Rails presentation

[Jamis Buck]

So I gave a presentation on Rails to the Utah Java Users Group last
night. It went very well, I think. There were three presentations--I
was sandwiched between JDO and Tapestry.

I've posted my presentation to my website, but there are a few
caveats:

1) it is 14 megs in size (I made use of screen capture videos). So,
be as kind to my host as possible. If anyone could mirror this,
or set up a bittorrent for it, etc...

2) it contains the presentation in both PDF and OOo formats. Either
way, you'll have to run the videos manually, because I'm sure the
links in the OOo version will be broken by packaging it up.

3) it is NOT a Rails tutorial--it was targeted at a roomful of Java
developers, most of whom had never seen a line of Ruby code in
their life. Thus, it is intended to demonstrate Rails' ease of
use, and how well it can perform "in the wild."

That said, you can view the presentation online, here:

http://ruby.jamisbuck.org/ujug-presentation/Ruby on Rails.html

Or, feel free to download the whole thing and take a look:

http://ruby.jamisbuck.org/ujug-presentation.tar.bz2

Enjoy!

- Jamis

 

[pat eyler]

So I gave a presentation on Rails to the Utah Java Users Group last
night. It went very well, I think. There were three presentations--I
was sandwiched between JDO and Tapestry.

The big questions then are:

1) How did it go?

2) For those of us that might be planning Rails presentations to
non-Ruby audiences, what should we watch for/think about?

 

[Curt Hibbs]

The big questions then are:

1) How did it go?

2) For those of us that might be planning Rails presentations to
non-Ruby audiences, what should we watch for/think about?

and

3) How many people were present in the audience?

Curt

 

[Jamis Buck]

The big questions then are:

1) How did it go?

Really well, I think. The audience was very considerate, the questions
were very appropriate, and they laughed in all the right places.
There were over 100 people there, too--the largest group I've ever
given a Ruby-based presentation to.

2) For those of us that might be planning Rails presentations to
non-Ruby audiences, what should we watch for/think about?

For me, I was trying to show a group of Java users how easy it is to
use Rails. So I avoided "teaching" Ruby syntax, and instead focused on
things like the code generation (which I demonstrated). They _really_
loved that Rails uses WordNet to suggest model and controller names
when there is a conflict. If I had it to do over again, I'd
probably use more little things like that in the presentation.

I also tried to focus on Rails' performance, which I think is where
most of their interest was. These people were all pretty focused on
Java--using it in their workplace, etc., so it's not very likely that
any single presentation would ever convert them to Ruby if all it does
is focus on "cool". So I tried to give some performance metrics for
Basecamp and 43 Things (thanks very much to DHH and Eric Hodel for
their assistance!), and I think that really opened their eyes.

One thing that tripped me up: I was demonstrating a simple AR model:

class Book < ActiveRecord::Base
belongs_to :author
validates_presence_of :name
validates_presence_of :isbn
validates_uniqueness_of :isbn
validates_format_of :isbn, :with => /^[- \d]+$/
end

I hadn't considered how foreign that might look to a non-Rubyist.
I got the question "are those method calls?" And how do you explain
that in two sentences? Method calls, in the middle of a class
definition? I tried, but I think I just boggled them. So, beware.

There were a few technical things I would do differently next time:

1) Don't try to be clever and type out method comments in the screen
capture videos. It might be nice for later, when you distribute
the video, but during the presentation it just makes it lag.

2) Keep the action in the videos as close to the top of the screen
as possible, so that people sitting WAY in the back can see it
better.

3) Use a very high contrast color scheme for the syntax
highlighting. Some of the colors were hard to read when
projected.

4) And last, something I just need to do better at in general when
giving presentations to large groups: repeat every question that
is asked of me, instead of just assuming everyone can hear it,
and that *I* heard it correctly.

- Jamis

 

[gabriele renzi]

Curt Hibbs ha scritto:

and

3) How many people were present in the audience?

4) how was the interaction with Howard Lewis Ship?

 

[Jamis Buck]

Curt Hibbs ha scritto:


4) how was the interaction with Howard Lewis Ship?

Howard was great. He's already been dabbling in the Ruby waters (he's
been known to post to this list, actually) so he was actually quite
curious about Rails. Don't expect him to change Tapestry to an ERb
style templating system anytime soon, though. I must say, I *am*
curious as to what it would take to plug a component-style model (ala
Tapestry) into Rails, though. I know others (on the Rails list) have
been asking for that, too.

I'm certainly satisfied with Rails' existing view system, though.

- Jamis

 

[gabriele renzi]

Jamis Buck ha scritto:

Howard was great. He's already been dabbling in the Ruby waters (he's
been known to post to this list, actually) so he was actually quite
curious about Rails. Don't expect him to change Tapestry to an ERb
style templating system anytime soon, though. I must say, I *am*
curious as to what it would take to plug a component-style model (ala
Tapestry) into Rails, though. I know others (on the Rails list) have
been asking for that, too.

