What do we want our home page to contain? #4191
Replies: 5 comments 9 replies
-
|
just my 2c: I think there will likely come a day when we want to de-pythonify our first-impression, generally make it gentler, and hype up the buzz around napari ... But I repeatedly find myself feeling like we're routinely pushing those specific things a bit too hard and too fast. The argument that is usually made is something along the lines of "we don't want to scare anyone off", and this definitely comes from a super well-intentioned and positive place. But, in my experience, every time I sit down with a relative python newbie and try to introduce them to napari (and I've been doing so much more recently at my local institution) we get about 5 minutes into doing something before some problem or detail comes up that, frankly, requires at least an intermediate knowledge of python, or requires diving into some middle-to-deep level napari API stuff, or flat out just breaks when we try a plugin 😂 ... I ultimately find myself thinking "yeah, but you can imagine how cool it would be... right!??", suggest they stick with Fiji, and walk away feeling a bit deflated. By contrast, when I sit down with people who I think would define themselves as at least intermediate level python coders (folks who have been routinely using numpy, matplotlib, scikit-image, pandas, etc...), those inevitable hiccups are very quickly explained... they have that look of "oh yeah, I've seen that elsewhere", and by the end of the session, they're jazzed, I'm jazzed, and I walk away feeling like we've really got something here. So, irrespective of our medium-to-long-term goals of where we want napari to end up, and which communities and experience levels we want to benefit (and I think we're all relatively aligned there), i think we always need to sell the product we currently have, and not the product we wish we had. For me, that "product" remains something that can really help an intermediate-to-advanced python coder – and help them help other non-coders in their sphere – but has a lot of potential to disappoint a newcomer (for now at least) outside of just using napari as a relatively non-interactive image viewer. |
Beta Was this translation helpful? Give feedback.
-
|
I'm 100% in agreement with Talley on this. As a very specific point, we should remove the bundle from consideration until it supports in-place update. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for starting this conversation, @DragaDoncila ! I agree that it would be nice to update the napari.org landing page . I am also +1 on @tlambert03 general statement. I also agree with @DragaDoncila 's assessment that it is difficult to know if napari is useful to you based on the napari.org landing page alone.
If we are in agreement that the current "target audience" (i.e., people who would most immediately be excited about napari) is people with some intermediate python experience (e.g., have written scripts) who want to interact with their images, here are some things I think set napari apart and might get people excited:
Even though I think we want to be clear that some python knowledge is useful for a good napari experience, I don't think the vast majority of users (especially the "intermediate" folks we are targeting) are going to be interested in the underlying libraries. I think we I think we should make sure there is space on the website to properly acknowledge the folks we depend on, but I don't think it needs to be the first paragraph on the landing page. Thus, I agree that we should update the introduction to focus what one can do with napari (i.e., what napari currently does well) as opposed to how napari does it. |
Beta Was this translation helpful? Give feedback.
-
|
I totally get where this is coming from—I can see the tremendous amount of support that @tlambert03 @jni and all of you offer on image.sc, zulip, etc. I know it's not easy. It has to be massively time and patience consuming to deal with the same questions, problems, etc. So, I think it's totally fine to be honest upfront that some willingness to learn/use python is a plus. I can say from personal experience that the overall "spin" of the opening lines would be (was) demotivating. Basically this To me this makes it sound like napari is like matplotlib (which I'd heard of before napari). For python, in python—not for me doing some image analysis. Also, an important fact that's missing there—to me—is that napari has a very nice GUI, with a unique and well-thought out UX. The layers and labels model and tools are fantastic! And there are a ton of plugins that give access to a ton of features via said GUI. When we finally get to an example with a GUI shown it's: |
Beta Was this translation helpful? Give feedback.
-
|
agreed @psobolewskiPhD! to be clear... I definitely wasn't trying to imply that our current home page is, in any way, optimal :) More than happy to put many of the implementation details (like vispy, qt, etc...) below the fold (or even leave off the home page). 100% agree with highlighting what napari is good at (if there was any point i was trying to drive home, it's that we should step back and determine what we're good at and highlight it...) I'm also generally for decreasing the wall-of-text phenomenon wherever it may occur. and one of those places is definitely currently the home page! |
Beta Was this translation helpful? Give feedback.
