Re: Oracle v. Google jury returns partial verdict, favoring Oracle

From: Snit <>
Date: Mon, 14 May 2012 18:00:05 -0700
Message-ID: <>

On 5/10/12 10:27 AM, in article, "joel garry" <> wrote:

> On May 9, 11:55 am, PÓl <Pó> wrote:

>> On 08/05/12 16:55, joel garry wrote:
>>> Disclaimers: Yes, I use google all the time, and yes I have an Android
>>> phone, and yes, I've been abused in commercial situations by Oracle,
>>> and yes, I gave up on linux advocacy years ago when I learned the hard
>>> way how bogus open software can be.
>> How can Open Source be "bogus"?
>> Unlike with closed software - if you don't like it, you can do something
>> about it, at least in theory.

> Sandman and Snit pretty much got what I was driving at.
> Of course that "bogus" comment is way overreaching, and does refer to
> two quite different things: the inappropriateness of expecting anyone
> to be able to do anything about it, and the effect on commercial
> software.
> I think the first is obvious by now, but I still see specific examples
> in the real world. One is my kid - I gave him Open Office (first on
> XP, now on whatever the latest hype is - of the dozen computers in my
> house, his laptop is the only one that can run Portal 2), school
> requires MS format. Hey, OO can write that, right? Well, no. There
> are various little fuckups.

Quite a few. And of the two businesses I suggested LibreOffice for, both have decided it it worth the cost to move to MS Office. It serves them better, even without the compatibility issues.

> Of course, you say, who would really give a good goddamn about some minor
> formatting issues? It turns out, his honors English teacher (who is totally
> hot - how come I only got wrinkly old nutcases when I was in school?) does.

And in many classes it is very important to use APA or MLA or a similar format.

> Now you say, OK, some minor point loss, what's the BFD? It turns out, this
> can mean the difference between Valedictorian or not. You expect him to fix
> it (he's scary-smart, but does have other obligations)? You expect me to fix
> it (he does!)? Even if I could, or knew someone who could, it's not getting
> done in the middle of the night with the paper due the next day. Another more
> obvious issue is hardware support, which is still the major issue it was 10
> years ago, I'm sure you guys could give all sorts of examples.

While the examples are somewhat older, I gave some very specific examples:

    <>     <>

Even though I provided the files I used so anyone could replicate my work, the response of the "advocates" in COLA was to say I somehow forged the results (Peter Köhlman repeats that accusation on a regular basis). Of course, nobody was able to show OpenOffice doing anything better than I showed (Ray tried and I posted his info).

> So yes, there may be minor differences, and they add up, and have
> major, often unpredictable, consequences.
> The second is more interesting, and is really what made me turn away
> from COLA. As someone mentioned, quality should be the real deciding
> factor, and I do think OSS had a major positive impact in numerous
> cases. Most software is indeed crap anyways (and getting worse, is my
> judgement from the social media examples out there, $100B facebook,
> shit).

Here is an example of doing what should be a simple task - rotating an image - in both LibreOffice and Pages: <>

The response was that this is a task that is not common... which, on OpenOffice and LibreOffice, it is not. Of course it is not - the UI for doing so is *horrible*. Usability matters.

> So what I'm thinking of is the Apache and EM stuff from Oracle. I'm
> not dinging Apache, actually I think it is a good example of how OSS
> can work for a commercial environment, making it easier to figure out
> operational issues, at least if you get far enough up on the learning
> curve, and given the milieu of the software world, there are plenty of
> people who do. But as I see it, the adoption by Oracle of linux
> created a large population of so-so cheap programmers creating some
> garbage. So you wind up with low quality perl and bash stuff, which
> is then ported to other unix without even trying to see if it works,
> and this crap being distributed to paying customers. Yes, that is
> entirely Oracle's fault, but it required a linux infrastructure with
> low barrier to entry to create the problem. I see this as the same
> problem with Windows: Dave Cutler's team did an excellent job that
> still stands up technically, it's all the cruft beyond the kernel
> that's the problem.
> Which brings us to what some Sun guy used to say, "the network is the
> computer." How data
> moves around, how people communicate, what groups of people do with
> their computers is really more important than any one installation.
> Or as someone else may have put it, "it's the app, stupid." That
> doesn't mean OS's are trivial - it means they need to get out of the
> way, not cause problems themselves.

And a well done environment can add a *lot* to the functionality / usability of a system. The example I use is OS X with is proxy icons, versioning system, easy finding of where a file is saved from within the window itself, media browser, application services, PDF services, etc. These things matter. A lot.

> The democratization that OSS provides is a two-edged sword: yes, many people
> can come up with good stuff, but many more people have no clue of design,
> engineering or quality. The Linus-centric model for linux itself works around
> this problem, but beyond that there is no control. Sturgeon's Revelation is
> over-optimistic.
> jg
> --
> is bogus.

Received on Mon May 14 2012 - 20:00:05 CDT

Original text of this message