The little Schemer" and "The seasoned schemer"?. )
thanks a lot.
"arnuld"
"arnuld" <·······@gmail.com> writes:
> point is: newbie to programming wants to start in Common LISP.(i like
> CLISP)
>
> i am on my own, no teachers, no courses, no previous background of
> programming, only mentor is ( are ) the people @ comp.lang.lisp, who,
> from my experience are very talented and helpful.
>
> where & how do i start in Common LISP?
"Practical Common Lisp", an introduction to Common Lisp
by Peter Seibel, available at http://www.gigamonkeys.com/book/
and in dead-tree form from Apress (as of 11 April 2005).
A quick guide to getting set up to learn Common Lisp, the #lisp way:
http://www.unmutual.info/startingwithcl.html
http://www.cliki.net
http://www.cliki.net/Education
http://www.cliki.net/CLHS
> ( one more thing, i did not read them but heard about them, how about
> "The little Schemer" and "The seasoned schemer"?. )
Very good. I read "The Little Lisper" which should work better with CL.
--
__Pascal Bourguignon__ http://www.informatimago.com/
"Logiciels libres : nourris au code source sans farine animale."
> "Practical Common Lisp", an introduction to Common Lisp
> by Peter Seibel, available at http://www.gigamonkeys.com/book/
> and in dead-tree form from Apress (as of 11 April 2005).
>
> A quick guide to getting set up to learn Common Lisp, the #lisp way:
> http://www.unmutual.info/startingwithcl.html
>
> http://www.cliki.net
> http://www.cliki.net/Education
> http://www.cliki.net/CLHS
> > ( one more thing, i did not read them but heard about them, how about
> > "The little Schemer" and "The seasoned schemer"?. )
>
> Very good. I read "The Little Lisper" which should work better with CL.
>
> --
> __Pascal Bourguignon__ http://www.informatimago.com/
well, i tried "Practical Common Lisp" but did not like it. how about
this path:
tLS -> tSS -> ACL -> PAIP
tLS : the little schemer ; tSS : the seasoned schemer
ACL : ANSI ommon Lisp by paul Graham
PAIP : Paradigms of AI : case-studies in Common Lisp by Peter Norvig
what do you say about this?
"arnuld"
"arnuld" <·······@gmail.com> writes:
> well, i tried "Practical Common Lisp" but did not like it. how about
> this path:
>
> tLS -> tSS -> ACL -> PAIP
>
> tLS : the little schemer ; tSS : the seasoned schemer
>
> ACL : ANSI ommon Lisp by paul Graham
>
> PAIP : Paradigms of AI : case-studies in Common Lisp by Peter Norvig
>
>
> what do you say about this?
I say choose the path with a scenery you like.
--
__Pascal Bourguignon__ http://www.informatimago.com/
"You cannot really appreciate Dilbert unless you read it in the
original Klingon"
> I say choose the path with a scenery you like.
> __Pascal Bourguignon__ http://www.informatimago.com/
>
ok, i did lots of work and finally came up with 2 sceneries:
tLS -> tSS -> SaAoP -> SICP --> not decided untill now
or
tLS -> tSS -> SaAoP -> ACL -> PAIP -> start developing web-apps
where SaAop = scheme and art of programming by Springer & Friedman
may you tell me how much time the process -- tLS -> tSS -> SaAoP --
will take?
thanks
"arnuld"
From: Hrvoje Blazevic
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <ds4iie$2og$1@ss405.t-com.hr>
arnuld wrote:
>>I say choose the path with a scenery you like.
>
>
>>__Pascal Bourguignon__ http://www.informatimago.com/
>>
>
>
> ok, i did lots of work and finally came up with 2 sceneries:
>
> tLS -> tSS -> SaAoP -> SICP --> not decided untill now
>
> or
>
> tLS -> tSS -> SaAoP -> ACL -> PAIP -> start developing web-apps
>
> where SaAop = scheme and art of programming by Springer & Friedman
>
> may you tell me how much time the process -- tLS -> tSS -> SaAoP --
> will take?
>
> thanks
>
> "arnuld"
>
Hmm... Interesting. Some years ago, I chose precisely your first path
(with the addition of SSICS as the first (zeroth :-) book).
As for how long will it take, that's dependent mostly on you and your
background. I can only say how long it took me:
tLS 3 weeks (with chapter 9 working up a serious dent on my ego)
tSS 3 weeks
SaAoP 3 months
SICP 3 months
Looks like 7.5 months altogether, but in fact, it was spread over
several years. Your mileage will vary, because I'm *not* a
mathematician, *not* a computer scientist, nor studying to become one.
I'm doing this for fun, just killing time while working as a master
mariner. And not to forget... I'm getting a bit long in the tooth... 50+ :-)
I would suggest though, to add HtDP to your list (probably as a zeroth
book). HtDP before tLS, because (again) of chapter 9, which I took as a
sort of the test: whether to continue reading all those other books, or
simply roll over and take up drinking as a pastime, instead of reading
CS books :-)
-- Hrvoje
Hrvoje Blazevic wrote:
> I would suggest though, to add HtDP to your list (probably as a zeroth
> book). HtDP before tLS, because (again) of chapter 9, which I took as a
> sort of the test: whether to continue reading all those other books, or
> simply roll over and take up drinking as a pastime, instead of reading
> CS books :-)
>
> -- Hrvoje
well, it seems you did not read my previous post, 25th message on this
thread , it is on the 1st page ( as this thread has 2 pages now,
messages from 1-25 are on 1st page from 26 to this one are on 2nd ,
this is how google manages the newsgroups posts). have a look at that.
1st of all - i do not want any FLAME-WARS here. if anybody tries to
have a flame-war with me on this issue i will simply refuse to talk.
i literally hate HtDP as i am doing it from last 10 months and not able
to go beyond section 15 AND this is *not* a technical problem. this has
something to do with my thinking pattern and analysing abilities which
do not resonate with HtDP. after searching whole "groups.google.com"
for more than 15 days, sometimes 8 hours at a stretch, i came to know
that intorduction to programming has nothing to do with scheme, it can
also be done well using Common LISP or python or Ruby or may be
something else. SCHEME is prefered because it is easy to express many
excellent ideas of CS in SCHEME than in other languages but problem is
that i worked as SALESMAN for 2 years. YES you read it correctly,i do
not like SCHEME because i was a SALESMAN. being a salesman taught me
that student life is dream-life, it is not real, it is not intended to
be. for the 1st time i went to sell "personal loan" i encountered lots
of difficulties because all of my beliefes i learned as a student were
just broken & selling broke them because they do not fit in real life,
the hard-life. i have studied for my exams sometimes 8 hours at a
stretch (8 pm to 4 am in the morning) and i did it for more than 3
years but trust me that is not hard-work, if you compare to my 2 hours
selling experience.
that is thing i am taliking about - scheme is used and was created
especially for academic purposes. scheme was created to teach CS
concepts to graduate CS students .i am not saying that scheme can not
be used to create softwares. i know that PLT scheme can be used to
create Desktop Apps. even DrScheme itself is a programme written in
scheme.what i am saying is : scheme has those beliefs and those
feelings of student-life or academic life and i have experienced it as
a user of HtDP. i literally hate student-life, it is a birthplace of
many wrong ideas about real-life. please also search for Kent Pitman's,
Shriram's and Matthias's posts on the different threads of
comp.lang.lisp.
also my search concludes that Common LISP was created to solve
real-life problems , to solve very complex problems we face in
producing high-quality and highly-functional softwares (which has lots
of function and features) and is much more powerfull than SCHEME which
is a sort of academic (SICP is an exception & that happened
accidently). Common LISP is a sort of real-life thing.
i do not know about you and others, but i can guarntee you from my
experience as CS grduate and salesman that there will be lots of people
loving HtDP and that will be so because they have exactly opposite
thinking-style as compared to me. if someone goes beyond part-3 of the
book easily or with some effort and likes the style in which book is
written then i recommend he should continue to use HtDP. HtDP will be
the best thing ever happened for that person, i am sure about it. i do
not feel fit into their style and literally felt frustrated.
also HtDP and tLS are written in exactly opposite styles, so may be i
will love that book.
hence it all depends upon that individual and on what he wants to do
with his life and this is not a technical issue.
thanks for your time
"arnuld"
From: Hrvoje Blazevic
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <ds7fmc$6ld$1@ss405.t-com.hr>
arnuld wrote:
> Hrvoje Blazevic wrote:
>
>>I would suggest though, to add HtDP to your list (probably as a zeroth
>>book). HtDP before tLS, because (again) of chapter 9, which I took as a
>>sort of the test: whether to continue reading all those other books, or
>>simply roll over and take up drinking as a pastime, instead of reading
>>CS books :-)
>>
>>-- Hrvoje
>
>
> well, it seems you did not read my previous post, 25th message on this
> thread , it is on the 1st page ( as this thread has 2 pages now,
> messages from 1-25 are on 1st page from 26 to this one are on 2nd ,
> this is how google manages the newsgroups posts). have a look at that.
