May 2003 - Oberon - mailman3.inf.ethz.ch

Re (4): [Oberon] PPP
by peter_easthope＠gulfnet.sd64.bc.ca 20 May '03

20 May '03

Edgar, > Pieter Muller <pieter.muller(a)alumni.ethz.ch> wrote: > For the escape mechanism I suggest URL encoding: > http://www.w3.org/International/O-URL-code.html Certainly a better idea than quoting each argument. Seems that this encoding might extend to Unicode. es> The only problem would be 0X. If 0X is encoded at %00 why is it a problem? P. http://carnot.pathology.ubc.ca/

1 0

garbage collector
by hurlimann 20 May '03

20 May '03

Whether it gives a possibility to make the garbage collector manually by programm an not automatic? Greetings Bruno Hürlimann

3 4

Re (3): [Oberon] PPP and overall Oberon coordination
by peter_easthope＠gulfnet.sd64.bc.ca 20 May '03

20 May '03

Chris, cg> Well, I don't/shouldn't have to know what kind of server system I am dealing with. True. GulfNet uses a server called FirstClass which uses a proprietary authentication protocol. I was wondering whether your server offers a proprietary authentication prior to PAP. Regards, P. http://carnot.pathology.ubc.ca/

1 0

Re: [Oberon] garbage collector
by eas-lab＠absamail.co.za 20 May '03

20 May '03

hurlimann wrote: > > Whether it gives a possibility to make the garbage collector manually > > by programm an not automatic? Patrik wrote: > You have two options: > > 1. disable and enable the GC by hand (this could kill the system, if you use > a lot of memory) > 2. allocate a memory block (with SYSTEM.NEW) and do whatever you > want with its contents > > You should only use untraced pointers in this block. Using normal pointers > could (and probably will) crash the GC. > Me too ! Because of crisis management here, my work-horse is loaded down: 10 days since I rebooted, and hundreds of e-mails, News, https have passed through (some are still 'packed' as viewers waiting to be filed). Now sometime a mouse or key response delay of half a second occurs. I'm guessing that 'forced' GC is often causing the delay ? Is the heap off-on loaded to/from the hard drive ? I think the RAM is only 20 Mb. I did HyperDocs.ClearCache with no noticable improvement. Perhaps unloading unused modules ? Or is this done automatically ? -- Chris Glur.

