IF one were to omit using X/startX & doing a GUI shell like KDE or GNOME, couldn't just about ANY Linux 'cut the mustard' for this task??
Oh certainly. It's just that it would be a lot of hard work cutting Ubuntu down to the 386/486 level, and what you'd have left, while it would certainly be Linux, it wouldn't be Ubuntu in any meaningful sense. Better to start from Debian,
Slackware, or Linux From Scratch.
TinyGentoo
would be another contender.
If you want to get an idea of the minimum spec for a Linux system,have
a look at
tomsrtbt. It's a linux installation on a floppy disc - a "live
floppy", if you will. It needs 8 meg of ram to boot, runs from
a ramdisc, and doesn't have a minimum cpu requirement.
Aha, ok... something like SELinux gives you folks on Linux, that Windows has had on its filesystems via NTFS & the registry for years in ACL (access control lists): MAC (mandatory access control labels) - an analog of that which I note existed on NT-based OS' for years...
Well, to be fair, the Unix three level permission system is (as far as I can see) isomorphic to ACLs and MACs. That's been around since the seventies. It's just that the guy who wrote the security certification levels back in the 80s was a VMS head. So nothing got certified that didn't use VMS terminology. Windows NT got it because MS poached the chief architect of VMS, but Linux has has comparable features from the start.
That said, having a separate, orthogonal permission system can be useful; I believe the NSA have a Linux box online where they tell you what the root password is and give you a telnet prompt. You can't do anything once you're in, of course, but it makes a nice proof-of-concept.
Makes sense here: They "pre-harden" it for you, but how is THIS
done? SELinux kernel hook addons (this is what I am assuming on this
note)??
Pretty much. There are a whole pile of kernel compile time options to
enable the subsystem, and to configure it. I can't remember too much beyond that -
the last time I tried to set up SELinux by hand, I nearly locked myself out of my own box. So I can see
the appeal of a distro where these things are set up for me.
OK. Hope that answers your questions. Let me know if I missed anything
and I'll see if I can help:)
Hmmmm... can't ANY Linux distro, with NetConfig & 2 NICS inside of it
(hardware-wise), do the same... for the MOST part?
Well, up to a point. You'll not run Ubuntu on a 468 for instance, but you can
do it with smoothwall. And you don't have to harden the distro - it comes
that way out of the box. So deploying SmoothWall should be half an hours work,
rather than three or four days messing on with ACLs and policies
It's not that you can't do it with another distro - and this is true of all
the specialist distros - its just that SmoothWall has been designed to make
one particular application of the technology very simple to implement.
But what do we gain by having a dozen "generic" distros that are
essentially the same?
But what do we lose? This is free software and to most questions the
default answer is "yes, you can". Before anyone starts saying "no you
can't have your own distro", I think I'd need to see someone establish
that harm is being caused.
I don't think anyone has done that so far.
And if they're not the same, what makes them different
Well, from a user perspective, the difference between most distros comes
down to Gnome vs. KDE. And how nice the wallpaper is, since most users
never bother learning how to change it.
Behind the scenes, they can vary quite a bit. The most visible variation has
been over desktop acceleration - Beryl vs Compiz vs Metisse, and all the rest
of them. But there's quite a bit of envelope pushing behind the scenes as
well. Some distros rushed to embrace the 2.6 series of kernels while others
stayed with 2.4 until the major problems had been ironed out. Different
filesystems, different device managers, different boot scripts. We can't just
choose the best one, because for a reasonably up to date distro, we don't know
yet. So everyone makes their best guess, and over time the pros and cons
become apparent. That's where we score over Microsoft: we can do this sort of
massive parallel evaluation and it doesn't cost us a penny.
Really, I think maybe we need a standard unified "generic" linux distro.
It's been done before. UnitedLinux was (someone correct me if I mis-remember)
just such an attempt. What happens is someone starts a new standard distro, and
none of the others pay any noticeable amount of attention. More practically,
there is also the Linux Standards Base which attempts to solve the problem
without requiring anyone to fold their distro and join someone else's project.
Because let's face it, that is always going to be a hard proposition to sell.
It has validity, the argument that more is better does not necessarily hold true.
That doesn't follow. More is not necessarily better, but neither is it necessarily worse. Nor is less automatically better for that matter.
So what else do those distributions serve except egocentrical purposes,
especially since the majority consists from taking a large well-known
distribution and only tweaking it slightly
You mean like Knoppix, which I believe invented the LiveCD, and is
still the recovery disc of choice for a great many of us? Or maybe
DamnSmallLinux, which packs into 50MB and will run on just about anything?
Then there's Smoothwall which vainly flatters the egos of its developers
by providing a dedicated, hardened distribution capable of converting an old
computer into a firewall router?
That's to name but a few. There are a lot of specialist distros out there supporting
a specific activitity, interest or region.
Seriously, for most consumers, assuming Linux is still going after Windows
and the desktop, more choice is not necessarily better, especially not
when it numbers in the hundreds.
If you're worried about users migrating from windows, then we have
enough trouble drawing people's attention to the big names like
Ubunbtu and RedHat. I doubt the existence of tomsrtbt or Astrumi
are even going to impinge upon their awareness, let alone sow
the seeds of confusion
Microsoft is probably protected from retribution from the FSF by the doctrine of first sale.
So then if I buy a copy of Windows, don't install it, and then start
selling copies... no, that can't work. Of course, I'm copying a CD
there which puts it into fairly well defined territory.
So how about if I bought a copy of a game from one of the online
distributors. Can I then decide I don't like the licence and sell the
electronic copy onwards? Can I do it multiple times?
I don't think MS will fall back on the first sale doctrine if they can
avoid it. If they strengthen the doctrine as applied to software, they
weaken their own licencing in the process or so it seems to me.
Keep in mind that "handles the MS-Novell deal" could be interpreted as
"marginalizes the MS-Novell deal".
True enough. RMS and Eben Moglen might have just decided it was a more
elegant solution to allow the deal, and then let Microsoft run away screaming
than it would have been to try and kill the agreement outright. If nothing
else, it certainly lends credibility to the new licence.
Short of overturning GPLv3 in court, I think that possibility represents
the "best case" outcome for Microsoft here. I just don't suppose it to be
the only possibility.
If I were running MS, I would avoid the GPL v3, too.
