Re: Getting Started

[Chris McGrath]

No problem here either :)

Chris

On 4 Jun 2010, at 17:01, mr_urf wrote:

> I've no issues :)
>
> On Jun 4, 11:56 am, Jose <jjdominguezve...@gmail.com> wrote:
>> Hi guys,
>>
>> just read Chris' mail to the ruby list (thanks for that!). Sounds
>> really good. Haven't done any functional stuff since college (a long
>> time ago!)
>> I do a bit of research in my spare time on pedagogy focused on CS. I
>> do not intend to use anything I see here for my research; am just in
>> for the experience, but if anyone is not comfortable with it please
>> let me know and I'll step out of the group.
>>
>> cheers,
>> José
>>
>> On Jun 4, 11:37 am, Chris McGrath <ch...@octopod.info> wrote:
>>
>>
>>
>>
>>
>>> On Jun 3, 11:04 pm, Paul Boxley <paul.box...@me.com> wrote:
>>
>>>> All sounds good to me :)
>>
>>>> My goals are really just to learn some more about functional
>>>> programming. I studied LISP at uni (probably more than I studied OOP
>>>> at uni, but that's not saying much) but only really scratched the
>>>> surface. I've already gone through some of the exercises in the book,
>>>> but I'm happy to start again. I'm sure it will enhance my
>>>> understanding to go through things with other people.
>>
>>> Yeah, me too. Also it'll be harder to keep putting off the hard
>>> exercises :)
>>
>>>> Just for the record, I've been using the paper book. It's £26 new on
>>>> Amazon:http://www.amazon.co.uk/Structure-Interpretation-Computer-Electrical-...
>>
>>>> I'd rather stick with the book but I'm happy to fall back on the PDF
>>>> version if there are noticeable differences. :)
>>
>>> Cool, Is it the 2nd edition book? Since getting an iPad I'm not so
>>> interested in paper books :)
>>
>>> Chris

[Chris McGrath]

On 4 Jun 2010, at 20:29, mr_urf wrote:

> I'd just like to say a big thanks to Chris for coming up with the
> study group idea by the way. It looks like it may be getting quite
> popular!
>

s/coming up with/remembering what Jim said at SRC and vaguely mentioning it on twitter/ :)

Going to give it another week to allow ScotRUG / Techmeetup people to join. Sound reasonable? I'll set up a wee survey during the week so we can figure out what time suits people. There are 9 so far which I think should be ok with just one group. If we get many more we may have to split into two.

Cheers,

Chris

[mr_urf]

I think we actually have around 27 now (unless you were just counting
the Irish contingent). I think that, for the most part we could
operate as a single group. However it may be an idea to split the
group into 3 or 4 smaller groups for the Skype calls to save on lag
and confusion. Another advantage of this approach is that we can have
them on different nights and people can join in the one that suits
them best, providing that there's a space? Perhaps the 4 founder
members (Chris McGrath, Paul Wilson, Paul Boxley and myself) could
each host a conf call? I'm just barfing ideas, let me know what you
all think ...

[Ulises]

[snip]

> operate as a single group. However it may be an idea to split the
> group into 3 or 4 smaller groups for the Skype calls to save on lag
> and confusion. Another advantage of this approach is that we can have
> them on different nights and people can join in the one that suits
> them best, providing that there's a space? Perhaps the 4 founder

[snip]

Sounds good. What about those that may miss a couple of calls here and
there? Will there be written guidelines or even minutes of these
calls?

Just a thought ...

U

[Steve Tooke]

On Tue, Jun 8, 2010 at 7:10 AM, Ulises <ulises....@gmail.com> wrote:
> Sounds good. What about those that may miss a couple of calls here and
> there? Will there be written guidelines or even minutes of these
> calls?

The group that Jim Weirich was running used to meet on YuuGuu. This
enabled desktop sharing, and allows people to call or skype in to
their web conferencing system, this should cut most problems from
hosting a large conference call on someone's adsl line!

http://www.yuuguu.com

Somebody dialed in on skype would often record a screencast of the
session for those who were unable to attend.