3 2

Re (3): [Oberon] PPP
by Edgar＠EdgarSchwarz.de 19 May '03

19 May '03

eas-lab(a)absamail.co.za wrote: > Level 2. > ... > Apparently with the existing n-o system, any login ID and password > would be specified in the "Dial = {" section of Oberon.Text ? > And here the 'phrases' are deliniated by <doubleQuote,newLine>, > so there might be a problem if <doubleQuote> was a char, of the > ID or password ? Perhaps '..."...' would work in such a case. > level 3. > ... > Apparently the PAP authentication login ID and password > are obtained/used in n-o from/via NetSystem.SetUser . > So apparently at this level "@" could be 'masked', as could > "%" and "/" ? Yes ! > Pieter Muller <pieter.muller(a)alumni.ethz.ch> wrote: > For the escape mechanism I suggest URL encoding: > http://www.w3.org/International/O-URL-code.html This means e.g: > For worldwide interoperability, URIs have to be encoded uniformly. To map the > wide range of characters used worldwide into the 60 or so allowed characters in a > URI, a two-step process is used: > * Convert the character string into a sequence of bytes using the UTF-8 encoding > * Convert each byte that is not an ASCII letter or digit to %HH, where HH is the > hexadecimal value of the byte > > For example, the string François > > would be encoded as Fran%c3%a7ois > > (The "ç" is encoded in UTF-8 as two bytes C3 (hex) and A7 (hex), which are then > written as the three characters "%c3" and "%a7" respectively.) > Pieter Muller <pieter.muller(a)alumni.ethz.ch> wrote: > > Bleibt nur noch eine Frage: Ist es zulässig das in NetSystem.SetUser > > einzubauen oder soll ich ein PPPMain.SetUser machen ? > It's fine by me to include it in NetSystem.SetUser. So this could be the way to go (There are also other options): NetSystem.SetUser learns about the % mechanism and decodes on storing. The only problem would be 0X. But do we expect this ever to appear in a password ? Cheers, Edgar -- edgar(a)edgarschwarz.de "http://www.edgarschwarz.de" "http://www.edgar-schwarz.de/forum/oberon" Running Active Oberon Make it as simple as possible, but not simpler. Albert Einstein --- start of oberon mail --- 0?Ahm;<P3m]eE^Y]\b=V`dUCM\^eU6e]di3fm1`7<0<P?HB80O0<0^c1X0OZH;[LO[KS9<M100lI 2@0>C4HI3kLK9K3[H1JH5kL3KK3KJI[;7kKMYN3;8_[LO;M;[>KX?19C;[M;;K1Y<MI3m98MY;MI 3mA@2bP2Gh6FiFFgBGfVG@NgdBGd2BjRfb2bbRgd>GjVFgNF@jfFnF@>gl>GjFffbB@6FgVG@bfg NfdjF@VDR2b`jFb2Bh6fi>gknFiBf6jC@2Rgm]eU]T5TR]\PL^`]\S=]V=]UU\P<]^5Td5]UY89J J3;K1I?1iN598WKI7;MCkKM;8O[I1iC5KIUS^TZU5_d5To\Qn4TPH@M;I19J;[L;;8Y;J;cC2Gd: g`>gb>gC2b`:gb2BbFFfVFgVf`BgbBF@:fl2BNBfgFGabfb6ejnFjFFFjfbNGVVFgFFObb6jC@hL