🚀 go-pugleaf

In article <4jt6s0$pem@neptune.ethz.ch>, oswald@inf.ethz.ch writes:
> 

. System 3 has been developed by the group of Prof. Gutknecht, which
. currently has five members and is a part of ICS. As you probably know,
. we're all busy with projects that further extend System 3. Our resources
. are just too limited to port System 3 to a lot of platforms and
. _support_ it there, i.e. keep it up to date for every release. The only
. official version in the future will be the PC native version, probably
. along with versions for MS Windows and Mac. It is a matter of time
. rather than of commitment. Prof. Gutknecht will certainly approve of
. other versions but will hardly pay somebody for them, least of all us
. guys in his group. We'd have to do the work during evenings and
. weekends, like Hannes Marais did for the Windows port.


Thank you for your answer. I think it clarifies at least a few issues,
and  I am really glad you wrote your comment. I am sure Oberon circles
will have  some comments here. From my own perspective, I was asking
myself how viable was System-3 as my own development platform. Now I
know it is not viable,  because good portability is one of my key
requirements. Thus, I need to  count on good old V4, even though its
development is not so vigorous anymore. (But, when it moves to Linz, it
becomes vigorous again -- see recent announce  about Linux port! Thank
goodness, there is Univ. Linz out there!)

Having said all this, I must say I am a bit surprised. As you wrote in 
your earlier posting:

> The differences really [between V4 and S3] aren't that big but 
> the text models are incompatible.

It they are not big, and if so many V4 ports exist, then why are you
saying now that things are difficult? From the depths of my ignorance it
seems to me, that most of hard work has already been done, such as 
porting the compiler and OS-specific bindings. I would think that this
work can be reused. Existing ports can be used as "launching pads"
to bootstrap new ones, or am I missing something here?

Now, I noticed what you said about manpower and funding and such. 
What I am saying here, is simply that good "launching pads" do exist, 
that not everything has to be done from scratch, and that from the point 
of view of disseminating your own ideas it would be advantageous 
to create more ports by reusing old stuff. As you said yourself 
"I'd be the first to welcome a wider use of S3" -- well, we are 
discussing just that!


. If you really need a specific port, you'll have to get a source license
. for S3 (easy), find somebody to do the port (difficult), and to keep the
. version up to date (very difficult). Come back when you have done so and
. we'll be happy to help you as much as we can. 

Myself -- not at the moment. I am not the right guy. But let me direct
your attention to Ofront. Please note that the same "launching pad" 
has been reused to support both V4 and System-3 "front ends". It shows
that such things are doable, and therefore I do not quite buy your 
arguments about how difficult it would be.


. This posting is not a flame against you or anybody else. It just
. describes the  status quo. In fact, I'd be the first to welcome a wider
. use of S3. 

I truly appreciate your comment and clarification that it offered.


Wojtek