If you haven't seen the original google group that Dave and Jim set up
its here: http://groups.google.com/group/wizardbookstudy

Steve

--
/tooky

[mr_urf]

Yuuguu looks like a great option and I'd certainly be keen to use it
if others agree. However, I still think we should split the group into
smaller study groups as it'll make discussions easier and make it more
likely that everyone will get a shout in. It'll also allow us to be
flexible about the day/time the confs are run. I'm not sure whether we
want to go down the route of assigning people to groups or just
setting up the confs and see how people attend.

Recording the confs is a great idea, even if only so that those of us
that attend can refer back to it at a later date. Can you record
direct from yuuguu or was other software used for this? Perhaps we
should set up a vimeo account for hosting?

Alan

On Jun 8, 7:17 am, Steve Tooke <steve.to...@gmail.com> wrote:

[mr_urf]

Just reading through the yuuguu site and see that, unless I'm reading
it wrong, the free account is restricted to 5 users per session
(http://www.yuuguu.com/pricing/yuuguu-pricing). Perhaps this should
dictate the sizes of our study groups? ;)

Alan

[Steve Tooke]

On Tue, Jun 8, 2010 at 7:38 AM, mr_urf <urfol...@gmail.com> wrote:
> Recording the confs is a great idea, even if only so that those of us
> that attend can refer back to it at a later date. Can you record
> direct from yuuguu or was other software used for this? Perhaps we
> should set up a vimeo account for hosting?

I think somebody who was connected via skype would record the session
with camtasia or something similar.

Steve

--
/tooky

[Paul Boxley]

On 8 Jun 2010, at 07:07, mr_urf wrote:

> Perhaps the 4 founder members (Chris McGrath, Paul Wilson, Paul Boxley and myself) could each host a conf call?

Happy to do this. If numbers drop we can always try something different.

Paul B.

[Alan Gardner]

Sounds good.

[Chris McGrath]

From my basic 2 minutes on google research, 5 appears to be a good number for a study group. I'm happy to use Yuuguu and have someone record the calls. Will still need to use skype to call in though, the dial in numbers are only for corporate plan afaict.

Later today I'll send out a link to a little google docs form that should help us break up into groups that are convenient. I don't think that membership should be set in stone. Some people will find times that are convenient now turn out not so convenient in the future, some will lose interest after a while etc. etc.

Regarding group leaders, I'm including some questions on the form about experience in lispy languages and SICP. I think people with some experience here would be good to moderate initially. I think we should then swap round the moderator so everyone has a chance / is forced to do it. Can decide when we see what mix of people we have.

Cheers,

Chris

[Paul Wilson]

On 8 Jun 2010, at 08:47, Chris McGrath wrote:

> From my basic 2 minutes on google research, 5 appears to be a good number for a study group. I'm happy to use Yuuguu and have someone record the calls. Will still need to use skype to call in though, the dial in numbers are only for corporate plan afaict.

From my zero minutes of research I agree with everyone else ;-)

>
> Later today I'll send out a link to a little google docs form that should help us break up into groups that are convenient. I don't think that membership should be set in stone. Some people will find times that are convenient now turn out not so convenient in the future, some will lose interest after a while etc. etc.
>

You're the man with the plan.

> Regarding group leaders, I'm including some questions on the form about experience in lispy languages and SICP. I think people with some experience here would be good to moderate initially. I think we should then swap round the moderator so everyone has a chance / is forced to do it. Can decide when we see what mix of people we have.
>

Both good points. It might also be handy for people to be able to switch groups, if they can't make a session for whatever reason. I also suggest we/someone/5 random people have some dry runs with the tech, before we launch into the study. 5 to 10 minutes of "Can everyone hear everyone? There's a terrible echo here? Can everyone see the screen?"

