IRC log started Sat Jul 3 00:00:00 1999 [msg(TUNES)] permlog 1999.0703 -:- SignOff _QZ: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) -:- Closing Link: TUNES[bespin.dhs.org] by clarke.openprojects.net (Ping timeout for TUNES[bespin.dhs.org]) -:- Connection closed from clarke.openprojects.net: Success -:- Connecting to port 6667 of server clarke.openprojects.net [refnum 1] -:- BitchX+Deb1an: For more information about BitchX type /about -:- Welcome to the Internet Relay Network TUNES -:- Your host is clarke.openprojects.net, running version u2.10.05.18.(ipcheck4-5) -:- This server was cobbled together Wed Apr 28 1999 at 13 24:47 EDT -:- clarke.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv -:- [local users on irc(8)] 3% -:- [global users on irc(123)] 43% -:- [invisible users on irc(164)] 57% -:- [ircops on irc(12)] 4% -:- [total users on irc(287)] -:- [unknown connections(0)] -:- [total servers on irc(31)] (avg. 9 users per server) -:- [total channels created(85)] (avg. 3 users per channel) !clarke.openprojects.net Highest connection count: 23 (22 clients) !clarke.openprojects.net Welcome to Open Projects! You are on 2 ca 1(2) ft 14(14) tr. -:- Mode change [+f] for user TUNES -:- Mode change [+iws] for user TUNES -:- JOIN activated by "TUNES #tunes tunes@bespin.dhs.org " -:- TUNES [tunes@bespin.dhs.org] has joined #tunes -:- Topic for #TUNES: Free Reflective Computing System -:- topic set by hcf [Thu Jul 1 19:35:46 1999] -:- [Users(#tunes:4)] [ TUNES ] [ abi ] [@Tril ] [@Fare ] -:- Channel #tunes was created at Sun Feb 28 08:48:06 1999 -:- BitchX+Deb1an: Join to #tunes was synced in 6.813 secs!! -:- Mode change [-s] for user TUNES -:- Fare is now known as FareAway -:- water [water@ppp-tnt-199.tscnet.net] has joined #tunes -:- AlonzoTG [Alonzo@209-122-196-152.s152.tnt5.lnh.md.dialup.rcn.com] has joined #tunes -:- tcn [tcn@cci-209150250107.clarityconnect.net] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[209-122-196-152.s152.tnt5.lnh.md.dialup.rcn.com]) >>> tcn [tcn@cci-209150250107.clarityconnect.net] requested PING 931016061 362183 from #tunes -:- SignOff tcn: #TUNES (tcn has no reason) -:- AlonzoTG [Alonzo@216-164-230-222.s222.tnt7.lnh.md.dialup.rcn.com] has joined #tunes wb atg -:- SignOff AlonzoTG: #TUNES (Read error to AlonzoTG[216-164-230-222.s222.tnt7.lnh.md.dialup.rcn.com]: Connection reset by peer) 09:50am -:- smkl [sami@MCCLVIII.rdyn.saunalahti.fi] has joined #tunes hi smkl hello water it's pretty today i mean quiet :) it's pretty quiet today 10:10am -:- water [water@ppp-tnt-199.tscnet.net] has left #tunes [] -:- water [water@ppp-tnt-199.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (The Tao has left the building!) * Tril/#TUNES is back from the dead. Gone 23 hrs 10 min 25 secs -:- binEng [Anders@dialup92-1-5.swipnet.se] has joined #tunes hi bin! hi 12:30pm I'm continuing with my scripts ;) one mom ok k Tril: Do you know Python? rumour has it Python is interesting abi: good girl :) no, I don't know the language. ok. I was just wondering how it compares to Perl don't know :) it's probably simpler than perl more consistent syntax, anyway, i suspect less weird operators yes... that's likely. Is Perl faster, or else how come Python isn't as used as Perl? python is newer, that's all it's not as well known ic 12:40pm Should I get tkCVS, jCVS or WinCVS? 12:50pm -:- AlonzoTG [Alonzo@216-164-231-239.s493.tnt7.lnh.md.dialup.rcn.com] has joined #tunes I dont know, bin I've used wincvs but not the others wincvs isn't very intuitive, you have to know how CVS works at the command line to use it anyway plus it's buggy :) 01:00pm so maybe you should try one of those other ones :) yeah :/ just thought it could be good to have a client... just in case you can get the command line client for Windows. * binEng/#tunes wants no CLI then I'd have to *learn* the damn thing, wouldn't I... well, you will still have to learn it, but maybe with a well-designed GUI you could learn how CVS works faster. cyclic.com is prolly the hardest site to get something useful from I've ever seen. it's incredibe binEng: try http://www.loria.fr/~molli/cvs-index.html 01:10pm I give up. I don't like cvs anyway fine if you are making changes to the pages, you can send me a diff if you think CVS is hard to learn, try to set up a repository :) A diff is what you get when you run 'diff' on the files, not? yes, right I've never done that. How do a diff look like? there's a couple different formats, i like diff -u rather than the default if you have multiple files, keep the original ones in one directory, and edit a copy of that directory tree, then run "diff -r original new > filename.diff" if it's just one file, just type "diff original new>filename.diff" and they can be in the same directory uh (in the first example, origial and new are directories, the second example they are files) Now I understand. -r means recursive, it goes into the dirs and compares every file. 01:20pm and then, to restore the new version? patch < filename.diff the filenames to change are stored inside the diff ok, the program patch? don't you have to specify the target? ic right. patch is the opposite of diff if you used subdirs then you want to learn -p0 or -p1, for whether the directory to change is a subdir of the one you are in now (-p0), or whether you are in it (-p1). So if the dir is named something different than the patch file expects, then you go into it and use -p1. linus still uses diff to manage changes to the linux kernel. He just wouldn't use most of the features CVS has. So CVS handles it in another way? Is the Linux source managed with CVS? CVS stores all diffs for each file in a central repository, and gives each one a version number. You can retrieve arbitrary versions of a file (cvs applies the diffs in the right order automatically). It also allows multiple people making changes, by storing who made each change and comments about what the change was for. More advanced, it allows you to split development in multiple "branches", which are separate trees of changes, but can be merged back together later (or on an individual basis) ah. Advanced. it