Next Hackfest: Tuesday, June 3rd

0 views
Skip to first unread message

Josh

unread,
May 29, 2008, 12:29:09 AM5/29/08
to Boston Ruby Group
Hey everyone,

It's about that time of month for another hackfest.

Same hackfest time, same hackfest place:

7-9pm

Food, beverages, wireless provided by thoughtbot.

41 Winter Street, 3rd floor
Map: http://tinyurl.com/3ch9pn
Park Street T stop

As for what to work, we're open for suggestions:

* Do you want feedback on your app or library?
* Trying to solve a particular problem you've been banging your head
against?
* Want to write some tool, library, or plugin to make your life
easier?

I don't have any crazy ideas to hack on yet. Do you have any?

Jared Nedzel

unread,
May 29, 2008, 11:14:42 AM5/29/08
to boston-r...@googlegroups.com, Michael Reich
Folks:

I'd love to get feedback on my application. I'm an experienced Java
developer, but new to Ruby and Rails. So while my application works,
I'm sure I'm doing many things the hard way.

BACKGROUND ON THE APPLICATION

The application that I'm working manages data from a Luminex bead
sequencer: http://www.luminexcorp.com/products/luminex_100IS.html

Our scientists are studying the effects of various drugs on the
expression of proteins in various biological samples.

They put samples into wells on plates. They have beadsets which contain
antibodies hybridized to microscopic beads. These bead sets are put
into the wells and the proteins hybridize onto the beads. The Luminex
machine then uses lasers to count the number of each type of beads that
have hybridized to proteins in the samples. The result is a formatted
text file that gives gene expression readings. This file contains
roughly 100 wells times 100 beads times 10 distinct statistics, so
roughly 100,000 data points per plate.

My web application currently allows the users to import the data file
and manage the following entities:

- project
- experiment
- plate
- well
- beadset
- bead
- tag (the antibody probe)
- treatment

In the future, I'll be adding more flexible query and output capabilities.

Compared to some of the presentations at the Boston Ruby Group, this
application is dirt simple, quick-and-dirty, and ugly. There is no
high-speed, low-drag clever coding here. But it may be interesting to
those looking at doing more basic development techniques on a real-life
application.

AREAS OF INTEREST

I'm sure we could spend far more time than we have go over the
application, but here's a few possible topics where I could use feedback:

1) general Ruby/Rails style. I'm still thinking in Java and my code
looks like it.

2) unit testing of active record objects. While I have some testing, it
just isn't done right. I've read a bit about testing active record
objects, but haven't gone down that road yet. It would be great if
someone could show me how to get started in the right way on this.

3) performance improvements. I've done some performance work on the
import of the Luminex data model. This import creates 10,000 rows. My
first cut took 5 minutes to import. I've got it down to 1 minute, but
it would be great if we could speed it up even more. The next
generation Luminex machine uses a 384 well plate and 1000 beads per
plate, so that will have a significantly larger dataset and performance
will be even more important.

4) Incorporate Ajax. Right now the management functions are somewhat
clumsy. For example, the user sees a list of projects in an HTML table.
To add a project, they hit the "New Project" link, which returns a
project form. They enter the name and description, hit save, and get
sent back to the HTML table. It would be nice to instead have the main
page be a spreadsheet-like form, with no need to navigate to the "New
Project" form to add a new project.

Jared


--
Jared Nedzel
Cancer Genomics Informatics
Broad Institute
7 Cambridge Center
Cambridge, MA 02142

617-324-4825
jne...@broad.mit.edu

Dan Croak

unread,
May 30, 2008, 10:46:54 AM5/30/08
to Boston Ruby Group
Jared,

Let's do it! How about you bring your app in on a laptop and we'll
hook it up to our big screen? Bring any adapters you might have.

If we finish giving feedback early, I'd love for us to begin messing
with JRuby.

Dan Croak
thoughtbot, inc.
organic brains, digital solutions

Jared Nedzel

unread,
May 30, 2008, 10:52:04 AM5/30/08
to boston-r...@googlegroups.com
Will do. I've got a Macbook and I have the DVI-to-VGA adapter. Btw, I
am still on Rails 1.x.

Hardbap

unread,
May 30, 2008, 10:57:46 AM5/30/08
to Boston Ruby Group
Hey guys.

