Re: Controlling the number of users in the database

From: Richard Kuhler <noone_at_nowhere.com>
Date: Tue, 23 Nov 2004 06:51:11 GMT
Message-ID: <z5Bod.58136$hN1.24144@twister.socal.rr.com>

DA Morgan wrote:
> Richard Kuhler wrote:

>> DA Morgan wrote:
>>> DanO wrote:
>>
>>>> I have been tasked with finding a way to control (limit) the number of
>>>> users that log into the Oracle database while using our proprietary
>>>> application product.  For example, a client may buy our product and
>>>> purchase a 100 user license.  We need a way to prevent the 101st user
>>>> from logging into the database.
>> <snip>
>>
>>> I can't imagine anyone buying your product.
>>> We purchase our operating systems flat fee basis. We don't pay extra
>>> when more people log into Linux, or Solaris or HP/UX or AIX. 
>>
>> Well, if it's good for operating systems, it must be the right answer 
>> for every piece of software on the planet.  Case closed.
>>
>>> We pay a per CPU price for Oracle. We don't pay more depending on 
>>> the number of connected users.
>>
>> Some people do.  Lots of Oracle products have a named user licensing 
>> option (including all database flavors).

>
> They are vanishingly few in the current market. And even though it is
> possible ... the option to license by CPU or site also exists.

So? Maybe the posters company has a licensing option for unlimited users as well. What does that have to do with the validity of having per user licensing?

>>> And the same model goes for all of our other software.
>>
>> Well, if you say so I guess I have to believe you but that's 
>> surprising to me given how common user based licensing is with many of 
>> the products I've had to use.

>
> Consumer based products such as those from Microsoft are a different issue.

So if the posters product is a database centric word processor then everything's fine? Why does it really matter? What are these black and white distinctions you seem to see that I don't?

>>> What makes you think we want to deal with a company that
>>> tries to create its own personal pricing model: 
>>
>> To make the company more profitable?  Isn't that ultimately why any 
>> business should buy software?

>
> Not at all. We purchase software to make US more efficient and to
> improve OUR productivity.

That's part of the profitability equation. If the gains don't outweigh the costs then they are making the company less profitable. RDBMS X takes more DBA effort to maintain than RDBMS Y. Is this enough information for you to make a purchasing decision?

> Very few business can claim to make $1 from
> software they've purchase.

What? I can't think of any project that I've ever worked on where the company didn't have a projection on the return on investment for software that they purchase. I suspect you've had to make the case for Oracle over some other RDBMS before. Certainly that rationale was based on lots of factors that influence the companies profitability, right?

>>> Certainly I would never recommend anyone go along with such a plan.
>>
>> Well, hopefully the people making the decisions are more interested in 
>> making money for their investors than drawing a line in the sand with 
>> regards to how you pay for software.

>
> No doubt they are trying to make money for their investors. And I,
> and just about every DBA and SA out here is trying to stay within a
> budget and not make extra work for ourselves. We already have full
> time jobs without having to play accounting games for someone's
> software license scam.

What extra work? I don't see how you leap to this conclusion that some licensing mechanism is going to hamper your productivity. It's simple FUD with no basis in fact. I can think of a relatively simple stored procedure implementation that would have absolutely zero impact on the installation time of the product (assuming that it's using stored procedures for other things) and require no DBA effort to maintain. But frankly, who cares if it adds three hours to the DBA's job every single day if it makes far more than that back somewhere else?

> So unless the company has zero competition ... and we all know the
> chance of that ... they won't last long or the model will change.

That's ridiculous. I don't see how you can believe this. You really have such incredible insight into the marketplace to predict the demise of a product, sight unseen, based on complete speculation regarding what it's licensing mechanism is going to mean to a DBA's productivity?

--
Richard Kuhler

Received on Tue Nov 23 2004 - 00:51:11 CST