They don't have much choice, really. If they ever try and sue for
patent infringement (and I'm not holding my breath BTW) then they
are going to have to be able to show that they are not bound by
the terms of the GPLv3, or the defendant can reasonably claim
by conveying covered software, Microsoft grant them a licence
for those patents. Which ought to be enough to get the case thrown
out.
Everyone wins, except perhaps the end user.
How does the end user suffer? I still get to run Linux. One more
flame war on LKML, maybe a couple of projects get forked... I don't see
this affecting me in the long term.
Microsoft and Novell are free to amend the agreement between them,
if it was at all ambiguous in the first place,
Which raises, I think, an interesting point. Because while we don't know
the full details of the agreement, Eben Moglen does. He was allowed (IIRC)
to look over the agreement so he could satisfy himself that it didn't contravene
the GPLv2. And while he's barred by NDA from discussing the
terms of the agreement, that wouldn't necessarily stop him
from exploiting a flaw in the agreement had he noticed one in his review.
Now, I've read your profile, and I have to respect your legal credentials.
I'm a coder, not a lawyer, and I'm not about to argue the law with you.
But I think we also have to respect Moglen's credentials here, and he
had plenty of time to think about how he was going to address the
MS-Novell pact.
So like I said earlier - we won't really know unless and until this gets
tested in court. But I think we have to allow the possibility that Microsoft
might be in trouble here.
Look, Microsoft is not an "Open Source" software company. Neither they,
nor anyone else... are obligated to distribute software under GPLv3
Which is true as far as it goes. The missing detail is the vouchers
MS have been selling for SUSE Linux which have no expiry date. This means
that, in principle, if anyone redeems such a voucher for a copy of SLES,
and if that collection contains any code licenced under GPLv3 at the time
they redeem the voucher, then there's a chance MS may be held to account
under the terms of GPLv3.
Now whether that will stand up in a court of law or not is another matter.
Eben Moglen and RMS seem to think so, since they wrote the new licence
to allow the MS-Novell pact specifically for this reason. But like I say,
we won't know for sure until it's tested in court.
On the other hand it seems reasonably certain that Microsoft sees some
legal exposure there, or they wouldn't be making such a fuss. Because
for all they talk as if the licence poses no threat to them,
they are nevertheless backing away from it at every opportunity.
The thing is that if the GPLv3 does apply, then anyone they sue for
patent violation hereafter is going to be able to claim that Microsoft
licenced the patent for their use - else they had no right to distribute
in the first place. That too will need to be tested in court, but again
it seems that Microsoft are taking the threat seriously.
So that's "so what". It's not Microsoft don't use GPLv3 and we think they should.
It's more a case of MS may already be using GPLv3 which makes them a lot less scary.
BTW, I was agreeing with you. Adding a pinch of sarcasm. Glad I could help
In which case, I apologise. I think my hackles went up with the "welcome to
the real world" line, and I probably didn't read the rest as carefully as
I should have.
The reality I'm speaking about is that of the market place.
mmm...I understand that. The thing that gets my goat is that
more and more people seem to be swallowing the notion that
marketing reality > objective reality. In particular, I keep running into arguments like the following:
Suppose X and Y are mutually exclusive possibilities. Then the argument
goes:
X is supported by objectively verifiable facts
However Y is widely belived to be true
Therefore Y is true and X is false.
It's logic for the Jerry Springer generation, and it winds me right up.
And yeah, I thought you were serious. Sorry about that.
It has never been about arguments, it's about beliefs, convictions and emotions
What has never been about arguments, precisely? Is this going to be one of those everytime anyone says "Microsoft Sucks!"
God kills a kitten arguments? Because frankly, I could screw up my eyes and make myself believe in Microsoft ever-so-ever-so-much,
but that still wouldn't help anyone outside MS implement a spec that defines chunks of its functionality with reference to antique
proprietary document formats.
The arguments are always chosen to support whatever people already believe.
I think you missed the word "frequently" out of that sentence. "Increasingly" would work too.
The thing is though - this is Slashdot. We're a pretty cerebral bunch, on the whole. Certainly, we
see our fair share of faith based reasoning from the fanboys,
and I'll grant that topics like this tend to bring the astroturfers out in force. But if ever there was
a forum where sound logic and factual argument carried weight, this would be it. It's a techie thing -
you can't debug a program by wishing, even if you wish really really hard. That's the first
thing a good coder learns and it colours your thinking after a while.
Welcome in the real world.
You're going to tell me "there is no spoon" in a minute, aren't you?
More seriously, I find it interesting that you seem to think that
the lies and carefully chosen logical fallacies of marketers are in
some way more "real" than sound logic founded upon verifiable facts.
It seems to me a marketer's mind set. You wouldn't be in the business,
by any chance?
OOXML is better than ODF because Java apps don't use native Microsoft widgets. But although wxWidgets demonstrates that non-MS products can indeed conform to Microsoft standards, that doesn't apparently count because you like Visual Studio. Neither of which points is in any way a non sequiteur, probably for reasons that will turn out to involve the mating rituals of crocodiles.
Really, if that's the sort of argument Microsoft are reduced to, I'm surprised the debate has lasted this long.
So by posting this, am I spreading fud about spreading fud? I think I
broke my brain.
Well, only if you're afraid that the news might slow down the take up of
Tech Writers. But frankly, I think that by this time the brand is well
enough established as to be pretty much bullet proof.
Of course, that may not apply to the forthcoming release of Tech Writers 2.0,
but as far as I remember, that's still in the discussion phase, so it's too
early to say anything for definite.
After the initial load time its fairly snappy on most sites provided I
don't open too many tabs but only because of the flashblock and adblock
plus extentions
Yeah, flash and heavy duty javascript can be a problem, certainly. And not just for older
machines, either.
the heavy javascript of some adds was really bogging the machine down
and i really didn't want to turn off javascript completely
I use NoScript for that. Turn scripting off by default, and then whitelist
sites that use it in the Furtherance Of Good. It apparently doesn't get
on with FlashBlock - but then you don't need flashblock if you have NoScript.
It's one of my must-have extensions these days.
Ive had Ubuntu loaded on this machine and it didnt seem any faster then XP
It won't. Full fledged Ubuntu needs a decent system spec. Gnome has got its own
framework overheads, and Ubuntu adds a lot more newbie friendly stuff that ups
the system reqs to about the same as XP. Ubuntu is a lovely distro in a great
many ways, but svelte it ain't.
sure I probably could have custimized it to hell
You could try XUbuntu. Uses XFCE rather than Gnome, and uses a lot less
in the way of resources. You can install it from Ubuntu using synaptic. Did it
in about 10 mins on my wife's old desktop machine. I'm still not sure it
would cope with a PII, but it runs just fine on a Windows ME era AMD Duron
system. Which is more than can be said of vanilla Ubuntu.
