You can subscribe to this list here.
2014 
_{Jan}

_{Feb}
(232) 
_{Mar}
(323) 
_{Apr}
(383) 
_{May}
(359) 
_{Jun}
(435) 
_{Jul}
(252) 
_{Aug}
(172) 
_{Sep}
(265) 
_{Oct}
(263) 
_{Nov}
(350) 
_{Dec}
(359) 

2015 
_{Jan}
(267) 
_{Feb}
(220) 
_{Mar}
(311) 
_{Apr}
(269) 
_{May}
(388) 
_{Jun}
(403) 
_{Jul}
(172) 
_{Aug}
(399) 
_{Sep}
(364) 
_{Oct}
(269) 
_{Nov}
(357) 
_{Dec}
(468) 
2016 
_{Jan}
(618) 
_{Feb}
(592) 
_{Mar}
(625) 
_{Apr}
(516) 
_{May}
(375) 
_{Jun}
(155) 
_{Jul}
(346) 
_{Aug}
(262) 
_{Sep}
(346) 
_{Oct}
(291) 
_{Nov}
(333) 
_{Dec}
(335) 
2017 
_{Jan}
(436) 
_{Feb}
(460) 
_{Mar}
(370) 
_{Apr}
(189) 
_{May}
(252) 
_{Jun}
(272) 
_{Jul}
(286) 
_{Aug}
(293) 
_{Sep}
(303) 
_{Oct}
(331) 
_{Nov}
(248) 
_{Dec}

From: Dimiter Prodanov <dimiterpp@gm...>  20171114 09:36:23

Hello, I will strongly support a tool integrated in GitHub or bitbcuket. I used to have projects on SF but they do not accept project deletions so we may end with a lot of "ghost" projects, which will be no better than the current situation. best regards, Dimiter 
From: René Kälin <renekaelin@gm...>  20171114 07:30:21

Hi Stavros It doesn't seem that I'm authorized to open a bugticket on https://sourceforge.net/p/maxima/bugs/. Is there another possibility or can you do it yourself? Thanks. On 13.11.2017 23:09, Stavros Macrakis (Σταῦρος Μακράκης) wrote: > Please report this (serious!) bug at sourceforge. > > Other, related, cases work correctly: > > assume(q<0,r<0)$ > > sign(q*r) => pos > sign(a*r) => pz > sign(a*2) => pz > sign(a*q*r) => nz > BUT > > sign(a*b*q) => pz SHOULD BE nz > > Thanks! > > On Mon, Nov 13, 2017 at 3:49 PM, René Kälin <renekaelin@... > <mailto:renekaelin@...>> wrote: > > Hello > > With Maxima 5.38.1 I get the following > > (%i1) assume(a<=0,b<=0);sign(a*b); > (%o1) [a <= 0, b <= 0] > (%o2) nz > > I'd expect pz... > > > > >  > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Maximadiscuss mailing list > Maximadiscuss@... > <mailto:Maximadiscuss@...> > https://lists.sourceforge.net/lists/listinfo/maximadiscuss > <https://lists.sourceforge.net/lists/listinfo/maximadiscuss>; > > 
From: Kris Katterjohn <katterjohn@gm...>  20171114 00:49:21