>
> 1st of all - i do not want any FLAME-WARS here. if anybody tries to
> have a flame-war with me on this issue i will simply refuse to talk.
Oops...
And just how did you get the idea that I'm trying to have a flame war
with you?
It is true that I did *not* read any previous posts in this thread...
just sort of glanced at the books you listed and the way you lined them
up, and realized that this was exactly the way I read them. That's why I
responded to *your* question, as to how long will it take to read them.
>
> i literally hate HtDP as i am doing it from last 10 months and not able
> to go beyond section 15 AND this is *not* a technical problem. this has
> something to do with my thinking pattern and analysing abilities which
> do not resonate with HtDP.
My first reaction to this was:
If you had to struggle for 10 months with first 15 sections of HtDP,
then your chances with tLS are less then "Little", but I realize that
this is overly rude. Therefore, following is a more moderate comment:
I will grant you that IMHO, first 8 sections of HtDP, are not very
interesting/exciting. However, the book does pick up the pace, and
becomes far more interesting after that. If you worked up to section 15,
you should have realized at least some of this.
The value of HtDP (at lest for me, after reading all those other books)
was that it puts a lot of concepts into perspective. It helped me to
align properly a lot of previously acquired, but somewhat scattered
knowledge.
However, the concepts that all those other books eventually deal with
are far more difficult than HtDP ever gets.
But, if you hate HtDP, then SSICS should work well as a replacement. It
*is* more fun to read then HtDP. Actually, I'm at a better position to
recommend this book to a beginner than any of the other books, because
this *was* my first Scheme book.
-- Hrvoje
1st of all thanks to TIM for sharing his valuable experience with me &
comp.lang.lisp:
1.) Others will disagree, but for me, the most important thing to
learn is to become familiar with data abstraction techniques and
writing algorithms.
2.) In the end, you will only learn by doing.
3.) Learn to debug using the best debugger in the world, that lump of
grey matter between your ears.
4.) Using a stepping/tracing debugger eliminates the need to really
understand
and think about what is going on, which is never a good thing.
Debuggers are very useful tools, but overkill at the learning stage.
5.) - Don't get caught up in doing things the way veryone else says you
ahve to do things. Look around, read different authors and find the
ones which work for you.
thanks TIM for your valuable advices.
---------- and now for Harvoje --------
> Oops...
> And just how did you get the idea that I'm trying to have a flame war with you?
i said *if anybody*. i did not say *Harvoje*. i said so because some
people try to start a flame war.
> I will grant you that IMHO, first 8 sections of HtDP, are not very interesting/exciting.
> However, the book does pick up the pace, and becomes far more interesting after
> that. If you worked up to section 15, you should have realized at least some of this.
YES, you are 100% right. and it speeds-up more after section 11.
> The value of HtDP (at lest for me, after reading all those other books) was that it puts
>a lot of concepts into perspective. It helped me to align properly a lot of previously
> acquired, but somewhat scattered knowledge.
i do know about that so can not comment on it.
> However, the concepts that all those other books eventually deal with are far more
> difficult than HtDP ever gets.
Amen to that. what i hated in HtDP is :
1.) - their too much focus on theoretical explanation of their "Design
Reciepes" sometime 4-5 pages long theoretical explanation. i could have
liked it if they spend 1and a 1/2 page of explanation and 3 and a 1/2
pages of codes of examples and exercises on small programmes, compilers
and interpreters.
2.) - their very clear (explicit) way of telling the design recipies
and recursions. i think i am not a kid, i am 25 year old man, hence if
someone gives me lots of examples of code and says: "find out
something", then i will like to figure out common-patterns and
different approaches the programmer have taken in creating those
programmes. in other words i do not like hand-holding. i can try to
find out myself and increase the sharpness of my analysing and critical
thinking.
