PullJosh wrote:

Hey Bookowl! Do you think you could find a solution to the jumping hover issue and fork my pen with the solution? I don't care if the solution requires JS… just get it to work please.

Firedrake969 wrote:

I'd suggest using x10, not 000webhost. The latter tends to be somewhat unreliable, and puts ads in your page.

GH pages works with custom domains (search for CNAME file) and is super easy for everyone to contribute to. Let's use that. (Oh, and they give crazy fast load times.)

BookOwl wrote:

PullJosh wrote:

Hey Bookowl! Do you think you could find a solution to the jumping hover issue and fork my pen with the solution? I don't care if the solution requires JS… just get it to work please.

Firedrake969 wrote:

I'd suggest using x10, not 000webhost. The latter tends to be somewhat unreliable, and puts ads in your page.

GH pages works with custom domains (search for CNAME file) and is super easy for everyone to contribute to. Let's use that. (Oh, and they give crazy fast load times.)

OK, I'll take a look.
I agree to using github pages.

PullJosh wrote:

Hey Bookowl! Do you think you could find a solution to the jumping hover issue and fork my pen with the solution? I don't care if the solution requires JS… just get it to work please.

Firedrake969 wrote:

I'd suggest using x10, not 000webhost. The latter tends to be somewhat unreliable, and puts ads in your page.

GH pages works with custom domains (search for CNAME file) and is super easy for everyone to contribute to. Let's use that. (Oh, and they give crazy fast load times.)

BookOwl wrote:

I am trying to experiment with convolution kernels for stuff like edge detection. Here is what I have so far: https://scratch.mit.edu/projects/76374468/
It doesn't seem quite right though - it seems very over reactive.
The kernel I am using is
[ 0 1 0
1 -4 1
0 1 0 ]
Could somebody else take a look at it, please?

Layzej wrote:

BookOwl wrote:

I am trying to experiment with convolution kernels for stuff like edge detection. Here is what I have so far: https://scratch.mit.edu/projects/76374468/
It doesn't seem quite right though - it seems very over reactive.
The kernel I am using is
[ 0 1 0
1 -4 1
0 1 0 ]
Could somebody else take a look at it, please?

It's over my head, but @gtoal has this hidden in some of his projects. You may want to take a peak inside one of his.

PullJosh wrote:

I'll create a Github organization. We need a name, though. What should I call it?

BookOwl wrote:

PullJosh wrote:

I'll create a Github organization. We need a name, though. What should I call it?

ScratchNPR-Website?
Did you see my post?

ChocolatePi wrote:

BookOwl wrote:

PullJosh wrote:

I'll create a Github organization. We need a name, though. What should I call it?

ScratchNPR-Website?
Did you see my post?

Why not just ScratchNPR? Also gh org names iirc never have hyphens and are CamelCase, while repo names have no uppercase letters and are snake-case

gtoal wrote:

studio question… it currently says “75 projects” but the display is a number of complete rows of 4, so there can't be exactly 75 showing - however there's no “next page” button anywhere! Where are the missing projects? (at least one example: my warholiser is not on the page anywhere)

Layzej wrote:

gtoal wrote:

studio question… it currently says “75 projects” but the display is a number of complete rows of 4, so there can't be exactly 75 showing - however there's no “next page” button anywhere! Where are the missing projects? (at least one example: my warholiser is not on the page anywhere)

Scroll down and more projects should appear…

ChocolatePi wrote:

BookOwl wrote:

PullJosh wrote:

I'll create a Github organization. We need a name, though. What should I call it?

ScratchNPR-Website?
Did you see my post?

Why not just ScratchNPR? Also gh org names iirc never have hyphens and are CamelCase, while repo names have no uppercase letters and are snake-case

BookOwl wrote:

PullJosh wrote:

I'll create a Github organization. We need a name, though. What should I call it?

ScratchNPR-Website?
Did you see my post?

BookOwl wrote:

PullJosh, I can't seem to find any issues. Could you please describe the problem to me?

gtoal wrote:

Also I notice that when you remove the mouse from the zoomed image, it fades gradually but when you move into the image it appears immediately. It would be better to fade in the same way as it fades out.

PullJosh wrote:

It seems that the issue is rather sporadic - sometimes it works, sometimes it doesn't.

BookOwl wrote:

I am trying to experiment with convolution kernels for stuff like edge detection. Here is what I have so far: https://scratch.mit.edu/projects/76374468/
It doesn't seem quite right though - it seems very over reactive.
The kernel I am using is
[ 0 1 0
1 -4 1
0 1 0 ]
Could somebody else take a look at it, please?

Layzej wrote:

brightness is often a negative number. You then set r, g, and b to brightness. Something is wrong there since you can't have a negative colour… can you?

Maybe normalize the result to 0-255? or pink if brightness >20, grey otherwise?

PullJosh wrote:

Also, I'm on Chrome (and I'd wager gtoal is too).