On Mon, Nov 13, 2017 at 01:01:32PM 0800, Robert Dodier wrote: > More importantly, keeping all of the code in one place, which is > administered by persons other than the package authors, means (1) > there is only one place to find the "official" copy of the package, > and (2) people other than the package authors can make changes > directly, without going through the authors. > > These are both very important considerations to me. (1) means there's > no confusion about what's the "right" version of a package. (2) means > that we don't need to bring authors into the development cycle if > someone else wants to contribute documentation, bug fixes, test cases, > or whatever, and we don't need to worry about authors going missing in > action and thereby orphaning their packages. > > (2) implies that the person who initiates a project doesn't have veto > power over anyone else's changes. I believe it's better that way. If > someone wants veto power, they can create their own independent > project. I appreciate you working on this, and I think it would be good to have a Maxima Forge of some sort. After reading your paragraphs above, I decided to think about what it would be like from pointofview of an outsider wanting to contribute a package to Maxima Forge. Here's the first story that played out in my head: === Say I write a package Foo. I spend time on it, get it cleaned up and then I decide to contribute it to Maxima Forge. It gets accepted and I'm happy to be a Maxima contributor. Oops, later on I find that Foo has some bugs; so I fix those up, submit some PRs and they get accepted. The process seems to be working. Now Foo is awesome and all is right with the world. Maybe it turns out that Foo is a fairly popular package. I'm proud of myself and I mention on my website and CV (or wherever) that I'm the author of Foo. Cool. Then, possibly quite a bit of time later, someone else submits a PR to "enhance" Foo on Maxima Forge without my knowledge. Then whoever is going through PRs at the time thinks "Great! I'll accept this so Foo is even better!" ...but, nope, it's not better. As the author, it's obvious that the PR shouldn't have been accepted. Would it have been obvious to the dev who accepted the PR if they reviewed it? Maybe, maybe not. So now the new, buggy Foo is sitting out on Maxima Forge. Maybe someone catches the problem soon, maybe it stays buggy for some time. Now I've been the "author" of a buggy package without even realizing it before. Even if I wasn't mentioning elsewhere that I'm the author of the package, it's built in to the directory structure! People would have to dig around the commit logs to see that it was someone else who was mucking around in my package that caused the bug. If I'm the one who notices the bug, then I'm very annoyed that I have to then spend time to fix what shouldn't have gotten broken to begin with. Either way, I'm upset that the Maxima Forge devs just accept PRs from random people without review. I make sure that my future packages Bar and Baz are kept under my control because I don't want to deal with this again in the future. ...which, of course, defeats the purpose of having Maxima Forge. === I'm not saying that the way it's currently done for the share packages is any better. I'm also not saying any other this is unique to Maxima, but I just wonder what other people (outside the project) would think about this when they're thinking of contributing. What do authors have to do to be kept in the loop without being flooded with stuff about other packages? What (and how much) do they have to do in order to prevent a PR from being accepted? If the author strongly disagrees with the change, do we still accept the PR from randomguy2017 just because it's there? I also don't think any sort of changelog would help catch or notice things after the fact either. Right now Maxima's changelogs don't even mention some bug fixes or enhancements, while they do mention things that I seriously doubt any user will ever care about. Again, I'm glad you're thinking about and working on this. I'm not trying to discourage or stop you, but I do think that doing it this way may be offputting to some (a lot of?) people. Then again, I don't have a better plan and I'm sure other ideas would be offputting to some people as well. Hopefully my rambling makes a little bit of sense. Please let me know if I misunderstood anything. Cheers, Kris Katterjohn 
From: Stavros Macrakis (Σταῦρος Μακράκης) <macrakis@al...>  20171113 22:10:05

Please report this (serious!) bug at sourceforge. Other, related, cases work correctly: assume(q<0,r<0)$ sign(q*r) => pos sign(a*r) => pz sign(a*2) => pz sign(a*q*r) => nz BUT sign(a*b*q) => pz SHOULD BE nz Thanks! On Mon, Nov 13, 2017 at 3:49 PM, René Kälin <renekaelin@...> wrote: > Hello > > With Maxima 5.38.1 I get the following > > (%i1) assume(a<=0,b<=0);sign(a*b); > (%o1) [a <= 0, b <= 0] > (%o2) nz > > I'd expect pz... > > > > >  >  > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Maximadiscuss mailing list > Maximadiscuss@... > https://lists.sourceforge.net/lists/listinfo/maximadiscuss > 
From: Jaime Villate <villate@fe...>  20171113 22:08:26

On 13112017 19:58, José Antonio Delgado Trujillo wrote: > I’ve installed Maxima 5.41.0 in my Mac. > I had to remove the param com.apple.quarantine > (https://sourceforge.net/p/maxima/mailman/message/35910588/) > > it seems works all OK except when i use Xmaxima and i have to write > the sign ^ (elevate). > The Xmaxima suddenly close and i see a error report. > ... > It seems that the code of ^ is out of bound and i don’t know why. > > Surprising if i write ^ in the other Launchers (RMaxima or wxMaxima) > it works fine. > > Perhaps it os something about Wish (what is that? — An application > under the Xmaxima runs? ) > HI, that seems like a tcltk bug in the Mac. Tcl/Tk8.5 is a bit old anyway, so my advice would be to first try to upgrade to Tcl/Tk8.6 rather than trying to find the source of the error. Wish is the interactive shell for Tk. There is also an interactive shell for Tcl, called tclsh. You should be able to launc any of those shells directly, but as I said, it would be better if you first upgrade them to 8.6. Best regards, Jaime 
From: Raymond Toy <toy.raymond@gm...>  20171113 21:28:51

