[boud] > Sorry, that's just something you'll have to dig in for yourself. It's not > worth my time at least to get you started on something like that: it will not > help me achieve my goals.
I guess it depends on the scope and ability of parallel works to give back to Krita. KOffice/Calligra was apparently not in the end closely related enough to bring more value from the shared portions than the cost of maintaining those portions for the different use cases. However this historical event may not be a definitive proof that it is not worth sharing Krita code with derivative products. As said many time, Krita focuses on drawing workflows and its restricted scope allows it to excel at it and gain market share and donations, in return funding further development. Photo editing and graphics design are larger markets but are even more entrenched as they are mostly done by paid people instead of having an active community of artists around it. But then again, as Krita gain more drawing related features that happen to also benefit the other workflows, it gets potentially significantly more attractive to those markets *if* there was some work on missing core features and an UX optimized for them. Bloating Krita into a jack of all trace raster and vector graphics design apps would most likely overwhelm the (already crowded) GUI and make it less attractive. >From that point of view, maybe having more generally reusable libraries within Krita would improve the situation. In no means should you even consider freezing their API or officially supporting them when used by 3rd parties. That adds development and maintenance cost that indeed prevent you from achieving your goals. But lowering the cost of new GUI experiments and alternate workflow can maybe help second or third parties implement toy GUIs and maybe contribute back some features that may eventually lower the cost of getting an official Krita derivative (Karbon14 new generation?) out of the door to tap into a new donation market. It seems a generally "low cost" avenue to take a bet toward getting closer to achieve your goals. I would love a better alternative to GIMP to exist as a companion (or even better, directly integrated) app for Digikam. I do some hobby photography and often have to use GIMP or, back then, my old PhotoShop 7 license to remove electric wires and similar changes. GIMP user interface is ridiculously illogical and unintuitive. Krita one is much better but lacks some photo manipulation features I like and I acknowledge it is not designed for the same kind of work. I guess I am not alone on being on the "edge" of being able to enjoy all the work invested in Krita but who have workflows it isn't optimized for (or lack features). [Kuntal] > So a couple us are trying to build a raster graphics editor which looks and > behaves similar to Photoshop with the help of Qt5. As everybody here said, it's a lot of work for a market that has been proven not to exist. I wont repeat what was said above, but just add 2 examples: The first one is the original GimpShop. Back in Photoshop 7 days, GIMP was mostly on par when it came to features beside non-RGB color systems. So someone just forked it to clone the Photoshop menu and tools layout. It was a novelty for a time, but didn't take Photoshop crown when it was technically close to be on par with it. Then someone usurped the project brand recognition to distribute malwares and the current "GimpShop" has nothing to do with the original fork. The second was called Pixel. This person made a shareware that really, really cloned Photoshop. It was ported to every operating systems under the sun[1], even the most obscure ones. You could pay a very small fee to get the version that didn't add random watermarks from time to time when you saved. The GUI was a 1:1 clone and all the features actually worked surprisingly well. Still, that was a dozen year ago and where is it now? Apparently people who want Photoshop will use Photoshop and no amount of time and love will fix it. Plus, the time it takes to maintain such large software is apparently requiring some sort of full time developers as demonstrated successfully by Krita (please everybody, consider donating to their ongoing fund raising). My advice would also be to contribute to Krita or a derivative to better tap into the other workflows (pick one and master it). Boudewijn and the other Krita contributors proved that you can compete with Photoshop if you have a laser focus on the needs of the group of users you target and love your work and your users. But as far as just cloning it and hope for the best, I think that avenue is totally hopeless. Cheers, Emmanuel Lepage [1] http://www.kanzelsberger.com/pixel/?page_id=5 On Sat, 22 Sep 2018 at 09:43, Boudewijn Rempt <b...@valdyas.org> wrote: > > On zaterdag 22 september 2018 15:38:03 CEST Luigi Toscano wrote: > > Andy B ha scritto: > > > Can you guys maybe now move this discussion to telegram or phabricator? It > > > seems that there is good debate and good will. > > > > I'm not sure why this discussion should be diverted from a mailing list to a > > messaging system (*) and a planning and development system like > > phabricator. > > Yes, that's kinda weird. Since Fixture isn't a KDE project, we can hardly > discuss it on Phabricator. Not that I think we'll need much more discussion > until something material happens that makes it interesting again. > > > Maybe (just maybe) it can be be moved to a more appropriate mailing list. > > I'm fine with Hellozee mailing the kimageshop mailing list asking for > information; that would probably be most appropriate. > > > > > (*) let's not use "telegram" as synonym for "messaging system", please. > > Amen to that. > > -- > https://www.krita.org