O;8Y;J;[L\]Ym\XU^PD\UIH19LUkK5;K;KK1IJ=;8i9IO30m98_KHW;83;87;J3[LI98O[I19MAc 6jC@2R9UXPl]b5T`=\c12bOf@XFFiRf`2gi2bCjBGjBAjBGjbC2bknfj`knFi^F@VFg2biFgaRF@ 6F@>f`>gbjb6f@O2BfFFkFVPLV^\Qn4T^02RP<X`IHUCgPd5=19D3:D1IH[;MAKIM;MCkH3;MCkK MCfnfc06627P=88HHUKI1iK5;M3KJMKI9k;[kL;;I1IJd]]l==[LOKKOYMCKH1YC;;MWJNW;MdUC ]\d]Zc]\b5T^\Qn44LZ_5TQ5^`Q^13;M19MAKJW;8IKI]359@5987kK[;KD<1i9KKHWkJ;;I?9;1 IHW;87kK[S=dWPPR\TR4TQe]TY8OY8198oI3cJIW;83I3KX?19DCKIYKIU;8KJMI;K;[L19?QKJ; ;MdU]]^\Q0=\\]^]e]YeUUU^XE_^L\XeWPl^bm=P6m]b5Td5=Ffi>f`2gb2bfFfaRf`jfd>gf2bT 2biFgcNfb>Gj2bZ:EV2bbjfanFbVFgNFMf@O2B@4]dU^`EW_lUgm^geUgMV^l]bm\_<Y^U^UE^^= \d=]_IHIk;OJ;[ZDIJ;7kK9KIM9JYKKIK3K8EAKJWcfFf`jfi2bbjbcZ3J4Ngg:GfBfkVFbFF@VF gBgbl]`]\b=\R=]\=]d=_\4TEEZ9M^P4]Qe^U5Tdm]PD\U5TUe]SabBF@FGgVFcnFifFfVGG2BZn F@ff`2G@BGdFf6hMC;ID^Qe]W]\Pl]V5TS5]QIH7;M;[LW;8[kL;;INgg0HJl=Y;J2BK2C@nFiL^ _5TQU]\m]g]\T5TS5]Q1`djVQ]Qn4TEEZ9UUP<\PT^gm]]L^d]\`5T`E^_M\UM^c5TYM^P\^cAMf Pn4TZ<Q3m]^e^UE^d5Td5=>Fd04TcU^b=]^m\P<]^U>6F@>gb6gjFFg>fb2bgXHc;M;kL1IMVFgN F@BGdHEYZAK9>388;[KHJ\Qn4TZ<10Ff`>Fd2BaVGj4Td5]QU^P<]cYKO;M1IHM;83jD7JBC:8IK IY;M;[L1iKU;89KJ?KJY;8YkK1I9A:BI98_;J;[L;CTRD@VfiT^XI34]U5_QU\UM\Y]]QU]Pd^QU ]e]\Pl]V9M4TR=_dI3ma6jC@J4FFl6ff2GfFFF2BjR6W;MU3JDi6Fg>igVfif@O2b6hMOKMBF@0H I<\cYAUSUL\c\TQmV_=]c14UD5]U5TRLbR4TYM^P\\^M<`dHEYZA4TQM>YkMO;85KNYciLXc4TX4 ]U5_Y4TQe]TI@_98A9J8;1iMAKJ7;J1IHUC@BGdFFgf@O2bk:gdBGjFFg2b`>G@4=19M:gbFF@>F dPd]<WCAFba>CA2b`j65I93k=598UKIW;L;kHYKJ]KIIKNMI:KH3mAX01ICPPT7P^04800P_1dWP DX\]\YE\d5T^]^b5T^m]S5]P\\Ye]U5T6E^Qm\UEWP<YcU^P\\c5Tj]^\M`cM^Ym\PT\QM^P<]^5 T>]\d1fVCQcYLKX?1QU=]^E_eE\Q]^U98O;I;[L1iLO;KI;8CkHAcb88Q:DQJC3KJM[;WJIYciD^ P\]QM\X]\^5To\Qn4T9U^WL^Pd\Ye]U5TR=_P\]U5Tdm=C[K7;K[;I;;8BG@VFg00W2WCijb6f`Y nF@BWYaa08MFVg=\i5Tdm=?kK19:Y:J:733bfinF@nFjRfb:G@nFhBgdnFg>gDZc6jdbBgY0PcYL 19K6Fijfi2b`:fgFGj2BjR6;98Kca0Kc`j69KI7kK9ci2bgjF@>GjnFiVFgNFGf@ZRfb2bgjFfVG @2GiD<hMOKM0Q9FM985JMY;89kK1iM;;8;;NQKI7;M19MAKJ\\f]\b5Tdm]P<\`II3[L1IJM;83; 8QKHW34ToH37:J;KIUkLI98;:I?KHUK3KH3KH;K98KHI9kI3CPFFbNf`L^S5]g=\bE_^T\U5TP4T P440059JY;MQ3`bB60PR\QR4]dU>00KiL7;J_3h;=kKUKMKk;O[H;[LDTP4TP4TB]^^e]Ye]W5T1 M\d==2bW:fb`6fd`^fb2bdBWQM>WKJK;LIC@6VP4^_M^c=]RU]UUUPD\eU^Pd]_U>WKJK3U[;198 4T1U]R]\bU>;JJMkLYKIC[Kg^@CJAUZRCPn>1:PY616a4@0:V0hoo?4:H12@04PWEPP8818@18H2 XP:`2>38838@38H38P38X38`38h3@12B12D12F12H12J12L12N1LPPLP0MPPMP0NPPNP0OPPO098 02@0NV:hoo?5H01hooo@6Pc501801P:X2nooQ107JK9C1Rdl2gb:GRnfa>GGJdgbFfnfkbddL]P4 TU4TSlY`U^P4TSLYU=?GJIcc1J01hCQ;M1X03:0JFj2GMnRVU^`eUYM^YeUUU\emUYe]]d]_U^UM ^_D^VM\bDVg<W^T^hU>580YJIa;M?JH9kI;;MW[;MJI_kDYKNIKIQZLOkH18Bc;L;3XC;CVV67jK M;MUkKIS02$