> But, if you hate HtDP, then SSICS should work well as a replacement. It *is* more fun > to read then HtDP. Actually, I'm at a better position to recommend this book to a
> beginner than any of the other books, because this *was* my first Scheme book.
now what is SSICS? is it "simply scheme by Brian Harvey"?
thanks
and may anybody explain me the meaning of following words:
IMHO, OTOH, AFAIK and some others like these. aftera all, i am a
newbie.
"arnuld"
> and may anybody explain me the meaning of following words:
>
> IMHO, OTOH, AFAIK and some others like these. aftera all, i am a
> newbie.
Google knows the answer :-)
Google search (www.google.com):
IMHO OTOH AFAIK
From: Hrvoje Blazevic
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <dsd71v$nis$1@ss405.t-com.hr>
arnuld wrote:
>
> now what is SSICS? is it "simply scheme by Brian Harvey"?
>
Yes.
-- Hrvoje
> > now what is SSICS? is it "simply scheme by Brian Harvey"?
> Yes.
>
> -- Hrvoje
1.) why you said SSICS is better for me?
2.) is it really a *good book*. i mean, is it worth to spend money on
it?
if yes, then how about 1994 edition, 1999 edition is very costly. i
will buy the one prefered by you.
-- arnuld
From: Hrvoje Blazevic
Subject: Re: only 2 simple questions : a newbie
Date:
Message-ID: <dsdf9f$cc2$1@ss405.t-com.hr>
arnuld wrote:
> 1.) why you said SSICS is better for me?
>
> 2.) is it really a *good book*. i mean, is it worth to spend money on
> it?
>
> if yes, then how about 1994 edition, 1999 edition is very costly. i
> will buy the one prefered by you.
>
You should reread what I said before (comparing HtDP & SSICS). Both are
classified as high school books, but where HtDP is IMHO by far more
systematic, SSICS is more fun. It starts faster, but in the end, both
books reach about the same level. SSICS draws some of it's style from
Brian's earlier Logo books.
Both books are *excellent* starter CS books, and I would recommend HtDP
as, all in all better book. (Now, after reading about 10 other Scheme
books.) However, the fact remains that SSICS was apparently good enough
to get me hooked on functional style, and to make me want to continue
reading all those other books.
1994 is the one I have. AFAIK :-), there's very little difference
between the two.
-- Hrvoje
"arnuld" <·······@gmail.com> writes:
> > > now what is SSICS? is it "simply scheme by Brian Harvey"?
>
> > Yes.
> >
> > -- Hrvoje
>
> 1.) why you said SSICS is better for me?
>
> 2.) is it really a *good book*. i mean, is it worth to spend money on
> it?
>
> if yes, then how about 1994 edition, 1999 edition is very costly. i
> will buy the one prefered by you.
>
I suggest you see if there is a library near you which has some of
these books. As I mentioned earlier, personal book recommendations are
only part of the equation - individual differences in style,
comprehension and learning rates are the rest. See if you can get the
various books from the library and start reading them - some will work
for you and some will not.
Once you find the ones which you really like, then buy them. This will
save you from spending money on books you proably will never
completely read.
Note that its not always the well known books or authors which may be
best for you. Back in the beginning of my exploration of computers, I
stumbled on a simple book called something like "The Microprocessor
Cookbook Vol 1". I cannot remember the author and I lent it to someone
and never got it back. This was a really great beginners book - it was
wonderful for people who had just purchased a PC and who didn't really
understand anything about it. One of the reasons it was good was
simply it was pitched at the right level and had just the right
mixture of information and ability to inspire the reader to want to
know more.
The point is I had never heard of the author and I purchased the book
for 20 cents from a garage sale. Many wouldn't like this book because
it simplified things too much. However, at the time and at the point I
was at with my understanding, it was perfect and really inspired me to
learn more. Nobody wold have recommended this book because it was
quite obscure and possibly because once you have gained some
knowledge, you wouldn't think it was worth reading. sometimes, once we
have learnt a bit, its easy to forget how difficult it was at the
beginning.
Tim
--
Tim Cross
The e-mail address on this message is FALSE (obviously!). My real e-mail is
to a company in Australia called rapttech and my login is tcross - if you
really need to send mail, you should be able to work it out!