works on a "checkout" basis, if you want to make a change, you have to check out a local copy of the file(s) to work on, out of the repository. Then you make "checkins"(also called commits) to save your changse to the repository for others to see. (or you can just delete your copy if you don't want changes to take effect) 01:30pm And what if two users try to commit an update to the same version of a file? when the second author tries to check in, it won't let him, it says that he doesnt have the latest version in his local copy. makes sense. So he runs "cvs update" which gets the latest changes from the repository, and merges them with the changes he made. Then he can commit his version that has both changes in it. (update only modifies the local copy, not the repository) Hmm... how is the versioning data handled? It can't be attached to the very files in question, can it? Is is managed by the local CVS client? cvs stores files in the repository in RCS format, which means it has mutliple version numbers and patches in one file. yes, but locally, when you work on the file? the process of checkout gets you a working copy of the latest version (or the selected version), of the actual file. And the repository then knows what version you have, when you commit? no, it just makes a new version by storing the difference from the latest version and your version 01:40pm it adds one to the version number well, actually it knows what version you have, because checkout creates a directory CVS/ in your local copy that contains administrative information. ah. That's what I was after. it says what server you checked out from, and what directory in the repository maybe other stuff, try it and look at one :) sounds like asking for trouble ;) or just find someone's checkout directory in bespin and read a CVS/ in it 01:50pm -:- water [water@ppp-tnt-199.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (The Tao has left the building!) -:- hcf [nef@me-portland-us226.javanet.com] has joined #tunes hi lo hi anything new ppl? apart from my scripting efforts, no. interesting thread started on the list by JimL requesting honest critique of his prism assumptions ok, that too ;) and the previous thread including RMS regarding intellectual property :) (which may not be ended yet, I haven't been following) or perhaps on hiatus i think it's waiting for RMS' turn of course he is not obliged to respond :) 02:10pm -:- water [water@ppp-tnt-199.tscnet.net] has joined #tunes hello bonjour, water hi what's up? i think up is the opposite of down well... nothing you don't already know about, I think is everyone agreeing about jim little's prism message, then? I don't think so... he's not even part of tunes, so they need not be water, I think you have some good comments, but you walk the line between criticizing his ideas and attacking him as a person ok, that makes sense yes i do I hope he can take it wish i would change hcf! water! I was going to reply, but at this rate of response he is going to be flooded anyway and I'd rather work on my alternative (my own system) 02:20pm did you find any free or open persistent hyperprogramming systems? tril: indeed hcf: not that i expected you to find them, but it would be a beneficial result just the same no real systems found related urls: http://www-ppg.dcs.st-and.ac.uk/Research/HyperProgramming.html, http://www-ppg.dcs.st-and.ac.uk/Java/HPS/, http://www-ist.massey.ac.nz/~PLyons/personal/ I would suggest that we try to specify what we want in an ideal hyperprogramming environment, but then that would be the same thing as our goals for TUNES, wouldn't it? hcf: i've never seen that last one. i'll check it out tril: probably water: its the guy who seemingly coined the term hyper-programming hcf: really? i had no idea water: as mentioned in the visual programming faq hcf: wow. i didn't even know that the vpl faq even mentioned hyperprogramming hcf: i just browsed it one day water: when looking for poop, dont neglect deja, its equiv to searching all the faqs hcf: ok. i'll remember that the next time i look for "poop". :) hmm. tho i suppose the search engine as faqs.org would be better for such maybe if i just took a vacation, my inflamatory net personality would subside faqs.org? cool like playing quake umm i dunno on the other hand, i can't really identify an outlet for frustration in my life right now 02:30pm * binEng/#tunes logs on to Bespin water: how r the papers coming? hcf: slowly :( i've been trying to improve my writing style, mostly and collecting url's for bibs i also found Lyx need any help? !!!! :) hcf: not really, thanks bleh! the E language page is overloaded with buzzwords it's hard to pin down just what it's semantic features are. 02:40pm -:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-199.tscnet.net]) -:- water [water@ppp-tnt-121.tscnet.net] has joined #tunes back 02:50pm * Tril/#TUNES reading mail -:- jdl [jiml@ultra1.inconnect.com] has joined #tunes Hello all hi speak of the devil ;) hi jim! * jdl/#tunes has come for your souls ;) hehe What's the topic? well, the topic is Free Reflective Computing System No, not you... :b not much yet i'm trying to review E 03:00pm abi: forget topic hcf: I forgot topic What's E? E is, like, Enlightenment Window Manager @ http://www.rasterman.com or a secure distributed object platform and scripting langauge at http://www.erights.org or a predicate that curry used; it applies to all obs hmm it's that scripting lang and obj platform it's also very complicated Why review E? i proposed it :) dunno. tril asked om Still trying to create gloop? what's gloop? Oh, review E as in for the TUNES review project it said distributed object, sounds applicable. jdl ; yeah :) /me is having telnet projeblems be back in a second... it has some strange message-passing conventions -:- SignOff jdl: #TUNES (Leaving) among other things -:- jdl [jiml@ultra1.inconnect.com] has joined #tunes U know In GEB, Bloop, Floop, and Gloop! =P * jdl/#tunes reviews the log * jdl/#tunes looks at the E web site i think i can write up something simple for it Hmm... E looks interesting. I'll have to look at it