Merge with ImageJ Ops? Or make an official ImageJ subproject?

[ctrueden]

One goal of ImageJ Ops is to provide a central, curated collection of image processing algorithms, which are easy to use from scripts. It would be very helpful to hear about where/how Ops falls short, and take steps to improve the situation. We do have plans to make Ops easier to use, and easier to extend.

Could we start a thread on the ImageJ Forum where we discuss these things, and figure out how to move the project forward for the whole community's benefit, rather than splintering it into side projects like this?

See also holgerbrandl/kip#1.

[benoalo]

Hi @ctrueden,

Thanks for your interest in the project. CIP project is pretty young and still maturing but getting help from IJ developpers at some point could be awesome! The project tries hard to ease access to script and image processing tools for a non programmer audience. It would also have benefit for prototyping. If you are around and interested to learn more about the project motivations and features we could chat about it at the next ImageJ Hackathon in Dresden.

I mention @holgerbrandl as I am sure he'd be interested in further discussion on the topic.

[ctrueden]

@benoitlo Chatting at the hackathon would be awesome. I really want ImageJ to be super easy to use from scripts—we are targeting Jupyter notebooks pretty heavily now. From my perspective it would be ideal if all the scripting juju and niceties were available to all users of ImageJ, and not requiring a dependency on a non-imagej-organization project. But developing and experimenting with these ideas in places like this does also make total sense to me... as long as the idea is to eventually migrate things upstream. What I want to avoid is multiple different tools for "easing scripting access" that are not mutually compatible, not highlighted in the core ImageJ documentation, not used in the core ImageJ tutorial scripts, etc.

A question to ask yourself is "do you want this project to have community impact"? If the answer is an honest "yes" then keeping it separate makes it much less accessible to the ImageJ community as a whole, for a variety of reasons. Whereas improving the core ImageJ scripting patterns, and updating the tutorials to use them, makes the ideas extremely accessible.

[benoalo]

Hi @ctrueden,

Thanks for your additional feedback. I am glad to hear that there are many projects geared toward ImageJ end users. All the IJ2 novelties really needs to percolate to them in some form. What form? I don't know, but I agree that experimenting is super important here as is, in the longer term, keeping the toolset consistent. In any case, I'd be more than happy to discuss this, and all the rest ;-), around a cup of coffee when you are in Dresden.

[imagejan]

@benoalo I discovered this only now, because of you recent activity on the ImageJ wiki (http://imagej.net/CIP), and I have to say I'm quite shocked because it looks like you're exactly replicating the functionality of ImageJ Ops. I might be wrong, but just after a quick look at the examples, it feels to me like a great waste of resources to have yet another (one-man) project in active development and not joining forces...

Was there any outcome of discussions at the recent hackathon that can be shared with the public, e.g. on the forum? I'd love to see things converge for the benefit of the community.

[benoalo]

Hi @imagejan,

Wow! Thanks for your sweet comments ;-). I understand your worries about effort dispersion. At the same time I disagree with you that CIP participate to that dispersion.

To tell you about the project, CIP goal is to remove the last hurdles that makes learning scripting so hard for beginners and to me this is a lot about user experience. With that in mind CIP focuses on a small set of functions, wrapping existing functions, adding consistent and flexible signatures, implementing unavailable, nevertheless usefull, options, hidding away some complexity (import, conversion, types, ...) and providing a user documentation. Certainly some of the described features are or will be in Ops todo list, in the meantime CIP makes them available to the curious and allows me to better support my colleagues and collaborators.

I hope I could make you less pessimistic about CIP project and effort dispersion. Regarding that last point having a roadmap and prioritizing tasks could probably help more people to support ... In any case I'd be glad to continue that discussion if you like and in the mean time don't worry too much and enjoy diversity of proposition in IJ community ;-)