> Note that its not always the well known books or authors which may be
> best for you. Back in the beginning of my exploration of computers, I
> stumbled on a simple book called something like "The Microprocessor
> Cookbook Vol 1". I cannot remember the author
I think that must be Don Lancaster. Also author of some electronics
books and of the fantastic "The Incredible Secret Money Machine".
> One of the reasons it was good was simply it was pitched at the right
> level and had just the right mixture of information and ability to inspire
> the reader to want to know more.
Yes, I agree.
1st of all thanks to TIM for sharing his valuable experience with me &
comp.lang.lisp:
1.) Others will disagree, but for me, the most important thing to
learn is to become familiar with data abstraction techniques and
writing algorithms.
2.) In the end, you will only learn by doing.
3.) Learn to debug using the best debugger in the world, that lump of
grey matter between your ears.
4.) Using a stepping/tracing debugger eliminates the need to really
understand
and think about what is going on, which is never a good thing.
Debuggers are very useful tools, but overkill at the learning stage.
5.) - Don't get caught up in doing things the way veryone else says you
ahve to do things. Look around, read different authors and find the
ones which work for you.
thanks TIM for your valuable advices.
---------- and now for Harvoje --------
> Oops...
> And just how did you get the idea that I'm trying to have a flame war with you?
i said *if anybody*. i did not say *Harvoje*. i said so because some
people try to start a flame war.
> I will grant you that IMHO, first 8 sections of HtDP, are not very interesting/exciting.
> However, the book does pick up the pace, and becomes far more interesting after
> that. If you worked up to section 15, you should have realized at least some of this.
YES, you are 100% right. and it speeds-up more after section 11.
> The value of HtDP (at lest for me, after reading all those other books) was that it puts
>a lot of concepts into perspective. It helped me to align properly a lot of previously
> acquired, but somewhat scattered knowledge.
i do know about that so can not comment on it.
> However, the concepts that all those other books eventually deal with are far more
> difficult than HtDP ever gets.
Amen to that. what i hated in HtDP is :
1.) - their too much focus on theoretical explanation of their "Design
Reciepes" sometime 4-5 pages long theoretical explanation. i could have
liked it if they spend 1and a 1/2 page of explanation and 3 and a 1/2
pages of codes of examples and exercises on small programmes, compilers
and interpreters.
2.) - their very clear (explicit) way of telling the design recipies
and recursions. i think i am not a kid, i am 25 year old man, hence if
someone gives me lots of examples of code and says: "find out
something", then i will like to figure out common-patterns and
different approaches the programmer have taken in creating those
programmes. in other words i do not like hand-holding. i can try to
find out myself and increase the sharpness of my analysing and critical
thinking.
> But, if you hate HtDP, then SSICS should work well as a replacement. It *is* more fun > to read then HtDP. Actually, I'm at a better position to recommend this book to a
> beginner than any of the other books, because this *was* my first Scheme book.
now what is SSICS? is it "simply scheme by Brian Harvey"?
thanks
and may anybody explain me the meaning of following words:
IMHO, OTOH, AFAIK and some others like these. aftera all, i am a
newbie.
"arnuld"
"arnuld" <·······@gmail.com> writes:
> Hrvoje Blazevic wrote:
> > I would suggest though, to add HtDP to your list (probably as a zeroth
> > book). HtDP before tLS, because (again) of chapter 9, which I took as a
> > sort of the test: whether to continue reading all those other books, or
> > simply roll over and take up drinking as a pastime, instead of reading
> > CS books :-)
> >
> > -- Hrvoje
>
> well, it seems you did not read my previous post, 25th message on this
> thread , it is on the 1st page ( as this thread has 2 pages now,
> messages from 1-25 are on 1st page from 26 to this one are on 2nd ,
> this is how google manages the newsgroups posts). have a look at that.
>
> 1st of all - i do not want any FLAME-WARS here. if anybody tries to
> have a flame-war with me on this issue i will simply refuse to talk.
