Received: from localhost (unknown [200.46.204.184]) by postgresql.org (Postfix) with ESMTP id 7D7779F9944 for ; Sat, 3 Nov 2007 16:19:53 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.184]) (amavisd-maia, port 10024) with ESMTP id 19472-08 for ; Sat, 3 Nov 2007 16:19:38 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from smtp158.iad.emailsrvr.com (smtp158.iad.emailsrvr.com [207.97.245.158]) by postgresql.org (Postfix) with ESMTP id B34A39F9921 for ; Sat, 3 Nov 2007 16:19:41 -0300 (ADT) Received: from relay5.relay.iad.emailsrvr.com (localhost [127.0.0.1]) by relay5.relay.iad.emailsrvr.com (SMTP Server) with ESMTP id 9E3536553A0; Sat, 3 Nov 2007 15:19:40 -0400 (EDT) Received: by relay5.relay.iad.emailsrvr.com (Authenticated sender: chander-AT-otg-nc.com) with ESMTP id 0B2CD64FC6C; Sat, 3 Nov 2007 15:19:40 -0400 (EDT) Message-ID: <472CC949.6000608@otg-nc.com> Date: Sat, 03 Nov 2007 15:17:29 -0400 From: Chander Ganesan User-Agent: Thunderbird 2.0.0.6 (Windows/20070728) MIME-Version: 1.0 To: Magnus Hagander CC: Dave Page , josh@agliodbs.com, pgsql-www@postgresql.org, Peter Eisentraut Subject: Re: Training events policy ... first test case References: <200711031703310000@114527600> <472CADE4.5000807@hagander.net> In-Reply-To: <472CADE4.5000807@hagander.net> Content-Type: multipart/alternative; boundary="------------090902060000020409010209" X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200711/76 X-Sequence-Number: 12854 This is a multi-part message in MIME format. --------------090902060000020409010209 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Magnus Hagander wrote: > Dave Page wrote: > >>> ------- Original Message ------- >>> From: Chander Ganesan >>> To: josh@agliodbs.com >>> Sent: 03/11/07, 16:31:33 >>> Subject: Re: [pgsql-www] Training events policy ... first test case >>> >>> Actually, I think that its not unusual for us to see a few enrollments >>> the week before a class... so listing classes that are "next week" is a >>> good idea - i'm sure this is a trend seen by all training companies.... >>> >> Here's a possibly crazy idea - how about we remove the 3 or 4 listings from /index.html altogether and replace them with a dynamically generated summary saying something like: >> >> "There are 24 training events in 9 countries scheduled over the next 6 months from OTG, EnterpriseDB, Command Prompt, 2nd Quadrant and others. View the complete schedule to find the PostgreSQL training you want." >> >> The numbers are easy to calcuate of course, and we could just grab 3 or 4 company names randomly. >> > > I like it. Takes away the whole problem, really. So +1 from me. > > I've got mixed feelings about it, for a few reasons: - The problem, as I see it, is really in culling from the list those companies that post training events to achieve one of the following: A. Gain "front page" status by posting events they have no intention of running (this definitely addresses that). B. Gan "overwhelming odds" status by listing a whole lots of events, the majority of which they never intend to run. I know of a number of different training companies (not all of which offer PostgreSQL training) that do the following: - List courses in many cities at once for lead generation. - Follow up with those leads by "consolidating" them into a single class they run in a single city. Consider that when you advertise on Google if you have three impressions per page (say 2 natural search results and a single paid result) your leads actually go up by more than a factor of three...the more impressions the better results, and the response is non-linear. The same thing occurs with the PG training page. Legitimate providers are unable to compete as effectively because they list courses they intend to run...and not a mashup of courses designed to generate leads. The proposal drives people to look at a training page, and limits those "front page" impressions. However it does not address the fact that when someone goes to the training page they have to somehow sift through an immense amount of courses, many of which may not run, and many of which are designed for lead generation, rather than to drive enrollment in a single class. The question is really, how do you weed out the legitimate courses from the not-so-legitimate ones? I think that there isn't a really "good" way, and hence some sort of restriction... Some ideas: - Have training providers provide the name of their instructors and a few references, then limit them to posting courses that those instructors would be able to teach (and an instructor can't be in two places at once). This however placed an undue burden on PGDG, since now you have to weed out instructors. - Have providers pay a non-trivial administrative fee to list a course ($1000 per course listing?) . A small fee would be a license to ill, a larger fee however would serve as a deterrent (and the fee is a donation..). This however places an undue burden on smaller companies or companies that are venturing into this area and thus stymies growth.* - Have providers pay a non-trivial fee for each course over the 8 per year free listings. Say 8 free listings per year, others are $1000 apiece. This would allow EDB's and OTG's to legitimately list courses they intend to offer, and prevent "spam" type postings.* My guess is that those folks that aren't legitimate will probably start to drop off when they find the cost of using bait and switch techniques becomes unfeasible. 8 a month works even better for those growing companies. They can offer lots of courses at the start of the year (back to back, whatever) and if they are successful they can pay the fee to continue the practice later in the year. I like the third one, it allows for community growth, and the regular courses offered by BNR, Modern, and others. If a company really plans to offer 30+ courses a year and run them, I'm sure that they wouldn't mind paying a bit extra...especially considering they are "cleaning up" running all those courses with the large frequency. In general, people, once they enroll or are in contact with a single vendor, don't go out of their way to search for something else unless their experience (taking a crappy course) drives them elsewhere. If I have someone enroll and tell them that they have to wait until next month, or fly to some other city for the course, they won't be happy...but more times than not they will most likely do it...since it saves them the work of looking elsewhere for training. It's a sad thing to say, but I think its the truth. An example for us was a Sony subdivision company. They went somewhere else for training and absolutely hated it...their training was someone with some unix admin experience, but almost no PG experience. Their course was originally scheduled in the LA area, but they ended up having to fly to a different city for the course. At the time, they were even reluctant to provide the name of the company, since they were considering pursuing legal action to get their money back When they looked to train other staff they called and gave us the third degree, asked for references, asked technical questions, asked about experience, etc. I think if customers did that right off the bat, the whole training industry would be a lot smaller.... -- Chander Ganesan Open Technology Group, Inc. One Copley Parkway, Suite 210 Morrisville, NC 27560 Phone: 877-258-8987/919-463-0999 http://www.otg-nc.com --------------090902060000020409010209 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 8bit Magnus Hagander wrote: 
Dave Page wrote:
  
