Re: [Fsfe-uk] BECTA discriminate against FLOSS?

[Chris Croughton]

On Sat, Jan 03, 2004 at 08:19:44PM +0000, ian wrote:

> On Sat, 2004-01-03 at 18:03, Chris Croughton wrote:
> > 
> > It's the old "No one ever got fired for buying IBM" problem.  When there
> > is an 'obvious' large supplier that is the safe option, irrespective of
> > whether there are cheaper or 'better' options, because "everyone does
> > it".  IBM itself is no longer in that position, but MS are very firmly
> > there.
> 
> I think you might be surprised. IBM's brand still holds a lot of weight
> which is why their backing for Linux is vitally important.

True, I should have said "no longer in quite such a good position".  IBM
now know that they can lose contracts, whereas they used to boast that
they had the US government in their back pocket.  MS know that they can
lose contracts but can afford to undercut everyone else on price to get
their product out there, and as a result seem very confused when someone
actually turns them down.

> On the issue of risk, (Alex's point), the onply risk in business is to
> take no risk.

Only true if you say "no risks at all".  For instance, a bakery might
take risks on a new type of bread (their speciality) but might be taking
no risks on accounting software at all (still using paper and pen,
even).  A software house might be willing to take risks in developing
software but not in Health and Safety.

> The highest risk stocks have overall outperformed the
> lowest risk ove the last Century, but some spread betting is needed ;-)

Overall maybe, but can you afford to bet in the first place?  And where
were they taking the risks?  The risks of using MS software are
well-known, but for most companies they are easily dealt with by
existing procedures.  If they don't need to change for some other
reason, why take the risk?

> Anyone that sells any MS products is effectively working for MS.

OK, in that way.  But I don't sell MS software (OK, I did buy it in the
case of Visual Basic, but that's because it is /the/ best GUI
development package in my opinion for what I want, no FLOSS equivalent
comes near it).

> Its very difficult to be in the IT business and on principle sell no
> MS product. The way I look at it is that its a necessary evil to stay
> in business that enables me to shift away from MS. It also gives a bit
> of even-handedness. We will supply what ever the customer wants
> because we know both systems and we can give even-handed advice. In
> fact, if all purchasers who had to buy MS products but preferred free
> software bought from companies like mine, a) we would be in a better
> position to promote FLOSS and b) It sends a message to other companies
> that there is a marketing edge in supporting FLOSS.

Agreed, and that's why I don't like total "anti-MS" stances.  What a
customer (or client, or friend) needs is something which will do the job
they want, and I am doing them no favours if I insist on only
considering one part of the products available.

> > But is FLOSS actually all
> > that financially rewarding? 
> 
> At least as good as selling MS based stuff - actually better if you can
> persuade people to buy it without spending a lot of money on marketing.

Er, yes, but isn't that just as true for non-open (but cheap) software?

> That is the killer for most people. The cost of promoting FLOSS against
> the behometh of MS branding can be prohibitively expensive. In the end
> its all marketing bo***ox :-). Most software licensing revenue goes into
> marketing and fighting court cases etc which is why its a pretty
> inefficient production model. Once people realise that the open standard
> is the thing to go for the marketing battle is won as far as FLOSS is
> concerned so prices tumble.

Open standards don't necessarily imply open source (even less free
software).  There's loads of proprietary software implementing TCP/IP,
for instance.

But if you don't advertise then a lot of places will see that as a
downside.  I do, even, when looking through the Yellow Pages I generally
ignore entries without at least a box advert.

> > I have yet to see a business model for
> > FLOSS which really supports developing the software, the RMS ideals seem
> > to be based on getting revenue for maintaining it which seems to me a
> > lot more shaky (for instance, if I have the source and can apply the
> > patches myself, or even just download the next version from the net, I
> > don't make any revenue for anyone).
> 
> I think on the major applications such as OS an Office suite, browsers
> etc - the only things MS makes any real money on - the development model
> is easily sustainable from relatively marginal input from large
> corporates and governments say largely through the universities. For
> specialist apps, I think the jury is still out, but just the OS and
> major productivity apps would be a major step forward. It doesn't have
> to be all or nothing and in fact I doubt it ever will be.