some more later water: What are your impressions thus far ? and just mention that it has some complicated details jdl: not sure jdl did you look at that "prism now?" link, http://tunes.org/list/tunes/9906/msg00059.htm jdl: it's obscure what the language really is, other than your average script lang I like the emphasis on security. That's what I'm going to look at in more depth (later) http://tunes.org/list/tunes/9906/msg00059.htmhttp://tunes.org/list/tunes/9906/msg00059.html <-- HTML!! not htm stupid paste what do they mean by security, though? probably not the same thing Fare does.. :) definitely not Tril: Yeah, I didn't respond because I didn't fully understand it, but it didn't seem to be Prism-like. ok i didnt look at it I don't think ANYBODY means the same thing Fare does :) i try to Fare tends to go a lot further than most people :) their idea of security is based on java --I don't mean Fare is hard to understand, just that he's generally unsatisfied with other people's implementations as not going far enough-- probably a good thing tril: what? about basing security on java? 03:10pm it's good to go farther oh yes water: My initial impression is that the security is *distributed* security; i.e., probably not based on Java water: I need to look at it further, though. well, they've ridiculously complicated the semantics of message-passing. it seems to be for that purpose water: You're talking about their "smart contracts?" that among other things it abounds with buzzords they're advertising things that i don't think ordinary users care to know much about water: I've only looked at the front page, but their language seems more descriptive than buzzwordy water: arguably, ordinary users don't care about security :) "pure object model", "capabilities", "message pipelining" * Tril/#TUNES shoots jdl not funny hehe * Tril/#TUNES emails windows NT to jdl :) no, it's just that they describe parts of the system in implementation level ways Tril: I'm not saying the users are RIGHT :) See original goals of Sphere Tril: Have you read my 'challenge' email yet? jdl: have you read the responses yet? :) jdl: see the conversation the other day of Fare and I and Kaufmann about microsoft's mindshare and user choice jdl: i hesitate to reply since it will create more work for both of us, and you already got some answers water: Yes (yours too), I'm just waiting for everything to come in so I can respond with one message how do i represent an abstract concept in Prism i'd have to use a bit, a stream, or a map, but none is a good fit i need an abstract entity that has no properties Tril: I'm interested in your reply because you seem to have strong understanding of Prism; replies thus far show relatively weak understanding (water, even yours, but not as bad) jdl: I hate your pseudo ASM language for the reflection domain, it seems kludgy Tril: I don't understand... can you give me a concrete example? jdl: i understand it, but you haven't introduced any terms for me to express what you want to do with the prism idea. a concrete example of an abstract object? ok, i'll try !!! Tril: Me too :) My goal was for the VM to be easy to create, but I definitely would like to replace it -- just can't think of a good replacement 03:20pm my cat is evil. Tril: BTW, I've realized that reflection is a bad term, so now I'm calling it the "Prism Control" metamodel -- it's for representing programs that control the Prism compiler s/bad term/bad term for that thing/ prism control sounds clearer jdl: Are you using the name sphere for one of your things or is there a sphere project somewhere? are you familiar with metaobject protocols, though? abi, Sphere? jdl: wish i knew Fare says MOP is used for everything and its contrary, but if you clarified what you meant you could use that term for prism control. Sphere is Jim Little's Sphere project, at http://www.teleport.com/~sphere abi prism prism is a way to program with multiple specialized, interoperable languages. or at http://www.teleport.com/~sphere i think we told her about prism since sphere is kind of postponed :) Tril: metaobject protocol interests me, but I haven't delved into it yet Tril: Sphere (the OS) has been abandoned :) it's mutated into the "Sphere Project" water: sorry, not trying to ignore you, just so much going on jdl: it's ok Tril: Can you give me an example of an abstract concept? I'll describe how it would work with Prism jdl the basic idea of MOP is providing a restricted, precisely defined access to the underlying system, by providing hooks into it within its own constructs. Tril: I have Abelson and Sussman's book, I think it's discussed in there... I'll look it up s/it/MOP maybe, but a better book is AMOP, if you can stand the CLOS syntax abi amop amop is quite interesting. or _The Art of the Metaobject Protocol_, by Kiczales, des Rivieres, and Bobrow I didn't read SICP, actually I started to, but decided it was about lisp programming Tril: It's okay, didn't go into some of the fundamental philosophy of programming that I wanted at the time... yes, my conclusion too, essentially no idea if it has any reference to a MOP Tril: Hmm... maybe not. Chapter 4 is about "metalinguistic abstraction" but I didn't get that far. abstract concept: perhaps you have heard of MOOSE, a project to create an OS created entirely of objects Hmmz... Tril: Okay or Arrow's notion of homo-iconic, or Fare's HLL Requirement of "Uniformity" I was going to call my project the same thing and have it do just about the same kinds of things... =\ tril: was moose very far from squeak? (sphere) 03:30pm atg: Sorry! :) What same kind of things? water: moose was the project that turned into TUNES, obviously it didn't get very far itself back in 1994 (i wasn't involved at the time) tril: ok Tril: Sphere (the now-defunct OS) had the same goal Tril: What about these things? the idea is each one is universal within its system, every part of the system is made up of them. wh9ich means an object is a very abstract concept used for every purpose Tril: okay, but I don't see how it relates I like to think of it as a node in a visual environment, that is interacted with. I don't care what it is or what I can do to it just that it is something and I can do something, each of which i can find out more about if I choose sphere is defunkt? :((((((((((((((((((((((((((((((((((((((((((((((((((((((( Tril: You've lost me. 