Talking of which, I strongly suggest that everyone invests in headphones, preferably with a mic. Having the audio feedback from speakers into someone's mic with 1 second delay is painful for everybody on the call. iPhone style headphones are adequate. At EdgeCase Edinburgh we bought a handful of inexpensive Plantronics [1] headsets which are working very well for remote pairing and Skype

[1] http://www.amazon.co.uk/gp/product/B000UNRNX4

[Alan Gardner]

+1 for headsets. I invested in one recently and it's much better for Skyping.

I would have been happy to join the test run on the tech but there's a good chance I'll be elbow deep in nappies within the week ;) Still, keep me informed of the when and I'll do my best.

Alan

[Paul Boxley]

All sounds good. I will have a look for a decent headset.

I'm not fussed whether the 'hosts' have an FP background or not, I didn't envisage them as teachers or even leaders, just nominated people who would host the conference calls. Don't mind either way, really :)

Happy to take part in a test call some time.

Paul B.

[Chris McGrath]

On 8 Jun 2010, at 09:03, Paul Wilson wrote:

>
>> Regarding group leaders, I'm including some questions on the form about experience in lispy languages and SICP. I think people with some experience here would be good to moderate initially. I think we should then swap round the moderator so everyone has a chance / is forced to do it. Can decide when we see what mix of people we have.
>>
>
>
> Both good points. It might also be handy for people to be able to switch groups, if they can't make a session for whatever reason. I also suggest we/someone/5 random people have some dry runs with the tech, before we launch into the study. 5 to 10 minutes of "Can everyone hear everyone? There's a terrible echo here? Can everyone see the screen?"
>

I was thinking the first session for each group would be along those lines. Phone and recording tech test plus everyone doing some kind of (hello-world) to make sure their dev env is working. A bit of a get to know each other chat and confirmation of what should be covered for the first 'proper' meeting. I suggest we completely steal the initial lesson plans from Jim's groups page here [1]

> Talking of which, I strongly suggest that everyone invests in headphones, preferably with a mic. Having the audio feedback from speakers into someone's mic with 1 second delay is painful for everybody on the call. iPhone style headphones are adequate. At EdgeCase Edinburgh we bought a handful of inexpensive Plantronics [1] headsets which are working very well for remote pairing and Skype
>
>
> [1] http://www.amazon.co.uk/gp/product/B000UNRNX4
>

Yes! Headset is a must I reckon.

Chris

[1] http://groups.google.com/group/wizardbookstudy/web/weekly-assignments

[Alan Gardner]

On Tue, Jun 8, 2010 at 9:39 AM, Chris McGrath <ch...@octopod.info> wrote:

I was thinking the first session for each group would be along those lines. Phone and recording tech test plus everyone doing some kind of (hello-world) to make sure their dev env is working. A bit of a get to know each other chat and confirmation of what should be covered for the first 'proper' meeting. I suggest we completely steal the initial lesson plans from Jim's groups page here [1].

Chris

[1] http://groups.google.com/group/wizardbookstudy/web/weekly-assignments

I think that this is a good approach. An initial introduction/test will allow each SG to get acquainted and agree rules of engagement. Small bites on the book, backed up by the lectures and exercises seems a logical and manageable way to proceed too.

[Ulises]

> I was thinking the first session for each group would be along those lines. Phone and recording tech test plus everyone doing some kind of (hello-world) to make sure their dev env is working. A bit of a get to know each other chat and confirmation of what should be covered for the first 'proper' meeting. I suggest we completely steal the initial lesson plans from Jim's groups page here [1]

+1 especially on confirming that our devenvs are up and running correctly.

U

[Rick Moynihan]

Hi all,

I heard about this on the techmeetup group, and am naturally
interested! I've owned the SICP book for a while, and have flicked
through it several times, but have never really sat through and
studied it.

I'm a little sad to see you're not doing the thing in Clojure (I run
the Dundee Clojure Dojo http://clojuredojo.com/ ), but understand the
motivation for sticking to scheme. Besides I'm sure learning some
scheme will be good for me!

