Shop OBEX P1 Docs P2 Docs Learn Events
JonnyMac obex contributions - missing? - Page 2 — Parallax Forums

JonnyMac obex contributions - missing?

2

Comments

  • rjo__rjo__ Posts: 2,114

    We don't know what we don't know... always possible, sometimes true:)

    I think if JonnyMac was generally motivated by the issue of attribution, he would have started a thread to explain his
    concerns. If that is the issue, then I would encourage him to be open about it.

    I think it is something else... possibly none of our business:)




  • Heater.Heater. Posts: 21,230
    When did the MIT requirement happen? I thought it had always been there. Or had it been left unspecified for ages?
  • I've worked with JonnyMac for twenty years, so I consider him both a professional colleague and a considerate friend. He's a tremendous contributor to Parallax and the community through code contributions, production projects of all types (that would be so interesting to share), DEFCON badges, and too much more to list.

    Jon's disappearance has nothing to do with being thin-skinned. He's also an actor and lives in Los Angeles, a place that sometimes requires a tough soul. I think he just got a bit fed up and we pushed him over the edge at the wrong time. We've all been there before.

    Jon is also very consistent in everything he does so give him a little break to do whatever he wants to do. We'll see him again soon, I think. I've exchanged a few messages with him the past couple of days over a code example to use the Propeller as an RTC, so he's busy.

    Ken Gracey


  • Ken Gracey wrote: »
    I've worked with JonnyMac for twenty years, so I consider him both a professional colleague and a considerate friend. He's a tremendous contributor to Parallax and the community through code contributions, production projects of all types (that would be so interesting to share), DEFCON badges, and too much more to list.

    Jon's disappearance has nothing to do with being thin-skinned. He's also an actor and lives in Los Angeles, a place that sometimes requires a tough soul. I think he just got a bit fed up and we pushed him over the edge at the wrong time. We've all been there before.

    Jon is also very consistent in everything he does so give him a little break to do whatever he wants to do. We'll see him again soon, I think. I've exchanged a few messages with him the past couple of days over a code example to use the Propeller as an RTC, so he's busy.

    Ken Gracey

    That's good to hear. I would certainly love to see him back here contributing. Everything I've ever seen from him has been of the highest quality and his support has been exemplary.
  • You guys have been so obsessed with the comment that you can't see what happened.

    Jon is an actor who works a world that is an illusion that captivates all who enter.

    Millions of people come to Hollywood every year looking for stars. All they find are chunks of Terrazzo stuck in the ground.

    Jon is probably off working on his greatest special effect yet, one that rivals the great Harry Houdini himself.

    So sit back, enjoy the show, and in the end you will be amazed and yet never really know how it was accomplished.
  • It's really a shame that the objects have disappeared. They were all good examples of Spin coding, clean, and well documented. I have backup copies of a few of them like the 1-wire driver and the WS2812 driver
  • Mike Green wrote: »
    It's really a shame that the objects have disappeared. They were all good examples of Spin coding, clean, and well documented. I have backup copies of a few of them like the 1-wire driver and the WS2812 driver
    In fact, the license notice in these files says the original copyright notice must be retained. That includes Jon's name so non-attribution isn't an option allowed by the license under which these were released.

  • jmgjmg Posts: 15,140
    David Betz wrote: »
    Lev wrote: »
    ....

    There was nothing wrong with JonnyMacs original insightful comment. ...
    Calling someone else's code a "train wreck" is an insightful comment?

    From my reading, he did not quite call anyone's personal code a "train wreck".

    The original comment was actually this, relating to translated code...
    JonnyMac wrote:
    I don't know what happens to Spin/PASM when it gets translated to C, but it always looks like a train wreck. This is why I develop in Spin.

    There is a partial point in there, in that ANY transpiling/translation is going to 'add entropy', but I'm not sure users are bothered by that.

    Even C flows within themselves can have entropy issues : the smarter the optimisers, the less source-line correlation there is, to the point it is common to see forum advice to disable optimise for debug. Of course, what you then debug, is now not quite what you ship...
    That is far from a 'language war', it is a simple reality of how MCU flows work. The originating language(s) is immaterial.


    My own approach to transpiling/translation is to always have a clear 'warning' TOP header, saying this is Auto-Generated Code.
    along the lines of
    // !! Translated/Auto-Generated  using : WizardName v1.2.3  on Some:date !!
    //  Reading:  Source/File/Dir
    //  Command line:   WizardName.exe  - Full/Source/File/Dir -abcde
    //  Manual Post Edited : N
    

    I also try to include the original source, as comments, where practical. General rule: Clarity is more important than size.

    In a completely different thread, Jon also said
    JonnyMac wrote:
    Please remove my name from any objects you modify and redistribute. I do not require attribution, and refuse to be held accountable errors that may be introduced by modifications that are not my own.

    I can understand that, but there is an audit trail in code that it is very useful to preserve.

    Perhaps the letter of his request can be met, and still keep an audit trail, by simply adding a carefully worded web-link line like this ?
    // Derived from code :  https://github.com/xxx/yyy
    
    His name does not appear in isolation, but users can find the original material.
    Anyone planning large changes would like to see the original material, plus it's useful to check if the Auto-generated code has missed bug fixes.

