January 2005 - Oberon - mailman3.inf.ethz.ch

Re (3): [Oberon] NIC in BB
by shark＠gulfnet.sd64.bc.ca 16 Jan '05

16 Jan '05

Doug, re: ambiguity of BB abbreviation. Hadn't thought of that. From now on I could use AOS or Aos to abbreviate BlueBottle and BB for Black Box. Thomas, tf> If your computer can run VMWare or Qemu ... No but late last week I installed crazy. With a decent network connection it is easy enough. Nice release! Thanks! tf> You can for example use AosPing.Ping 129.132.134.50 ~ to Ping the bluebottle server without DNS lookup. Or you can use the graphical FTP Browser ( WMFTPClient.Open ~). They both work. Thanks. The display panning ability offers convenient access to AosOberon. Ie., start Aos, pan right (left, up ...), bring the main menu, start Oberon, unpan, return the menu. Whenever Oberon is needed just pan to it with one click. (Chris G.: this works quite well. You gain by installing Aos and still have Oberon a click away.) Can anyone tell me how to do this startup in AosConfig.XML. The example, AosDisplayWM.Install Oberon 900x700;System.Init Oberon is on the home screen whereas I am aiming for an adjacent screen. Thanks, Peter Desktops.OpenDoc http://carnot.pathology.ubc.ca/

2 1

Tool for incremental progress vs. immediate WOW.
by easlab＠absamail.co.za 16 Jan '05

16 Jan '05

Soren Renner wrote: > > 1. I do admit that functional languages have some claim here as well, > > especially = Ocaml. It is just that I would prefer it to run on top of Oberon. Patrick.Hunziker(a)unibas.ch wrote: > There is a functional language available in Oberon: > - Scheme.Mod implements an interpreter for the Scheme language (is > in the Gfx package) > There are some examples about how to use it, hidden in the Gfx > documentation. > - Vinci.Mod utilizes Scheme.Mod for graphical purposes as an example. Yes there's a mass of potential gems lying hidden; which just need a bit of polishing to add great value to the total Oberon package for potential new users. ------- Soren Renner wrote > > > > [...] > > Look at it like this: on the one hand, we have what is clearly the > > best OS and language project in the world*1, yet almost noone > > will pay any attention to it. Stefan Salewski wrote: > Yes, ETH-people have done great work to develop this system, and > they were very successful all the years to hide it from the world. And > I think they will successfully continue this tradition by preventing > writing down any documentation. > > Please excuse my sarcasm. >From all my harping on this same topic for 5 [or is it 10] years I've distilled the following ideas [also by observing other collaborative systems which work better]: * there's entropy: stuff get dirty and broken - unintentionally. * let's try 'negative entropy': stuff 'evolves' - unintentionally. There are these massive piles of stones in various parts of the world, which have evolved over centuries from people who [for religious reasons] throw a stone on the pile, when passing. I.e. a 'method of capturing even small contributions'. The massive pile evolves unintentionally. The collaborative systems which which I know well which work better than oberon(a)inf.ethz.ch depend on a 'boss-man' who co-ordinates the contributions. This is less than ideal, because the 'idea' [like the stone pile] should survive and grow, independant of any individual. I guess linux has acheived this ? By moving up the heirarchy of control one achieves leaverage. Ie. 'never mind about doing coding', rather concentrate on developing systems which will ensure that coding effort is captured and adds value. Eg. when a new contribution is made, what about naming conventions and clashes with existing material; how is it integrated into the existing ...etc. These considerations need to be formulated, instead of each contributor making his own. Remove creativity from tasks which can be formalised. Perhaps this is the difference between coding and software engineering ? ------- > > Translating James Carlson's published C-code > > implementation > > might be good ? BTW a tool-set to *assist* C to > > Oberon code > > translation would be of great value to the Oberon > > community. > > John M. Drake wrote:- > I agree. Good idea. Don't know why nobody's ever > done this. Linux has p2c utility: Pascal to C; but not c2p. When I investigated I found: because Pascal is clean [canonical] it can be directly translated to C. Because C is a dog's-breakfast it can't always be directly translated to Pascal. There's a problem, because of the wow-factor: people want immediate automagical answers, like the 9th root of 87.436; instead of a 'tool' which 'helps' you create a solution. Like a pencil & paper doesn't replace your brain, but it's a powerfull tool. So a tool-set to HELP translate some of the mass of C source code to Oberon would be of great value. An example of a valuable tool which we have is the FontEditor. == Chris Glur.

