Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Software GNU is Not Unix Programming Silicon Graphics IT Science Technology

PLplot Notes Its 10,000th Commit 66

iliketrash writes "From the PLplot development team is the announcement of their 10,000th commit: 'PLplot is a cross-platform software package for creating scientific plots that has been in continuous development since its inception 17 years ago. On May 23, 2009 the PLplot developers quietly celebrated our ten thousandth commit since our initial software repository was populated back in May 1992. This longevity puts PLplot in some select company amongst open-source software projects. We may even be unique within this group because all PLplot development has been done by volunteers in their spare time. The enthusiasm for PLplot development continues; we have averaged more than 100 commits per month over the last year which is double our 17-year average, and we are looking forward to the celebration of our next ten thousand commits!'"
This discussion has been archived. No new comments can be posted.

PLplot Notes Its 10,000th Commit

Comments Filter:
  • congrads to everyone who develops PLplot
    • Re: (Score:1, Funny)

      by Anonymous Coward
      congrads to everyone who develops Slashdot for your consistent failure to
      +----------+
      | FIX YOUR |
      |  FUCKIN' |
      |   CODE   |
      +----------+
          |  |
          |  |
        .\|.||/..
  • Mazel Tov! (Score:1, Funny)

    by Anonymous Coward

    I'm plotzing!

  • Commit? (Score:1, Informative)

    by Anonymous Coward

    What on earth is a commit?

    • Re:Commit? (Score:5, Funny)

      by QuantumG ( 50515 ) * <qg@biodome.org> on Thursday June 04, 2009 @10:08PM (#28217993) Homepage Journal

      It's like a patchset but not thought out as much.

    • Re:Commit? (Score:4, Informative)

      by MichaelSmith ( 789609 ) on Thursday June 04, 2009 @10:13PM (#28218031) Homepage Journal
      This is a list [sourceforge.net] of examples..

      Quite a good record too. I went looking for howlers to link to but they seem to be doing a good professional job of tracking changes to their code.
    • What on earth is a commit?

      Turn in your geek card.

    • Re: (Score:3, Insightful)

      by R.Mo_Robert ( 737913 )

      A better question is: what on earth is PLplot?!

      ...or maybe I'm the only person on Slashdot who had no idea what this software was before finding this article on the home page.

      • No. I'm with you. I thought gnuplot was the first-choice package for creating scientific plots. But this one seems to be some kind of library to create plots. I'm not sure why one would want to write scripts just to create plots. I'm exploring it though.
        • My first-choice package is matploblib, which is made in Python.

        • I'm not sure why one would want to write scripts just to create plots.

          One good reason is to make graphs of a consistent style (same fonts, sizes, axis types, etc.) which can then be generated from different data sets. Publications look crappy when plots of equivalent data are inconsistently presented.
          Scripting is how I produce graphs in gnuplot (and Matlab). Modifying the analysis or changing the datasets then re-running the scripts gives comparable plots.

  • Well done to the PLplot team.

    The example graphs don't look so pleasant though. The (default?) colour scheme is excellent for a semi-lit astronomical dome (doesn't ruin your night vision) but I put those in front a business board without a fair bit of work on the aesthetics - wouldn't want the company directors to start throwing chairs, would ya?.

    • s/but I put/but I wouldn't put/
    • Re: (Score:3, Insightful)

      The example graphs don't look so pleasant though. The (default?) colour scheme is excellent for a semi-lit astronomical dome (doesn't ruin your night vision) but I put those in front a business board without a fair bit of work on the aesthetics

      Agreed. When it comes to your papers, you really want the best looking plots, and the examples on the PLPlot site don't even use anti aliasing! Check out the commercial competition [sigmaplot.com]. Mathematica generates pretty plots too, and some amazing mathematical graphics [wolfram.com]. Hell, even a recent Gnuplot seem to do a better job [sourceforge.net] at plotting.

  • by morrison ( 40043 ) on Thursday June 04, 2009 @10:59PM (#28218281) Homepage

    BRL-CAD [brlcad.org] is about to cross 35,000 commits [ohloh.net]. Emacs has more than 85,000 [ohloh.net]. GCC has about 12,000 unique over 150,000 commits [ohloh.net].

    That rounds out the three oldest continuously developed repositories [ohloh.net] with preserved revision history.

    • by QuantumG ( 50515 ) *

      I've tried about 10 times to get interested in BRL-CAD. It's just so awesome in theory, and then you run into the cruft and back away slowly. tcl-tk? I don't have that much hair left. I think the last time I looked at it I just gave up on the editor and hacked the ascii version of the .g files directly. Now they're talking about getting rid of the g2asc and asc2g tools because you don't need them to convert between platforms anymore. If you just wanna make raytraced images you're better off going with

      • by morrison ( 40043 )

        I'm certainly biased, but it's more than awesome in theory. The project has a lot of momentum and development going on not only towards addressing your concerns, but a lot of others as well. Most everything being worked on now is towards making the GUI much better.

        That includes dropping Tk for the modeler (going to Qt now that it's LGPL). That includes implementing full BREP/NURBS support and multiple-representation geometry (no longer just unevaluated wireframes until render). That includes a fresh i

    • The KDE project is approaching commit 1 million :) I wonder if they will get there before converting to git, though,.
      • by morrison ( 40043 )
        The numbers referred to for the other projects are 'unique' trunk/mainline commits only. Per that same measure, KDE sits at a little over 125,000 [ohloh.net] commits. Raw commits to all branches is vastly different for everyone.
      • Atheism is not a religion, it is the absence of religion.
        Agnosticism is the absence of decisiveness.

        Heh, atheism sure sounds like a religion sometimes, given how fervently some atheists push their belief that there is no god. Some seem almost as bad as evangelical Christians.

        Agnosticism is merely logic at play. Anyone who isn't an agnostic is deluded. Claiming to know with certainty that there isn't a god is just as unscientific as claiming that there is a god (Richard Dawkins' beliefs notwithstanding).

        Of course, the mass media has confused the definition of agnosticism to the point of uselessness,

        • Atheism is not a religion, it is the absence of religion. Agnosticism is the absence of decisiveness.

          Heh, atheism sure sounds like a religion sometimes, given how fervently some atheists push their belief that there is no god. Some seem almost as bad as evangelical Christians.

          I have heard of such atheist, but never met any. As long as that is the case, I'd rather defer judgment.

          Agnosticism is merely logic at play. Anyone who isn't an agnostic is deluded. Claiming to know with certainty that there isn't a god is just as unscientific as claiming that there is a god (Richard Dawkins' beliefs notwithstanding).

          I think even Richard would claim that the God hypothesis is simply completely unsupported. He tends to use a colorful language involving alfs and so on, but I think that is the essence.

          (Full disclosure: I'm an agnostic atheist.)

          As am I. Which entitles me to mock both :P

          • by kelnos ( 564113 )

            Agnosticism is merely logic at play. Anyone who isn't an agnostic is deluded. Claiming to know with certainty that there isn't a god is just as unscientific as claiming that there is a god (Richard Dawkins' beliefs notwithstanding).

            I think even Richard would claim that the God hypothesis is simply completely unsupported. He tends to use a colorful language involving alfs and so on, but I think that is the essence.

            Yes, I'd agree that the god hypothesis is completely unsupported, but how is the no-god hypothesis supported either? You can pull out things like Occam's Razor in defense of the no-god hypothesis, but you can't definitively *prove* that there is no god. Any possible proof you come up with can be met with the (annoying but reasonable in this context) rebuttal of "it's that way because God is omnipotent and wants it that way."

            I haven't actually read The God Delusion, but I've picked up a copy and will hop

            • Yes, I'd agree that the god hypothesis is completely unsupported, but how is the no-god hypothesis supported either?

              Well, in my opinion I don't have to. In my mind, I just have to prove that any given god (pick one) cannot exist. And they all, invariable, makes certain testable claims, typically miracles. As any miracle hypothesis always ends up disproven, you can conclude that gods non-existence from that.

              But it all depends on what you *mean* with proof. Facts about the physical world can never be proven any more than you could prove that the next beer can you drop will actually fall to the floor instead of hanging in

              • by kelnos ( 564113 )

                Yes, I'd agree that the god hypothesis is completely unsupported, but how is the no-god hypothesis supported either?

                Well, in my opinion I don't have to. In my mind, I just have to prove that any given god (pick one) cannot exist. And they all, invariable, makes certain testable claims, typically miracles. As any miracle hypothesis always ends up disproven, you can conclude that gods non-existence from that.

                That sounds reasonable, assuming there actually are testable claims for every god/religion. I'm not really sure either way. Of course, this doesn't rule out some other "being" that we haven't thought up yet, but why over-complicate things...

                But it all depends on what you *mean* with proof. Facts about the physical world can never be proven any more than you could prove that the next beer can you drop will actually fall to the floor instead of hanging in thin air. Yet, since I rather like beer, I would rather not drop the can.

                Heh, true in an absolute sense, but not in any real-world practical sense. We take it as a given that gravity exists, and works, even though we don't completely know how. If we can't rely on some things as being "truth," then I don't think we can really live meaningf

    • by kelnos ( 564113 )
      Er, I don't quite understand why 10,000 is such a big deal. Xfce is at 29,994 [xfce.org], and that's just for version 4, which was started around... 2001 or 2002 or so? Version 3 adds another 1181 commits [xfce.org] to that, and I'm sure v2 and v1 would add a bit more if their history was still around.

      I'm not saying this to brag (hell, KDE has over 125k, and I'm sure GNOME is comparable to that); I doubt it'll be a big deal when Xfce hits 30k commits, even. I'm just wondering why PLplot (arguably something of an obscure pro
      • by morrison ( 40043 )

        It's not a big deal except to the PLplot devs as an arbitrary line being passed. Kudos to them for making ten thousand measured changes to their code, hope they make it to the next decimal place up. Probably not unlike the mild nod of appreciation you might feel when you hit 100,000.

        It's a rather meaningless metric in general simply because there are massive differences in commit styles across projects and even within the same project. Even less meaningful than the meaningless lines-of-code metric. As a

        • by kelnos ( 564113 )
          Yeah, true. I guess I was just pointlessly whining about how I don't think this article is /. front page-worthy. Not that such whining is ever useful, but...
  • I hate to be a jerk, but the example plots [sourceforge.net] are not of the quality I would be proud to publish in a paper. I wish there were more of an open-source tradition among graphic artists.
  • Hmm, reminds me a bit of R [r-project.org], the plotting part of it, at least. There are a few examples of the kind of plotting you can achieve in R here [addictedtor.free.fr].

    • by Coryoth ( 254751 ) on Friday June 05, 2009 @12:06AM (#28218671) Homepage Journal

      Hmm, reminds me a bit of R, the plotting part of it, at least. There are a few examples of the kind of plotting you can achieve in R here.

      There are a plot of quite good open source plotting tools out there. I would consider R useful in as much as it provides powerful data crunching tools to distill your data into something essential to plot, but for plotting alone it is merely adequate. GNUPlot [sourceforge.net] is actually surprisinly powerful and flexible with professional output if you're willing to take the time to learn all it can really do (it's default output can be rather underwhelming). Going outside the box of what GNUplot does easily can be an exercise in extreme contortions of an already slightly arcane language however. matplotlib [sourceforge.net] is one of the best straight plotting tools out there, with a good mix between simple high level plotting, and sane easy to manage low level drawing tools, and good looking default output. CairoPlot [wordpress.com] is nice for very pretty charts, but is not as flexible as one might like. I'm sure there are more, but I don't happen to know them offhand.

      • by flynt ( 248848 )

        Merely adequate is not really correct. R's basic "graphics" package is not bad, but you simply must look at the "lattice" and "ggplot2" packages. There are entire books written on the two latter packages by their respective authors. Also see the book "R Graphics" by Paul Murrell, which will introduce you to the "grid" package, a low-level plotting package in R (upon which lattice and ggplot are based). I've made some pretty interesting graphics with it using minimal code, and combined with R's data mani

    • Comment removed based on user account deletion
  • In what way is this better than GNUPlot?

    • It has more commits, so you know that it works better!
    • Re: (Score:3, Informative)

      I'm no expert, and this is how I understood about the differences between PLplot and GNUPlot.

      GNUPlot is modelled like an interpreter. It works by interpreting the input as a stream of commands. If you are to embed GNUPlot in your own application, you need a separate process for GNUPlot, and you construct the GNUPlot commands from your data and send them to the plotter process. On the other hand, PLPlot is a library with multiple language bindings, making it easier to embed in applications (you just call the

    • In what way is this better than GNUPlot?

      Well, it has a decent license, for one thing.

    • In what way is this better than GNUPlot?

      It has what graphs crave. It has electrolytes.

  • by sigmet ( 1569971 )
    Yet it still looks like ass? it doesn't look like something that has been in development for 17 years but more like something that was developed 17 years ago unfortunately.
  • Veusz [gna.org], my scientific plotting package, is up to revision 1009, and I'm virtually the single author and a volunteer. It has been in development since 2003. The output, IMHO, looks quite a bit nicer than PLPlot.

  • And given that releases are roughly every 3 months, it exceeds 10,000 commits per minor release.

    Source [celinuxforum.org]

  • SigmaPlot (Score:3, Informative)

    by solanum ( 80810 ) on Friday June 05, 2009 @09:10AM (#28221925)

    As mentioned in a comment above, the paid for competition here is SigmaPlot. I remember first using Sigmaplot over ten years ago and the output of that version was head and shoulders above this. In fact there isn't any open source competition for Sigmaplot. Grace used to style itself as a SigmaPlot alternative, but hasn't been updated in a very long time and as never a match anyway.

    As a scientist who primarily uses a Linux desktop I am fed up with rebooting to Windows just to run Sigmaplot (or SPSS for that matter - whilst there are plenty of stats software for unix, there are no easy to use but powerful GUI based packages for those of us that have to use stats every day but aren't statisticians). I'd happily pay for a Linux version of SigmaPlot, but I'd much rather use open source software. It's not about the money (my employer pays) it's about being able to access my own files and results in five years time. Unfortunately there are some areas where open-source doesn't come close to proprietory software and this is one of them.

    • Re: (Score:3, Informative)

      by xiox ( 66483 )

      What sort of plotting do you do? Can you give me some idea of what the most useful plots missing from Veusz [gna.org] are? I know it doesn't do 3D plots, but I don't find them generally useful (except for volume renderings, etc.).

      • Hmm...well, your program looks really nice. I hadn't heard of it before. But a fairly critical feature is the ability to do statistical analysis on data sets and curve fitting. As far as I can tell, your program doesn't do that. Grace does, but it's interface is rather beastly. These days I use R directly, but it doesn't have a nice GUI like SigmaPlot. Personally, I think R requires you to know a little bit more about what you are doing than SigmaPlot, which is a good thing. But it would be nice to have a p

        • by xiox ( 66483 )

          It does do curve fitting (fitting functions to data). It doesn't do enough statistical analysis however.

          You can call python code to do it, though it should have a UI for doing it.

          Thanks for your comments.

      • by solanum ( 80810 )

        I'll check out Veusz (I hadn't come across it before), from the screenshots it looks like it does most of the plot types I use. The main things for me are to be able to put multiple plots on the same axis, not always the same type (ie line over bar etc), multiple y axes, and the ability to manipulate individual elements easily (e.g. change the colour/style of a single bar/line in a multiple bar/line graph, alter font/font size). That sort of thing. I use stats software to do stats, so I don't care about t

    • I'm personally fond of gnuplot. The output is quite nice if you use a vector drawing backend like eps or svg (much better than the PNG demos on the homepage would suggest), and integration with latex via the epslatex terminal is perfect for my needs. Matplotlib output is almost as good (and getting better all the time) and is arguably easier to use, especially for those familiar with Matlab. The most popular open source plotting packages among my colleagues seem to be gnuplot and xmgrace.

      I downloaded and

Keep up the good work! But please don't ask me to help.

Working...