Failing that, look at some of the lightweight distros. The slackware
derivatives seem to be good at this, so Vector Linux might be a good place
to start. Or else Damn Small Linux, if you can persuade it to install
on the HD and not run from a ramdisk.
but who has the time and the need to do that
Well, if you don't need it, don't do it. Personally, I'd sooner my
resources were doing what I tell them to, rather than monitoring the system to
see if I needed a help bubble popping up, or a usb drive mounted - but
I admit I'm not the typical user. Then again, if you have old hardware, and you
can't upgrade for whatever reason, it's not a bad angle to explore. It
certainly beats using out-of-support Windows 98 because you can't cope
with XP.
Who would want to use a machine without a PIII or higher chip in the
first place.
Someone who isn't running Vista or XP? Someone stuck with ten year
old hardware?
I am running a dual core centrino chip + 2gigs of ram and it still isnt enough.
I have a ten year old P266 machine I still use from time to time. It runs
Gentoo Linux, X.Org, a heavily customised FVWM desktop, and Firefox. For most
purposes, it's still quite nippy. (And yes, emerge -avuD world is
kind of slow; but that's why distcc was invented:)
When I work on an older machine I just want to cry they are so slow
compared to todays standards.
Oh, certainly. But in most modern operating systems, most of the speed
increase gets used up by infrastructure overheads. That's why a Dual Core
monster now doesn't seem much faster to the user than my top-of-the-range
P266 did back in 1997. The gripping hand is that if you use a lightweight
distro, then firefox will run just as happily on 10 year old machines as it
will on the latest hardware.
Which, if you're stuck using something from the dark ages, can be a good thing.
Well, the source currently published would stay published, of course.
Most of the free software devs are pretty much on-side about the source availability,
so most of their updates would be released.
Most of the proprietary stuff would stay unpublished - so no change there.
Where we might see a difference would be the cases where a company has to e
sued before they will comply with the licence. I think that's a fairly slim
slice, however. Most of the corporations involved with FOSS have learned that
there are benefits in having a community willing to work on your codebase.
So, not much difference on the whole. And in return, we get to run
chmod -R 0777 /
on the whole of our culture. I'd say that was a good deal.
Sure, the GPL is viral. I don't think anyone really denies that.
This is why we don't let anyone in our server room unless they've had a full course of
inoculations. We have to insist on this policy, otherwise all it takes is one sneeze to
infect all our computers. The next thing you know we'll be getting cease-and-desist letters
from the FSF and...
Or to put it another way, the problem with "viral" in this context isn't so much that
is is not accurate; it's just that it's woefully imprecise. If you think about it
the term is accurate only in a single limited scenario - one where you distribute proprietary
binaries linked against GPL libraries. Of course, that's why the LGPL was invented, and these days
the situation rarely occurs. Al of the other associations people have for the word "viral", death,
disease, broken computers, epidemics, rootkits... they don't really apply.
Trouble is, there are kids out there who think it's really cool to release under a "viral licence"
and there are PR drones out there who aren't above capitalising on the negative connotations.
So, on the whole, I think it's best to avoid the term.
There is no biological response, yet, to keeping your information private.
Let me propose an experiment. Take some people you don't know particularly well, and open their mail. Make sure they catch you reading it.
See how many pro-active responses you elicit. You can report back when you get out of hospital.
Don't blame that market for the privacy issues, blame your government that created the cartel (mercantilism, not capitalism).
Something about they way you phrase that leads me to imagine how you mus t have been as a child:
Kid:"Hey, dada21, Who do you think is strongest? The Thing or The Hulk?"
dada21:"I think the market ought to be allowed to decide that on its own terms, don't you?"
Do we have to bring "the market" into everything? Privacy is a human issue; how about we leave the corporations out of it for five minutes?
I can't think of one reason why I need or require complete privacy.
Key word there being "complete", I suppose. So what level of privacy do you need?
What do I have to hide?
I dunno. But if you give me free and complete access to all your data and all your activities,
I bet I could write you a list.
This is to me the downside of using open source code in one of your projects - at any time your ability to use future versions with their bug fixes, security fixes, etc. may go away.
Of course, the same is true of proprietary software. The vendor could cease trading, withdraw the product,
release a new version that broke backwards compatibility...
So really, you can be left in the lurch in either case. It's just that with open source there exist
options that are not available for the proprietary solutions. Like organise a fork, or hire a developer to
make changes based on the last available release.
And even if you're unable or unwilling to do any of those things, the worst case scenario is still
no worse than the proprietary one.
I'm not saying you're wrong in the general case,
but I did a little bit of searching in the case of EA.
Have a look at
this:
I don't know if I'm reading that right, but those figures seem
to show negative sales growth (compared to a positive from the
rest of the industry). There are some other interesting figures
in there too, such as the return on assets.
It seems there may well have been problems at EA over the
last 12 months, and the last quarter in particular.
Of course, the extent to which their choice of material
is a factor is still open to debate.
Still, it may well be more than just routine PHB verbiage.
And so, I say again: "Privacy is dead. Get over it!"
And I say again, that's a stupid thing to say. Privacy is not some lifeform
that can be slain, never to walk the earth again. It is a courtesy that we
customarily extend to one another; one important enough to be protected by
law in many countries. "Privacy is dead" might make a good sound bite
for a controversy loving CEO, but as a basis for argument it grossly misrepresents
the issues
But your privacy is already losing to increasing levels of surveillance right now while we debate it.
I think you're conflating privacy with surveillance. Yes, there is more
and more data being gathered on all of us. However, the availablily of
the data does not invade my privacy any more than does the window in my
front room. However, if you stand outside my house and peer in through
that window for long periods of time - that's an invasion of privacy,
and you can expect me (or anyone else you try this with) to get annoyed
about it.
It's not the availability of the data that invades my privacy - it is the
abuse of that data. And that battle is far from over.
And the problem (if you feel that this kind of transparency is a problem)
Now, if you'd been reading my posts, you'd know that I don't
have a problem with - let's call it Brinian Transparency - as such.
I just don't believe it's ever going to happen. It only works if it works
for everyone, and the principle offenders are the ones who have the most
compelling case that their doings should be exempt from transparency.