    That should keep everyone happy ?
  • Here is the WS2812 OBEX upload from Oct 2016. It is the complete upload as well as being a little newer that the one Mike Green posted above.
  • David BetzDavid Betz Posts: 14,511
    edited 2017-08-02 21:29
    jmg wrote: »
    David Betz wrote: »
    Lev wrote: »
    ....

    There was nothing wrong with JonnyMacs original insightful comment. ...
    Calling someone else's code a "train wreck" is an insightful comment?

    From my reading, he did not quite call anyone's personal code a "train wreck".

    The original comment was actually this, relating to translated code...
    JonnyMac wrote:
    I don't know what happens to Spin/PASM when it gets translated to C, but it always looks like a train wreck. This is why I develop in Spin.
    That message was posted after my response. In any case, the code was not auto-translated with spin2cpp. I translated it myself. In fact, my translation isn't the most beautiful code possible. It certainly lacks sufficient comments. That's why I didn't complain too loudly. Mainly, my post was an awkward way of taking personal responsibility for the "train wreck" and not putting the blame on spin2cpp.

    There is a partial point in there, in that ANY transpiling/translation is going to 'add entropy', but I'm not sure users are bothered by that.

    Even C flows within themselves can have entropy issues : the smarter the optimisers, the less source-line correlation there is, to the point it is common to see forum advice to disable optimise for debug. Of course, what you then debug, is now not quite what you ship...
    That is far from a 'language war', it is a simple reality of how MCU flows work. The originating language(s) is immaterial.


    My own approach to transpiling/translation is to always have a clear 'warning' TOP header, saying this is Auto-Generated Code.
    along the lines of
    // !! Translated/Auto-Generated  using : WizardName v1.2.3  on Some:date !!
    //  Reading:  Source/File/Dir
    //  Command line:   WizardName.exe  - Full/Source/File/Dir -abcde
    //  Manual Post Edited : N
    

    I also try to include the original source, as comments, where practical. General rule: Clarity is more important than size.
    I believe that spin2cpp does insert something like this into its output.
    In a completely different thread, Jon also said
    JonnyMac wrote:
    Please remove my name from any objects you modify and redistribute. I do not require attribution, and refuse to be held accountable errors that may be introduced by modifications that are not my own.
    Then he should modify the license that he puts in his source files because it requires that his copyright notice including his name be retained in derived works.

    I can understand that, but there is an audit trail in code that it is very useful to preserve.

    Perhaps the letter of his request can be met, and still keep an audit trail, by simply adding a carefully worded web-link line like this ?
    // Derived from code :  https://github.com/xxx/yyy
    
    His name does not appear in isolation, but users can find the original material.
    Anyone planning large changes would like to see the original material, plus it's useful to check if the Auto-generated code has missed bug fixes.

    That should keep everyone happy ?
    In fact, that is exactly what I did. Well, the "Derived from code" part. His stuff isn't in GitHub so I couldn't include a GitHub URL and I'm not sure how to include an OBEX URL. I think I just named the specific file I used as an original source.
  • PublisonPublison Posts: 12,366
    edited 2017-08-02 22:04
    David Betz wrote: »
    I wonder if JonnyMac took all of his code out of OBEX a long time ago when the requirement for using the MIT license was put into place? Maybe it wasn't a recent move.

    Jon used MIT on everything I have ever downloaded.

    His OBEX programs vanished about July22-24. They were his to delete.

    Sad nonetheless.

  • Cluso99Cluso99 Posts: 18,066
    Ken Gracey wrote: »
    I've worked with JonnyMac for twenty years, so I consider him both a professional colleague and a considerate friend. He's a tremendous contributor to Parallax and the community through code contributions, production projects of all types (that would be so interesting to share), DEFCON badges, and too much more to list.

    Jon's disappearance has nothing to do with being thin-skinned. He's also an actor and lives in Los Angeles, a place that sometimes requires a tough soul. I think he just got a bit fed up and we pushed him over the edge at the wrong time. We've all been there before.

    Jon is also very consistent in everything he does so give him a little break to do whatever he wants to do. We'll see him again soon, I think. I've exchanged a few messages with him the past couple of days over a code example to use the Propeller as an RTC, so he's busy.

    Ken Gracey


    I sincerely holy Jon returns.

    His code snippets were really professional and helped so many people. While I didn't read many Nuts and Bolts articles, those I did were again very professional, interesting, and were extremely beneficial to the prop and the community.

    I also loved to see his little posts on what he did in Hollywood. Would have loved seeing a lot more.