------- Original Message -------
From: Chander Ganesan <chander@otg-nc.com>
To: josh@agliodbs.com
Sent: 03/11/07, 16:31:33
Subject: Re: [pgsql-www] Training events policy ... first test case

Actually, I think that its not unusual for us to see a few enrollments 
the week before a class...  so listing classes that are "next week" is a 
good idea - i'm sure this is a trend seen by all training companies.... 
      
Here's a possibly crazy idea - how about we remove the 3 or 4 listings from /index.html altogether and replace them with a dynamically generated summary saying something like:

"There are 24 training events in 9 countries scheduled over the next 6 months from OTG, EnterpriseDB, Command Prompt, 2nd Quadrant and others. View the complete schedule to find the PostgreSQL training you want."

The numbers are easy to calcuate of course, and we could just grab 3 or 4 company names randomly.
    

I like it. Takes away the whole problem, really. So +1 from me.
I've got mixed feelings about it, for a few reasons:

- The problem, as I see it, is really in culling from the list those companies that post training events to achieve one of the following:
    A.  Gain "front page" status by posting events they have no intention of running (this definitely addresses that).
    B.  Gan "overwhelming odds" status by listing a whole lots of events, the majority of which they never intend to run.  I know of a number of different training companies (not all of which offer PostgreSQL training) that do the following:
       - List courses in many cities at once for lead generation.
       - Follow up with those leads by "consolidating" them into a single class they run in a single city.

Consider that when you advertise on Google if you have three impressions per page (say 2 natural search results and a single paid result) your leads actually go up by more than a factor of three...the more impressions the better results, and the response is non-linear. 

The same thing occurs with the PG training page.  Legitimate providers are unable to compete as effectively because they list courses they intend to run...and not a mashup of courses designed to generate leads.

The proposal drives people to look at a training page, and limits those "front page" impressions.  However it does not address the fact that when someone goes to the training page they have to somehow sift through an immense amount of courses, many of which may not run, and many of which are designed for lead generation, rather than to drive enrollment in a single class.

The question is really, how do you weed out the legitimate courses from the not-so-legitimate ones?  I think that there isn't a really "good" way, and hence some sort of restriction...  Some ideas:

- Have training providers provide the name of their instructors and a few references, then limit them to posting courses that those instructors would be able to teach (and an instructor can't be in two places at once).  This however placed an undue burden on PGDG, since now you have to weed out instructors.
- Have providers pay a non-trivial administrative fee to list a course ($1000 per course listing?) .  A small fee would be a license to ill, a larger fee however would serve as a deterrent (and the fee is a donation..).  This however places an undue burden on smaller companies or companies that are venturing into this area and thus stymies growth.
- Have providers pay a non-trivial fee for each course over the 8 per year free listings.  Say 8 free listings per year, others are $1000 apiece.  This would allow EDB's and OTG's to legitimately list courses they intend to offer, and prevent "spam" type postings.

My guess is that those folks that aren't legitimate will probably start to drop off when they find the cost of using bait and switch techniques becomes unfeasible.

8 a month works even better for those growing companies.  They can offer lots of courses at the start of the year (back to back, whatever) and if they are successful they can pay the fee to continue the practice later in the year.

I like the third one, it allows for community growth, and the regular courses offered by BNR, Modern, and others.  If a company really plans to offer 30+ courses a year and run them, I'm sure that they wouldn't mind paying a bit extra...especially considering they are "cleaning up" running all those courses with the large frequency.

In general, people, once they enroll or are in contact with a single vendor, don't go out of their way to search for something else unless their experience (taking a crappy course) drives them elsewhere.  If I have someone enroll and tell them that they have to wait until next month, or fly to some other city for the course, they won't be happy...but more times than not they will most likely do it...since it saves them the work of looking elsewhere for training.  It's a sad thing to say, but I think its the truth.  An example for us was a Sony subdivision company.  They went somewhere else for training and absolutely hated it...their training was someone with some unix admin experience, but almost no PG experience.  Their course was originally scheduled in the LA area, but they ended up having to fly to a different city for the course.  At the time, they were even reluctant to provide the name of the company, since they were considering pursuing legal action to get their money back

When they looked to train other staff they called and gave us the third degree, asked for references, asked technical questions, asked about experience, etc.  I think if customers did that right off the bat, the whole training industry would be a lot smaller....  
-- 
Chander Ganesan
Open Technology Group, Inc.
One Copley Parkway, Suite 210
Morrisville, NC  27560
Phone: 877-258-8987/919-463-0999
http://www.otg-nc.com
--------------090902060000020409010209--