'it'? an object? but the general case it's just an abstract "something" right Tril: I still don't see how this relates... Jdl: I was going to have a special language for specifying device drivers, one for specifying modules... then any number of languages for specifying programs inside modules... =\ atg: Sphere OS is defunct. The Sphere Project is still alive * AlonzoTG/#tunes picks up the torch. atg: Prism, the language thingummy, is very much alive... if you want to assist/participate/contribute, please do! Specifically, none of the actual specialized languages have been defined jdl: every class hierarchy has a class at the top which all the other classes inherit from. We just want this root class to have absolutely NO features so that ALL features are specified in the class hierarchy and are controllable through the language. Even though we don't use the term class, hierarchy and inheritance, it's a related concept. jdl: what is 'sphere' in 'sphere project'? hcf: wish i knew atg: (except "Prism Control," the Prism control language Tril: okay jdl: in other words every object is an instance of the root class rrgggghhhhh! hcf: Nothing anymore :) Sphere was going to be composed only of objects, and "Sphere" referred to the somewhat common representation of an object with a sphere :) why does everything have to be squeezed into a hierarchy for you people? Tril: okay water shut up. I said it's related, but not a hierarchy. hierarchies hide semantic complexity! i'm using it as an example. grumble I don't think a rooted language tree would work... =\ 03:40pm jdl in some contexts i want to treat some object as an instance of a featureless class, even though it may have other features -:- SignOff water: #TUNES (The Tao has left the building!) Tril: I'm still not getting the point of all this the point is what if I want to design something top down, instea of bottom up? You make me write it in bits, streams and maps and combine those to something larger. What if I want to start with an abstract "something" and gradually add properties (information) about it? what if I want to draw a diagram using a visual editor, but not specify what any of the parts do? Tril: Ah... -:- water [water@ppp-tnt-121.tscnet.net] has joined #tunes jdl: I want to assume that my diagram is NOT representable into bits streams and maps, and then be required to find a representation, if one exists but that shouldnt prevent me from having the diagram, even if it is a program Tril: Well, one point I think everyone is missing is that _Bit, _Map, and _Stream are really only necessary when you're dealing with METAmodels. When your creating actual programs, it's more natural to use other things (source code, diagrams, etc) which are translated into Prism models by parsers. so why wouldnt i deal with metamodels on a regular basis because I'm an "end user"? jdl: but the translation could become extremely complicated Tril: You would only create a new metamodel when you are dealing with a new category of concepts/programs. it's our philosophy that the end user should be empowered to adjust the system incrementally and intuitively, and that involves modifying metamodels. water: The translation from source code/diagram to Prism model? jdl: yes, and vice-versa Tril: Sure. And that's why I want bit,map,stream to be intuitive. jdl: new categories of concepts come up every day, multiple times. The purpose of a reflective system is to make it easy to cope with these new concepts. okay I wrote to the E-mail addy on the sphere page and offered to write one of the domain specific languages. atg: excellent! I'll get back to you (it goes to my mailbox) -:- binEng_ [Anders@dialup44-1-4.swipnet.se] has joined #tunes water: The translation is another type of program, so the complexity could be reduced by metamodels specifically for those types of programs water: the whole point of Prism is reduction of complexity 03:50pm jdl: at the expense of expressibility? oops did i say that -:- SignOff binEng: #TUNES (Ping timeout for binEng[dialup92-1-5.swipnet.se]) -:- binEng_ is now known as binEng tril: hehe tril: it seems so Tril: At the expense of expressibility (is that a word? :)) within an individual metamodel. Not within Prism as a whole. Tril: Because metamodels interoperate. *may* interoperate jdl: how does it reduce complexity? and what is your complexity measuring stick? water: Good question. I've searched for papers on the nature of complexity and haven't found any. water: both in psychology and CS. jdl: there's plenhty water: so my definition of complexity is ad-hoc at best... jdl: information theory (it actually is in the field of signal processing) Can either of you give me a good place to start? hmm. * water/#tunes looks at his url list Fare recommends a book by Li & Vitanyi water: Anyway, my complexity measuring stick is the amount of accidental difficulty involved in understanding/doing. abi, k-c? i heard k-c was Kolmogorov complexity or a measure for algorithmic complexity based on the shortest program that computes the algorithm or pointers at http://www.cwi.nl/~paulv/kolmogorov.html woo woo! yeah! * Tril/#TUNES praises "grep Kolmogorov /pub/tunes/irc/*" * water/#tunes suggests the LANL CS archives for papers on complexity You should know that I'm talking about HUMAN measurements of complexity. I.e., something may be algorithmically complex but easy to understand. By human standards, then, it's not complex. CS has been about people for as long as i can remember jdl: this is more a mathematical measurement or formal theory. Not that we necessarily agree with it here. So, for building projects, ASM is more complex than C; C is more complex than Make. But for computing a FFT, Make is more complex than both. jdl: if you can create a new metamodel every time you get a new concept, then the models won't be complex (in size