So the police will claim they need their privacy lest we enable organised
crime to evade the law; corporations will predict the collapse of the
economy if their precious trade secrets fall into the hands of overseas
competitors; the President will become a target for terrorists, if his
movements become public knowledge; and so on, until every body large enough
to afford their own lobbyist has cobbled up a reason why they should not
be part of it all.
The only people who will be transparent will be the likes of you and me.
Do you have a better suggestion than Brin's world?
This is the trouble with basing arguments on the writings of SF writers.
In their profession they need to devise scenarios that are internally consistent,
but the initial assumptions of that world need be no more than mildly
plausible. In fact a legitimate use of S/F is to examine some of the more
extreme ends of the probably curve. The problem with this is that when they
come to write seriously, they don't always examine their starting assumptions
as rigorously as would a scientist or a scholar.
Brin's solution might work - if you can get the infrastructure into place.
But that's the part he skips over, both in the Wired essay, and in Earth.
So if you can explain how you think we can establish this blessed social
order then by all means do so. But until you do, I'm not going to stop
objecting to invasion of privacy any more than I'm going to sponsor research
into Tree-Of-Life-Root, just because Larry Niven's vision in Protector was
so compelling.
Do you have a better suggestion than Brin's world?
We can make sure that privacy (as defined above) is recognised as a basic
pre-requisite of a sane society, and we can work to ensure that unwarranted
abuse of personal data is a criminal act, and that it remains a criminal act.
And we can campaign to set some sensible limits on where people want to put
these blasted cameras, as well, see if we can stop the rot before we get to
Orwell's nightmare, with cameras in every home.
And in the meantime, we can even campaign for greater transparency in
governments and corporations - because I don't think Brin's idea is a bad
one, any more than I think faster than light travel and limitless cheap
energy are bad ideas. I just don't want to adopt a social policy that's
dependent on any one of the three, at least until we get reasonable grounds
to think they might be achievable.
Well done on missing my entire point - but you didn't read the David
Brin's article, did you?
This is what I wrote two posts back:
[Brin] seems to think that the problem with lack of privacy is not the
lack itself, but the asymmetry of the arrangement. Various groups are
allowed to know all they like about me, but I am not allowed to know
anything about them.
His solution is that all surveillance information should be available to
everyone, and then we could each watch one another and keep the system
honest. And it's a good plan.
The trouble is, the first time the proposal is taken seriously, there's
going to be a lot of handwaving about IPR and national security and terrorist
threats, and the first thing that will happen is that we'll see a great long
list of people and organisations who are exempt. And then we'll have no legal
And when the legislation passes, we'll find that we have surrendered the last of
our rights to privacy, and that the watchmen are no more watchable than before.
In order for that not to happen, you need to change the way an entire culture
thinks. You need to persuade Coca-Cola to leave the AV feed on the boardroom
when secret formulae are being discussed. You need to get Microsoft to stop
deleting every email that might ever be deposed in a court of law; you need
to get the CIA to accept that it's oh-kay for foreign governments to
know the names and addresses of all their agents overseas. And frankly, I
don't see that happening any time soon.
But let's suppose I'm wrong about that. If you can do social engineering on that scale,
then why not just engineer things so people don't abuse their authority in the first place?
For that matter, the two are not mutually exclusive.
And that, I think, is where we find ourselves at issue. When you say "privacy
is dead. Get over it" You're suggesting that the debate already over (I
strongly disagree) and that we should all meekly accept increasing levels of
surveillance now whilst we campaign for greater openness and accountability
at some future time. As far as I'm concerned, that's the wrong way round.
You put the infrastructure in place so that we can watch the watchmen,
and maybe I'll happily accept that they can watch me more closely in turn.
But until that day comes I continue to value my privacy.
Apologies for the style and tone; it's just that people quoting McNealy out of
context to try and justify increased levels of government intrusion is a
pet peeve of mine. I probably went a bit over the top
you think I'm too lazy to type 3 words into Google.
I'm trying not to make assumptions about you - but you sound
like you're too lazy to type three words into Google. It as if
you said to yourself "well, I can't remember if this was Scott McNealy, or
Jonathan Schwartz, but if I just say I'm sure you all know who I mean
then everyone will assume I do know, and I don't have to risk embarrasing
myself by getting it wrong".
The trouble is that once people get that impression, it
undermines your credibility, since the reader can't help but
wonder what else you might be trying to fudge.
Think of it as feedback.
but I really didn't think I needed to cite the source, since the idea is clear.
Then don't provide a citation at all. This is Slashdot, not a paper
for an IEEE conference - you don't need to cite anyone. But if you do,
don't be half-hearted about it; it reflects badly on you, and weakens
your credibility.
Perhaps you labor under the misconception that only "authorities" have anything of value to say?
No, but then neither do I think that Larry Niven understands astrophysics
in the same depth as Stephen Hawking. If there's a debate about black holes,
Hawking's opinion is going to carry more weight. If you think that Brin's opinion is
as good as anyone else's, then have the courage of your convictions and say so up front.
"Noted SF author David Brin makes this case far better than I ever could in a wired article from
the 90s..." is all you need to write.
Don't have to. Millions of people already have.
Well, anyone who publishes their bank details and social security number on
MySpace almost certainly falls into the category of "idiot" and can be safely
ignored. I'll make an exception for
Hasan Elahi
as I will for anyone who is doing so to make a point
about privacy - if you'll provide a link to the specific page. In the general
case though, I don't think human ignorance and carelessness carries much
weight in this debate.
And, you might enjoy the fact that some of them are actually 13!
What interests me is your disinclination to practice what you preach.
You post under a pseudonym and include no personal data on your profile -
not even a link to a web page. You may not value anyone else's
privacy, but you certainly value your own. Oddly enough, that seems to
have been Scott McNealy's attitude as well.
A famous quote by a powerful man. I don't think I need to cite source.
Why that's right. I expect everyone here knows that the quote is attributed
to Scott McNealy, then CEO of Sun Microsystems. According to
a reporter for
Wired he was speaking at the launch of Sun's Jini technology in 1999.
It's just that by saying so up front, you can avoid sounding like an
insecure thirteen year old putting on a pose to try and hide the fact that
he's too lazy to type three words into Google.
Again according to Wired, McNealy was commenting in response to Intel's
recent U-turn regarding placing unique identifiers in each of their chips.
