Hi guys,
Greetings from an extremely green and nice rural location in Galicia (northwest Spain), where I'm spending a few weeks of vacation. I really need some relax after 1.6.1 and all the hard work.
Thank you all for your great support and ideas. Just let me take some time to digest everything before getting back with something useful to say
I just got ridiculed on the SBIG list for recommending PI.
I don't think
you got ridiculed, Max.
Acknowledging that documentation is something that PixInsight is lacking, there are hundreds of happy PI users and lots of great images being produced with PI. Just take a look at the gallery here and see quite a few APODs with images processed with PI.
Yes, documentation is necessary. But, what kind of documentation? Do you really want to have 1000+ more pages with technical descriptions of every tool implemented in PI? There is extensive literature on image processing that covers this aspect much better than anything we can write here. On the other hand, a number of critical PI tools are being published as open-source products, so anyone interested can explore the source code directly. Most new tools will also be open-source and some existing ones will also be released in the same way in future versions of PI. The general tendency is that only the PixInsight Core application will remain closed-source in the future.
Some applications have documentation of the type "Click the 'select image' button to select an image to process", or "click the Ok button to accept the dialog box". While this kind of documentation may be very easy to produce, I won't waste development time writing it because I think it is useless.
What I think we really lack is (not in any particular order):
- Written documentation on the user interface (including the graphical and command-line interfaces). This is very important IMO and also a lot of work.
- A general guide a la RBA, who has made a great work. RBA's guide with corrections and some extensions (basically made by me and other PTeam members) will be available in the PixInsight Core application through an integrated documentation system. This will be implemented during the 1.6 cycle.
- More written image processing examples and tutorials. Thorough tutorials and worked examples are the really useful documentation items IMO. Much more than dumb "click x to do y" documentation.
- More,
much more video tutorials. These are as important as written tutorials, and have the advantage that are much easier and faster to produce than written documentation. Videos are IMO the best and most effective learning tools available, as Harry has clearly demonstrated.
- Written reference documentation for the development frameworks: PixInsight Class Library (PCL) and PixInsight JavaScript Runtime (PJSR). The PCL has its own documentation which is reasonably complete, although needs more work. The PJSR documentation does not exist and this is a serious problem that I must address ASAP.
- Written development tutorials. We need at least a tutorial that explains how to design and write a basic PixInsight module. The same for a basic PJSR script. Then we need a general description of PixInsight's software architecture and how the different components of a module interact with the PixInsight Core application, at the high, object-oriented level.
Saying that PixInsight is worthless because it doesn't have a manual is destructive criticism denoting a very limited understanding of what PixInsight is and how it is evolving.
