Re: oracle server for learning purposes?

From: DA Morgan <damorgan_at_psoug.org>
Date: Mon, 23 Jul 2007 06:56:05 -0700
Message-ID: <1185198965.393844@bubbleator.drizzle.com>

hjr.pythian_at_gmail.com wrote:
> On Jul 23, 7:13 pm, Niall Litchfield <niall.litchfi..._at_gmail.com>
> wrote:

>> On Jul 23, 6:53 am, hjr.pyth..._at_gmail.com wrote:
>>
>>> If you exclude all use except A, there is no need to point out that B
>>> and C are not allowed. Once you state that B and C are not allowed, on
>>> the other hand, you are stating that A is not all-encompassing.
>> I think, and I'm only married to a lawyer and she doesn't practice in
>> california which is the relevant jurisdiction, that that's an argument
>> rather than a rule/custom in law.

>
> Indeed it's an argument. Which is what lawyers are good at.
>

>>> Put another way: you only clarify when clarification is needed. That
>>> the Corporation sought to clarify a 'thou shalt only use this product
>>> to do A' statement tells us (me, at least) that their first sentence
>>> is not proscriptive of all uses except A.
>> lawyers only clarify when clarification is needed? That must be the
>> reason for all those email disclaimers then - clearly written because
>> clarification is needed.

>
> Lawyers, even your wife, will call that a non sequiteur. A disclaimer
> disclaims. And hopefully in clear and concise language, because the
> less clear and less concise it is, the more wiggle room there is for
> your opponents to make a case.
>

>>> Besides which, there's more to it than the reading of two sentences.
>>> You have to read the entire license, not two sentences. And the list
>>> of 'you may nots' included in part 3 of the license is further
>>> evidence that the license **as a whole** is not proscriptive of non-
>>> prototyping use, whatever one or two sentences may purport.  And there
>>> remains the fact that I think we're agreed on: no definition of
>>> 'prototyping' is offered, so pretty much anything bar commercial use
>>> of the software could be construed as being allowed.
>> But the only part of the license that you seem to be offering is the
>> one line about third party training. There doesn't appear to be any
>> other segment of the license that mentions learning/home use at all.

>
> Then you haven't been reading what I've read.
>
> Part 1 (using the references mentioned above by Hans) contains two
> sentences which contain within them an unnecessary clarification that
> seems to negate the flat prohibition of everything bar prototyping
> that sentence one seems to assert.
>
> Parts 1 and 3 contradict each other, because the list of 'may nots'
> does not claim to be 'for example, but not limited to'
>
> And there is no definition of 'prototyping' that precludes the use of
> the software for solo home learning purposes, whereas reasonable
> people might well suggest that if you're going to prototype an
> application for RAC, for example, you will first need to learn what
> RAC is, how to install it, how to configure it and so on.
>
> That's three arguments against the proposition you started with, not
> one, and I've made all three of them repeatedly.
>
> There is in fact a fourth, which is that the license explicitly seeks
> to prohibit commercial use of the software in a third-party training
> environment; it explicitly prohibits the commercial use of the
> software in a development environment; and it explicitly prohibits the
> use of the software in a commercial database management/internal data
> processing environment. That's commercial trainers, developers and
> DBAs dealt with, and establishes a pattern of what the license seeks
> to prohibit, namely commercial exploitation of the software by anyone.
> Home learning use does not fall within these explicit and clear
> targets of the license (and my third point above rather suggests that
> it positively falls within the explicitly included uses).
>

>>> I have two barristers who'll back that three-part interpretation
>>> (because they gave it to me first!).
>>> But really, that's neither here nor there, since the only place such
>>> arguments carry real meaning is a courtroom. What I *don't* want left
>>> on the record is statements of 'but of course, you aren't licensed to
>>> learn with EE'. Or 'Clearly, using the software is not legal'. There's
>>> no 'of course' about it and 'clearly' there's room for considerable
>>> legal argument, at least.
>> My take is that there is usually room for legal argument about nearly
>> everything, unfortunately. I am not as convinced as you - though I
>> don't have the barrister's opinions on my side either - that basing a
>> case that because a similar activity is specifically excluded and my
>> activity isn't excluded I'm OK is the stringest one to be making
>> against a license that mutliple times gives only one permitted usage
>> and excludes all others.

