Re: native Oracle-port on Linux -- Anyone care for a Business Plan

In article <67cqc4$ctk_at_bgtnsc03.worldnet.att.net>,

        Nilanjan Bhowmik <nilu_at_worldnet.att.net> wrote:

>Josue Emmanuel Amaro wrote:
 

>> Lets face it, management at Oracle will simple answer with "show me the money".
 

>> Believe me, if someone can come up with a business plan with some hard
>> numbers, maybe someone will listen.
 

>If we can create a solid business plan and may be small presentation
>( anyone here has pull in ORACLE ? ) to them I think they will consider
>it. We may need to involve Caldera in it - they may be trying it by now
>but it does not harm to give them support for it.

I can't speak specifically for Oracle, but I had a long conversation with someone at Progress about their DBMS's lack of Linux support. Amongst the comments to arise:

• Technical porting issues are not the obstable, support is the biggie We would have to be able to guarantee *huge* revenue to change their minds on this
• Most DBMS makers are trying to *reduce* the numbers of different ports/versions, they getting choked by the large numbers of SKUs they now muct deal with and support
• Progress has no objection to Linux per-se, they just object to having to make and support yet one more native binary product, and they won't support anything that runs under emulation. If SCO and Linux could run the same native binary Progress would (they told me) have no objection to supporting both platforms with that binary.

These comments are amongst the things that got me involved in the 86open project (http://www.telly.org/86open). The fact is that SCO, Linux, SolarisX86 and the BSDs all have binary formats that are oh-so-similar but just different enough to be a pain, and this diversity presents a significant obstacle to anyone looking to port to Unix-on-Intel.

One more point specifically on Josue's comment:

The HUGE difficulty in presenting a business plan regarding the acceptance of Linux, is that there is *still* no really accurate way to indicate to software vendors just how many Linux users are out there, let alone how many would be willing to buy an expensive DBMS. The IDGs and other research firms on which the industry and its media depend, won't admit Linux exists until someone pays they to do it. Linux doesn't even show up on current breakdowns of the Unix market -- how is an ISV to know just how big this market is? Furthermore, just how much of this market is prepared to spend significant cash on applications, considering that it hasn't spent much on the OS and significant utlities.

Yes, I know Caldera sells a version of Linux for $399, and businesses who buy that clearly indicate a willingness to pay for software -- but how much of the whole Linux world is that?

These are questions that must be asked and answered before many ISVs will get involved -- not just Oracle.

-- 
  Evan Leibovitch, Sound Software Ltd, located in beautiful Brampton, Ontario
 Supporting PC-based Unix since 1985 / Caldera & SCO authorized / www.telly.org
When the only tool you have is a hammer, every problem comes to look like a nail