>>>>> "Robert" == Robert Dodier <robert.dodier@...> writes: Robert> On Mon, Nov 13, 2017 at 8:06 AM, Raymond Toy <toy.raymond@...> wrote: >> I don't really see the point of a PR. We don't even review core >> maxima changes, so having someone actually approve a PR for something >> not core seems unnecessary. And if the approval is basically >> automatic, why bother? Just let everyone commit however and whenever >> they want. The effect is almost the same. Robert> Well, using PR's will give us a very slight review process  I am Robert> thinking, for example, that I would give a cursory review for changes Robert> from persons with whom I'm unacquainted. My limited thinking on this is that we update the subrepo version at the time and version the author suggests. No reviews, no blocking of PRs or anything. It is frustrating if you have lots of changes coming and you have to wait for PR requests to be reviewed, even if it's just a rubberstamp approval. Robert> More importantly, keeping all of the code in one place, which is Robert> administered by persons other than the package authors, means (1) Robert> there is only one place to find the "official" copy of the package, Robert> and (2) people other than the package authors can make changes Robert> directly, without going through the authors. Robert> These are both very important considerations to me. (1) means there's Robert> no confusion about what's the "right" version of a package. (2) means It's open source. There's never really the "right" version; there's just a version that happens to be included with Maxima Forge. Some authors may be very protective of their code; they make the code public but want control over the changes. (This has already happened before in Maxima share code.) Robert> that we don't need to bring authors into the development cycle if Robert> someone else wants to contribute documentation, bug fixes, test cases, Robert> or whatever, and we don't need to worry about authors going missing in Robert> action and thereby orphaning their packages. But you do get to worry if the author decides not to publish as part of Maxima Forge. Then what? Delete the code and all related history? (This has heppened to me on a different project. But it was a legitimate request from the author, so I deleted the code. But it's still in the history.) Who actually owns the code now? With the current share directories, I guess we're acknowledging they're not part of maxima core and that the authors have given permission to include this particular version. I think we are quite sloppy about this for the share directories. Robert> (2) implies that the person who initiates a project doesn't have veto Robert> power over anyone else's changes. I believe it's better that way. If Robert> someone wants veto power, they can create their own independent Robert> project. Not if it's all in Maxima Forge. They get to clone everything. Anyway, I'm not planning on blocking any progress. Feel free to proceed however you see fit. I just wanted to make some suggestions on disentangling ownership and work from Maxima Forge and Maxima. My ideas may end up being even more work; I just don't know.  Ray 
From: Robert Dodier <robert.dodier@gm...>  20171113 21:01:40

On Mon, Nov 13, 2017 at 8:06 AM, Raymond Toy <toy.raymond@...> wrote: > I don't really see the point of a PR. We don't even review core > maxima changes, so having someone actually approve a PR for something > not core seems unnecessary. And if the approval is basically > automatic, why bother? Just let everyone commit however and whenever > they want. The effect is almost the same. Well, using PR's will give us a very slight review process  I am thinking, for example, that I would give a cursory review for changes from persons with whom I'm unacquainted. More importantly, keeping all of the code in one place, which is administered by persons other than the package authors, means (1) there is only one place to find the "official" copy of the package, and (2) people other than the package authors can make changes directly, without going through the authors. These are both very important considerations to me. (1) means there's no confusion about what's the "right" version of a package. (2) means that we don't need to bring authors into the development cycle if someone else wants to contribute documentation, bug fixes, test cases, or whatever, and we don't need to worry about authors going missing in action and thereby orphaning their packages. (2) implies that the person who initiates a project doesn't have veto power over anyone else's changes. I believe it's better that way. If someone wants veto power, they can create their own independent project. best Robert Dodier 
From: René Kälin <renekaelin@gm...>  20171113 20:49:21

Hello With Maxima 5.38.1 I get the following (%i1) assume(a<=0,b<=0);sign(a*b); (%o1) [a <= 0, b <= 0] (%o2) nz I'd expect pz... 
From: José Antonio Delgado Trujillo <jadtmat@gm...>  20171113 19:58:35

I’ve installed Maxima 5.41.0 in my Mac. I had to remove the param com.apple.quarantine (https://sourceforge.net/p/maxima/mailman/message/35910588/ <https://sourceforge.net/p/maxima/mailman/message/35910588/>;) it seems works all OK except when i use Xmaxima and i have to write the sign ^ (elevate). The Xmaxima suddenly close and i see a error report. —— Process: Wish [1006] Path: /usr/bin/../../System/Library/Frameworks/Tk.framework/Versions/8.5/Resources/Wish.app/Contents/MacOS/Wish Identifier: com.tcltk.wish Version: 8.5.9 (8.5.9) Build Info: tcl118000000000000~199 Code Type: X8664 (Native) Parent Process: ??? [1004] Responsible: Wish [1006] User ID: 501 Date/Time: 20171113 19:47:47.499 +0000 OS Version: Mac OS X 10.13.1 (17B48) Report Version: 12 Anonymous UUID: 3509BB9796A6B3AB2C064A16004AEB6A Time Awake Since Boot: 9300 seconds System Integrity Protection: enabled Crashed Thread: 0 Dispatch queue: com.apple.mainthread Exception Type: EXC_CRASH (SIGABRT) Exception Codes: 0x0000000000000000, 0x0000000000000000 Exception Note: EXC_CORPSE_NOTIFY Application Specific Information: *** Terminating app due to uncaught exception 'NSRangeException', reason: '[__NSCFConstantString characterAtIndex:]: Range or index out of bounds' abort() called terminating with uncaught exception of type NSException ——— It seems that the code of ^ is out of bound and i don’t know why. Surprising if i write ^ in the other Launchers (RMaxima or wxMaxima) it works fine. Perhaps it os something about Wish (what is that? — An application under the Xmaxima runs? ) Un saludo, José A. 
From: Raymond Toy <toy.raymond@gm...>  20171113 16:06:53