>
> You are again only working on one of the points I've made. The whole
> thrust of my blog piece and the posts I've made here is: the license
> has to be taken as an entire document, not one or two points out of
> context. And as a whole, the document lists a number of things you may
> not do, but does not say 'these are examples only, and what is not
> permitted is not limited to this list'. It's a fairly standard bit of
> legalese that is easy enough to drop into any agreement, after all:
> "You may not use the software to, for example but not limited to, (a)
> blah blah blah..." That they chose not to include such a get-out
> clause is instructive.
>

>>> The only thing that is clear to me, therefore, is that no-one is
>>> breaching any term of the license in downloading the software and
>>> using it for non-commercial, solo home learning purposes... or that,
>>> if they are in breach of the license, it's not clear-cut and obvious
>>> that they are. In those circumstances, absent the Corporation proving
>>> the point with a test case, there's really nothing to get worried
>>> about. Suggestions to the contrary remind me a lot of Microsoft's
>>> claim that Linux infringes numerous unspecified patents: worrying,
>>> frightening, causing concern... but fundamentally, mere FUD until the
>>> company proves otherwise.
>> As I did say in my blog, such was also my understanding - I was
>> prompted to go and re-read the license in another context - namely the
>> development part of the discussion, but not withstanding this the non-
>> commercial part of the license that I thought was there just isn't. If
>> suggesting that people read the license of software they download to
>> make sure that they can use it as they assume they can is FUD then I'm
>> probably guilty as charged.
>>
>> I also note that the license was amended in Sept 2005 (some years
>> after I last read it) just before the release of Oracle XE. That
>> product is most definitely and without any FUD free for home (and
>> indeed commercial use). I believe that you and I have similar views
>> about it's suitability for learning the dba role, but it's definitely
>> free to use without any restrictions on purpose  (there are hardware
>> restrictions for those unfamiliar with the product).

>
> And it's useless for learning Data Guard, RAC, ASM etc etc etc.
>
> Learning of which, as I say, would be (could be, I suppose) a
> prerequisite for developing a prototype for it.
>

>> And yes if Oracle were to amend the license to specifically allow
>> private study purposes I'd very much welcome it. Equally if they were
>> to post here, or preferably on an Oracle owned site a similar
>> clarification that showed I was wrong that would be fine. As I said in
>> my blog though in the end you'll need to read the relevant licenses
>> yourself and come to your own conclusions about what is permissible -
>> it should go without saying that anything I (or Howard and his lawyers
>> or anyone else on the thread) say here is opinion only.

>
> And that's fine: 'opinion only'. If your original piece had said,
> 'there's an argument to be made...', fair enough. It doesn't though.
> It says, "[the claim that] you can download the software and learn for
> free is in fact not true". A bald statement of purported fact... when
> it's not fact at all, but your opinion (and one you're entitled to
> hold, naturally enough).
>
> My rebuttal piece ends, "I say that's a misreading of the license and
> ignores context and clues contained elsewhere in the license that,
> read all together, make it clear home learning is a perfectly proper
> and legitimate use of the software." ...and the 'I say' at the start
> of that makes a bit of a difference, I'd suggest.
>
> We are agreed, I think, that the license is an unholy mess and it
> needs to be clarified. We should be careful what we wish for, though:
> I would hate a license that allowed anything at all for XE, crippled
> as it is for would-be DBAs, and banned pretty much anything for the
> 'proper' software. But I suspect that's what we might inherit...

In a practical sense I suspect the license does precisely what it is intended to do. It gives Oracle the ability to make the software available to a wide variety of individuals and organizations and at the same time yank on the leash if something happens that they view as an abuse.

They can't put it out there with no license and they don't want their legal beagles to tighten down the screws so tightly that we would all feel obligated to get a legal opinion before going to OTN.

Contracts are almost always a compilation of clauses; each one dealing with avoiding a previously identified error in judgment. The way this license reads my presumption would be that Oracle is perfectly happy having a large stick in plain sight and doesn't have to pick it up very often.

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu (replace x with u to respond)
Puget Sound Oracle Users Group
www.psoug.org

Received on Mon Jul 23 2007 - 08:56:05 CDT