Re: MISC-d Digest V00 #12

[Greg Alexander <galexand@xxxxxxx>]

>Wayne Morellini writes:
>
> > -Why would one want to know anything about zombie vapourware? I lost my
> > 
> > Toas' OS is not vapor ware and still has features that Linux people could 
>
>I know all about TaOS. I used to push them after the Byte series
>articles wherever I could. Nifty technology, will never make it
>because their they have no clue about business, and even if they had a
>ghost of a clue now it would be too late anyway, since now OpenSource
>OSses are out there.

Yeah, I was thinking the only way to get an upstart into the market these
days is to go open source.

> > --- themselves for.  Amiga though working on some nifty stuff keeps getting
> 
> > interference, court challenges, and take overs that has held it up for 6-8 
> > years, thanks a lot Guys.
> > 
> > -proprietary) system whatsoever. Then why not BeOS (which has no point
> > either, but be my guest).
> > BEOS has been a shipping product for a long time too.  Is the Palm OS vapor
> 
> > ware? what abour Linux, doesn't make sense.
> 
>Anything which doesn't come with source is essentially dead. Free
>binaries alone don't cut it anymore. I'm surprised so few people are
>yet grokking it.

I tihnk there's a huge dividing line between people who have most of their
software experience with commercial tools that are routinely "good
enough" and those who have had their experiences with free/open tools that
are often not nearly good enough but always somehow hackable.  The former
group has a long history and appears to do most of the hardware
development today (as is evidenced by the requirement of proprietary
software to do FPGA design, a limitation most engineers seem to not even
realize is a limitation) and doesn't even realize that there are other
ways to do software.  The open source kids, however, are in the opposite
situation: they have seen how much more efficient it is to share
everything and they hardly even consider writing anything and not
releasing source code and they have strong reservations about using
anything they can't understand.  The former group is under the illusion
that the internals of the software they use are too complex for them to
ever understand so they don't consider it a problem that they're using a
black box.  The latter group is convinced that every piece of software
they could eventually understand if they had the source code and that
people who don't distribute source code are almost always hiding low
quality (whereas a lot of open source projects show off low quality to
encourage people to fix it).  I tend to think that the latter group is a
lot more right than the former group, if not just because the latter group
realizes that the former group is closed-minded, whereas I suspect the
former group has absolutely not even the slightest inclination that the
latter group is closed-minded, they figure it's just a fad...I think if
you told them that people who learned from open source software are
typically more experienced they wouldn't be surprised, but if you told
them that these more experienced programmers and engineers would never
touch their crappy proprietary software with a 10ft pole, they would stare
at you in disbelief.  I'm clearly a member of the latter group.  :)
	I'm posting this to the list because I think there are a lot of
talented engineers here from all sides and I want to know what you thin.k
	As a side note, does anyone here have any suggestions for how I
can develop with low-cost FPGAs or CPLDs on a Linux box without using any
closed-source software?  A lot of people have made boards with open source
software to upload the bitstreams, but I can't find any chips (let alone
cheap boards using those chips) that publish their full bitstream specs
and I have 0 tolerance for proprietary place&route software.  I figure
some people here have probably went through the same search and
frustration.

<snip>
> > -Technical superiority doesn't matter anymore. Availability of source
> > -and distribution copyright decides everything these days, because on
> > -the long run it also ensures technical superiority. Linux is highly
> > 
> > Yes in the long run, patents run out and Linux can catch up, that is 
> > correct.  But on the matter of superiority, I am very frustrated -with my 
>
>The good part of Linux is that it is noncommercial, and hence not
>subject to lawsuits. I don't care a fig what is patented/copyrighted
>or not, because I can implement it and put it on the web. Especially
>with budding new anonymous distribution infrastructure publishing
>becomes irreversible, since not giving you a point of attack in
>meat/legalspace. It is quite amusing to see the CEOs and lawyers
>scurry, and not get anywhere.

It is kind of worrisome.  DeCSS is out there, I can get it, you can get
it, we can all get source and descriptions of the algorithm, so they were
a success at defeating patent law, horrible licensing law, and the very
essence of proprietariness, etc....but some of the authors and other
involved parties have had their computers raided and are facing lawsuits.
We have supreme power over information, but unless you join a militia you
don't have too much say over what the govt/corps do to your body or
pocketbook.  In other words, we win, and nothing they can do will make
them win...but they sure can make us lose at the same time (I guess only
an asshole would do his best to make sure everyone loses in a situation
where only himself needs to lose, but...it's America).

> > PC, performance and quality suffers (not mentioning my recent problem with 
> > registering unmade mouse clicks while I'm scanning menues).  The software 
> > and the architecher of the PC, a problem.  Custom hardware (below good unti
>l 
> > PC and software redesigned.
> > 
> >  > with the exception of Windows and QNX, no one in their right mind would
> >  > release a new operating system under closed source.  Windows can do it
> > 
> > Palm
> 
>It's a small-footprint OS. Nothing that Linux can't handle.
> 
> >  > time.  It's a combination of the Commodore effect with the right product
> >  > being at the wrong place/time.
> > 
> > Well Commodore could have kept investing in the Amiga, they had it over 
> > Apple, but Apple invested.
> 
>It's all history now.
>
> >  > Furthermore, the new Amigas being designed are going to be based on open
> >  > standards.  While a proprietary motherboard will likely be used (after
> >  > all, how many people do YOU know of who are raking in the cash selling
> >  > CHRP PowerPC motherboards or Crusoe borads that fit the ATX
> >  > specification?), they will use industry standard expansion slots, such a
>s
> >  > PCI, Firewire, and USB.
> > 
> > Funny the new Amiga was going to be available in a Crusoe ATX motherboard 
> > with standard pc parts too (including new generation ATI chipset).
>
>I call a standard motherboard what Asus sells in quantities. So far,
>Transmeta's offering is not even a standard: it's vaporware.
> 
> >  > The operating system that is going to be used will almost certainly be
> >  > POSIX compliant, so porting applications to it will be trivial.  An X
> >  > server is likely to be a given as well.  Even device drivers.  And yes,
> > 
> > -So what's the difference to a vanilla Linux box, just a realtime
> > -nanokernel? They've got no chance at all, even if they were entering
> > -the market now.
> > Great Object oriented real time Kernal with great efficencies.
> > 
> > - I'm a bozo, so I stick with Linux. It has kept me very happy all these
> > years.
> > 
> > In case other people haven't realised, QNX and the Toas OS (too some extent
>) 
> > have a very Misc flavour about them.  They are practically minimial and 
>
>Which is probably why they are doomed.
>
> > scallable, OO, fast efficent etc.  Maybe Linux will change direction to thi
>s 
> > one day.  Ohh Be-OS is available as a free download from their site.  You 
>
>It will, when the need arises. BeOS is not free, just the download
>is. I will not invest in anything which can be pulled from under me,
>either by stopping development or pulling binaries off the net.
>
> > only need good drivers for good hardware, you can get Dev sys at cheap pric
>e 
> > and develope better drivers.  As far as changing the OS code, well it's 
> > pretty good as it is why change it, their on top of it.  Linux has not quie
>t 
> > caught up yet.
>
>Users are not purists, they are users. 98% could hardly care less
>about the architecture, as long as the system more or less does what
>it is expected to do. If you don't believe me, ask a few Windows users.

Agreed 100%.  I don't even try to explain my desire to design a MISC
processor in FPGA and implement a tiny ForthOS on it to my Windows-using
friends.  "What can you do with such a machine that you can't do with your
current desktop?" "I can restrain the vomit reflex."  Their response to
the vomit reflex is to take some dramamine.