>>>>> "Robert" == Robert Dodier <robert.dodier@...> writes: Robert> On 20171111, Raymond Toy <toy.raymond@...> wrote: >> Can we choose something other than sourceforge? Robert> Well, SF has the advantage that we already have Maxima there. We don't Robert> need to make up new user ids, and I feel like there's a certain mental Robert> continuity in being "not too far away". >> I mean github, bitbucket, and even gitlab all support continuous >> integration and delivery, if desired. Not so with sourceforge. >> Then the package authors can easily enable continuous >> integration/delivery for their package if desired. Robert> Github etc are all good, and if we were to pick a new host ab initio I Robert> would choose Github. But continuous integration doesn't seem like a Robert> strong enough feature to overcome continuity with SF. I'm ok with having Maxima Forge on SF. But really, SF has fallen way behind. No CI/CD, commits don't close bugs when fixes are commited, no pull request system, no review tool. (Not that the latter really matter since no ever reviews anything.) >> Hmm. Or maybe just leave Forge on sourceforge, and have individual git >> packages to come for wherever. That might be easier for Maxima devs. [snip] Robert> I looked into git submodules and read a few blog posts about it, but I Robert> couldn't figure out how that works. It does seem more complicated than Robert> having a single repo. If I can't figure it out, I think it's an Robert> unreasonable burden to put on wouldbe authors (not to mention, on Robert> myself). I'll poke around and see what it's like. I have a project now where it would be nice if it imported another project into its sources to maintain a local copy. Robert> Approving a PR is (I believe) also very simple for MaximaForge admins. I don't really see the point of a PR. We don't even review core maxima changes, so having someone actually approve a PR for something not core seems unnecessary. And if the approval is basically automatic, why bother? Just let everyone commit however and whenever they want. The effect is almost the same. Sorry for being cynical on this. But I don't think I'm being overly pessimistic. No one reviews maxima commits. I don't think PRs are any different, except someone will probably just click the "yes" button without looking at the changes.  Ray 
From: Daniel Volinski <danielvolinski@ya...>  20171113 12:13:34

Hi Jaime, This is perfect, muito obrigado! Daniel En lunes, 13 de noviembre de 2017 01:15:47 IST, Jaime Villate <villate@...> escribió: On 11112017 14:29, Daniel Volinski via Maximadiscuss wrote: > Given a numerical calculation like the result of rk, rkf45 or Eulix, I > would like to obtain the numerical integral function of the result. > ... > rksol:rkf45([x*(1y),y*(1x)],[x,y],[2,2],[t,0,10,0.01], > absolute_tolerance=5E8,report=true)$ > ... > Is there a Maxima package that solves this problem? I'm sending in an attachment a sample program to solve that problem. load("rk_int.mac"); rksol: rk_int([x*(1y),y*(1x)],[x,y],[2,2],[t,0,10,0.01])$ plot2d([[discrete,map(lambda([u],part(u,[1,2])),rksol)], [discrete,map(lambda([u],part(u,[1,3])),rksol)]], [style,[lines,2]])$ Regards, Jaime 
From: Jaime Villate <villate@fe...>  20171112 23:15:55

On 11112017 14:29, Daniel Volinski via Maximadiscuss wrote: > Given a numerical calculation like the result of rk, rkf45 or Eulix, I > would like to obtain the numerical integral function of the result. > ... > rksol:rkf45([x*(1y),y*(1x)],[x,y],[2,2],[t,0,10,0.01], > absolute_tolerance=5E8,report=true)$ > ... > Is there a Maxima package that solves this problem? I'm sending in an attachment a sample program to solve that problem. load("rk_int.mac"); rksol: rk_int([x*(1y),y*(1x)],[x,y],[2,2],[t,0,10,0.01])$ plot2d([[discrete,map(lambda([u],part(u,[1,2])),rksol)], [discrete,map(lambda([u],part(u,[1,3])),rksol)]], [style,[lines,2]])$ Regards, Jaime 
From: Gunter Königsmann <gunter@pe...>  20171112 14:59:04

There was a bug in wxMaxima that caused this. If you use a newer version of the program this problem should no more occur. Kind regards, Gunter. 
From: Jaime Villate <villate@fe...>  20171112 12:02:50

On 11112017 14:29, Daniel Volinski via Maximadiscuss wrote: > Hi All, > > Given a numerical calculation like the result of rk, rkf45 or Eulix, I > would like to obtain the numerical integral function of the result. > ... > Did anyone solve this problem before? Yes, my students have to do that all the time. > Is there a Maxima package that solves this problem? Not that I know, but you just need a simple sum() command to solve the problem. > Any suggestions? Use Simpson's rule. For example, consider the equation 'diff(y,x,2) = y. The numerical solution in the interval 0<= x <= 3, can be found with: (%i1) r: rk([z,y],[y,z],[0,1],[x,0,3,0.1])$ Then Simpson's method for the integral of y(x) from 0 to 3 goes like this: (%i2) n:length(r); (%o2) 31 (%i3) A: 0.1*sum(r[2*i1][2]+4*r[2*i][2]+r[2*i+1][2], i, 1, floor((n1)/2))/3; (%o3) 1.989994688553148 Let's compare it with the exact result: (%i4) float( integrate(sin(x),x,0,3)); (%o4) 1.989992496600445 If n were an even number, you would have to add 0.1*(r[n1][2] + r[n][2])/2 But you can always choose an increment of the independent variable (0.1 in this case), such that the length of the integration interval divided by that increment is an even number, so the length of the result will be even. I think it's not worth it using more complicated interpolation of the points, since those points already have a numerical error. Improving the value of those points can be done by decreasing the independent variable increments, which will also improve the result of Simpson's rule. I guess rk could give a much better estimate of the integral, using the values of the derivatives that it computes at each point in the integration interval. I'll think about it. Regards, Jaime 
From: Gunter Königsmann <gunter@pe...>  20171112 08:18:46