2 1

Programming in Oberon, A set of WinWord documents:
by Stefan Salewski 16 Jan '05

16 Jan '05

http://www.oberon.ethz.ch/ wrote: > Latest News > 21 Dec 2004: Prof. Niklaus Wirth delivers Part 4 of Programming in Oberon. > Whereas Part 1 to 3 were derived from Programming in Modula-2, Part 4, > entitled "Object-oriented Programming", is new. http://www.oberon.ethz.ch/wirthPiO/ wrote: > Programming in Oberon - A derivative of Programming in Modula-2 (1982) > by Niklaus Wirth > A set of WinWord documents: > Table of Contents > Preface > Part 1 > Part 2 > Part 3 > Part 4 > Appendix: Syntax, Keywords, Standard functions > Part 4 is new and was added on Dec. 21, 2004 and at the same time a slightly > revised version of the Appendix was loaded. > Part 4 is not yet part of the PDF document mentioned below. Is Microsofts WinWord (.doc) format the new form for publishing scientific papers at ETH or will Part 4 become available as PDF or other format soon? Best regards Stefan Saleswski

1 0

Oberon Digest, Vol 16, Issue 5
by keller＠creacare.com 16 Jan '05

16 Jan '05

>they were very successful all the years to hide it from the world. >And >I think they will successfully continue this tradition by preventing >writing down any documentation. We do not hide Oberon from the world: Here the proove http://www.google.com/search?hl=en&q=active+oberon&btnG=Google+Search a sposoed link for active oberon at google! Thanks to Patrik Reali.

2 1

VBE 3.0 in Bluebottle
by Jack Johnson 15 Jan '05

15 Jan '05

I have a motherboard with onboard video (SIS315-based) and when the video initializes it comes up in the lowest supported refresh rate, which always seems to be interlaced. I dug around a few of the modules and I see that I can specify a vertical refresh rate using a VBE 3.0 string, and it *seems* to imply that a 4-digit Init value can specify a VBE 3.0 setting, though I still see the long xInit string. Can anyone point me in the right direction? I'd like to have a string that I can use when booting from the CD for testing various hardware without doing an install. Something like 1024x768x16 @ 75Hz would be great. -Jack

1 0

Functional languages
by Soren Renner 15 Jan '05

15 Jan '05

I do know about the implementation of Scheme in Oberon. The Vinci graphics package is based on it. I don't think of Scheme as a functional language. It certainly isn't much like the ML languages. That aside, the Scheme interpreter is just that: an interpreter, using boxed types. It isn't compiled to native code, doesn't do type inference, and so on. I admit that I haven't programmed in Ocaml [yet], and that some of the arguments in its favor are valid if the choice is between Ocaml and C or C++, but lose force when comparing Ocaml and Oberon. Nevertheless I'd like to have an ML-style language integrated with Oberon. But it might be a while before I have the courage to try to do it myself. sr

1 0

Re: Oberon Digest, Vol 16, Issue 11
by Soren Renner 15 Jan '05

15 Jan '05

> your renderer works great on WinAos. > You can use PELinker.Mod to create a standalone demo for windows (just add all > the modules and files called by your renderer to the file Oberon.Link that is > somewhere in the distribution, take care of adding them in the right loading > order, rename the executable's name, and Link everything with PELinker.Link ). > And do not forget to give us a feedback about your experience. > > BTW: - is your presentation from oct 22 available ? > BTW: - a release with a HowTo to create own renderable objects, a short > description of what is does and how, would be greatly welcomed and certainly > would significantly lower the threshold for early adopters of your fascinating > approach, including me. > > Greetings > Patrick > Patrick: Thank you for the suggestions. I am working on a greatly improved demo with a little documentation. Since I do not (and refuse to) run Windows, I hope that you yourself can generate the standalone executable from what I send you. Then we could perhaps generate interest among the mob. [Bitter experience has moderated my natural hopefullness]. If we could get a blurb on Slashdot, that would do a lot. Please continue your encouragement as it will help keep me on a work kick. Perhaps someone at ETHZ will provide a little web site for us. Or could we use Google's blogspot? sr

1 0

Re: %XX/W3C-encoding not applicable to ISP login
by easlab＠absamail.co.za 14 Jan '05