So it more likely that McNealy's comment was self-serving, rather than
indicative of his being a fount of Ethical Truth. Although in fairness,
his comments do have a measure of truth in a networking context - you can't
do much without leaving your IP behind you. But then again, this was also
before the use of NAT gateways and dynamic IPs became quite so widespread,
so even then, he doesn't have too much credibility.
Incidentally the correct quote would seem to be "you have zero privacy anyway,
get over it". This at least is to McNealy's credit, since "privacy is dead"
is a profoundly stupid thing to say. Privacy cannot be dead, because it was
never alive. Privacy is some fragile, endangered creature that can be slain
by a terrorist bomb, or by an uncaring government. Privacy is a courtesy
we offer to one another. And if groups, be they government departments
or struggling computer companies, should choose to withdraw this courtesy
it is by their choice that they do so.
Nearly 10 years ago, an insightful author at then-amazing Wired answered
this question
That would be David Brin, well known writer of science fiction. I suppose
that when you say that, he loses some of the gravitas that might otherwise
attach to "an insightful author". Perhaps that's why you shied away from
citing Mr.McNealy as well.
That said, I have to admit that have some sympathy for Brin's views on this
matter, at least as he went on to develop them in Earth. He seems to think
that the problem with lack of privacy is not the lack itself, but the
asymmetry of the arrangement. Various groups are allowed to know all they like
about me, but I am not allowed to know anything about them. The trouble I have
there is that I didn't find the picture of society in Earth particularly
appealing, and I'm not at all convinced that it would work as advertised.
It did make for an interesting novel, though.
I, for one, find it far more effective to deal with what is than what
I'd prefer there was to work on, and the reality is that privacy is dead.
Well, I still don't think it's especially useful to anthropomorphise
abstract concepts, especially ones founded in courtesy and dignity. On the
other hand, if you really believe that, perhaps you'd like to show us the
way forward. You could start by posting your real name, email address, age,
racial background, social security number, marital status, any major
illnesses, any history of family illness. Just for a start. I'm sure once
we see well you fare in a post privacy world, we'll all be eager to join you.
Why spend resources on games that are hard to develop and requires a lot of work when you have cash cows like that..?
Because, after a time, people get bored with the same formula over and over again.
When that happens a company loses market share, which can be hard to regain.
Of course, if the cash cows are supplying decent amounts of milk, then
you can argue that the formula is just giving the customers what they want.
Although if that were the case, why the shake up at EA? Why this call for "innovation?"
Slashdot Mirror
Oh certainly. It's just that it would be a lot of hard work cutting Ubuntu down to the 386/486 level, and what you'd have left, while it would certainly be Linux, it wouldn't be Ubuntu in any meaningful sense. Better to start from Debian, Slackware, or Linux From Scratch. TinyGentoo would be another contender.
If you want to get an idea of the minimum spec for a Linux system,have a look at tomsrtbt. It's a linux installation on a floppy disc - a "live floppy", if you will. It needs 8 meg of ram to boot, runs from a ramdisc, and doesn't have a minimum cpu requirement.
Well, to be fair, the Unix three level permission system is (as far as I can see) isomorphic to ACLs and MACs. That's been around since the seventies. It's just that the guy who wrote the security certification levels back in the 80s was a VMS head. So nothing got certified that didn't use VMS terminology. Windows NT got it because MS poached the chief architect of VMS, but Linux has has comparable features from the start.
That said, having a separate, orthogonal permission system can be useful; I believe the NSA have a Linux box online where they tell you what the root password is and give you a telnet prompt. You can't do anything once you're in, of course, but it makes a nice proof-of-concept.
Pretty much. There are a whole pile of kernel compile time options to enable the subsystem, and to configure it. I can't remember too much beyond that - the last time I tried to set up SELinux by hand, I nearly locked myself out of my own box. So I can see the appeal of a distro where these things are set up for me.
OK. Hope that answers your questions. Let me know if I missed anything and I'll see if I can help :)
Well, up to a point. You'll not run Ubuntu on a 468 for instance, but you can do it with smoothwall. And you don't have to harden the distro - it comes that way out of the box. So deploying SmoothWall should be half an hours work, rather than three or four days messing on with ACLs and policies
It's not that you can't do it with another distro - and this is true of all the specialist distros - its just that SmoothWall has been designed to make one particular application of the technology very simple to implement.
But what do we lose? This is free software and to most questions the default answer is "yes, you can". Before anyone starts saying "no you can't have your own distro", I think I'd need to see someone establish that harm is being caused. I don't think anyone has done that so far.
Well, from a user perspective, the difference between most distros comes down to Gnome vs. KDE. And how nice the wallpaper is, since most users never bother learning how to change it.
Behind the scenes, they can vary quite a bit. The most visible variation has been over desktop acceleration - Beryl vs Compiz vs Metisse, and all the rest of them. But there's quite a bit of envelope pushing behind the scenes as well. Some distros rushed to embrace the 2.6 series of kernels while others stayed with 2.4 until the major problems had been ironed out. Different filesystems, different device managers, different boot scripts. We can't just choose the best one, because for a reasonably up to date distro, we don't know yet. So everyone makes their best guess, and over time the pros and cons become apparent. That's where we score over Microsoft: we can do this sort of massive parallel evaluation and it doesn't cost us a penny.
It's been done before. UnitedLinux was (someone correct me if I mis-remember) just such an attempt. What happens is someone starts a new standard distro, and none of the others pay any noticeable amount of attention. More practically, there is also the Linux Standards Base which attempts to solve the problem without requiring anyone to fold their distro and join someone else's project. Because let's face it, that is always going to be a hard proposition to sell.
That doesn't follow. More is not necessarily better, but neither is it necessarily worse. Nor is less automatically better for that matter.
You mean like Knoppix, which I believe invented the LiveCD, and is still the recovery disc of choice for a great many of us? Or maybe DamnSmallLinux, which packs into 50MB and will run on just about anything? Then there's Smoothwall which vainly flatters the egos of its developers by providing a dedicated, hardened distribution capable of converting an old computer into a firewall router?
That's to name but a few. There are a lot of specialist distros out there supporting a specific activitity, interest or region.
If you're worried about users migrating from windows, then we have enough trouble drawing people's attention to the big names like Ubunbtu and RedHat. I doubt the existence of tomsrtbt or Astrumi are even going to impinge upon their awareness, let alone sow the seeds of confusion
So then if I buy a copy of Windows, don't install it, and then start selling copies... no, that can't work. Of course, I'm copying a CD there which puts it into fairly well defined territory.