1 0

Re (2): [Oberon] PPP and overall Oberon coordination
by eas-lab＠absamail.co.za 19 May '03

19 May '03

cg> 1. there is possibly a loginID and loginPassword to the ISP *before* entereing the ppp level. Peter Easthope wrote: > In general, yes this is possible. > What is this server system you are dealing with? Well, I don't/shouldn't have to know what kind of server system I am dealing with. Provided that it complies with the applicable RFCs. But you finally have stimulated the suspicion that I've been talking (to more than one person) about a different level than what they are disagreeing with me. Here's how I (currently) see the situation: Level 1. Calling modem negotiates with ISP modem. No problem here, but I'd like pointers to info. of how this is implemented. Level 2. Serial data is passed to ISP, which traditionally included client's login ID and password. These 'phrases' can apparently include any ascii chars except <lineterminators> - which apparently parse these tokens. Apparently with ppp usage this level of authentication is seldom used, and apparently nor does my present system. Apparently with the existing n-o system, any login ID and password would be specified in the "Dial = {" section of Oberon.Text ? And here the 'phrases' are deliniated by <doubleQuote,newLine>, so there might be a problem if <doubleQuote> was a char, of the ID or password ? This is the level that I have been inappropriately arguing about !! level 3. Once ppp starts, if PAP authentication is used (most common currently), the login ID and password are negotiated after the LCP negotiation stage. Apparently the HDLC Framing facility at this level, makes 'character breaking' possible, so that any 'abnormal chars' (eg. "@") in the login ID and password can be masked. Apparently the PAP authentication login ID and password are obtained/used in n-o from/via NetSystem.SetUser . So apparently at this level "@" could be 'masked', as could "%" and "/" ? ------------------ > OK. Please describe your best plan for > organization of authentication parameters in Oberon. Since I have the "@" char in my PAP authentication login ID I will test that the prevously proposed methods to handle this work OK. Which I previously mistakenly rejected. Meanwhile other(s) should check and correct my description above, so that in future we can be 'reading from the same score' ! -- Chris Glur.

1 0

Re: [Oberon] Systems that work
by eas-lab＠absamail.co.za 19 May '03

19 May '03

Rob Solomon top-posted: ]Why not just install WinMe. You have the CD, don't you? ]Why did the local microcomputer store force the upgrade? Cos they're in business. I think M$ gives "can't resist" conditions. Do you know about the reported offer that Gates made to IBM that got him in the door ? At 12:40 PM 5/18/2003 -0400, you wrote: >Folks, > >A story: >Several months ago my main hardisk started to show signs of total failure. >I did a lot of CD backups and then decided to have it replaced. >The local Microcomputer store essentially forced me to upgrade my Windows >ME operating system to Windows XP which I did (reluctantly). > >Now neither my Canon Multipass C5000 nor my Jaz drive work! By >downloading some none Canon software I can at least print to the C5000 >but >the scanner is non functional. It turns out that Canon can not get a >reliable device driver working under XP. Now note, the hardware is >unchanged on my fax/copier/printer the cables are the same, the CPU is the >same. It is only the operating system that has changed and yet the maker >of the hardware who had a perfectly viable driver for Windows ME can not >get it to work under Windows XP! (The Jaz issue I have not fully >investigated since I have a scsii to USB converter on the cable. I will >probably have to buy a scsii board.) > >This Microsoft dominance is another reason for sticking with the Oberon >language and the various Oberon operating systems. At least one can "see" >what is going on internally. > >We need the Oberon alternative. Hang in there! > >-Doug Danforth > Linux is the one that's increasingly threatening M$. There's no 'hanging in' to be done, as if waiting for a spectacular event to arrive. It's just a matter of refining/cleaning: I'm amazed at how little n-o has been 'cleaned' since I started with V2.3.2: still dud underline char in Oberon10.Scn.Fnt, no documentation for various modules ....etc. OTOH the multi-partitions facility solved a serious limitation, while retaining the (apparently faster) flat file system. It took me 18 month usage to discover how to use the drop-down menu frame for Desktops.OpenDoc ^ in the SystemTrack, to solve various problems. I don't think it has been documented ? That's why I don't intend to 'upgrade' and face another wastefull learning curve. re. Windows XP and latest/fastest hardware: it still seems slower than my n-o 2001 on scrap 486 hardware for what counts for me: inet fetching text following by save, search, cutNpaste. re. Windows 'marketting', my ISP is becoming congested (why don't they give engaged, instead of allowing a call charge, and keeping you in the ppp negotiate queue - to probably time out): and when I phone the help desk, it's always: " no we have no problem at our side" " what version of Windows are you using ? " Yet with a different machine, OS, modem and telco line I still can't connect. But to the neighbouring town I can connect. The next day all is working - no admission of problems, always: " no we have no problem at our side", " what version of Windows are you using ? " -- Chris Glur.