Seems like posting your message worked. Kind regards, Gunter. 
From: Edward Montague <quintin9g@gm...>  20171112 04:36:07

Thanks , I've now bookmarked that page . I had a massive clean out of my email a wee while ago , hence I'm somewhat disorganized. On 12 November 2017 at 17:00, Kris Katterjohn <katterjohn@...> wrote: > On Sun, Nov 12, 2017 at 04:54:01PM +1300, Edward Montague wrote: > > Let me know if you receive this . > > Got it. If you want to check to see if a message went through, you can > check the mailing list archive for it after you send it: > > https://sourceforge.net/p/maxima/mailman/maximadiscuss/ > > It seems to update every few minutes. This way you can just send the > email you want and check that it went through without having to ask for > replies and/or send test messages first. (I think I gave you this > advice when we went through this in the past.) > > Kris Katterjohn > 
From: Kris Katterjohn <katterjohn@gm...>  20171112 04:00:37

On Sun, Nov 12, 2017 at 04:54:01PM +1300, Edward Montague wrote: > Let me know if you receive this . Got it. If you want to check to see if a message went through, you can check the mailing list archive for it after you send it: https://sourceforge.net/p/maxima/mailman/maximadiscuss/ It seems to update every few minutes. This way you can just send the email you want and check that it went through without having to ask for replies and/or send test messages first. (I think I gave you this advice when we went through this in the past.) Kris Katterjohn 
From: Edward Montague <quintin9g@gm...>  20171112 03:54:11

