FAQ
Hello,

There is a PostgreSQL Community Conference happening in Portland,
Oregon in a couple of months. It would be great if we could get a
Catalyst developer who preferred PostgreSQL to give a talk... Here is
the announcement of the call for talks:

The second annual PostgreSQL Conference: West is being held on October
10th through October 12th 2008 in the The Native American Student &
Community Center at Portland State University.

We are currently accepting papers and you can submit your talks here:

http://www.postgresqlconference.org/west08/talk_submission/

We have already seen submissions on Tsearch2 as well as pgTap. Do you
have something you would like to share about PostgreSQL? Now is the
time!

This year West will be providing its proceeds to the Postgresql.us.

Sincerely,

Joshua D. Drake


--
The PostgreSQL Company since 1997: http://www.commandprompt.com/
PostgreSQL Community Conference: http://www.postgresqlconference.org/
United States PostgreSQL Association: http://www.postgresql.us/
Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate

Search Discussions

  • Jay Shirley at Aug 20, 2008 at 12:14 am

    On Tue, Aug 19, 2008 at 3:41 PM, Joshua Drake wrote:
    Hello,

    There is a PostgreSQL Community Conference happening in Portland,
    Oregon in a couple of months. It would be great if we could get a
    Catalyst developer who preferred PostgreSQL to give a talk... Here is
    the announcement of the call for talks:

    The second annual PostgreSQL Conference: West is being held on October
    10th through October 12th 2008 in the The Native American Student &
    Community Center at Portland State University.

    We are currently accepting papers and you can submit your talks here:

    http://www.postgresqlconference.org/west08/talk_submission/

    We have already seen submissions on Tsearch2 as well as pgTap. Do you
    have something you would like to share about PostgreSQL? Now is the
    time!

    This year West will be providing its proceeds to the Postgresql.us.

    Sincerely,

    Joshua D. Drake
    I'm local, and definitely prefer Pg even though I find myself stuck
    with MySQL rather too frequently. I don't know if I can commit to a
    talk right now, but I can help out getting you a speaker if I can't do
    it myself.

    However, one of the problems you may find from the Catalyst side of
    things is that we tend to forget the particulars of various databases
    because DBIx::Class, SQL::Abstract and SQL::Translator take a lot of
    the pain from being cross-database compatible out of the mix.

    Happy to give a talk on painless scaffolding using some various CRUD
    and DBIx::Class stuff, that works very well with PostgreSQL (my
    current fascination is REST-based stuff, and am quite fond of
    Catalyst::Controller::DBIC::API)

    Anybody else have any thoughts?
  • Joshua Drake at Aug 20, 2008 at 1:35 am

    On Tue, 19 Aug 2008 16:14:43 -0700 "J. Shirley" wrote:


    I'm local, and definitely prefer Pg even though I find myself stuck
    with MySQL rather too frequently. I don't know if I can commit to a
    talk right now, but I can help out getting you a speaker if I can't do
    it myself. Thanks!
    However, one of the problems you may find from the Catalyst side of
    things is that we tend to forget the particulars of various databases
    because DBIx::Class, SQL::Abstract and SQL::Translator take a lot of
    the pain from being cross-database compatible out of the mix.
    Right, that is the constant issue with all the toolkits. My mind is
    something that provides an introduction to Catalyst (which is good for
    Catalyst) but also, "Why use Catalyst with PostgreSQL?". What kind of
    cool features can Catalyst (and is associated libs) expose that
    developers might be missing.
    Happy to give a talk on painless scaffolding using some various CRUD
    and DBIx::Class stuff, that works very well with PostgreSQL (my
    current fascination is REST-based stuff, and am quite fond of
    Catalyst::Controller::DBIC::API)
    That would be very cool!
    Anybody else have any thoughts?
    Sincerely,

    Joshua D. Drake



    --
    The PostgreSQL Company since 1997: http://www.commandprompt.com/
    PostgreSQL Community Conference: http://www.postgresqlconference.org/
    United States PostgreSQL Association: http://www.postgresql.us/
    Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate
  • Ferruccio Zamuner at Aug 20, 2008 at 10:15 am

    Joshua Drake wrote:
    Right, that is the constant issue with all the toolkits. My mind is
    something that provides an introduction to Catalyst (which is good for
    Catalyst) but also, "Why use Catalyst with PostgreSQL?". What kind of
    cool features can Catalyst (and is associated libs) expose that
    developers might be missing.
    I hope to listen you Joshua at PgDay.it 2008 about this topic.


    Bye, \fer
  • Gene Selkov at Aug 20, 2008 at 7:04 pm

    On Tue, 19 Aug 2008, Joshua Drake wrote:

    However, one of the problems you may find from the Catalyst side of
    things is that we tend to forget the particulars of various databases
    because DBIx::Class, SQL::Abstract and SQL::Translator take a lot of
    the pain from being cross-database compatible out of the mix.
    Right, that is the constant issue with all the toolkits. My mind is
    something that provides an introduction to Catalyst (which is good for
    Catalyst) but also, "Why use Catalyst with PostgreSQL?". What kind of
    cool features can Catalyst (and is associated libs) expose that
    developers might be missing.
    The coolest feature in postgres is its extensibility. Tsearch2, which
    someone mentioned in an earlier post is based on a user-defined data type.
    I have done some work on the postgres GIST a decade ago, which enabled
    applications like Tseach2, among other things. There is no full-text
    search capability in the postgres engine, but it doesn't have to be
    implemented in the engine. Two Russian astronomers came along and
    figured out how to index arbitrary vectors with a GIST-based data type,
    then they used those vectors to index documents for full-text search
    That's not a very straightforward method, but it works better than the
    built-in search methods in other databases. This goes to say that there is
    almost nothing one cannot do with postgres; however, it is mostly used to
    mimic what everyone else does with databases, with the only
    conspicuous advantages being its zero price, high performance and
    convenience.