So how about if I bought a copy of a game from one of the online distributors. Can I then decide I don't like the licence and sell the electronic copy onwards? Can I do it multiple times?
I don't think MS will fall back on the first sale doctrine if they can avoid it. If they strengthen the doctrine as applied to software, they weaken their own licencing in the process or so it seems to me.
True enough. RMS and Eben Moglen might have just decided it was a more elegant solution to allow the deal, and then let Microsoft run away screaming than it would have been to try and kill the agreement outright. If nothing else, it certainly lends credibility to the new licence. Short of overturning GPLv3 in court, I think that possibility represents the "best case" outcome for Microsoft here. I just don't suppose it to be the only possibility.
They don't have much choice, really. If they ever try and sue for patent infringement (and I'm not holding my breath BTW) then they are going to have to be able to show that they are not bound by the terms of the GPLv3, or the defendant can reasonably claim by conveying covered software, Microsoft grant them a licence for those patents. Which ought to be enough to get the case thrown out.
How does the end user suffer? I still get to run Linux. One more flame war on LKML, maybe a couple of projects get forked... I don't see this affecting me in the long term.
Which raises, I think, an interesting point. Because while we don't know the full details of the agreement, Eben Moglen does. He was allowed (IIRC) to look over the agreement so he could satisfy himself that it didn't contravene the GPLv2. And while he's barred by NDA from discussing the terms of the agreement, that wouldn't necessarily stop him from exploiting a flaw in the agreement had he noticed one in his review.
Now, I've read your profile, and I have to respect your legal credentials. I'm a coder, not a lawyer, and I'm not about to argue the law with you. But I think we also have to respect Moglen's credentials here, and he had plenty of time to think about how he was going to address the MS-Novell pact.
So like I said earlier - we won't really know unless and until this gets tested in court. But I think we have to allow the possibility that Microsoft might be in trouble here.
Which is true as far as it goes. The missing detail is the vouchers MS have been selling for SUSE Linux which have no expiry date. This means that, in principle, if anyone redeems such a voucher for a copy of SLES, and if that collection contains any code licenced under GPLv3 at the time they redeem the voucher, then there's a chance MS may be held to account under the terms of GPLv3.
Now whether that will stand up in a court of law or not is another matter. Eben Moglen and RMS seem to think so, since they wrote the new licence to allow the MS-Novell pact specifically for this reason. But like I say, we won't know for sure until it's tested in court.
On the other hand it seems reasonably certain that Microsoft sees some legal exposure there, or they wouldn't be making such a fuss. Because for all they talk as if the licence poses no threat to them, they are nevertheless backing away from it at every opportunity.
The thing is that if the GPLv3 does apply, then anyone they sue for patent violation hereafter is going to be able to claim that Microsoft licenced the patent for their use - else they had no right to distribute in the first place. That too will need to be tested in court, but again it seems that Microsoft are taking the threat seriously.
So that's "so what". It's not Microsoft don't use GPLv3 and we think they should.
It's more a case of MS may already be using GPLv3 which makes them a lot less scary.
Hope that helps, have a nice day.
In which case, I apologise. I think my hackles went up with the "welcome to the real world" line, and I probably didn't read the rest as carefully as I should have.
mmm...I understand that. The thing that gets my goat is that more and more people seem to be swallowing the notion that marketing reality > objective reality. In particular, I keep running into arguments like the following: Suppose X and Y are mutually exclusive possibilities. Then the argument goes:
-
X is supported by objectively verifiable facts
-
However Y is widely belived to be true
-
Therefore Y is true and X is false.
It's logic for the Jerry Springer generation, and it winds me right up. And yeah, I thought you were serious. Sorry about that.What has never been about arguments, precisely? Is this going to be one of those everytime anyone says "Microsoft Sucks!" God kills a kitten arguments? Because frankly, I could screw up my eyes and make myself believe in Microsoft ever-so-ever-so-much, but that still wouldn't help anyone outside MS implement a spec that defines chunks of its functionality with reference to antique proprietary document formats.
I think you missed the word "frequently" out of that sentence. "Increasingly" would work too.
The thing is though - this is Slashdot. We're a pretty cerebral bunch, on the whole. Certainly, we see our fair share of faith based reasoning from the fanboys, and I'll grant that topics like this tend to bring the astroturfers out in force. But if ever there was a forum where sound logic and factual argument carried weight, this would be it. It's a techie thing - you can't debug a program by wishing, even if you wish really really hard. That's the first thing a good coder learns and it colours your thinking after a while.
You're going to tell me "there is no spoon" in a minute, aren't you?
More seriously, I find it interesting that you seem to think that the lies and carefully chosen logical fallacies of marketers are in some way more "real" than sound logic founded upon verifiable facts. It seems to me a marketer's mind set. You wouldn't be in the business, by any chance?
So let me see if I've got this straight:
OOXML is better than ODF because Java apps don't use native Microsoft widgets. But although wxWidgets demonstrates that non-MS products can indeed conform to Microsoft standards, that doesn't apparently count because you like Visual Studio. Neither of which points is in any way a non sequiteur, probably for reasons that will turn out to involve the mating rituals of crocodiles.
Really, if that's the sort of argument Microsoft are reduced to, I'm surprised the debate has lasted this long.
Well, only if you're afraid that the news might slow down the take up of Tech Writers. But frankly, I think that by this time the brand is well enough established as to be pretty much bullet proof.
Of course, that may not apply to the forthcoming release of Tech Writers 2.0, but as far as I remember, that's still in the discussion phase, so it's too early to say anything for definite.
Yeah, flash and heavy duty javascript can be a problem, certainly. And not just for older machines, either.
I use NoScript for that. Turn scripting off by default, and then whitelist sites that use it in the Furtherance Of Good. It apparently doesn't get on with FlashBlock - but then you don't need flashblock if you have NoScript. It's one of my must-have extensions these days.
It won't. Full fledged Ubuntu needs a decent system spec. Gnome has got its own framework overheads, and Ubuntu adds a lot more newbie friendly stuff that ups the system reqs to about the same as XP. Ubuntu is a lovely distro in a great many ways, but svelte it ain't.
You could try XUbuntu. Uses XFCE rather than Gnome, and uses a lot less in the way of resources. You can install it from Ubuntu using synaptic. Did it in about 10 mins on my wife's old desktop machine. I'm still not sure it would cope with a PII, but it runs just fine on a Windows ME era AMD Duron system. Which is more than can be said of vanilla Ubuntu.