  • If you actually read Jon's "train wreck" statement without bias to either language, it's pretty obvious he wasn't pointing out anyone's code, but rather the result of the code conversion in general. The comments regarding "starting a language war" were pointless and could have been taken as a personal attack on his generalized opinion which may have sent him walking since so many language debates are pointless in themselves; suum cuique I say. Maybe he felt it better to walk away then to attempt to back his point.
    Jon is extremely knowledgeable about the Propeller and its usage, has helped countless people on the forums over the years, and has several proven Propeller based products in the field. I for one hope he is simply busy with other priorities as I look forward to any new Propeller related project/concept he has to share. His SPIN code is all over my laptop and his Neopixel code is the only code I use for them.
  • Instead of wondering what happened to Johnny Mac you can just send him a Twitter and ask. ( I don't have Twitter)
  • Heater.Heater. Posts: 21,230
    I don't have twitter. This forum was good enough.
  • No! Twitter is for Presidents. Let the man be, if he returns he returns, otherwise others will take his place in the forums. He has taught / inspired enough of us, so more can step up and learn by contributing. For now, the speculation in this thread is turning into a geeks episode of The View.
  • Cluso99Cluso99 Posts: 18,066
    No twitter, Facebook or any other social media things. Just viber with immediate family and FaceTime. This forum only, but looking for a good fpga forum. I would rather be doing than wasting time ;)

    Hope to see Jon soon.
  • DigitalBob wrote: »
    Instead of wondering what happened to Johnny Mac you can just send him a Twitter and ask. ( I don't have Twitter)
    I tried sending him email but didn't get a reply.

  • A Jon McPhalen is contributing to the F*c*b**k page Blockly for Microprocessors (Aug 2).

    P'raps the JonnyMac persona has been written out.
  • I have always found JonnyMac to be very helpful and understanding and I admire his very well polished and professional coding style. He inspired me to move away from coding and programming to software engineering.
  • I sincerely hope that JonnyMac reconsiders his decision to remove the objects.

    I also have learned a great deal about the propeller - and more importantly the correct way to do things - from him.

    I needed a special version of his WS2812 object that would control 4 individual light strips. I could have used 4 instances of his object, but that would have required 4 cogs.

    So I modified it to handle the 4 strips in the same object.

    It took just one try... really! The first time I ran it, it worked. This included changes to PASM code. (I didn't change any of the fancy stuff like color wipe, just the basic set methods, etc.).

    Not a major change... but made much easier due to his clear comments and well thought out code.

  • Reading that post, it seems obvious to me it wasent the train wreck comment that offended but the fact that it was insinuated he was trying to start a language war. I hope Jon comes back, his coding is good, and far over my head.
  • MikeDYurMikeDYur Posts: 2,176
    edited 2017-08-04 22:05
    I can't believe he would leave the forums over the mild communications that were involved.

    As far as I know JonnyMac preferred spin and PASM. I wonder if it could have something to do with spin taking a back seat to Blockly in education, and he has no interest or time in going down the blocky road.

    Just saying..

    Going incognito.

    BTW: Maybe there is something that will be published in the near future.
  • JasonDorieJasonDorie Posts: 1,930
    edited 2017-08-04 22:07
    He has made recent contributions to the Blockly Facebook group, so that's probably not it. :)
  • Seems like he was in a bad mood (or something) and overreacted to a fairly minor thing.
    Maybe there is more to it than what is visible in the thread?

    Just seems like an extreme reaction to something unworthy of such a reaction.
  • All actors need an audience and love some drama to be equally involved in the proceedings. I remember a post of his where he was cross and suggested that the people who implemented the Ardunio I2C library should have been water boarded. I found that totally offensive and believe that I alerted Ken Gracey personally to the post where it was quickly moderated to remove these comments he had made. That was over fours years or so ago. Sure he is technically correct that specifications were shortened from the original Philips (NXP) implementation. Hopefully he will come back and in part further knowledge around here.

    Regards
    Mike.
  • And talking about his acting having to do with this, and needing drama will bring him back.

    Everyone does realize this is public and other people out there can read it?


  • Heater.Heater. Posts: 21,230
    Let's not generalize about actors. Or anyone else for that matter.

    There are plenty of drama queens that are not actors. There are plenty of "normal" actors.

    Jon seems to be a very special actor. How many actors are also into nerdy technical stuff?

    Yes, this is a public space and everyone can read it. That's the point of the thread. The Free and Open Source software that was, is not anymore. Curious minds want to know how/why this can happen.
  • xanaduxanadu Posts: 3,347
    edited 2017-08-05 17:53
    Ha, easy for you to say, you're anonymous. Your name isn't all over this. Your other jobs and personal life aren't being pulled into it.





  • Heater.Heater. Posts: 21,230
    edited 2017-08-05 18:13
    I'm not sure that anonymity or otherwise is pertinent to what I posted above.

    I certainly did not say anything negative about Jon. Quite the opposite. A man for whom I have great respect.

    My point was that Free and Open Source software, published in OBEX, has disappeared. No matter who put it there, or if we know who they are or not. No matter what reason. This is potentially a bad thing.

    Presumably it was put there for use by and education and pleasure of everybody. Now it's gone. That may cause distress to those that use it.

    Potentially people have published projects, tutorials, forum posts that link to that code. Now all of that effort is broken.

    This may or may not be a big deal. I have no idea how many dependencies on those OBEX projects there are out there. It was a big enough deal to cause the OP to start this thread though.

    Frankly, I think anyone who has copies of those objects and thinks they are good worthwhile things, which I'm sure they are, has a duty to put them back into OBEX. The license allows for this.

Sign In or Register to comment.