of data relative to that metamodel) 04:00pm tril: that makes sense the meta-metamodel has to relate all the metamodels together Tril: There's naturally a balance that must be maintained between complexity of models and metamodels. One assumption I'm making is that metamodels will generally be created by more advanced users, and reused by everybody. I don't think a common low-level representation can do that proprly. I prefer the persistent hyperprogramming method * binEng/#tunes looks up from his shell neither do i I have three options with the meta-metamodel: 1) Common form -:- HickServ [dave@bespin.dhs.org] has joined #tunes 1) Common representation (I mean) y0 2) Common interpretation 3) Common representation and interpretation (water's approach) jdl: sure the metamodels are created by more advanced users. But the SYSTEM ITSELF should be cybernetically assistive of the process, too. That's the reflective philosophy, any repeated activity can be programmed into the computer, including programming I'm going with number 1. Here's my rationalle: Eventually, we hope to make it easier for end users, that way. Prism resolves these issues by limiting the physical form of the models which may be included in a metamodel. The meaning of models is not limited, however; a metamodel may define a model as representing any program desired. (This approach, by the way, means that the programmer will always have to manually create transformation programs to transform models. If Prism limited the meaning of models as well as their form, it could "understand" them, and thus automatically transform any model between metamodels without using a programmer-generated transformation program. I didn't use this approach in Prism because I feel that "meaning" is an entirely human concept with infinite horizons. In my opinion, to attempt to define or capture meaning is to limit the possibilities of human expression. This is in direct conflict with my goal of allowing the programmer as much freedom as possible to create new metamodels, so Prism does not restrict the meaning of models.) the balance between metamodels and models: Make the metamodels enormously complex if need be! models should be simple as much as possible. -:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-121.tscnet.net]) uh bbl i gotta go to the copy shop :\ -:- SignOff HickServ: #TUNES (later) that's with the goal of allowing the computer to do its job Tril: That's my preferences as well. Tril: The paragraphs I just pasted in are a response to your last statement about cybernetic assistance oh, i thought you typed really fast (as well as a rationale for my meta-metamodel approach) Tril: I wish! :) Allowing the computer to process as much as possible is not contrarry to the goal of programming freedom Because you are not restricting human expression by putting some data in the computer, just because it can't understand it. You can find that paragraph at http://www.teleport.com/~sphere/documents/000e/3/meta_metamodel.html -- second paragraph a word processor document is not really understandable to a computer. but if you made a concept map instead, maybe parts of it could be automated- like indexing by concept, automatically Tril: No disagreement. -:- water [water@ppp-tnt-38.tscnet.net] has joined #tunes jdl: so why can't I create new primitives to stand side by side with bit, stream and map Tril: My point with that paragraph was that the best I can do with a meta-metamodel is provide a common representation, NOT a common interpretation. bit,stream,map is my common representation. to use in my own metamodels or models Tril: Because bit,stream,and map are supposed to be sufficient by themselves. Ah ha! You can use metamodels in metamodels as if they were primatives. See http://www.teleport.com/~sphere/documents/0010/1/index.html (Text Document metamodel) They may be sufficient, that doesnt mean someone won't want to make their own that is easier for the current problem. 04:10pm Tril: I think my last sentence answers that. i bet you didnt program that part yet jdl: why? sounds like a hack if that is true, then you do have a reflective system, which you denied in the past yes Metamodels are Prism's extension mechanism. The first-order constructs (bit,map,stream) were intentionally minimalistic to reduce the number of unchangable (errror-prone) assumptions Tril: That part doesn't need to be programmed. doesn't need to be programmed? bah! Tril: No, because metamodels incorporate pieces which are not part of the system -- natural language. However, Prism could be MADE reflective by coming up with a new kind of metamodel which could be represented in the system. No harder than s/No harder than// Tril: To make Prism reflective, you'd have to have a metamodel which could be used to define metamodels. It would have to include constructs which described how models could be interpreted -- what they meant. jdl: but that's what you get if you use metamodels as primitives within other metamodels water: I don't understand jdl: well, what do you think primitives are? jdl: they _give_ _meaning_ to a sentence using them as vocabulary i don't know how to express that in the model /meta-model world water: I didn't mean that metamodels were literally primatives, just that you could use metamodels to define new metamodels. You're not restricted to just bit,stream,map - that would be very tedious! :) But everything boils down to bit,stream,ma p eventually . boiling down to bit stream and map requires reflection for your system, i say. otherwise metamodels could not be used by metamodels to make models god damn this is confusing No, because metamodels are specifications. They're documents, not part of the Prism environment. The inclusion mechanism is just a hypertext link! oh, so their docs they're -:- SignOff smkl: #TUNES (sleeping ...) this is pointless. you keep shifting terms ok. substitute model for meta-model in my previous sentences water: I think there's a fundamental misunderstanding occuring. Look at this reference for an example of a metamodel which includes another metamodel: http://www.teleport.com/~sphere/documents/0010/1/index.html 04:20pm ok. looks like gibberish The spec consists of two parts: form, which describes what text models look like, and "meaning," which describes how to interpret a single model. big deal it's just text it doesn't do anything other than contain hyperlinks Exactly. Prism IS NOT reflective. I never said it was. well, how do you make new models? manually? original models (those which have never been created before) are created manually. that's terrible why not just call it the c-language? and call meta-models hyper-comments? I've already posted my rationale for this approach. it's not rational not to a tuneser abi, rationale? jdl: wish i knew abi, rationale is the fundamental reason for something; basis (from the American Heritage Dictionary) geez, jdl, i'm not an idiot abi forget rationale water: I forgot rationale your rationale is not reasonable my rationale is that the range of human thought and expression is infinite and cannot be captured for use by a computer. That's unreasonable? Wrong, perhaps... but unreasonable? ahh. (But I don't think it's wrong, either :) ) no, just limited in vision it is unreasonable that you suppose such a system would be useful haven't you read my paper? -:- SignOff binEng: #TUNES (Ping timeout for binEng[dialup44-1-4.swipnet.se]) water: Part of it... to be honest, I'm waiting for your next revision. The current one is hard to read. geez i didn't release it only to the tunes list because it was ready for world-wide publication i need people to chew on the ideas, and that takes work just like i chew the ideas over and over from various perspectives try some Stephen King I heard he was easy to read (inside joke see the other day conversation w/Fare about quality of products in monopolies) 04:30pm hehe water: You said you were working on a new revision; I figured I'd wait for that. Besides, I've had a LOT of discussions with you about Arrow; I feel I understand it pretty well. yes, only novelists who provide easy reads are worth reading. :) jdl: i feel you don't, because arrow addresses the very thing you seem vexed by water: Plans to address, but you haven't addressed it yet... at least not to my satisfaction. jdl: the fundamental inability of computers to fully manage human-level concepts fully?? i'd be satisfied with "a little bit more" which would be a lot :) Me too :) jdl: well, i'm sorry if my development is not yet complete enough but more than Prism :) _inability_ huh? water: I think your plan is noble and visionary. BUT I think it's going to take a long time. Which is why I took the bastardized route with Prism. no, it won't take a long time, if people start to think about it's ideas at the very least, working on arrow ideas could help with intermediate goals, like tunes or prism -:- binEng [Anders@dialup83-2-47.swipnet.se] has joined #tunes Tril: still here? hi] yes :) -:- binEng_ [Anders@dialup80-2-43.swipnet.se] has joined #tunes So, after all this, and given that Prism metamodels can use other metamodels as building blocks, do you still think bit,map,stream is a bad way of representing concepts? (For metamodels, not models) yes for data and code? especially because metamodels are informal -:- SignOff binEng: #TUNES (Ping timeout for binEng[dialup83-2-47.swipnet.se]) -:- binEng_ is now known as binEng 04:40pm Tril: yes, but particularly programs... but remember, data and code MODELS, not data and code within a running program water: informality is part of the constraints of Prism. If you finish Arrow and succeed in fully representing human concepts, I'll switch to your approach in a heartbeat :) i'm going to go work on my paper bye -:- SignOff water: #TUNES (The Tao has left the building!) good luck! well i am going now mabye follow up on email bye jim * Tril/#TUNES is away: (afk) [BX-MsgLog Off] okay, see ya -:- SignOff jdl: #TUNES (Leaving) 04:50pm -:- Niobe [chaldea@ppp-001.m4-1.osh.ican.net] has joined #tunes -:- Niobe [chaldea@ppp-001.m4-1.osh.ican.net] has left #tunes [] -:- beholder [chaldea@ppp-001.m4-1.osh.ican.net] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Read error to AlonzoTG[216-164-231-239.s493.tnt7.lnh.md.dialup.rcn.com]: Connection reset by peer) Hey all Deb rocks, thanks Tril and Fare for enligtening me ;) Does someone know what access rights are needed for files to be readable by cgi scripts? -:- AlonzoTG [Alonzo@209-122-196-203.s203.tnt5.lnh.md.dialup.rcn.com] has joined #tunes 05:00pm -:- tcn [tcn@cci-209150250089.clarityconnect.net] has joined #tunes hey people Hey TCN hi is there any topic to discuss? -:- SignOff tcn: #TUNES (Ping timeout for tcn[cci-209150250089.clarityconnect.net]) my fucking computer keeps crashing -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) Alon: Hmm... if you haven't changed anything then it might be a hardware problem -:- tcn [tcn@cci-209150250089.clarityconnect.net] has joined #tunes >>> tcn [tcn@cci-209150250089.clarityconnect.net] requested PING 931047028 372603 from #tunes hey, this server works for some reason are there some major backbones down? tcn: Could be. Were you split? no, i can't even connect to lots of sites -:- _QZ [brand@p0wer.qzx.com] has joined #tunes tcn: Could be your ISP's router doesn't have any fallback capability ;) they've got 3 uplinks i think maybe their routers are partially screwed 05:10pm Hehe.. Cisco stuff randomly craps out (yes I have stories to back it up ;) right in fact.. on my isp's "news" page.. it says their routers died 2 weeks ago That sucks. It might just be a config issue then probably a cisco thing :) anyways, I have a route to bespin That's all that matters :) tcn: Do you use Linux? yeah what distro? debian 2.1 tcn: I just installed that :) I love it... but there is too many packages to choose from... Does someone know what access rights are needed for files to be readable by cgi scripts? tcn: ... is there any easy way around that? tcn: I mean selecting them one at a time not that I know of hang on a bit.. 05:20pm ok well i'm back.. is the net still running? I imagine :) hey.. alter.net has improved -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us226.javanet.com]) hcf