I'm certainly satisfied with Rails' existing view system, though.

eheh.. but I was thinking of HiveMind vs Needle

 

[jason_watkins]

I also tried to focus on Rails' performance, which I think is where
most of their interest was. These people were all pretty focused on
Java--using it in their workplace, etc., so it's not very likely that
any single presentation would ever convert them to Ruby if all it does
is focus on "cool". So I tried to give some performance metrics for
Basecamp and 43 Things (thanks very much to DHH and Eric Hodel for
their assistance!), and I think that really opened their eyes.

Would you mind elaborating on the figures in the slides a bit, with the
sort of comments you'd make in the talk?

I'm considering advocating Rails as an alternative to PHP for some
projects I'm likely to be involved it. We'd be using it backed by
postgres (on linux). The audience is a couple of guys familiar with
java, php and perl. I believe I can articulate the developer benefits
clearly, but I'm not really a sysadmin and so can't go into the detail
of what a statement like "43 things has load averages of 0.5 at
6.6pages/sec on dual processor 2ghz xeons" really means. I mean I can
look at the homepage size and do back of the envelope math to say "ok,
that's roughly sustaining 8mbit" but I don't really have any framework
to compare that to what PHP, Perl, or ASP would sustain in that
situation.

 

[Jamis Buck]

Jamis Buck ha scritto:


eheh.. but I was thinking of HiveMind vs Needle

The topic didn't come up, actually. I had to leave early, though,
so I didn't get much chance to talk with him face-to-face.

- Jamis

 

[Ryan Davis]

The big questions then are:
1) How did it go?
2) For those of us that might be planning Rails presentations to
non-Ruby audiences, what should we watch for/think about?

We can and should answer these questions for ourselves for the
south.seattle.rb meeting. We obviously have a lot to learn about giving
smooth/impressive presentations.

 

[Eric Hodel]

--Apple-Mail-11--536214559
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII; format=flowed

Would you mind elaborating on the figures in the slides a bit, with the
sort of comments you'd make in the talk?

I'm considering advocating Rails as an alternative to PHP for some
projects I'm likely to be involved it. We'd be using it backed by
postgres (on linux). The audience is a couple of guys familiar with
java, php and perl. I believe I can articulate the developer benefits
clearly, but I'm not really a sysadmin and so can't go into the detail
of what a statement like "43 things has load averages of 0.5 at
6.6pages/sec on dual processor 2ghz xeons" really means.

Actually, page 14 should read "~200k hits/day", and that's 6 pages/sec
per FastCGI process (we're running 30 total).

each box does ~100k hits/day of dynamic content without any noticeable
load on the servers. (Load isn't noticeable until it hits at least 2,
once you get to 4 you really start to feel things, because your
commands get laggy and you have to nice things).

Basically, we could probably run the whole site off of one box for a
few more months of growth.

I mean I can
look at the homepage size and do back of the envelope math to say "ok,
that's roughly sustaining 8mbit" but I don't really have any framework
to compare that to what PHP, Perl, or ASP would sustain in that
situation.

Our homepage is almost entirely pulled from memcache, we can serve
*lots* of homepages.

--
Eric Hodel - (e-mail address removed) - http://segment7.net
FEC2 57F1 D465 EB15 5D6E 7C11 332A 551C 796C 9F04

--Apple-Mail-11--536214559
content-type: application/pgp-signature; x-mac-type=70674453;
name=PGP.sig
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
content-transfer-encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (Darwin)

iD8DBQFCFnMAMypVHHlsnwQRAjBtAKCQSQaSAWEXMAU7zCl+k7AGsG2yRACggdHV
zhx0lzMumFsLmiAijiAbw8s=
=RP0Z
-----END PGP SIGNATURE-----

--Apple-Mail-11--536214559--

 

[Jamis Buck]

Actually, page 14 should read "~200k hits/day", and that's 6 pages/sec
per FastCGI process (we're running 30 total).

each box does ~100k hits/day of dynamic content without any noticeable
load on the servers. (Load isn't noticeable until it hits at least 2,
once you get to 4 you really start to feel things, because your
commands get laggy and you have to nice things).

Basically, we could probably run the whole site off of one box for a
few more months of growth.

Thanks for the clarification, Eric. Sorry for the misrepresentation.
I'll fix the presentation and upload it again.

Our homepage is almost entirely pulled from memcache, we can serve
*lots* of homepages.

Very, very cool.

- Jamis

 

[Jamis Buck]

Would you mind elaborating on the figures in the slides a bit, with the
sort of comments you'd make in the talk?

I'd be happy to answer specific questions. Trying to narrate each
slide would take more time than I'm willing to spend on this right
now, though.

One thing I learned while I was giving the presentation: the load
number is the average number of processes waiting for a resource.
Didn't know that before.

- Jamis

 

[jason_watkins]

Thanks Eric. That's much more the sort of numbers I expected.