rakeda

Thanks for your feedback Bob, the website is currently optimized for viewports smaller than 1920x1080p but I do plan support for the future.

Thank you for the advice, will incorporate smoother zooming, theres a trick between feeling snappy and responsive and smooth but sluggish.

There might be, but that would be missing out on half of the fun. Plus the real development challenge wasn't the frontend, but the database strategy, caching strategy, and edge availability. As with all my projects they serve as a way for me to learn.

Yup, if there is a bit more support ill begin to add in the mid-tier paid services, but I feel like its in a good spot for now. Working on terrain right now!

From what I put on reddit:

"The globe itself is just a black ball that hides anything behind it. On top of that we draw the grid (lat/lng) by looping through coordinates and connecting dots on the surface (same math youd do for platpoints on a map, just wrapped around the sphere). The coastlines and borders come from free geo data files and just drawn on the sphere, anchored.

The atmosphere and a majority of the affects are just tiny shaders (maybe 10 lines of code each) which gives it a strong 3d affect.

Everything else, data points, flights, events, is just coordinates associated with database entries and drawn ui components.

So if you tried it before and went down the rabbit hole of 3D and triangulation, you were probably overthinking it the same way I did at first. The wireframe aesthetic actually works in your favour because you never have to fill anything in"

It was less about performance and more about the visual storytelling. The globe itself isnt actually a 3d model, all of this are rendered in three.js as two dimensional wraps, but the performance is hit with the amount of data.

That is high praise. Thank you.

Satellite GIS for all vessels is pretty pricey. Waiting to see this project grow before investing in that realtime data. For now we have coastline GIS for vessels.

Some layers have additional enrichment, data linking everything internal external is a major goal for the project!

Military installations currently need to be sanitized and cleaned out. Will be working on later today. The source is unreliable and much of the data can be stored internally instead of public dataset.

Mobile has been a common complaint that I have received. Planning on conquering this soon! (90% of all initial traffic is mobile).

On it right now, should see the change live in 10 minutes! Was supposed to only show when imagery was selected.

I'm routing and caching a majority of the data into my own database so that all usage and rates are through my service. There are some client based api calls that are made (mostly around existing enrichment) but 95% of traffic hits the db my workers populate.

Would love to take a deeper dive into what you feel would improve your experience.

Thank you for that insight, indeed k8s uses pascal case. I have opened an issue to investigate further and resolve. https://github.com/kubenote/KubeForge/issues/10

The color of the fields is not based on right or wrong but based on typing. Green is string, red is boolean, orange is object, purple is array, etc. I can definitely make that more apparent. https://github.com/kubenote/KubeForge/issues/11

As for the empty object, there are a lot of objects are that incomplete that I need to go through and apply manual override definitions for (https://github.com/kubenote/KubeForge/blob/main/src/componen...). Thank you for pointing out the two above. https://github.com/kubenote/KubeForge/issues/12

For the fields that exist but aren't feasible for users to interact with, currently I'm puling in the schemas raw. but I do want to create a preferred fields override with commonly edited and changed fields to reduce clutter. https://github.com/kubenote/KubeForge/issues/13

Thank you for the awesome feedback!

There is a lot to unpack here! Thank you for the feedback!

1. For the errors on downloading the schema, that is something that I need to bake through and improve. (https://github.com/kubenote/KubeForge/issues/5)

2. To minimalize scrolling around for deployments I want to incorporate a "minimize" feature on nodes so that users can free up space on the graph. (https://github.com/kubenote/KubeForge/issues/6)

3. Interconnectivity is built in but currently only with inline data on nodes. Adding the capability to object refs does make sense. In the future I also want to create generic "string nodes" as a first pass at templating via the node graph. (https://github.com/kubenote/KubeForge/issues/7)

Thank you for kicking the tires and giving me feedback on how to improve! Did you find the concept refreshing? And what would you like to see in it for it to be something that you would actively use?

Created the issue, there is was a bit of a design flaw preventing me from doing this off the bat, but I do think it is important and would expedite understanding from the demo. Thank you!

https://github.com/kubenote/KubeForge/issues/4

Please let me know if there are any features you enjoy or if there is something you would like to see added!

I have a extended future roadmap, depending on what the community would like to see, but adding in deployment from the ui and cluster management/analytics is also on there. However that goes a bit beyond my current skillset, so I wanted to launch small and build.

Are there any other features that you felt are missing or that you think would complement KubeForge?

Thank you! I'll double check the schema objects for metadata node, most likely some issues with how I have it defined.

For the overlapping nodes, I'll also throw an issue for both of those in GH :)

Did you like anything in particular or have a feature in mind that would make this better to you?

--- edit ---

For the metadata node warning, it was stating that the node is created but does not have values inside of it.

--- edit 2 ---

resolved the get.kubefor.ge/latest issue with: https://github.com/kubenote/KubeForge/issues/1 and removed undefined + improved warnings with: https://github.com/kubenote/KubeForge/issues/2

Thank you for the insight, for the initial version I wanted something that laid down the groundwork for defined schemas. I also wanted to make sure that I handled data conversion from yaml -> json object -> yaml correctly.

I love the idea to be able to import helm url's (and possibly edit + rehost), the next features are going to be around naming required/most used fields for simplification and decluttering the node components, and adding in support for helm and kustomize with template variables.

Was there anything else that you noticed that you would like to see or anything that you saw that you enjoyed?