Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!ira.uka.de!fauern!lan!roell From: roell@informatik.tu-muenchen.dbp.de (Thomas Roell) Newsgroups: comp.unix.sysv386 Subject: Re: Legality Status of Thomas Roell's X11R4 server Message-ID: <5953@tuminfo1.lan.informatik.tu-muenchen.dbp.de> Date: 7 Dec 90 08:52:40 GMT References: <5918@tuminfo1.lan.informatik.tu-muenchen.dbp.de> <1990Dec6.173229.7100@jdyx.UUCP> Sender: news@lan.informatik.tu-muenchen.dbp.de Distribution: comp Organization: Inst. fuer Informatik, Technische Univ. Muenchen, West Germany Lines: 140 In-reply-to: tpf@jdyx.UUCP's message of 6 Dec 90 17:32:29 GMT (this letter might boar you, but I think this letter should be open, so everybody can read it, and thing about it) >> >> I heard from someone that Tom Friedel posted here an article (which was >> canceled by himself) in which he stated that the binaries and sources would > >no, someone else cancelled. One of you fans, I presume. > Maybe, but I don't find this a good behavior, too. That could be not fan of myne, who takes away the chance from me to add some comments to this article. Whoever this article canceled: NEVER DO THIS AGAIN ! >No, here is why posted, and I debated about it for several weeks, and I >consulted with a few poeple here on the net. If I did the wrong thing I am >truly sorry. I appreciate your bringing the issue into the open. Why didn't you talk with me ????? >I have been working in Unix 8386 graphics for a few years, and yes, I am >also working on an X11R4 server. I have to pay hundreds of dollars a year >for 'support' contracts, which means a few days later I get an answer to >my question that more often than not is not very helpful. > >I also discovered that the STREAMS ldterm module was broken, but I could >only guess. It might have been the 'ptem' module. It would have been so >nice and saved me so much time if I could have pulled up the AT&T source >and checked it out. > >I also had to learn how to do video memory mapping, vt switching, and a few >other things with header files, poor documentation and much trial and error, >at great time expense. > >So the reason I posted I think was to cry about how unfair it all seemed. >I sure didn't do it to make myself popular. Hmm, ok I think it's time to tell the REAL whole story. Perhaps you missed some of my e-mails, perhaps my english was not clear enougth, or perhaps I should have told you the whole story. My server war running long before I had any contact to AT&T sources (excpet the standard includes ..) -- about in early june it was quite finished. So I had all the trouble you had. Even more: I had no 'support' contracts. I had not the resources you had. Do you know how hard it is to get data books for video chips ??? Did you pay around $1000 to make a PD-X11R4-server happen just for idealism ??? Ok, ok I worte this server for myself, too, but the high costs remain. I had also to learn by trial and error. Debugging a graphics application via a seriall terminal, cause your console gets scrambled is no fun ! And the documentation I got form ISC is rediculous. Just call it good instinct. And having only a 110MBytes HD for all sources. I don't want to cry here, cause writing this part of software was quite funny, and I learned much. BTW, even you did send me earlier some patches to help me... (You may ask, why I didn't give away my server in june ? The reason is simple: I had to write some important exams (aug/sep/oct). So there was not the time to fully test the server. And there were some bugs! There were also some extensions, which I wanted to do.) Around early june I experimented with a german keyboard, which has some charecters on it that are included in the ISO-Latin-2 charecter set. The codes of these chars are above 127. So I got to the problem that ldterm stripped the eigth bit. I even dissabemled the ldterm and pterm driver, to dectect the bug. I got it in ldterm. I also got a workaround: The eigth bit is only stripped, if OPOST is enabled. This was REAL reverse engeneering !!!! Some weeks later there was a discussion in comp.unix.sysv386, why VPIX didn't run under my first server. I instantly knew by looking at some include-files that ldterm was the reason. One of the persons (who's name I NEVER will tell), who urgently wanted to run VPIX with xterm told me that he had the sources for the STREAMS package of AT&T (here I wrote something differntly to you, but I wanted you NOT to know that somebody has sent me these soucres, cause I promised him nobody to tell his name). There was a long discussion with my friends and some people from the net whether I should simply include the ldterm driver in my X11R4 distribution. Most thought that this would be illegal, but the only way. If I sent the diffs to a UNIX vendor, there might be patches, but can you really wait for them ?. A ldterm driver without a STREAMS package would make no sence. So all I was about to do was to give away simply a bugfix. >>2) Tom Friedel did this posting because he works for a company that also want >> to sell an X11R4 server for VGA. He didn't tell me this fact, when he >> reported me his problem. I also assume that he hab a look at my beta-test > >I did NOT report to you this problem. I have never reported any problems to >you. I posted to the comp.windows.x group, and you replied. This was my BIG mistake! Because you helped me earlier with a bugfix, I thought it would be fair to give away some of my knowledge to you. (In February I a guy sent me some patches - I should treat them as confidental - and I remeber that his name was Tom Fridel. Maybe I remeber something wrong ?) I assumed, 'this is a good man, he helped me, why not help him before he spends the same time as you working on the same problem ?' When you sent me your letter, where you said that you think it's unfair to include the ldterm driver to the distribution I dropped the ldterm driver immediately. >> code. So I assume, he wants to do everything to drop my PD-Server. > >No, I won't say another word. I just wanted people to know. I'm very sorry that I said these hard words. But I was reported that you canceled this article by yourself. This made me very angry, cause I indeed wanted to answer and clarify this situation. But this behavior (which I assumed) looked like you didn't want me to answer. As I read the announcement of your work I was sure that your work collides with myne. So it was clear for me that you would do anything to drop my server (this was also influenced be the rought tone in your letter). Please forgive me that I get you wrong. >No, ours is shipping now on ega,vga,svga,TARGA,8514 with Motif 1.1 on ESIX, >ISC and SCO Unix. With a 341 server around the corner. >( Contact me and I'll get you in touch with the right people ) Ok, this is my very LAST PD PROJECT. There will be just one new release including the XView 2.0 toolkit. Thenafter I'll write software only for cash. So come on, I'm free !! If you pay me enought, I'll write software for you. It's not worth the trouble writing free software ! I also wanted to do a 8514 and TIGA (34010) X-server for free, but you got me to a point to get my fingers off from PD-software. But there is a last question I have for you: As I read in the annoucement of your product, you used GCC and GDB for developing this. You used FREE SOFTWARE !!!!!! Didn't you read the GNU copyleft ? Why couldn't you just accept that I'm a idealist ? - Thomas PS: The company Tom F. works for is named MetroX. -- _______________________________________________________________________________ Mail: Thomas Roell (c/o Daniel Hernandez) Inst. f. Informatik / Technische Universitaet M"unchen Arcisstr. 21 / 8000 Munich 2 / Fed.Rep. of Germany E-Mail (domain): roell@lan.informatik.tu-muenchen.dbp.de UUCP (when above fails): roell@tumult.{uucp | informatik.tu-muenchen.de} -------------------------------------------------------------------------------