Skip to main content

Actual Documentation



  • Jim McAndrew

    Have you watched the 5-part Dronelink 101 series?


    And so if I miss or forget a detail on a component or command I go re-watch them?

    How many times?

    Those are not actual documentation.

    Should tell  DJI to "document" their SDK API with videos.? Or would you rather have a concise, complete manual that you can read and refer to?

    Is python documented with videos?

    No . In fact every programming language has a canonical reference document that defines it.


    I did actually purchase your product, because I see great promise. But you need a programmer's reference manual to make that happen.


    BTW the way I did read your 5 part series - I was blown away by the of detail for each component and command. NOT

    0 examples

    0 list of components and what they do

    0 explanation of what all the parameters mean

    If I am mis-stating the facts, please correct . me. I am just pointing out how it is.

    Don't shoot the messenger - I can help the way it is.

  • Eric Goldstein

    James -

    This is Eric (I created the videos) so I'm jumping in here a little bit to help Jim.

    First of all, we appreciate your support in purchasing Dronelink and want you to get all you can out of it.

    While Dronelink was designed as a way to "program your drone", it's not as complex a "programming language".  So, we haven't yet released documentation in line with Python or a full API

    The documentation was intended to be a starting point for users.   But, the videos are the tool to help people learn to use Dronelink initially and on a going-forward basis.  The videos cover almost all of the core functionality and parameters that exist within Dronelink and also has sample examples.  The core components (Path, Orbit, Map, Destination Component, and heading) are covered.   The basic parameters are either explained or are relatively straight forward (e.g. radius, altitude, rotations, etc.).

    For most of the Drone software solutions that are out there, we find that people learn the basics through the videos and then master things over time and flying.   If people get stuck, we find that people do go back and re-watch videos (or post questions in the forums). 

    So, while I understand you're looking for a "user manual" with a specific explanation of each parameter, that doesn't exist at this time.   Watch the videos.   Create some missions and use the repos and compoennts.  Get out and fly.   Only by doing that will you start to get a feel for the power of Dronelink.

    If you have any other specific questions, post them in the forum and they will be answered.

    If there are features that you need, watch the roadmap or make suggestions on the forum.

    Good luck  and again, we appreciate your support.


    II get it I'm wasting my time. Your convinced it's documented, and like flat earthers - no sensible argument is going to convince you otherwise. On one hand you say how different and powerful drone link is - yet "For most of the Drone software solutions that are out there.."  so  really you're just like them, and should do as they do.

    I thought Jim was into "innovative software products", so why are you using other products as an example? Do you think that James Gosling would have let Java go out the door without a reference manual? After all it only had 40 keywords at that time. Surely we can just make a cassette tape that explains each keyword.

    I did not know you are a python programmer - oh no wait your a marketing guy. (not meant to be offensive, we need you just as much as engineers) So by what measure do you claim it's not as complex as anything else? I don't expect you to understand complexity theory but complexity is not measured by the number of elements you have but by the permutations and combinations that are possible. Java only has 51 keywords (now), yet try and figure out a 100K line Java program. It is vitally important that you completely understand each element of a language including side effects.

    Like I've said dronelink has great potential - I think it IS an innovative software product. But you also need to innovate elsewhere too. Pascal and Modula were innovative too.


  • Eric Goldstein


    I didn't say it was "fully documented", our goal was to create training that allows people to learn to use the software effectively and provide a community of users to push its capabilities further.  

    (btw, not that it matters, but my background is in computer programming before I went to the dark side of Marketing and Product Management.  I also am 99% confident that the earth is round).

    Most of our users (and the users of Autopilot that Jim and I worked on previously) preferred the videos to the written documentation.   We saw that in how people absorbed and learned from the info.  While we had discussed (and I actually drafted an early version which needed a good deal of wordsmithing) creating a complete user manual, we decided to focus on the videos initially.

    Instead of complaining that there is not a document that is put together the way that you want it to be presented, can you list actual things, plans, features, use cases, etc. that you're trying to accomplish?   We can either explain it here or point you to a video to watch.  


    I did just that 7 hours ago. No reply.

    as for your hand waving concerning Auto pilot, this is exactly what I am asking for.

    Maybe most people DID want a manual.

    Instead of complaining that my request for a manual is unreasonable, why not commit to writing one?