1 0

Re. e-ink and Oberon
by eas-lab＠absamail.co.za 18 May '03

18 May '03

Douglas G. Danforth wrote: > I have a dream and that dream is to have Oberon be the underlying OS for > pen based displays yet to be produced by e-ink of Cambridge, MA. > Why would Oberon be a candidate ? Because it has a high low-size*hi-speed product. But why can't this be achieved by others ? This has puzzled me: Is the file structure important ? The duplication of code in many modules, does not help the size. I guess duplication comes from each 'student' doing a full project ? eg. how many times has the 'hex2ascii' routine been rewritten ? Partitions.Mod contains a lot of code which would also be in Hex.Mod . Perhaps it doesn't matter with dynamic loading of modules ? Are they able to be swapped out ? Does e-ink intend to have a Hard-Drive ? I never got an answer to the question of why S3 code always uses var.pars instead of functions which are easier to read. Are these much more efficient. The syntax doesn't exclude functional style programing, but when I reported a 'failure' when eg. using Cost := Biggest( Cheapest(Shoes)) I was recommended to use the var. pars 'work around'. > Since I am a physicist by training I have at least 35 notebooks full of > calculations. I would like to replace those books by a single one > capable of holding thousands of pages that can be clean, neat, > searchable and capable of transmission to others and of high enough > quality for submission for publication. > > Needless to say the Oberon display and mouse functionality would be > replaced with gester (and possibly voice) recognition. > What's 'gester' ? Long hours of mouse usage has given me injury problems - which I previosly thought a myth. Note that the hand position for mouse usage is at one extremity of its 'swivel'. A Joy-stick is in the middle (most relaxed) of the swivel range. I want a 5 key joystick system ! N-o has generated many 'notebooks' full of info for me recently. Multiple partitions makes the info-glut managable. This can't be done with a hand held. You have to start from the human: ten fingers and size of fingers ....etc. The chair and table (for productivety) is as important as the computer ! Note book style can never be 'hi-volume-productive'. You don't want to tell me that you do serious calculations without sitting down COMFORTABLY ? Which negates the note book size. > Does anyone else share this dream? Does anyone else share this dream > strongly enough to kick around what it would take to form a company to > sell such a device? > I have dreams that involve n-o, but somewhat different from yours. I recently had a play with a pen/touch-screen based Motorola A008 cell-phone: e-mail access, Fax, SMS and WAP, handwriting recognition ..etc. It's not a new development, yet there seems little demand for them. I was impressed, but what the market wants is often very different from what we want. -- Chris Glur.

3 3

Aos on an old dual Pentium
by Paul Reed 14 May '03

14 May '03

I tried the "New Year" Aos CD on an old dual P120 system which I saved from a customer who was about to throw it away. At first I had problems because the Award BIOS (v4.50G, 1994) won't boot off the CD and NOBOOT.EXE didn't seem to work (Native wouldn't boot using NOBOOT.EXE either, but from a boot diskette Native was OK - is this a known problem?). Anyway, I installed Native (Jan 2003) and then installed Aos the long way, off the CD (extracting the zips off the mounted CD is slooow) and got the following trace. By the way, in AosInst.Tool I had to do "FileSystem." instead of "OFSTools.". Any ideas? I have no desparate need for this to work, but it IS the only multiprocessor machine I can readily lay my hands on. I am happy to try things out to see if the problem can be fixed, but by all means tell me my customer was right to be throwing it away... :) Paul Reed Aos 31.12.2002 Intel MP Spec 1.1 Processor 0, APIC 11, ver 5.2.1, features 000007BF, ID 0 Processor 1, APIC 11, ver 5.2.1, features 000007BF, ID 1 P0 starting P1, P0 timeout on P1 {P0 Unexpected interrupt 80002F70 00 00 00 00 46 32 00 00 00 00 00 00 90 2F 00 80 ....F2......./.. 80002F80 B0 00 E0 FE 00 00 00 00 C4 2F 00 80 04 00 00 00 ........./...... 80002F90 03 00 00 00 B0 2F 00 80 C2 7A 11 00 08 00 00 00 ...../...>...... 80002FA0 82 30 00 00 FE 08 00 00 3E 00 00 00 .0......>... System hangs with no further trace.

1 0

Re: [Oberon] OE03 ( final announcement )
by Thomas Frey 13 May '03

13 May '03

> First Name: Thomas > Surname: Frey > Address: Ursprungstr. 103; 3053 Münchenbuchsee > Phone: 01 632 0853 > eMail: frey(a)inf.ethz.ch > > For the ones that give a presentation or demo, please send me > a final description and how much time you think you need. There > will be a printout of the final program including title and abstract. > > Title: New User Intreface Ideas for Bluebottle > Time: 20 - 30min > Abstract: Introduction to the new features of the Bluebottle GUI. User and programmer perspective. --Thomas

1 0