For the interest of others there is SICP in Clojure, but sadly it's
nowhere near complete, with almost all of the text and examples still
being specific to scheme:

http://sicpinclojure.com/

Anyway, I just wanted to say hi, and register my interest. Do you
know when you're planning the sessions for, what time, days and how
frequently?

R.

[Denis Hennessy]

Actually, I think there is a subset who are going to try the exercises in clojure instead of scheme. Since we're partitioning the group anyway, maybe it would make sense to have a breakdown based on implementation language?

As an aside, I was quite impressed with Jim's implementation in Ruby 1.9 but I'm betting he did that _after_ the scheme version.

/dh

ps For multi-way skype calls, headsets are critical

[Alan Gardner]

It seems fair to me that each SG should decide their language/implement of choice. Perhaps we could add a section to the head of each group in the google doc to say which language the SG intends to use, Chris? That way people can join an SG based on their preference.

Personally I'm certainly interested to learn Clojure (it's on my horizon-spanning todo list) but I'd prefer to stick to Scheme for the first pass through SICP.

What's everyone else's feelings on this? Should each SG just decide their language/implementation of choice or should we split off and have separate google groups for Scheme and Clojure to easy use and admin?

Alan

[Alan Gardner]

Also, I've just noticed that most of the beginning of this thread is missing from the google group. Not sure how that happened. Pretty sure it wasn't me :S

Anyway, is everyone else seeing the same thing? If so I can copy the missing part to another discussion thread or page for reference.

Alan

[Chris McGrath]

On 8 Jun 2010, at 12:58, Denis Hennessy wrote:

> Actually, I think there is a subset who are going to try the exercises in clojure instead of scheme. Since we're partitioning the group anyway, maybe it would make sense to have a breakdown based on implementation language?
>

Was thinking it to, have added a question to the about to be sent form

Cheers,

Chris

[Matt Wynne]

There are a few criteria we could use to gather SGs - implementation language and maybe even physical location.

In lieu of a better idea, shall we start a google docs spreadsheet with everyone on it, and we can start putting things like preferred language on it and see if we can use that to gather groups together?

[FreeAgent]

Hi all,

(I agree with the spreadsheet idea.)

Why would you want to use Clojure?

Scheme is a teaching language. Clojure is a twenty-first century LISP.
Since it runs on the JVM and integrates easily with Java code, it has
all of Java's libraries and tools available. And Clojure's story on
concurrency is second-to-none.

Bluntly, Clojure may help you to get a new job, in a year or two, as
it gets more traction. I'm aware of a company in Dublin with a Clojure
group doing "ten-percent-time projects". (I'm not referring to
Google.)

I'm not trying to knock the people who want to use Scheme. Clearly,
there's great value in learning the ideas in SICP through Scheme.

In practise, the Clojure sub-group will read the book and get the
ideas, skimming over the details of Scheme; but when it comes to the
exercises, we'll do them in Clojure.

We can cover the same sections of SICP, at the same time as the
others.

Tutorials on Clojure:
http://java.ociweb.com/mark/clojure/article.html
http://clojure-notes.rubylearning.org/


Ciao,

Kevin.

[Chris McGrath]

On 8 Jun 2010, at 15:09, Matt Wynne wrote:

There are a few criteria we could use to gather SGs - implementation language and maybe even physical location.

In lieu of a better idea, shall we start a google docs spreadsheet with everyone on it, and we can start putting things like preferred language on it and see if we can use that to gather groups together?

Just sent an email to a google docs form to gather some info up. Will share it out to everyone who adds their name to the list.

