Sorry for not responding to my original post. I forgot to enable midrange-L
delivery when I posted that comment and was wondering why nobody was
responding :-)
Let me summarize some thoughts based on what has been discussed. Note I am
not trying to make money for IBM on some of these because they DO need to
pay a price to get new blood on the machine. Their negligence in the past
DOES come with a cost for the current/future. They need to do some "loss
leaders" to accomplish that task. With all of the enhanced processing with
Power7 it shouldn't be that far of a stretch to get a couple hundred
developers on a P30 for "The Hobbyist" scenario below.
---The Hobbyist---
Jim Franz mentioned the multiple developers per LPAR approach, and I think
there is a place for that. A price of $10/month would be reasonable here.
5GB of DASD. No services offered by the hosting company other than applying
updates on a regular basis. Remember that we are wanting poor college kids
to hop onto this bandwagon, and $10 is something they might not think twice
about. Again, IBM/Vendor xyz is making zero money on this and instead it is
an investment for the future.
---The Single Person Entrepreneur---
Next level would be somebody wanting their own tiny LPAR where they could
have more control over things like applying their own digital certificates
or creating an Apache instance with many virtual host entries or running
more than just RPG (i.e. PHP and Java). A LPAR like this would have 2GB
memory, 50GB DASD, and would cost $50 to $100. Again, IBM isn't making
money here (or maybe they break even) - they are investing for the future.
No services offered by the hosting company other than applying updates on a
regular basis. Cap the CPW at 500.
---The 5 Person Shop/Business---
By putting limits on the "tiny LPAR" you get people to upgrade into
something bigger where you *can* actually make money. This would have 4GB
to 6GB of memory, 250GB to 500GB HD, and cost $100 to $500. Cap the CPW at
1000. This would be the final tier that I would be concerned about because
at this point we have introduced a growth path that has a very low cost
entry point (i.e. $10/month) all the way up to the point of somebody that is
obviously making money and can afford something more expensive like
$2k/month.
I am looking forward to see what IBM comes up with because guys like Steve
Wills have publicly stated it is a big topic on their minds.
Aaron Bartell
http://mowyourlawn.com
http://mowyourlawn.com/blog/
midrange.com
