[openbeos] Re: Hi from the PetrOS camp.

  • From: "Michael Phipps" <mphipps1@xxxxxxxxxxxxxxxx>
  • To: openbeos@xxxxxxxxxxxxx
  • Date: Fri, 09 Nov 2001 01:13:36 -0500

>An operating system at its lowest level is a program. I've written both
>programs and small operating systems for dedicated tasks. If someone is
>putting their cards on the table, you have to question why. Insights and
>pitfalls on how to, and how not to design an OS have been well travelled
>roads. Ask any CS undergrad to point you to the pile full of OS books
>that compare and contrast different shared memory schemes, threading
>systems, and synchronization systems occur in all of the popular OSes. 

Perfectly true. Most OS books (that I have read, anyway) are also short on
real life information. :-/ Academics write text books. :-) That doesn't mean
that you should avoid them. But remember that they have a particular 
set of experiences.

>If you're claiming that maybe we need a push in the business world, you
>might want to pick up some basic books on marketing, and all of the
>various techniques to lie, cheat, and steal to sell x number of widgets. 
>It's been made pretty clear to us that the MIT license is what we're going
>with. Any attempt to hit a business angle on something that we don't
>already have done is nothing more than a distraction. 

I think that Peter would like the ability to (MAYBE) sell OBOS with his kernel.
All the power in the world to him. Works for RedHat, works for him.

>> There's more to software design than just coding.
>That was neither expressed nor implied, and if you had read the email I
>sent instead of snipping out what was convenient for you, you might've
>realized this. [FYI, I've spent the past 3 months at my dayjob working on 
>IEEE Std 830-1998 SRS requirements and functional specs for a huge
>software project that I've been working on, and UML data models and
>block diagrams will have to be prepared as well. ]

I feel your pain. :-)

>This guy comes out of of thin air and toots his own horn about his
>commercial OS, and low and behold it makes news on http://www.osnews.com
>(Thanks Eugenia!), which just bought him some free advertising without
>actually doing anything.

Actually, he and I have been talking for a few days. He brings help to the 
in terms of code (from an experienced OS guy) and wants nothing in return that 
wer weren't already doing. In fact, I thought that he was very honest and 
announcing who he was and what he had in mind from the beginning. He certainly 
could have signed up for the mailing list, lurked until we had an R1, then 
taken the code
and done as he pleased. If he is an opportunist, he is the dumbest one I have 
ever met.

>Take a look at the FAQ
>"Who can join? 
>Anyone! If you are a programmer, then let us know what your interests and
>skills are and we'll find something for you to dive into. If you are a
>writer, there are technical and user documents that will need to be
>written. If you have web design or graphics skills, those are of great use
>too. If you just like to write email, then help spread the word to others. 
>If you think you have any kind of contribution to make, by all means
>contact us and we'll find something! A project such as this can't have too
>many helping hands."
>It does indeed say "any" contribution, but the tone of the FAQ is that
>skills are needed, and that implies that people are using their skills to 
>produce output. Nothing in there about strategic partnering or licensing.
> > Peter....(the other one) > > > 

Agreed. And there have been a few people who I just could *NOT* find
any work for. I have encouraged them to hang out and wait for a beta, but
if you can't write documentation or code, there is not a whole lot we can do
with you right now. AFA the partnering and licensing thing, well, you know. That
is Helmar's bent, too. And, from a perspective, they are right. I think that 
there is
a time and place for that sort of thing. I want this OS to become a dominant
factor in the OS market someday. For that to happen, there will need to be 
some money (if for no other reason then to entice people to write the apps that
people need that no one wants to write - an accounting package, for example).
That is a ***LONG*** way off and those decisions and thoughts don't need to 
even be considered. When we are debating how many features to put into an R4, 
maybe those questions should be asked. There is no point of putting the cart 
the horse. Don't bust a sweat over this stuff. It is not a part of the 
future of OBOS. 

Let's get back to coding and give Peter the same chance to prove himself, and 
same way as the rest of us prove ourselves - quality and quantity of code. :-)


Other related posts: