Hi Fabio
Welcome to PixInsight Forum.
Thank you so much for your nice words. I appreciate a lot your wish to help. As you probably have figured out, I am aware of the CN thread. Unfortunately, I never post on any public forum except this one. That's a price I have to pay for my involvement in this project.
The main problem here is not very original: time. Contrarily to what has been said, I cannot stop active PixInsight development because the platform still lacks a lot of key features. To put you in situation, these are the main pending tasks, by approximate order of priority:
- An image calibration system. We are implementing this as a set of JavaScript scripts. This task is relatively easy, but time consuming.
- Image analysis tools. We definitely need at least a tool to acquire PSF sizes and shapes, both manually (point'n shoot) and automatically (by star detection similar to StarAlignment).
- Project files. PixInsight projects will be XML files that will allow you to store the state of a running PixInsight instance, including all images and their processing histories, all icons, workspaces, settings, open scripts, etc., to recover it later. This is a difficult and delicate task, given the complexity of the platform.
- More generic image processing tools. For example, we still don't have a generic convolution tool, or feature detection tools. This is a shame.
- Improved development frameworks. The upcoming version 1.5.8 introduces important improvements to the JavaScript runtime (PJSR). We still lack a debugger for the PJSR, though. The core PCL/C++ framework is now nearly 5 years old and also needs some revamping. We need scripts to automate trivial parts of module development.
- Image editing and drawing tools, including a layered composition system with vector graphics generation capabilities. Many users want these features (partly due to their experience with other applications), and it certainly will make PixInsight a much more versatile platform. On the other hand, these features will open PixInsight to other fields of technical and creative imaging, which is the way to go to ensure that PixInsight will survive as a live project in the long term.
- Mosaic and panorama construction tools. The StarAlignment tool has excellent mosaic construction capabilities (it is urgent to show them with a video, BTW), but we need a dedicated mosaic/panorama construction tool.
- Improved core application GUI. The GUI must be in constant evolution to be more dynamic, more flexible.
And this is just a partial review. I am the guy behind this, but I also must do things like user support and some administrative tasks. For example, I must revamp the website ASAP because it is showing things that are two years old (look the screen shots section - another shame).
I do not stir up your compassion, by no means, mainly because I do just what I want and I extremely enjoy PixInsight development. It is just that I do what I can, and I try to do it well, that's all.
Having said that (sorry for the lengthy story), I agree with all of you who say that we need documentation. We indeed need to say what PixInsight does and can do, and in many cases we are failing in such a basic task. However, in my opinion the traditional "user manual" concept does not apply to PixInsight. The main reason is that PI is too complex, evolving and fast growing, so a manual would become obsolete too often. In my opinion, PixInsight documentation must consist of:
- Video tutorials. This is the best way to communicate knowledge about PixInsight, and where I am currently investing all of my documentation efforts and time. Something that requires hours of writing to explain it confusely, can be shown in a few minutes with a video, and in a much clearer and direct way. Harry Page (not Larry!
) is doing a wonderful work. I am in debt with him for his videos, absolutely.
- Written tutorials and processing examples. We have a bunch of these on the website, but some of them are now obsolete. An example is the LRGB tutorial that gave rise to the CN's thread. That tutorial is three years old and is mostly obsolete --in fact, I disagree with many things I wrote in that tutorial. For example, now I would *never* apply some selective corrections, especially the last one to apply an arbitrary color correction (just to make the image appear similar to the widely accepted vision of the object). My vision of astrophotography has changed since then, and we have now wonderful tools to achieve much better results with a much more respectful attitude toward the data.
- Basic descriptions for each tool. This is a work in progress. Descriptions will be included in the main application distribution in HTML format, and will be available from the Process Explorer window.
- PixInsight Wiki. This is a project that was suggested on this forum, and I want to start it as soon as possible. I have to enable PHP5 on our server to install a good wiki software, but I must make sure the existing scripts (including the whole licensing system) will continue to work well.
That's how I see it. I'll welcome all opinions and points of view on this topic, positive, negative or otherwise. Thanks for opening this thread!