Let me know if you receive this . This is my interpretation of a scholarly article on Variational Iteration Methods , as applied to certain non linear differential equations . The paper can be found at : https://scholar.google.co.nz/scholar?q=Application+of+He%27s+Variational+Iteration+++Method+++++++++for+Solving+Seventh+Order+SawadaKotera+Equations+.&hl=en&as_sdt=0&as_vis=1&oi=scholart&sa=X&ved=0ahUKEwjp9eiJ8bXXAhUI5yYKHdtoA7gQgQMIJDAA If you have any difficulty with that link , do a google search for : Application of He's Variational Iteration Method for Solving Seventh Order SawadaKotera Equations and click through the first item , scholarly articles , the appropriate pdf should then be apparent as the first item in the list . Click on [pdf] to view and subsequently download . Failing that I maybe able to use my google drive as a repository . This is code :  /* VIMSSKb.mac Copyright (C) 2009 gedcda72 sciwise@... */ ""$ " VIM for Seventh Order SawadaKotera Equations ."$ ""$ /* Results from a paper by : Hossein Jafari , Allahbakhsh Yazdani , Javad Vahidi , Davood D , Ganji Application of He's Variational Iteration Method for Solving Seventh Order SawadaKotera Equations . 2008 I have no political , religious , economic or academic affiliation with the authors . The second iteration expression isn't very clear in the original pdf format ; a ' + ' character was found to be necessary before the (1/135) factor . */ ""$ " Initiating function . "$ ""$ u0:(4/3)*(k*k)*(23*(tanh(k*x)^2))$ es0:u0$ ""$ " First iteration ; from pdf ."$ ""$ eq:(1/9)*(k^8)*(sech(k*x)^2)*t*(2176896*cosh(2*k*x))$ es1:es0 + eq$ ""$ " Second iteration ; from pdf ."$ ""$ eq:(1/27)*(sech(k*x)^8)*(k^14)*(t^2)*((63285766566144*cosh(2*k*x)+ 1077248*cosh(4*k*x)+24832*cosh(6*k*x)  12544*cosh(8*k*x))+ (1/3)*(k^6)*t*(2544812032  2746548224*cosh(2*k*x) + 305070080*cosh(4*k*x) + 41746432*cosh(6*k*x)  5619712*cosh(8*k*x) )+ (1/9)*(k^12)*(t^2)*( 50980192256  23855104000*cosh(2*k*x)  55593402368*cosh(4*k*x) + 17983078400*cosh(6*k*x)  1258815488*cosh(8*k*x) ) + (1/135)*(k^18)*(t^3)*(238459436400640 + 291359575506944*cosh(2*k*x)  82083401695232*cosh(4*k*x) + 10956730007552*cosh(6*k*x)  563949338624*cosh(8*k*x) )) $ es2:es1 + eq$ ""$ "  "$ ""$ " Seventh order equation. "$ ""$ eq3:'diff(u(x,s,k),x,7)+21*u(x,s,k)*('diff(u(x,s,k),x,5))+42*('diff(u(x,s,k),x,1))*('diff(u(x,s,k),x,4))+63*('diff(u(x,s,k),x,2))*('diff(u(x,s,k),x,3))+126*u(x,s,k)^2*('diff(u(x,s,k),x,3))+378*u(x,s,k)*('diff(u(x,s,k),x,1))*('diff(u(x,s,k),x,2))+63*('diff(u(x,s,k),x,1))^3+252*u(x,s,k)^3*('diff(u(x,s,k),x,1))+'diff(u(x,s,k),s,1)$ eq3; eq3:'diff(u(x,s,k),x,7)+21*u(x,s,k)*('diff(u(x,s,k),x,5))+42*('diff(u(x,s,k),x,1))*('diff(u(x,s,k),x,4))+63*('diff(u(x,s,k),x,2))*('diff(u(x,s,k),x,3))+126*u(x,s,k)^2*('diff(u(x,s,k),x,3))+378*u(x,s,k)*('diff(u(x,s,k),x,1))*('diff(u(x,s,k),x,2))+63*('diff(u(x,s,k),x,1))^3+252*u(x,s,k)^3*('diff(u(x,s,k),x,1))+'diff(u(x,s,k),s,1)$ /* integrating thru w.r.t x . */ ""$ " Seventh order equation , neglecting outside derivative of x "$ ""$ eq3:('diff(u(x,t,k),x,6)+21*(u(x,t,k)*('diff(u(x,t,k),x,4))+('diff(u(x,t,k),x,1))*('diff(u(x,t,k),x,3))+('diff(u(x,t,k),x,2))^2)+63*(2*u(x,t,k)^2*('diff(u(x,t,k),x,2))+u(x,t,k)*('diff(u(x,t,k),x,1))^2)+63*u(x,t,k)^4)+'diff(u(x,t,k),t,1)$ ""$ "  "$ ""$ " Non  Linear Integrator , using eq3 "$ ""$ NLInt(u0):=block(local [u,eq3,t,x,s,us], define(u(x,t,k),u0), us:subst(s,t,u0), define(u(x,s,k),us), eq3:('diff(u(x,s,k),x,6)+21*(u(x,s,k)*('diff(u(x,s,k),x,4))+('diff(u(x,s,k),x,1))*('diff(u(x,s,k),x,3))+('diff(u(x,s,k),x,2))^2)+63*(2*u(x,s,k)^2*('diff(u(x,s,k),x,2))+u(x,s,k)*('diff(u(x,s,k),x,1))^2)+63*u(x,s,k)^4) + 'diff(u(x,s,k),s,1), eq3:ev(eq3,nouns), assume_pos : true, eq3 : u(x,t,k)  integrate(eq3,s,0,t), eq3:ev(eq3,nouns), eq3:expand(trigreduce(trigsimp(ratsimp(eq3)))), eq3 )$ ""$ "  "$ ""$ kill(u0)$ u0(x,t,k):=(4/3)*(k*k)*(23*(tanh(k*x)^2))$ ""$ " First Iteration "$ ""$ eq3:u0(x,t,k)$ eq3:NLInt(eq3)$ es1:expand(trigreduce(trigsimp(ratsimp(es1))))$ is( equal(eq3,es1) ); ""$ " Second Iteration "$ ""$ eq3:NLInt(eq3)$ es2:expand(trigreduce(trigsimp(ratsimp(es2))))$ is( equal(eq3,es2) ); ""$ " Defining functions f1 , f2 from es2 and eq3 respectfully and compiling these . "$ ""$ define(f1(x,t,k),es2)$ define(f2(x,t,k),eq3)$ compile(f1,f2)$ ""$ " Multiplot of f1(x,t,k) and f2(x,t,k) , via draw and vtk "$ " k = 0.1 , [x,100,100] [t,0.2,0.2] "$ ""$ ""$ " Exact answer u(x,t,k) = 4*k*k*(2  3*(tanh(k*(x(256*(k^6)*t/3))^2)"$ ""$ eq:k*(x(256*(k^6)*t/3))$ eq:(4*k*k/3)*(2  3*(tanh(eq)^2))$ kill(u)$ define(u(x,t,k),eq)$ compile(u)$ wxplot3d(u(x,t,0.1), [x,100,100], [t,0.2,0.2], [plot_format,gnuplot], [gnuplot_pm3d,true], [gnuplot_preamble, "set hidden3d"])$ wxplot3d(f2(x,t,0.1), [x,100,100], [t,0.2,0.2], [plot_format,gnuplot], [gnuplot_pm3d,true], [gnuplot_preamble, "set hidden3d"])$ wxplot2d([u(x,0.1,0.1),f2(x,0.1,0.1)], [x,100,100], [plot_format, gnuplot])$ wxplot3d(u(x,t,0.1)  f2(x,t,0.1), [x,100,100], [t,0.2,0.2], [plot_format,gnuplot], [gnuplot_pm3d,true], [gnuplot_preamble, "set hidden3d"])$ wxplot2d([u(x,0.1,0.1)f2(x,0.1,0.1)], [x,100,100], [plot_format, gnuplot])$ ""$ " Third Iteration "$ ""$ eq3:NLInt(eq3)$ ""$ " 3rd done "$ ""$ define(f3(x,t,k),eq3)$ compile(f3)$ wxplot3d(u(x,t,0.1)f3(x,t,0.1),[x,100,100],[t,0.2,0.2], [plot_format,gnuplot],[gnuplot_pm3d,true], [gnuplot_preamble,"set hidden3d"]); /* define(f3(x,t,k),eq3)$ compile(f3)$ wxplot3d(f3(x,t,0.1), [x,100,100], [t,0.2,0.2], [plot_format,gnuplot], [gnuplot_pm3d,true], [gnuplot_preamble, "set hidden3d"])$ wxplot2d([u(x,0.1,0.1),f3(x,0.1,0.1)], [x,100,100], [plot_format, gnuplot])$ wxplot3d(u(x,t,0.1)  f3(x,t,0.1), [x,100,100], [t,0.2,0.2], [plot_format,gnuplot], [gnuplot_pm3d,true], [gnuplot_preamble, "set hidden3d"])$ wxplot2d([u(x,0.1,0.1)f3(x,0.1,0.1)], [x,100,100], [plot_format, gnuplot])$ */ ""$ " fin "$ ""$ 
From: Edward Montague <quintin9g@gm...>  20171112 03:51:04

I believe we did . The reply and the forward selection is an occasional problem , also I may not of used the option of seeing my own posts in the list ; this is something that gmail has as a choice . Anyway now that's clarified , again , I shall send the message I was intending to . On 12 November 2017 at 16:37, Kris Katterjohn <katterjohn@...> wrote: > I got this one and also another one from earlier today. > > Didn't we go through this same sort of thing a couple of years ago? > > Kris Katterjohn > > On Sun, Nov 12, 2017 at 04:11:55PM +1300, Edward Montague wrote: > > Just trying this address. > 
From: Kris Katterjohn <katterjohn@gm...>  20171112 03:37:32

I got this one and also another one from earlier today. Didn't we go through this same sort of thing a couple of years ago? Kris Katterjohn On Sun, Nov 12, 2017 at 04:11:55PM +1300, Edward Montague wrote: > Just trying this address. 
From: Edward Montague <quintin9g@gm...>  20171112 03:12:03

Just trying this address. 
From: Edward Montague <quintin9g@gm...>  20171112 01:32:52

Is this the correct address for messages , or should I use the mailman at xmission . 
From: Roland Salz <salz.germany@gm...>  20171112 00:31:00

From: macrakis@... [mailto:macrakis@...] On Behalf Of Stavros Macrakis (Sta???? ?a??????) Sent: Sunday, November 12, 2017 12:40 AM To: Roland Salz Cc: <maximadiscuss@...>; Robert Dodier Subject: Re: [Maximadiscuss] Fixing the context mechanism A very small amount of data is being added. Mostly it is shared with the data for every other declaration. The plist just includes a pointer into this global data structure. If you do not understand this very basic fact, you should not be touching this code!!! Of course I won’t touch code I don’t understand. Don’t worry. I don’t have rights to mess in Sourceforge’s Maxima anyways. But I would not even do that in my own local copy. What can I do to understand this very basic fact? How can I translate what I see in the plist into something understandable? At the moment I have no idea. But it would interest me a lot. Could you possibly give me one or the other hint where to search? I did not work with pointers in Lisp yet. Is it general Lisp knowledge which I’m lacking? Or specific Maxima implementation knowledge? How can I find out, what this data sharing here is, and the circularities, circular substructures you mention? Best, Roland The data structure has many circularities, and it doesn't print in a useful way (the #'s are Lisp's way of showing circular substructures). This is one of the trickiest pieces of code in Maxima, and is completely undocumented. I know it took its author a long time to get it right, and I suspect that he removed the comments because they reflected multiple inconsistent versions of the code. Folklore says that he removed the comments because if there were comments, that might encourage someone else to modify the code... which he thought would be a bad idea. DO NOT TOUCH unless you have studied it thoroughly. Of course, if you DO study it thoroughly, and come to understand it well, I would recommend that your first step be to document it before modifying it. s 
From: Stavros Macrakis (Σταῦρος Μακράκης) <macrakis@al...>  20171111 23:45:00

You need to tell us more about your environment and your goals, and then we should be able to help you. Which front end to Maxima are you using? wxMaxima? XMaxima? or are you using the Tex output of commandline Maxima? What version of Maxima + front end are you running? What is your purpose in sharing an image of the output (rather than something that can be computed with)? How much does formatting matter to you? By the way, was Maxima recommended by the instructor? If so, it would be very interesting for us to know which OU course this is. Thanks! s On Sat, Nov 11, 2017 at 2:38 PM, R C <rc12002@...> wrote: > Is there a configuration option that controls printouts? > > > A full screen display of a polynomial function generates a printout which > is about 2 cm^2. > > > I'm taking an OU maths course  the help desk advice is to take a screen > print. > > > Regards, R. Clanchy > > > > >  >  > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Maximadiscuss mailing list > Maximadiscuss@... > https://lists.sourceforge.net/lists/listinfo/maximadiscuss > > 
From: Stavros Macrakis (Σταῦρος Μακράκης) <macrakis@al...>  20171111 23:40:24

Roland, A very small amount of data is being added. Mostly it is shared with the data for every other declaration. The plist just includes a pointer into this global data structure. *If you do not understand this very basic fact, you should not be touching this code!!!* The data structure has many circularities, and it doesn't print in a useful way (the #'s are Lisp's way of showing circular substructures). This is one of the trickiest pieces of code in Maxima, and is completely undocumented. I know it took its author a long time to get it right, and I suspect that he removed the comments because they reflected multiple inconsistent versions of the code. Folklore says that he removed the comments because if there were comments, that might encourage someone else to modify the code... which he thought would be a bad idea. DO NOT TOUCH unless you have studied it thoroughly. Of course, if you DO study it thoroughly, and come to understand it well, I would recommend that your first step be to document it before modifying it. s On Sat, Nov 11, 2017 at 5:18 PM, Roland Salz <salz.germany@...> wrote: > > Original Message > > > From: Robert Dodier [mailto:robert.dodier@... > <robert.dodier@...>] > > > Sent: Saturday, November 11, 2017 9:15 PM > > > > Thanks for your reply. > > > > Can anyone tell me why storing an assumption a>0 creates in the > symbolplist of a this enormous amount of data, that seem to be of extreme > redundancy and irrelevance? Every new assumption will add the same amount > of data (twice, as I will explain later)! > > > > (%i1) declare(a,integer); > > (%o1) done > > (%i2) :lisp(symbolplist '$a) > > > > (DATA (((KIND $A $INTEGER) CON $INITIAL)) +LABS > (1)) /* This is ok. */ > > > > (%i2) assume(a>0); > > (%o2) [a > 0] > > (%i3) :lisp(symbolplist '$a) > > > > (+LABS (1) DATA > > (((MGRP $A > > (0 DATA > > (((MGRP $A > > (0 DATA > > (((MGRP $A (0 DATA ((# CON $INITIAL) (#)))) CON > $INITIAL) > > ((MGRP (0.5772156649015329 DATA ((#) (# CON $GLOBAL) > (#))) > > (0 DATA ((# CON $INITIAL) (#)))))))) > > CON $INITIAL) > > ((MGRP > > (0.5772156649015329 DATA > > (((MGRP (0.5772156649015329 DATA ((#) (# CON $GLOBAL) (#))) > > (0 DATA ((# CON $INITIAL) (#))))) > > ((MEQP $%GAMMA > > (0.5772156649015329 DATA ((#) (# CON $GLOBAL) (#)))) > > CON $GLOBAL) > > ((MGRP (1.618033988749895 DATA ((#) (# CON $GLOBAL) (#))) > > (0.5772156649015329 DATA ((#) (# CON $GLOBAL) > (#))))))) > > (0 DATA > > (((MGRP $A (0 DATA ((# CON $INITIAL) (#)))) CON $INITIAL) > > ((MGRP (0.5772156649015329 DATA ((#) (# CON $GLOBAL) (#))) > > (0 DATA ((# CON $INITIAL) (#)))))))))))) > > CON $INITIAL) > > ((KIND $A $INTEGER) CON $INITIAL))) > > > > The next problem is that the context mechanism stores all Maxima > properties and assumptions in two places, in the affected variable’s plist > and in that of the context variable, here *initial*. This is inviting > inconsistencies. While the plists of the variables are fairly consistent, > those of the context variables are corrupted by several effects, probably > because all the more recent programming around properties simply did not > take them into account. So facts(context) and properties(context) both give > wrong results, facts more than properties. But without consistent plists of > the context variables I think it will not be possible to repair that. > Making the plists of the context variables consistent seems too much work. > Quite a number of fixes would have to be done in this package. And Richard > warned about its style. > > > > To give you an example: If I declare a property in initial, then create a > new context cont1, switch to it and remove the property there, it will be > removed from the variable’s plist, but not from the plist of initial. Thus, > it stays in properties(initial) and facts(initial), too. If I go back to > initial and declare the same property again, it will show up in initial’s > plist a second time, etc. etc. > > > > Without consistent context plists and functions facts(context) or > properties(context) it will not be possible to gain an overview of what the > different contexts contain. This was a main idea of my functions. Without > this information I don’t think the context mechanism is usable. > > > > Best regards, > > Roland > >  >  > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Maximadiscuss mailing list > Maximadiscuss@... > https://lists.sourceforge.net/lists/listinfo/maximadiscuss > > 