-
|
I completely agree with the three bullet points @kevinyamauchi has highlighted as "things we're good at that we want to show off". Definitely agree also that a screencap/image of the viewer itself should be front and center, before we start talking about the benefits of napari's close integration with the scientific python ecosystem. I really appreciate your perspective on this @psobolewskiPhD - thanks for sticking around to give it! I think we should make it clear that experience working with python and other Python packages is extremely useful, but maybe highlight also that there is a ton of community support out there for people who may only have a little experience but are interested and willing to "get their hands dirty" |
Beta Was this translation helpful? Give feedback.
-
|
I like the Twitter feed idea. Regarding what napari can do, the @napari_imaging Twitter feed is actually excellent. I also want to second @psobolewskiPhD's suggestion from Zulip to include the community calendar widget on the home page side bar, or maybe side by side with the twitter feed in the main page. |
Beta Was this translation helpful? Give feedback.
-
|
Following on from @tlambert03's comments of what napari is now, I also would like to figure out whose attention we would like to capture. I made this point lightly in a group meeting a few months ago, and promised to write something about it and I never did 😅, but I would very much like to expand our Python audience to non-bioimaging disciplines. There's been a few isolated cases of users in astronomy or remote sensing, and even computer vision, but by and large, napari usage is only strong in bioimaging circles. I think a lot of this is attributable to network effects, but I do wonder what missing features (if any) are missing to make napari useful more broadly. So, whatever examples we put up, I would love it if they were from diverse disciplines. |
Beta Was this translation helpful? Give feedback.
-
|
I am super +1 to expanding our user base from different fields, and there's a lot of useful features already in place, but I'm also cautious of not being misleading... I don't know if we can quite say |
Beta Was this translation helpful? Give feedback.
-
|
Some possible headings/sections (also inspired by @kevinyamauchi's point about annotation, where I agree we do well and will continue to do better):
|
Beta Was this translation helpful? Give feedback.
-
|
@tlambert03 I hear all your points on "scaring off people" and I absolutely agree about grounding ourselves in the product we have now rather than the one we want down the road. @kevinyamauchi I've heard from multiple users that they were especially excited by napari when they saw an image go from 2D to 3D. I think a high res GIF near the top of the home page could illustrate this well, potentially as part of this "What you can do with napari" that @jni has outlined above. Until copy has been agreed upon to reorganize the first section of the home page, I was imagining this short paragraph that could be inserted before "Installation" bringing a little more clarification to how napari's currently being used:
But I'm curious on everyone's thoughts about curtailing the "Installation" section. Do you feel that we have enough resources within the "napari quick start" and "Contributing guide" to shorten this section and link out to those aforementioned pages? Generally I'd opt in favor of making the page more scrollable, and thereby digestible to first-time visitors, but I understand that this 'best practice' may not necessarily be relevant to napari's primary audience, as we've been discussing. @melissawm |
Beta Was this translation helpful? Give feedback.
-
|
I like these suggestions! Not sure about "expanding napari's applications to geospatial, cell biology, and beyond." As of now, we don't really have much for geospatial applications in napari hub. Perhaps just "expanding napari's applications to various scientific domains"? Installation/getting started could be a short paragraph with 3 hyperlinks to (1) Python package installation (2) bundled app installation (3) development installation. Each could go into more details on "best practice" and we could free up the space on the home page. |
Beta Was this translation helpful? Give feedback.
-
|
I like your rewording of that last sentence; more domain-agnostic too. |
Beta Was this translation helpful? Give feedback.
-
Currently, the landing page of napari.org looks a lot like a README.md (unsurprisingly, since I believe it is the README). This can be off-putting for users who are brand new to the website and can't get a sense of whether napari is right for them before being greeted with install instructions and some fairly technical details.
We (the docs working group) would like to update this landing page to expose the value propositions of napari and provide users with a gentler introduction to what napari can do for them.
Some things we've discussed are:
Before we can start working on changing the landing page however, we need to get some consensus for what the community thinks this page should contain and what they'd like to see (or really not like to see) when first interacting with our website. I've therefore opened this discussion for everyone to add their ideas!
Pinging @napari/core-devs and the community at large!
Beta Was this translation helpful? Give feedback.
All reactions