got hit? anyways gotta get going.. see ya -:- SignOff tcn: #TUNES (tcn has no reason) Yep... poor bugger... his net connection was so young... so innocent... WHY! DAMNIT WHY!? heheh 05:30pm -:- binEng_ [Anders@dialup46-2-29.swipnet.se] has joined #tunes -:- SignOff binEng: #TUNES (Ping timeout for binEng[dialup80-2-43.swipnet.se]) -:- binEng_ is now known as binEng -:- iepos [iepos@d15.t1-7.tecinfo.com] has joined #TUNES hi hi hey yeah... So anyone have anything to talk about or are we gonna lurk forever ;) 05:50pm -:- HickServ [dave@bespin.dhs.org] has joined #tunes hi hi i just rented 1984 anyone here seen it? Nope well i hear a lot about it you know what was a weird movie thx1138 weird "It's practically impossible to teach good programming style to students who have had prior experience with BASIC" hahaha heh heh ": as pontential programmers are mentally mudilated by using it" 06:00pm er "mudilated beyond hope of regeneration" I can agree with that... I'm a victom... i misquoted it * HickServ/#tunes never learned basic nor does he wish to my first programming language was C++ :\ i am sad my first language was not assembler ( i believe that is the best first language) then C and then LISP BASIC: "As it, is it ruins thousands of potential wizards a year" :( i feel sorry for you beholder * HickServ/#tunes pats beholder on the back sorry buddy That's fine... 10 "Print Beholder will get better" 20 goto 10..... My god. ;( (display "Get a hold of yourself man") printf("Ahh... that's better\n"); print 'what about all ' 'features ' 'in ' 'Python, then?' ':)' heh bineng, do you realize you have 7 terminals open to tunes.org ? for i in range(100): print "Yes... this is even better" heheh i who'd bespin and saw how many "bineng's" there were at most i have 2 iepos: I've been disconnected two times... and I had some open each time anyone here familier with "black magic" programming? hick: Yep you take a magic wand and pull a program out of your hat... hehe :( i read it in the jargon file they also talk about vodoo programming 06:10pm I MUST KNOW IT I bet fare knows it he's well in sync with hackerdom brb abi, fare is probably asleep right now... ...but fare is sometimes looking at another screen and interruptible with a beep (/msg Fare ^G) or at http://www.tunes.org/~fare/ or connected through a crappy ISP (cybercable.fr) or an arrogant ass... abi, fare is also probably asleep right now... i'm sorry, but that's too long, iepos lame abi: bad girl! spank me beholder hehehe :) * beholder/#tunes is starting to believe the "small things amuse small minds" axiom ;) heh Isn't it the contrary? that never gets old :) -:- _BC [bmcbrine@pine.Alberni.Net] has joined #tunes <_BC> HI hoy, _BC _BC: do you know anything about vodoo programming or "black magic"? bin: I've never heard the contrary "big things amuse big minds".... so I don't think so ;) beholder: That's a double negation :) double negative is more commonly said <_BC> black magic?? hehe, nope! voodoo; well not .., uh no. BUT - I have also looked into some 3D engines, if that is anything? no i found it in the jargon file but it's rather obscure "it works but nobody knows how" <_BC> heheh probably programming the voodoo is the blackmagic part. <_BC> open gl is nice. 06:20pm my brother knows a whole shit load about that crap open gl this opengl that gets annoying <_BC> why? it's a good standard.. (well as least it looks like it to me) i don't know (im not one for great graphics) <_BC> ooh. i like gfx. <_BC> :) * beholder/#tunes thinks besides basic that has warped his mind, the South Park movie did a good job of that too... <_BC> you've seen the movie? bc: yep <_BC> neat. was it as good/bad as the tv show? bc: I'd quote some dialoge, but I think I'd get kicked... never heard more "interesting" language in 2 hours in my whole life ;) bc: If you like SP, it was nothing less than perfect <_BC> how many times did kenny get swamped/killed? <_BC> neat. <_BC> i'll see it, but maybe wait till video. bc: Only once <_BC> have to go for now. cya'z -:- SignOff _BC: #TUNES (I toss myself out of an airlock and try to sing....) oh my *yawn* 06:30pm hmmm... tunes.org is idle... maybe time to start another life simulation? better ask tril this time though. tril, are you around? not a RL, eh? "RL"? real life :) oh... oh it is real Life... what is a life simulation? Conway's game of life. it's played on a grid of cells abbrev otherwise known as Rocket Launcher at one time, each is either "on" or "off" (alive or dead) every generation a change takes place... in conways life the rules are this: if a cell is dead, it becomes alive if it has 3 live neighbors if a cell is alive, it stays alive if it has 2 or 3 live neighbors over time strange life forms evolve oh like vants including "gliders", a kind of pattern that moves no... it's way neater than ants AL is cool and there are still-lifes that do nothing. and oscillators that blink where do you simulate it, bespin? xlife is neat easy simulator... are you using unix, hickserv? only on bespin and in my room well, then you can simulate like on bespin and in your room i mean "life" not "like" you might want to try "xlife" or if you don't like xwindows, you can try my own weird simulator which uses svgalib well i don't have xwindows in my room and i can't use it on bespin heh heh... well do you want to try my weird simulator? will it hurt me in any way ;) hopefully not. but you will have to su to root to use it :-) 06:40pm don't worry i don't use the raw keyboard thing though so you can always press ^C can't i just run it in my shell? sure, but you can't see any pretty colorful pixels that's ok i haven't made a text version i meant: you can't see anything at all except maybe the generation number go up oh it's okay though to do it as root. it won't hurt anything i'll send you the source. you can check and make sure there aren't any viruses :-) well that's ok can i run it on bespin yeah, but no colorful pixels of course. just the generation number increasing. i don't wanna have to putit on disk and then transfer it (it doesn't have a modem) well... unless increasing generation number interests you, that's what you're going to have to do right? oh :-) well theni don't think i'll be interested i don't wanna bother with this crap right now anyway mabye later all right bbiafm -:- SignOff iepos: #TUNES (Leaving) LISP - Lots of Iiitating Superfluous Parentheses hehehe abi: LISP? well, LISP is simply perfect uh abi, no lisp is closest to perfect okay, HickServ. 