Indeed, but the point is how do I live while I'm writing my killer app?
And how do I make money after it if someone else can take my code and
support it?  A few people are lucky in finding an employer who will pay
you for developing something free, but there aren't many of those
around.  And if I'm working full-time to get money to live then I'm not
going to be writing much code "on the side".

> > Red Hat is not a good example (it
> > seems to me that they are breaking the spirit of FLOSS even if not the
> > letter of the licence).
> 
> Models will evolve and we are in a volatile state of flux, so the exact
> details of what will settle out with stability is difficult to predict.

True, but 'now' is when it's important.  Now is when companies need to
be weaned away from proprietary code, when they need to see an advantage
in letting their programmers write free software (heck, many programmers
are still locked in to contracts which dn't even allow them to write
free software on their own time).

> > But they don't really want to change anything, except to pay less money.
> 
> They don't even want to pay less if its not their money! Resistance to
> change is natural, it just becomes inevitable when early adopters start
> to get competitive advantage. When its change or lose your job, the mind
> becomes a bit more focused.

But that is still a long way ahead.  How many companies even think about
the next 10 years?  What company now (or UK state organisation) would
fire people for choosing MS over FLOSS?  The place I've recently been
working has had people trying to get OO.o installed instead of the
latest version of MSWord, and the company won't have it ("too much
risk").

> Its more complex than that. The freedoms are to an extent interdependent
> and free as in free beer is supported by lowering inconvenience and
> admin costs from free as in free to use and free as in free to develop
> also has potential to lower costs. In the end it all boils down to a
> more efficient economic model and where this is the case it will thrive.

But not as far as the users are concerned.  Sure, there may be a
knock-on effect but they don't care about that, or even think about it.
That's buried in the 'mechanics' of producing software, and there are
thousands (and more!) of managers who don't understand that.  Certainly
those who make decisions (largely the bean-counters these days) don't
understand software production, they are more likely to be concerned
about their "IP rights".

> > The only sort of 'free' they are really interested in is "as in beer",
> > and large organisations tend to distrust that 'free' things are any good
> > ("anything free is worth what you pay for it"), or think that there's a
> > catch somewhere ("there's no free lunch" -- as in someone taking it over
> > and making it proprietary again).
> 
> Yeav but there is around 12 billion a year going to MS which can be cut
> out of the equation. That's nothing to do with free lunches, its just
> improving production efficiency.

Try explaining that to the sort who say "If we pay a lot for it then it
must be worth it".  OK, you probably have -- did you make a dent?

> Betamax/VHS is the most misquoted example and rarely that applicable. On
> the VHS/beta argument you could just as easily say, yeah, even if Linux
> is worse than Windows it'll get adopted because its in more industry
> players long term advantage to do so.

That's the thing you need to prove.  Current experience for most
companies is that the established system is the best (and they have
billions of pounds sunk in it to 'prove' it).

If you don't like VHS/Betamax, try DAT/DCC.  Both of which lost out to
recordable CDs when the latter became cheap, both lots of early adopters
lost (DAT is still around, just, but it's difficult to find the
equipment now).

> >  Those who waited to see which would
> > win out gained because they didn't have to change.  The same is
> > happening with DVD +/- R/RW, whichever one becomes standardised the
> > "early adopters" will in the main lose.
> 
> No, the early adopters that get it wrong lose,

Yup, with DVD that will be most of them.  4 options, only one wins...

> those that get it right win hand over fist which is why VCs *expect*
> to lose quite a proportion of their investments for the one that goes
> real big.

Do tey so much now?  I thought they were being a lot more cautious in te
current economic climate.  It's why we had the DotCom boom and
subsequent bust.

> <SNIP>
> > If there were a "file of case studies" to which people could be pointed
> > then that would indeed help, but I don't see one.  Don't expect the
> > potential buyers to go hunting, though.
> 
> I don't, but its rather more efficient to go and talk to them and
> actively sell them product than to spend the time writing case studies
> ;-)

Surely it's more efficient to do it once and then re-use it for
subsequent times.  What sort of Unix programmer are you?  <g>

Chris C