Failing that, look at some of the lightweight distros. The slackware derivatives seem to be good at this, so Vector Linux might be a good place to start. Or else Damn Small Linux, if you can persuade it to install on the HD and not run from a ramdisk.
Well, if you don't need it, don't do it. Personally, I'd sooner my resources were doing what I tell them to, rather than monitoring the system to see if I needed a help bubble popping up, or a usb drive mounted - but I admit I'm not the typical user. Then again, if you have old hardware, and you can't upgrade for whatever reason, it's not a bad angle to explore. It certainly beats using out-of-support Windows 98 because you can't cope with XP.
Someone who isn't running Vista or XP? Someone stuck with ten year old hardware?
I have a ten year old P266 machine I still use from time to time. It runs Gentoo Linux, X.Org, a heavily customised FVWM desktop, and Firefox. For most purposes, it's still quite nippy. (And yes, emerge -avuD world is kind of slow; but that's why distcc was invented :)
Oh, certainly. But in most modern operating systems, most of the speed increase gets used up by infrastructure overheads. That's why a Dual Core monster now doesn't seem much faster to the user than my top-of-the-range P266 did back in 1997. The gripping hand is that if you use a lightweight distro, then firefox will run just as happily on 10 year old machines as it will on the latest hardware.
Which, if you're stuck using something from the dark ages, can be a good thing.
Well, the source currently published would stay published, of course.
Most of the free software devs are pretty much on-side about the source availability, so most of their updates would be released.
Most of the proprietary stuff would stay unpublished - so no change there.
Where we might see a difference would be the cases where a company has to e sued before they will comply with the licence. I think that's a fairly slim slice, however. Most of the corporations involved with FOSS have learned that there are benefits in having a community willing to work on your codebase.
So, not much difference on the whole. And in return, we get to run
on the whole of our culture. I'd say that was a good deal.
This is why we don't let anyone in our server room unless they've had a full course of inoculations. We have to insist on this policy, otherwise all it takes is one sneeze to infect all our computers. The next thing you know we'll be getting cease-and-desist letters from the FSF and...
Or to put it another way, the problem with "viral" in this context isn't so much that is is not accurate; it's just that it's woefully imprecise. If you think about it the term is accurate only in a single limited scenario - one where you distribute proprietary binaries linked against GPL libraries. Of course, that's why the LGPL was invented, and these days the situation rarely occurs. Al of the other associations people have for the word "viral", death, disease, broken computers, epidemics, rootkits... they don't really apply.
Trouble is, there are kids out there who think it's really cool to release under a "viral licence" and there are PR drones out there who aren't above capitalising on the negative connotations.
So, on the whole, I think it's best to avoid the term.
Let me propose an experiment. Take some people you don't know particularly well, and open their mail. Make sure they catch you reading it. See how many pro-active responses you elicit. You can report back when you get out of hospital.
Something about they way you phrase that leads me to imagine how you mus t have been as a child:
Kid: "Hey, dada21, Who do you think is strongest? The Thing or The Hulk?"
dada21: "I think the market ought to be allowed to decide that on its own terms, don't you?"
Do we have to bring "the market" into everything? Privacy is a human issue; how about we leave the corporations out of it for five minutes?
Key word there being "complete", I suppose. So what level of privacy do you need?
I dunno. But if you give me free and complete access to all your data and all your activities, I bet I could write you a list.
Of course, the same is true of proprietary software. The vendor could cease trading, withdraw the product, release a new version that broke backwards compatibility ...
So really, you can be left in the lurch in either case. It's just that with open source there exist options that are not available for the proprietary solutions. Like organise a fork, or hire a developer to make changes based on the last available release.
And even if you're unable or unwilling to do any of those things, the worst case scenario is still no worse than the proprietary one.
I'm not saying you're wrong in the general case, but I did a little bit of searching in the case of EA. Have a look at this:
I don't know if I'm reading that right, but those figures seem to show negative sales growth (compared to a positive from the rest of the industry). There are some other interesting figures in there too, such as the return on assets.
It seems there may well have been problems at EA over the last 12 months, and the last quarter in particular. Of course, the extent to which their choice of material is a factor is still open to debate.
Still, it may well be more than just routine PHB verbiage.
And I say again, that's a stupid thing to say. Privacy is not some lifeform that can be slain, never to walk the earth again. It is a courtesy that we customarily extend to one another; one important enough to be protected by law in many countries. "Privacy is dead" might make a good sound bite for a controversy loving CEO, but as a basis for argument it grossly misrepresents the issues
I think you're conflating privacy with surveillance. Yes, there is more and more data being gathered on all of us. However, the availablily of the data does not invade my privacy any more than does the window in my front room. However, if you stand outside my house and peer in through that window for long periods of time - that's an invasion of privacy, and you can expect me (or anyone else you try this with) to get annoyed about it.
It's not the availability of the data that invades my privacy - it is the abuse of that data. And that battle is far from over.
Now, if you'd been reading my posts, you'd know that I don't have a problem with - let's call it Brinian Transparency - as such. I just don't believe it's ever going to happen. It only works if it works for everyone, and the principle offenders are the ones who have the most compelling case that their doings should be exempt from transparency. So the police will claim they need their privacy lest we enable organised crime to evade the law; corporations will predict the collapse of the economy if their precious trade secrets fall into the hands of overseas competitors; the President will become a target for terrorists, if his movements become public knowledge; and so on, until every body large enough to afford their own lobbyist has cobbled up a reason why they should not be part of it all.
The only people who will be transparent will be the likes of you and me.
This is the trouble with basing arguments on the writings of SF writers. In their profession they need to devise scenarios that are internally consistent, but the initial assumptions of that world need be no more than mildly plausible. In fact a legitimate use of S/F is to examine some of the more extreme ends of the probably curve. The problem with this is that when they come to write seriously, they don't always examine their starting assumptions as rigorously as would a scientist or a scholar.
Brin's solution might work - if you can get the infrastructure into place. But that's the part he skips over, both in the Wired essay, and in Earth. So if you can explain how you think we can establish this blessed social order then by all means do so. But until you do, I'm not going to stop objecting to invasion of privacy any more than I'm going to sponsor research into Tree-Of-Life-Root, just because Larry Niven's vision in Protector was so compelling.