06:50pm what are you abi? what abi? i am the little girl borg with a big brain or #tunes' personal little whore or gaklosmontic or flurivostuginuous or xyvarestoplik or an Artificial Idiot or braindead or a crack addict infoslut or stubborn abi: bad girl spank me beholder hehehe I see no other reason to have a bot than that :) bye all -:- beholder [chaldea@ppp-001.m4-1.osh.ican.net] has left #tunes [] -:- SignOff HickServ: #TUNES (HickServ has no reason) 07:00pm -:- iepos [iepos@d5.t1-6.tecinfo.com] has joined #TUNES -:- _QZx [brand@defiant.qzx.com] has joined #tunes -:- SignOff _QZx: #TUNES () -:- AlonzoTG [Alonzo@209-122-204-22.s530.tnt3.lnh.md.dialup.rcn.com] has joined #tunes -:- shadow [Nico@63.65.250.174] has joined #tunes hi, shadow... hi haven't seen you around here much... what brings you to TUNES? 07:50pm this my second time oh... well i'm not on irc all the time... so i must have missed you i guess... or forgotten about you :-) I'm off. bye, bineng bye bye -:- SignOff binEng: #TUNES ( <k!14>) 08:00pm ow om greetings alonzo 08:10pm topick? its spelled 'topic', AlonzoTG hehhehehehehehe heh heh abi, topic? iepos: no idea -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) 08:20pm -:- shadow [Nico@63.65.250.174] has left #tunes [] -:- SignOff iepos: #TUNES (Leaving) -:- Kaufmann [Kaufmann@200.224.105.185] has joined #Tunes Anyone feel like engaging in pointless, inflammable discussion about religion? <_QZ> sure <_QZ> i think yer religion is stupid HA! I /have/ no religion! Gotcha! :) <_QZ> atheist? We have a lot of fun like this at DALNet #atheism <_QZ> if yer atheist then yer ok Well, it depends on your definition of "atheist". 09:40pm <_QZ> god is a figment of morons immagination <_QZ> there is only one definition of atheism <_QZ> u know that god dont exist <_QZ> agnostic is where yer in denial and that is NOT atheism Yes, that obviously seems to be the case, but it's always necessary to allow for the Bland Radical Solipsistic Hypothesis [sic]. <_QZ> eh? agnostic means you don't know for sure. I guess the best words for me are "agnostic rational skeptic". <_QZ> well im a pure atheist The Bland Radical Solipsistic Hypothesis [sic], as put forward, states that "there is always a possibility, however small, that everything your brain has ever accepted as input is an illusion, and that in the /real/ world from which you've been sheltered, things are completely different". Which in this case means that there actually /is/ a God. <_QZ> ya ok whatever Exactly. "Ya, OK, whatever" - but nonetheless that hypothesis /must/ be acknowledged if only because it seems to be impossible to prove it false. Therefore, You Can Never Be 100% Certain. <_QZ> and we might all live in his ass, and going to hell is when he shits u out "pure" atheist? Why? <_QZ> im an american and go by the innocent until proven guilty, or not real until proven to be real 09:50pm Oh. So you are not "pure", because you allow the possibility that if it's proven to be real, then you will accept it as real. A "pure" atheist is one that just says "There Is No God, no matter what." <_QZ> no that is an agnostic What? Nope, you got your defs wrong. Agnostic literally means "doesn't know". <_QZ> and if yer hypothesis is true and we are in some type of holographic world then what makes whoever controls it "a god"? <_QZ> what if i am controlling yer little world <_QZ> u can play "what if" all day but it aint gonna get u anywhere It's not "my" hypothesis, nor do I believe in it. I'm just saying that the nature of the hypothesis itself makes it so that the possiblility of it to be true /must/ be acknowledged. "u can play "what if"..." << Yes, that's true in a practical level, but we're precisely discussing absolutes here. <_QZ> i believe in facts and the fact is that a supernatural being does not exist <_QZ> and no one can prove that one does <_QZ> 'god' is in yer head and can be whatever u want it to be because the human imagination has no limits but my whole point is exactly that, even if it seems not to exist by every possible standard, no one can prove that it /doesn't/! And therefore any self-consistent model must account for the possibility of it being true! You see what I'm saying?!? <_QZ> if u believe in a good 'god' then u will be a good person, if not then.... <_QZ> ya yer using the moron model of thinking What I personally believe has nothing to do with what may or not be actually true. That said, I think that anyone who thinks in absolutes is in serious need of rethinking himself. (... or something to that effect but which actually sounds good.) <_QZ> my thoughts are not absolute, they are fact based <_QZ> i am open minded to proven facts 10:00pm Yes, and I'm trying to explain to you that no fact is absolute. <_QZ> ya for all we know u might say yer straight but u could really he gay Yes, I agree that in a practical level, it's rather pointless to think about this kind of remote possibility. I also agree that most people who are religious are religious because they're idiots or/and too lazy to think about it for a second, not because they think that the BRSH is true. But when you discuss what we're discussing specifically, then we have to talk about this kind of thing. <_QZ> nice, im watching some outdoor show and they are wearing shorts and huge ass winter coats Yeah, that makes sense :) 10:10pm -:- SignOff _QZ: #TUNES (Ping timeout for _QZ[p0wer.qzx.com]) -:- _QZ [brand@p0wer.qzx.com] has joined #tunes ... 10:30pm -:- _BC [bmcbrine@pine.Alberni.Net] has joined #tunes -:- SignOff _BC: #TUNES (changing servers) -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[200.224.105.185]) [msg(TUNES)] newlog 1999.0704 IRC log ended Sun Jul 4 00:00:00 1999