14 Jan '05

--- easlab(a)absamail.co.za wrote: > > No ! The ISP doesn't even see "%40". > > So the very concept of 'escaping' [outside of the > > n-o client] is inapplicable. > > The 'translation/disguise' is only internal to n-o. > > The whole mention of 'escaping' is inappropriate > > and misleading. John M. Drake wrote:- > I not sure you're understanding me. > > Ok, I'll try and explain again. > > Step 1: Change NetSystem.SetUser to "unescape" any > time it sees "%". > > Step 2: Use the "Escape" program to replace all > occurences of @ with %40. > > So then, NetSystem.SetUser will get the string with > "%40" in it, but then make it a string with "@" in > it BEFORE PPP sends it to your ISP. > Yes exactly: that's why I wrote ' The ISP doesn't even see "%40". ' And usually %XX/W3C-ENcoding would be DEcoded by some application on the other side of the world -- as James Carlson pointed out. You proposed a gearbox/translator which can 'hide' any character, [except "%"] and could pretend to fix the problem. I've used a band-aid/dirty-hack to 'hide' "@" and deliberately want to emphasise that nothing is fixed until the RFC is satisfied. I'm thinking that if we want to still use the 'read the tokens off a string' approach, we could use white-chars as the 'token terminators', instead of ":" and "@". Apparently linux does this. Perhaps userID & password may not contain white-chars ? > Alternatively you can change the code of PPP > to use another mechanism for setting the user > name and password. I can't help you there as > I don't use the NO PPP. But perhaps that's > the best longterm solution. > Yes, OK. But the PPP needs more fixes: the FSM apparently locks in bad-state(s) instead of resetting, for unexpected events. Translating James Carlson's published C-code implementation might be good ? BTW a tool-set to *assist* C to Oberon code translation would be of great value to the Oberon community. > > Thanks, it's a bit more complicated: > > Ver-2001 was patched 'from the beginning'. > > Now I need to somethimes email from a portable > > machine, and I don't want to repartition my > > note-book. > > So I want to use my DOSBased V2.3.6. n-o, which has > > a > > different syntax to Ver-2001 re. ppp arguments. > > So what syntax does DOSBased NO use for setting > the user id and password? > v2.3.3 StandAlone's documentation reads: NetSystem.SetUser dialup:<username>@DIAL ~ As does V2.3.6 DOSbased's documentation. > > Your mention of "NetSystem.SetUser" helped me go > > directly > > to my previous patch, and applying it to V2.3.2 > > [which source > > I have] which enabled login for DOSBased V2.3.6. > > NewsArticles fetch tested OK. > > Do you have the source for DOSBased NO? > Only some; so I used StandAlone Ver 2.3.2 which I patched to transform "$" to "@", which allowed me to login, but there's still a problem with the email password. I'm exhaused with/by the whole matter. == Chris Glur.

2 1

Re: Oberon Digest, Vol 16, Issue 5
by Soren Renner 14 Jan '05

14 Jan '05

> Personal rant here. While I'm still a fan of > original Oberon "tiled viewers" I'm beginning to > be weary (wary?) of the corded mouse setup. CHORDED mouse. Think of a piano. Mouse CHORDS == interclicks. Corded mice have cords. That's most mice, but not an interface issue. While I am on this subject. let me mention that: the Xmas Bluebottle release is beautiful, but I would LIKE the mouse chords (interclicks) to COME BACK, that is, for interclicks to work in the non-embedded-Oberon-window part of Bluebottle, because while I continue to use Oberon as my Bluebottle IDE, I can see that it would be fun to move to using a native Bluebottle dev environment, and without mouse interclicks such a move is almost unthinkable. Here is another idea: could Bluebottle be LNO-ized? So that a Bluebottle installation could run on a Linux box? Then it would be easy to release one of my amazing fractal-voxel-thingy demos for Linux. Not that it would be difficult to port to LNO itself -- I would just have to rewrite some code. But I really like the Active Oberon features. Here is another idea: embed the fractal-voxel-thingy renderer IN BLUEBOTTLE. Not for a large, extremely slow raytraced desktop and window system: no, only for small, extremely fast raytraced icons / buttons/ border elements/ decorations/ skin components. Look at it like this: on the one hand, we have what is clearly the best OS and language project in the world*1, yet almost noone will pay any attention to it. On the other hand, we have the first fruits of a new approach to 3d realtime rendering, yet EVEN FEWER people will pay any attention to it*2. Perhaps by multiplicatively leveraging these two obscurities we can break through to the other side. All that would be required is that there be a really nice demo (but the Sierpinski flythrough was probably almost good enough: some of you must remember it) and that it be in a form that dozens, hundreds, or thousands of interested parties (programmers, mostly) ACTUALLY SEE. Perhaps someone on this list could take a demo that I write and port it to WinAOS. Could it be distributed then as a single executable file (standing alone?) Is anybody interested in the voxel-renderer for skinning Bluebottle? 1. I do admit that functional languages have some claim here as well, especially Ocaml. It is just that I would prefer it to run on top of Oberon. 2.But thank you, University of Chicago CS department for hosting my talk on October 22nd ["What is Realism in Realtime Graphics"] and permitting me to fulfill my lifelong dream of baffling and alienating a whole roomful of people at once. sr "cygnus inter anates"

4 4

Executable for Soren's Renderer using WinAos on Windows
by Patrick.Hunziker＠unibas.ch 14 Jan '05

14 Jan '05

= Soren Renner wrote about his renderer... Hi Soren your renderer works great on WinAos. You can use PELinker.Mod to create a standalone demo for windows (just add all the modules and files called by your renderer to the file Oberon.Link that is somewhere in the distribution, take care of adding them in the right loading order, rename the executable's name, and Link everything with PELinker.Link ). And do not forget to give us a feedback about your experience. BTW: - is your presentation from oct 22 available ? BTW: - a release with a HowTo to create own renderable objects, a short description of what is does and how, would be greatly welcomed and certainly would significantly lower the threshold for early adopters of your fascinating approach, including me. Greetings Patrick ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

1 0