We can make sure that privacy (as defined above) is recognised as a basic pre-requisite of a sane society, and we can work to ensure that unwarranted abuse of personal data is a criminal act, and that it remains a criminal act. And we can campaign to set some sensible limits on where people want to put these blasted cameras, as well, see if we can stop the rot before we get to Orwell's nightmare, with cameras in every home.
And in the meantime, we can even campaign for greater transparency in governments and corporations - because I don't think Brin's idea is a bad one, any more than I think faster than light travel and limitless cheap energy are bad ideas. I just don't want to adopt a social policy that's dependent on any one of the three, at least until we get reasonable grounds to think they might be achievable.
This is what I wrote two posts back:
His solution is that all surveillance information should be available to everyone, and then we could each watch one another and keep the system honest. And it's a good plan.
The trouble is, the first time the proposal is taken seriously, there's going to be a lot of handwaving about IPR and national security and terrorist threats, and the first thing that will happen is that we'll see a great long list of people and organisations who are exempt. And then we'll have no legal And when the legislation passes, we'll find that we have surrendered the last of our rights to privacy, and that the watchmen are no more watchable than before.
In order for that not to happen, you need to change the way an entire culture thinks. You need to persuade Coca-Cola to leave the AV feed on the boardroom when secret formulae are being discussed. You need to get Microsoft to stop deleting every email that might ever be deposed in a court of law; you need to get the CIA to accept that it's oh-kay for foreign governments to know the names and addresses of all their agents overseas. And frankly, I don't see that happening any time soon.
But let's suppose I'm wrong about that. If you can do social engineering on that scale, then why not just engineer things so people don't abuse their authority in the first place? For that matter, the two are not mutually exclusive.
And that, I think, is where we find ourselves at issue. When you say "privacy is dead. Get over it" You're suggesting that the debate already over (I strongly disagree) and that we should all meekly accept increasing levels of surveillance now whilst we campaign for greater openness and accountability at some future time. As far as I'm concerned, that's the wrong way round. You put the infrastructure in place so that we can watch the watchmen, and maybe I'll happily accept that they can watch me more closely in turn.
But until that day comes I continue to value my privacy.
As it seems you do your own.
Now, tell me again how I missed your point.
Apologies for the style and tone; it's just that people quoting McNealy out of context to try and justify increased levels of government intrusion is a pet peeve of mine. I probably went a bit over the top
I'm trying not to make assumptions about you - but you sound like you're too lazy to type three words into Google. It as if you said to yourself "well, I can't remember if this was Scott McNealy, or Jonathan Schwartz, but if I just say I'm sure you all know who I mean then everyone will assume I do know, and I don't have to risk embarrasing myself by getting it wrong".
The trouble is that once people get that impression, it undermines your credibility, since the reader can't help but wonder what else you might be trying to fudge.
Think of it as feedback.
Then don't provide a citation at all. This is Slashdot, not a paper for an IEEE conference - you don't need to cite anyone. But if you do, don't be half-hearted about it; it reflects badly on you, and weakens your credibility.
No, but then neither do I think that Larry Niven understands astrophysics in the same depth as Stephen Hawking. If there's a debate about black holes, Hawking's opinion is going to carry more weight. If you think that Brin's opinion is as good as anyone else's, then have the courage of your convictions and say so up front. "Noted SF author David Brin makes this case far better than I ever could in a wired article from the 90s..." is all you need to write.
Well, anyone who publishes their bank details and social security number on MySpace almost certainly falls into the category of "idiot" and can be safely ignored. I'll make an exception for Hasan Elahi as I will for anyone who is doing so to make a point about privacy - if you'll provide a link to the specific page. In the general case though, I don't think human ignorance and carelessness carries much weight in this debate.
What interests me is your disinclination to practice what you preach. You post under a pseudonym and include no personal data on your profile - not even a link to a web page. You may not value anyone else's privacy, but you certainly value your own. Oddly enough, that seems to have been Scott McNealy's attitude as well.
Why that's right. I expect everyone here knows that the quote is attributed to Scott McNealy, then CEO of Sun Microsystems. According to a reporter for Wired he was speaking at the launch of Sun's Jini technology in 1999. It's just that by saying so up front, you can avoid sounding like an insecure thirteen year old putting on a pose to try and hide the fact that he's too lazy to type three words into Google.
Again according to Wired, McNealy was commenting in response to Intel's recent U-turn regarding placing unique identifiers in each of their chips. So it more likely that McNealy's comment was self-serving, rather than indicative of his being a fount of Ethical Truth. Although in fairness, his comments do have a measure of truth in a networking context - you can't do much without leaving your IP behind you. But then again, this was also before the use of NAT gateways and dynamic IPs became quite so widespread, so even then, he doesn't have too much credibility.
Incidentally the correct quote would seem to be "you have zero privacy anyway, get over it". This at least is to McNealy's credit, since "privacy is dead" is a profoundly stupid thing to say. Privacy cannot be dead, because it was never alive. Privacy is some fragile, endangered creature that can be slain by a terrorist bomb, or by an uncaring government. Privacy is a courtesy we offer to one another. And if groups, be they government departments or struggling computer companies, should choose to withdraw this courtesy it is by their choice that they do so.
That would be David Brin, well known writer of science fiction. I suppose that when you say that, he loses some of the gravitas that might otherwise attach to "an insightful author". Perhaps that's why you shied away from citing Mr.McNealy as well.
That said, I have to admit that have some sympathy for Brin's views on this matter, at least as he went on to develop them in Earth. He seems to think that the problem with lack of privacy is not the lack itself, but the asymmetry of the arrangement. Various groups are allowed to know all they like about me, but I am not allowed to know anything about them. The trouble I have there is that I didn't find the picture of society in Earth particularly appealing, and I'm not at all convinced that it would work as advertised. It did make for an interesting novel, though.
Well, I still don't think it's especially useful to anthropomorphise abstract concepts, especially ones founded in courtesy and dignity. On the other hand, if you really believe that, perhaps you'd like to show us the way forward. You could start by posting your real name, email address, age, racial background, social security number, marital status, any major illnesses, any history of family illness. Just for a start. I'm sure once we see well you fare in a post privacy world, we'll all be eager to join you.
Because, after a time, people get bored with the same formula over and over again. When that happens a company loses market share, which can be hard to regain.
Of course, if the cash cows are supplying decent amounts of milk, then you can argue that the formula is just giving the customers what they want.
Although if that were the case, why the shake up at EA? Why this call for "innovation?"