Is the hackfest open to all? Even newbies to Ruby/Rails (a refugee
from .NET)?

Thanks.
Mike

Josh

unread,
May 30, 2008, 12:41:06 PM5/30/08
to Boston Ruby Group
Absolutely!

Myself, I hadn't worked with Ruby too much when I started coming to
the hackfests 6 months ago (damn, has it really been that long???).
I've learned a lot through the hackfests, and it definitely helped
accelerate my learning.

Matthew Krom

unread,
May 30, 2008, 1:14:58 PM5/30/08
to boston-r...@googlegroups.com
I like this idea a lot; I'm planning to attend my first hackfest now.

Jared, upgrading to Rails 2.x should be simple; it's one thing we can all look at.  Which leads me to a question--how is your automated test suite?

Matt



--
matthe...@gmail.com
617 852 5130

Jared Nedzel

unread,
May 30, 2008, 3:08:33 PM5/30/08
to boston-r...@googlegroups.com
It's embarrassingly lame and one of the things I'd like feedback on.

Matthew Krom wrote:
> Which leads me to a question--how is your automated test
> suite?

--

Dan Croak

unread,
Jun 4, 2008, 10:56:11 AM6/4/08
to Boston Ruby Group
Thanks to all who showed up last night. From what I can remember, the
attendees were:

* Greg Sterndale
* John Nagro
* Josh Nichols
* Matthew Krom
* Mike Breen
* Mike Burns
* Oliver Scheck
* Tom Shealy
* Dan Croak
* Jared Nedzel

Jared shared the app he was working on, and we:

* provided feedback on his data import code
* discussed alternative schemas
* suggested making use of named routes
* discussed testing options for models

We made good use of the Campfire chat room for the first time in a
while (https://thoughtbot.campfirenow.com/fd31a). With a bigger group,
I think people appreciated having a backchannel.

Some of the things discussed:

* FasterCSV - http://fastercsv.rubyforge.org/
* The default value Hash syntax - Hash.new {| hash, key | block }
* Named routes - http://api.rubyonrails.org/classes/ActionController/Routing.html
* The upper limit of MySQL db columns - 4096 columns per table
* Passing a locals Hash in partials - http://api.rubyonrails.org/classes/ActionView/Partials.html
* Shoulda - http://thoughtbot.com/projects/shoulda
* Boston getting it's ass kicked on http://railscities.com/ (which
pulls from the top 100 on http://workingwithrails.com)

Our time focused on testing got cut a little short. Would anyone be
interested in focusing the next hackfest solely on testing Rails apps?

Dan Croak
http://thoughtbot.com
organic brains, digital solutions

Michael Breen

unread,
Jun 4, 2008, 2:05:51 PM6/4/08
to boston-r...@googlegroups.com
>> Would anyone be interested in focusing the next hackfest solely on
testing Rails apps?

I'd be interested in that.

Sean Hussey

unread,
Jun 4, 2008, 3:09:51 PM6/4/08
to boston-r...@googlegroups.com
Yes! I'll try to get my whole dev team to attend. (If you're reading
this, teammates, clear your calendar.)

Sean

W Greene

unread,
Jun 4, 2008, 5:55:07 PM6/4/08
to boston-r...@googlegroups.com
I would be interested in testing, too. At the hackfest, there was a small discussion about the costs and benefits of testing, but the discussion got cut short before it got too deep. Maybe we could revive that discussion by talking about the following, preferably in a Rails context:

How much testing is too much?
What testing stategies can end up being quagmires?

Dmitry Amelchenko

unread,
Jun 4, 2008, 6:38:00 PM6/4/08
to boston-r...@googlegroups.com
I've got an application that badly needs some testing implemented. I presented it few month ago at Boston Ruby group -- http://cleantogether.com

The application is shaping up, but since it's an after work effort, the testing is always been put off until after (and i admit -- it's wrong).
So, if people are interested in taking up a challenge and see how much of testing coverage can/should be done, let me know.

Dmitry A.


2008/6/4 W Greene <gree...@yahoo.com>:

Jared Nedzel

unread,
Jun 5, 2008, 7:28:34 AM6/5/08
to boston-r...@googlegroups.com
Folks:

Thanks for all the feedback on Tuesday night. I've got a bunch of work
to do and some things to think about.

Reply all
Reply to author
Forward
0 new messages