    So if you really want to highlight postgres's advantages, focus on
    its extended behaviors.

    But linking that to Catalyst will appear contrived, unless we can think of
    a nice and complete application that uses these things as independent
    components. Think of something that would be difficult to do without one
    or the other.

    I am currently working on porting one of my database applications to
    Catalyst. It has nothing postgres-specific in it, besides the fact that if
    I were to use anything else (mysql, etc.) I would spend more time being
    annoyed than making progress. However, it has a fairly sophisticated data
    backend running in postgres; it has a fairly sophisticated user interface
    done with yui, and there's Catalyst in between.

    Here's one component that may be worth showcasing (after some adaptation).
    Take a look:

    http://cci.uchicago.edu:8000/plot/tree/roach

    (if it bitches about "communication errors", wait a bit and reload the
    page. The server is so weak I am really afraid to publish this link)

    The idea is that there's a large tree structure in the database (about
    7000 nodes), and there is a mass of data linked to the nodes directly or
    indirectly. Think of the tree as a fancy container for things. The user
    has two basic needs: (1) navigate the tree and (2) look into things
    hanging on it. Looking into things, in most cases, is easier said than
    done, as it may require expensive processing. Another interesting notion
    that may be of general interest is branch aggregates, the simplest
    examples of which are the count of things hung on a branch and the mean
    value of the common attribute, but it can be much more complicated than
    that.

    I will not go into detail explaining the meaning of this application (it's
    for surgical education -- highly specific), but it should suffice to say
    that there's a tree, there's data linked to the tree, and there can be
    multiple views of the tree itself and the data it contains. The data is in
    postgres, served with Catalyst, rendered in the browser with YUI, and
    because the backend portion of the rendering process is expensive, the
    view loads itself dynamically on user's request.

    This is not a good example of easy and painless development, but I have
    solved so many typical problems while building it that it could be used to
    represent a sophisticated web technology, if properly generalized.

    Here's a list of fairly difficult things that it can illustrate:

    1. Storing a tree (nested-segment representation is used)

    2. Tree traversal (I have developed (and borrowed) a couple dozen
    utility queries for doing things with the trees).

    3. Some very involved DBIx::Class queries

    4. A combination of DBIx::Class subqueries with pass-through queries
    used where the task is beyond DBIC's ability.

    5. Tree rendering with YUI

    6. Thomas Pietrzykowski's AJAX controller is used to call the
    applications controller actions from the browser

    http://thomas.pietrzykowski.de/blog/?p=1

    7. External processes are called by the backend to process the data (R,
    gnuplot, etc.)

    This would be the kind of example I would like to use, but instead I had
    to spend months figuring things out by trial and error.

    If there is any interest in this kind of thing, I can spend some more time
    to generalize it and make it publishable.

    Alternatively, I could collaborate with someone to demonstrate the use of
    extended data types in postgres. I myself will not be able to attend any
    conferences outside UK for a year or so.

    Regards,

    --Gene
  • Joshua Drake at Aug 20, 2008 at 7:50 pm

    On Wed, 20 Aug 2008 13:04:52 -0500 (CDT) Gene Selkov wrote:
    Right, that is the constant issue with all the toolkits. My mind is
    something that provides an introduction to Catalyst (which is good
    for Catalyst) but also, "Why use Catalyst with PostgreSQL?". What
    kind of cool features can Catalyst (and is associated libs) expose
    that developers might be missing.
    The coolest feature in postgres is its extensibility. Tsearch2, which
    someone mentioned in an earlier post is based on a user-defined data
    type. I have done some work on the postgres GIST a decade ago, which
    enabled applications like Tseach2, among other things. There is no
    full-text search capability in the postgres engine, but it doesn't
    have to be implemented in the engine.
    Well as a note, 8.3.x has Tsearch integrated. It is no longer user
    defined. Your point about being user extensible is certainly true.
    So if you really want to highlight postgres's advantages, focus on
    its extended behaviors. Agreed.
    But linking that to Catalyst will appear contrived, unless we can
    I am not sure about that. The point of the talk would be, using these
    features in Catalyst.

    Here is an example. I was under (the false) assumption that Hibernate
    can't use natural or composite keys. In fact, I thought that
    Catalyst/DBIx was the only one that could.

    I am going to bet that a lot of developers think that a integer based
    key is required (which is really bad), when in fact it isn't. That
    would be a perfect thing to show the capabilities of, including why you
    would do such a thing.

    Alternatively, I could collaborate with someone to demonstrate the
    use of extended data types in postgres. I myself will not be able to
    attend any conferences outside UK for a year or so.
    That was definitely some very cool items.

    Joshua D. Drake

    Regards,

    --Gene

    _______________________________________________
    List: Catalyst@lists.scsys.co.uk
    Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
    Searchable archive:
    http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site:
    http://dev.catalyst.perl.org/

    --
    The PostgreSQL Company since 1997: http://www.commandprompt.com/
    PostgreSQL Community Conference: http://www.postgresqlconference.org/
    United States PostgreSQL Association: http://www.postgresql.us/
    Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupcatalyst @
categoriescatalyst, perl
postedAug 19, '08 at 11:41p
activeAug 20, '08 at 7:50p
posts6
users4
websitecatalystframework.org
irc#catalyst

People

Translate

site design / logo © 2022 Grokbase