Here's the link again: https://spreadsheets0.google.com/viewform?formkey=dGtTUVBERWh6NGdiN3RVX2FqclFBSlE6MQ

Chris

[Rick Moynihan]

On 8 June 2010 15:31, FreeAgent <kno...@gmail.com> wrote:
>
> Tutorials on Clojure:
> http://java.ociweb.com/mark/clojure/article.html
> http://clojure-notes.rubylearning.org/

For those interested in trying Clojure you can find a comprehensive
set of introductory Clojure videos and presentations in my delicious
links here:

http://delicious.com/InkyHarmonics/video+introduction+clojure?setcount=100

Specifically I'd recommend people look at the following, at least the
following (in order):

1) Are we there yet. A deconstruction of Object Oriented Time.

In this talk Rich, presents the rationale & philosophical motivations
underlying Clojure. An excellent talk for Clojure curious! (No code
level detail):

http://www.infoq.com/presentations/Are-We-There-Yet-Rich-Hickey

2) After this Checkout out the Introduction to Clojure (in Ten, ten
minute videos). This is especially recommended for those new to Lisp
dialects (though it includes a lot of Clojure subtleties if you
already know Common Lisp or Scheme):

Either way its a good bite sized intro:

http://www.youtube.com/view_play_list?p=AC43CFB134E85266

3) From here you're probably best to checkout the following chat
between Rich Hickey and Brian Beckman. Rich provides a near flawless
explanation of Clojure's homoiconic nature, how it differs from other
Lisps, and also digs into how its design is optimised for concurrency
and parallelism:

http://channel9.msdn.com/shows/Going+Deep/Expert-to-Expert-Rich-Hickey-and-Brian-Beckman-Inside-Clojure/

4) From here, you should be free to dive into any of the other videos
or links listed at

http://delicious.com/InkyHarmonics/clojure?setcount=100

I hope you find this is useful.

R.

[Alan Gardner]

OK, I have to admit that I'm torn now. On one hand I don't want to add another layer of complexity to the SICP course by having to translate the exercises and examples into a language I'm only just learning, but on the other I want to kill two birds with one stone and learn Clojure at the same time as I do the SICP. :S

Either way, learning Clojure just jumped pretty far up todo list!

Alan

[Matt Wynne]

The way I'm looking at it, I know virtually nothing about scheme or clojure, so it probably won't make any difference if I jump straight into clojure.

[Alan Gardner]

That's the side that I was starting to come down on myself. However, it does mean that every time we encounter an area where Clojure doesn't map to Scheme we will have to, as novices in both languages, find the appropriate alternate solution. Personally I think that that will add additional complexity. It will also require constant context switching from learning a new language to learning the fundamentals of programming, two related but entirely different tasks (as far as our brains are concerned at any rate). I appreciate that we will have to do the same with Scheme to a certain extent, however I think there will be less friction due the course being written this way.

Ultimately I'm just trying to decide whether or not I'm willing/able to take on that extra burden this time through, or whether to learn first in Scheme and then later in Clojure.

Alan

[Rick Moynihan]

Yes, there is a dillema, and I'm sorry for confusing the issues
further by enticing people to the Clojure camp!

If you want to learn Clojure and don't already know a LISP then SICP
is almost certainly not the best way to learn Clojure (though I'm sure
it is possible).

If you want to learn SICP and know niether a Lisp nor Clojure then
you're almost certainly best using Scheme.

If you know Clojure/FP or another Lisp/Scheme and want to learn SICP
then its "probably" not too hard to port the code from SICP into
Clojure... though if you don't know scheme some of the semantics might
be a hurdle....

Anyway, I'm personally happy to use either Scheme (with which I'm not
very familiar) or Clojure (with which I am (I am using Clojure
commercially))... However even I'm not sure if going the Clojure
route is a great idea, as I can imagine things like the metacircular
evaluator chapter getting more than a little mind melting to port.

For any wishing to dive into Clojure via SICP, then I'm willing to
assist with the Clojure specific details... though bear in mind my
name is Rick, not Rich! ;-)

R.

[Alan Gardner]

"or Clojure (with which I am (I am using Clojure
commercially))"

I think the nested parentheses would have given you away anyway ;P

Thanks for the advice. Will stick to Scheme.

Alan

iPhone + sausage fingers - patience = brevity + typos ;)