>
> i literally hate HtDP as i am doing it from last 10 months and not able
> to go beyond section 15 AND this is *not* a technical problem. this has
> something to do with my thinking pattern and analysing abilities which
> do not resonate with HtDP. after searching whole "groups.google.com"
> for more than 15 days, sometimes 8 hours at a stretch, i came to know
> that intorduction to programming has nothing to do with scheme, it can
> also be done well using Common LISP or python or Ruby or may be
> something else. SCHEME is prefered because it is easy to express many
> excellent ideas of CS in SCHEME than in other languages but problem is
> that i worked as SALESMAN for 2 years. YES you read it correctly,i do
> not like SCHEME because i was a SALESMAN. being a salesman taught me
> that student life is dream-life, it is not real, it is not intended to
> be. for the 1st time i went to sell "personal loan" i encountered lots
> of difficulties because all of my beliefes i learned as a student were
> just broken & selling broke them because they do not fit in real life,
> the hard-life. i have studied for my exams sometimes 8 hours at a
> stretch (8 pm to 4 am in the morning) and i did it for more than 3
> years but trust me that is not hard-work, if you compare to my 2 hours
> selling experience.
>
> that is thing i am taliking about - scheme is used and was created
> especially for academic purposes. scheme was created to teach CS
> concepts to graduate CS students .i am not saying that scheme can not
> be used to create softwares. i know that PLT scheme can be used to
> create Desktop Apps. even DrScheme itself is a programme written in
> scheme.what i am saying is : scheme has those beliefs and those
> feelings of student-life or academic life and i have experienced it as
> a user of HtDP. i literally hate student-life, it is a birthplace of
> many wrong ideas about real-life. please also search for Kent Pitman's,
> Shriram's and Matthias's posts on the different threads of
> comp.lang.lisp.
>
> also my search concludes that Common LISP was created to solve
> real-life problems , to solve very complex problems we face in
> producing high-quality and highly-functional softwares (which has lots
> of function and features) and is much more powerfull than SCHEME which
> is a sort of academic (SICP is an exception & that happened
> accidently). Common LISP is a sort of real-life thing.
>
> i do not know about you and others, but i can guarntee you from my
> experience as CS grduate and salesman that there will be lots of people
> loving HtDP and that will be so because they have exactly opposite
> thinking-style as compared to me. if someone goes beyond part-3 of the
> book easily or with some effort and likes the style in which book is
> written then i recommend he should continue to use HtDP. HtDP will be
> the best thing ever happened for that person, i am sure about it. i do
> not feel fit into their style and literally felt frustrated.
>
> also HtDP and tLS are written in exactly opposite styles, so may be i
> will love that book.
>
> hence it all depends upon that individual and on what he wants to do
> with his life and this is not a technical issue.
>
> thanks for your time
>
> "arnuld"
>
Just a couple of observations.
- The language you use to learn to program in is to a large extent
irrelevant. You can learn to program using any language. Some
languages like scheme make the expression of some programming
concepts easier than other languages, but as yo point out, everyone
is different in their cognitive processes and what suits one may not
suit another. For example, when I was learning, the language used
was Pascal and C and my favorite source was Knuth's Art of
Programming. Its probably not surprising that for me, the most
valuable course I did (the one which really cemented things for me),
was an introductory course in assembler programming. Different
strokes for different folks.
- Others will disagree, but for me, the most important thing to learn
is to become familiar with data abstraction techniques and writing
algorithms. With these skills, you can approach most problems,
break them down and implement the solution. There are a few
tricks/techniques which are very valuable, such as looking at a
specific problem and working out what is the general class of
problem it represents and then solving for the general case etc.
- I found reading books on data abstraction and algorithms more useful
initially than reading books about programming in some language. In
fact, a lot of my programming skill was learnt away from
computers. For example, when learning about different sorting
techniques, I would try out the algorithms by hand with a deck of
cards etc.
- Personally, I found alearnt a lot by implementing common data
structures by hand, even if the language had built-in
libraries/classes/whatever which did the same thing. In the end, you
will only learn by doing.
- If your using some sort of high level IDE, don't use the
debugger. Learn to debug using the best debugger in the world, that
lump of grey matter between your ears. When learning, the sort of
bugs you will run into are syntax and basic logic errors. Using a
stepping/tracing debugger eliminates the need to really understand
and think about what is going on, which is never a good
thing. Debuggers are very useful tools, but overkill at the learning
stage.
- Don't get caught up in doing things the way veryone else says you
ahve to do things. Look around, read different authors and find the
ones which work for you.
HTH
Tim
--
Tim Cross
The e-mail address on this message is FALSE (obviously!). My real e-mail is
to a company in Australia called rapttech and my login is tcross - if you
really need to send mail, you should be able to work it out!