ART feature requests and discussion

[spidermonkey]

Great, look forward to trying it, thanks for your work…

[eyedear]

@europlatus

A Reset button for the whole tool (maybe next to the power button). This would reset all changes made in the tool to default and negate the need to reset each individual slider.
Going one step further and in the spirit of keeping ART streamlined and less cluttered, why not remove the reset buttons for each slider, and instead use a double click and/or keyboard modifiers for the reset functions? This would remove a lot of graphical elements for a cleaner interface

any chance these feature would be considered for implementation in the near future. it would make the reset process faster

[agriggio (Alberto)]

Hi, a global reset for each tool is already in place… What version are you using? (Removing individual resets is not going to happen though)

[eyedear]

i just download the latest build 1.7.1-91 and found the global reset. So a mouse double click to reset the slider is not happening ?

btw thanks for the amazing work

[agriggio (Alberto)]

Hi,

@eyedear

So a mouse double click to reset the slider is not happening ?

If someone provides a patch, I’ll be happy to accept it, but I’m not planning on doing this myself, sorry (it seems it would require a bit of gtk tinkering)

[eyedear]

Noted thanks

[stuntflyer (Mike)]

@agriggio

You’re right, there’s no such thing in ART, and no plan to add one. It just works differently. You can simulate RT spots by using a combination of area and color similarity masks (moving the spot in RT would correspond to picking a different point for the color similarity mask), but it certainly won’t have the convenience of RT. If you find that you prefer that kind of interaction, ART is probably not the right choice.

Honestly, after trying the way you described, I do find that it’s implementation of using area and similarity to simulate RT spots works perfectly. I guess I just needed to know the best way to go about it.

Thanks!

[stuntflyer (Mike)]

@agriggio Can we get another color added to the background color selector for the preview window which is somewhere between black and middle-grey?

Thanks

[photomar (Charles)]

I have a request to make. In the Exposition tab, there are plenty of ways to adjust the luminosity: Exposure, Tone Equalizer, Tone Curves, etc. However, the traditional trio of global Lightness, contrast and saturation adjustments has been removed from the Exposition tab. True, the contrast and saturation sliders can be found at the top and bottom of the Tone Curves panel, but they are awkwardly placed. The traditional trio can be found in the Color Tab in the Lab panel. It is a bit of a pain to have to switch back and forth between the Exposition tab and the Color tab to effect those lightness adjustments. Of course, we can use the Tone Equalizer sliders to achieve the same result, but it requires messing about with all 5 sliders instead of the single lightness slider.
Is there any chance you might consider bringing back the lightness slider to the Exposure Tab?

At any rate, a huge thank you for your great work.

[agriggio (Alberto)]

Hi,

@photomar

Is there any chance you might consider bringing back the lightness slider to the Exposure Tab?

No
Cheers

[photomar (Charles)]

Too bad , I tried… thanks anyway!

[stuntflyer (Mike)]

When editing multiple photos at a time, I have to activate the clipped shadow and highlight warning when I start on each photo. If I close one of the photos and reopen it, I have to reactivate the clipped warning again. Would it be a good feature to only have to click once per editing session, regardless of how many photos I open or close?

[photomar (Charles)]

Not sure where to post that question: A friend of mine whom I introduced to this software asked me if bitbucket.org was secure from viruses. He’s using Windows. I could not find any useful information on the web except from a major incident on Feb 5 2020. Does anybody have additional info regarding this?

Thanks!

[elGordo (Gord)]

The Feb 5 2020 incident is the only one I am aware of that was specific to Bitbucket. Previously, there was an attack on hosted Git repos at GitHub, GitLab and Bitbucket in 2019:

Git Ransom Campaign Incident Report - Atlassian Bitbucket, GitHub, GitLab -...

But at the end of the day, if you want collaboration on an open source project, you pretty much have to pick a place to host the repo. Microsoft (GitHub) and Atlassian (Bitbucket) both have many large enterprise customers and would suffer major reputation damage if they couldn’t keep their customers’ repos safe, so I think you can be pretty sure that they put a lot of effort into keeping the platforms clean.

[agriggio (Alberto)]

@elGordo

so I think you can be pretty sure that they put a lot of effort into keeping the platforms clean.

On the other hand, please remember that the software comes with no warranty. If someone is uncomfortable downloading and/or running it, (s)he should just refrain from doing so I’m afraid.

[ggbutcher (Glenn Butcher)]

One thing that would help (and I need to start doing it too) is to post md5 checksums with binary releases. That way, the pedigree of the binary with respect to what the developer intended can be established. Even if one questions the posted checksum, it can still be conveyed in other channels.

[elGordo (Gord)]

@agriggio

@elGordo

so I think you can be pretty sure that they put a lot of effort into keeping the platforms clean.

On the other hand, please remember that the software comes with no warranty. If someone is uncomfortable downloading and/or running it, (s)he should just refrain from doing so I’m afraid.

Absolutely, if a potential user is uncomfortable with the origin of a package, they should not download it, in particular if their concern is well-founded. While this thread is about ART, it’s important to note that the concern is not specific to ART.

@ggbutcher

One thing that would help (and I need to start doing it too) is to post md5 checksums with binary releases. That way, the pedigree of the binary with respect to what the developer intended can be established. Even if one questions the posted checksum, it can still be conveyed in other channels.

Providing a checksum is a very good idea. However, MD5 has been deprecated, and a stronger algorithm would be better.

[gaaned92]

I suspect that the checksum only permits to be sure that what you have downloaded is identical to what is on the remote site.
But it doesn’t permit to be sure that what is on the site was really uploaded by the owner of the repository and not by any hacker.
It cannot be considered as a signature.

[agriggio (Alberto)]

What I can do is post the sha256sums here for the “official” tarballs, if that helps.

[ggbutcher (Glenn Butcher)]

Checksums posted to the same site as the file would be subject to that vulnerability. However, if I sent the checksum via a different channel, say, as a post to this forum, that vulnerability would be reduced to compromising both the file at one server and the database serving this forum. A better channel would be a direct email exchange, preferably with encryption. None of this is perfect, but it does put burden on the hacker that can be made to be onerous…