  • Jim McAndrew

    I'm glad you enjoyed the documentation I created for Autopilot. When I first created that app my dad (a retired 747 captain) showed me his 5 inch think flight operations manual and flamed me for not having "real documentation" for it either lol. In the end it took hundreds of hours to create Flight School for Autopilot but the app needed it at the time. In the beginning there were no written or video tutorials, forums, in-app tooltips, and the UI had no advanced mode toggle. With Dronelink I made the decision to start with all of these things instead (tutorials, forums, tooltips, advanced toggles) due to the popularity of these formats over Flight School from my Autopilot experience (many, many users telling me Flight School was way too long and hard to understand). Having said that, the goal is to eventually created "real documentation" but the reality is there are only so many hours in the day. It is always a choice between documenting existing functionality and adding new functionality, and there is always a tension between which one takes higher priority. If there were zero resources available to help people get started, creating any documentation would be better than nothing and it would be highest priority, but that simply isn't the case.

    The situation is even more complicated from my perspective because we are also furiously trying to satisfy our enterprise clients (who contribute the majority of our revenue, allowing us to make Dronelink free for non-commercial use). The highest priority for enterprise clients is not Flight School style documentation, but rather an open-source SDK (with documentation!), which we have been rolling out on GitHub over the last few weeks. I suppose if you want to complain about anything it should be the lack of documentation on those repos, but what I am lacking in docs there I am trying to make up for in open-source!

    Anyway, your opinion has been heard loud and clear and I appreciate your support for my work. It sounds like maybe you have a programming background so I encourage you to clone the repos on GitHub and take a stab at creating something cool with the Dronelink SDK!


    First of all I am not complaining at all. I'm informing you that you need to document your programming language for it to reach its full potential.

    Flight school is way more that you need in terms of features and fancy formatting. Your a bit confused between learning and reference. If you thought the reference part of flight school was for learning then I can see why you were disappointed with the results. You can't learn chemistry from the periodic table, but you can't do chemistry without it.

    You need both and one does not fill in for the other.

    Reference is not for learning it's for after you've learned how to use it and now need to fill in the specific details about how things work. I never said nor implied that the videos were worthless for learning, I did say they are worthless for reference, and we need to understand that distinction clearly.

    For one thing, the man page, if you will, serves as the effective functional spec for the components, command or parameter.

    You should write the man page before you write the code. At this point other than reading the code (which is prone to error as it can be hard to determine the exact context that the code runs in) I don't know of anyway to recover this information myself. A simple text wiki would suffice.

  • Jim McAndrew

    It's a great goal (to write the reference manual before the actual product), but the reality is you are the first person to ask for it. Everyone else just demands new features. We will get there but it will take time.


    The thing is it's for the coder too - it sets the goal clearly, since you probably don't have anything in your process where you write down what the feature is to be. If you do it for each one it is not one giant job.

    It's not that I am asking for that, I am suggesting that if you did that you would at least have input to possibly write a man page someday.

    As you have guessed I am a Software Engineer. Right now I am working for a Silicon Valley start up and I am tasked with bringing order to the chaos, which means some semblance of an SDLC. Ya, we argue about tabs.

    My experience has indicated that a clear definition of the new feature up front really helps and a man page would work for that. I realize with only a couple of guys it's not as important, but you , might find it helps


    Have you tried the Open Drone Map ISO? I want to start an easy affordable stitching service.

    Does it work?

  • Jim McAndrew

    I haven't tried it, but it seems like you could easily integrate it with the Dronelink SDK given the recently released Asset Manifest feature:



    I'd like to second more detailed documentation.  There's almost no description of what the components do and what the options mean under each component that I can find.  I had a mission fail this past weekend because I planned a path using a starting waypoint with a camera command to position the gimbal at -90 for nadir shots and a final waypoint with a command to position the gimbal at 0 for landing.  I should have noticed in the plan preview, but I didn't and when I flew the mission the gimbal slowly rotated from -90 to 0 along the entire path... interpolation was set by default, i wasn't sure what that meant and didnt change it. Afterwards I fixed the problem by just trying the different options in that field, but it would have been helpful to be able to go to the docs and see that was the expected behavior... also a tooltip in the app would be helpful.

  • Jim McAndrew

    Understood. Until then, seeing is believing - you should always run a mission preview before flying.

  • Peter Ku

    I definitely would prefer to have written documentation. 
    Videos are for learning, written documentation for quick lookup. There is no real discussion about these two facts.

    More over: It would be soo great, if a flight plan could be shown in printed form. I can scroll through such a document and (as an example) search for all places, where the gimbal is moved, where the poi changes etc. pp. 1000 times quicker than in the "user friendly" web application :-) Searching for mistakes or errors would be way easier. 

    I tried to export a plan and read the text file, but it is in chinese :-)

    Thank you!


Please sign in to leave a comment.