lphilpot

Then again, as I recall you Texans got to see Comet McNaught (C/2006 P1) in the daytime during early 2007 while farther east we were totally overcast for the entire apparition. And that's way cooler than an eclipse. 🙂

Every time I think about the crowds I lose all enthusiasm for the trip. After all, it's only 4 minutes out of an entire week... But we're paid up, so might as well go.

Ah ha... I'll blame it on trying to pack for our eclipse trip while replying... 🙂

It shouldn't be hard. Offhand I don't recall the syntax for creating a symlink on Windows, but it's not hard. Generically, the process would be something like: Make sure no Affinity app is running. Move the actual folders to wherever there's room, i.e., the new location (or copy then delete the original, just to be safe) In the original location, create a symlink(s) with the same exact name as the folder(s) you moved / deleted, but pointing to the new location instead Now when Windows, Photo, whatever, accesses those folders it will actually be accessing the ones in the new location, but it'll never know the difference. This technique is used all over the place by unix / Linux (in fact by many OSes).

At least you didn't say "wax"... 😛 In my world, that was for newspaper and quick-print types. LOL Letraset, Chartpak, Zip*a*Tone, Amberlith, Rubylith, acetate overlays, PMT cameras (shooting stuff from the VGC catalog) ... ah, the "good" old days. But I did enjoy the "construction" aspects of making presentation mock-ups. I even went as far as making embossing "dies" from matboard so I could emboss the mock-ups.

More than once when I worked in desktop support I had to help folks with email issues on their phones. They were sitting at their desk, 27" monitor(s) in front of them, Outlook open full-screen and complaining that this huge email was difficult to read on their phone. SMH 🙄

Possibly my comment apparently had a part to play in this discussion (?), so let me clarify. I personally try to be a realist and I tend to not get excited in general, but I'm not chronically pessimistic. I've tended to, in output terms, "under-promise and over-deliver" (if possible) while in intake terms, "under-expect and over-receive". The latter is a conscious thing, intended as a buffer against disappointment from unrealistic expectations. While I agree in general with @debraspicher about refraining from unwarranted classifications, so to speak (if I understand correctly) I also agree with @loukash in a general sense. However I didn't take his comments as specifically descriptive of me, but rather his view in general.

I dunno ... after 6+ decades of dealing with vendors, businesses, people in general and everything else I've learned that more often than I'd like (in that context) pessimism is often realistic. And in the long run, less disappointing.

Exactly. I'll take realism -- even if it superficially looks pessimistic at times -- rather than rose-colored glasses. I don't like surprises. My ultimate advantage over many here is that I'm not making my living with Affinity. I feel for those who are in an uncomfortable (to say the least) situation at the moment. The principles and fallout still apply to me, just not as potentially painful nor disruptive. And again, I hope it all works out, but I'll have to see it with my own four eyes. 🙂

I think it could depend on the level and nature of the AI services being offered. For 'from-scratch' AI content generation, yes, there will be an ongoing cost which could require continuing revenue / licensing. But what about much smaller scale AI logic attached to 'everyday' editing capabilities: masking, fill, error-fixing, etc.? It seems those capabilities could be supported under a perpetual license, as (from what I understand as a non-AI-expert) they're ultimately more logically oriented. Once algorithmically-seeded, they would seem to be less reliant on continual 'feeding'. I, for one (and I've seen the same expressed by others here), would greatly welcome AI-assisted editing but have no interest in wholesale AI-based content generation.

It doesn't work for layer masks and it creates a selection. What we need is a way to see the pixel layer while editing the mask (i.e., after Alt+click on Windows). It's been a while, but IIRC PS (used to?) shows a mask as a translucent red overlay through which you see the pixel layer. E.g., red == black and as you apply white it goes transparent and the pixel data becomes visible through it. Quick Mask mode initially looks like that but nothing shows through and a selection is created (which is unneeded).

FWIW as a landscape photographer, personally I'd like to see in Photo: Editing of masks with their pixel layer visible; otherwise the mask editing process is blind to either the pixel layer or the mask itself. This could be as simple as making the mask a translucent color over the pixel layer (with a hotkey toggle). AI features (please note I'm limiting AI to editing, not content creation) -- Improvements to inpainting / content-aware fill and / or generative fill of some sort Better, more intelligent masking AI-based sharpening Adjustment brushes (e.g., as in LR / ACR), preferably non-destructive to whatever they're adjusting Better, more parametric selections by color A non-destructive, re-editable gradient tool usable on masks / pixel layers; yes, I can (and have) used the existing gradient tool on a fill layer but its clumsy A true scripting / macro language with document and app hooks; IOW, the ability to record macros that can automate tasks but with dynamic variables. That could also enable the growth of a plugin ecosystem. The ability to toggle a layer off / on with the same keystroke Add some kind of visual flag to indicate that a (and which) layer is soloed Support tiled document windows in the UI Make the tools panel a first-class dockable-anywhere window (like the other panels) A customizable status bar, where we can specify which app / image information variables to include Make the drop-down menus left align with their parent by default (like virtually every other app) Make the UI font size (and even better, the font itself) a preference Make it possible to copy, paste, cut, move, etc., selections on a pixel layer without creating a new layer Fix whatever is causing the massive (2x, sometimes more) bloat in *.afphoto files when initially saved Remove (or make a preference) the presence of the default snapshot.

The problem is, these "pledges" say nothing specific. For instance, the perpetual licensing -- Is it for all products, all levels, all support plans, all extras, etc.? Forever? Or just a "lite" version one step above free, with minimal capabilities? We don't know because it doesn't say. Pledges are meant to sway the receiver to believe intent. Nothing more. The proof will be in the pudding and unfortunately we won't know that for years in a positive sense. However, in a negative sense we may learn far sooner than we'd like.

Any time marketing-speak like "empower everyone", "vast range of templates", "anyone can take an idea and create something beautiful" and such gets thrown around, run the other way. It's just code for, "we're dumbing down the product". Anything worth using is worth learning. I'm not advocating needless complexity but, e.g., you can't write an operating system with batch files (to reference another area).

...and usually free versions aren't worth using. You have to pay anyway. As someone already pointed out, TANSTAAFL. That's one of the big points here -- Affinity is at serious risk of being dumbed down for general consumption, rather than being enhanced for professional (level) usage. There are plenty of free tools already. But despite the damage control follow-up announcement, I'll believe it when I see it. That's just SOP during takeovers -- Promise nothing will change, then pull the rug out. I hope I'm wrong, but Serif's trust level right now is near absolute zero.

I've been a low-vis supporter of GIMP for years, but eventually moved to Affinity for those very reasons. It appears possibly things are changing now a bit (more openness, more acknowledgement of the Real World, less semi-snarky "it'll be ready when it's ready" attitude). I'd love to be proven wrong, but I think in the current context GIMP may have done relatively too little for far too long to claw it's way up to being a viable alternative for even advanced amateurs, much less professionals (at least in the photo editing space). Again I'd like to be wrong, but ...

That's the one fact that can be taken to the bank. NONE of this is primarily for the users' benefit. Yes, there may be beneficial side-effects along the way, but it's All About the Money. 100%. Canva -- and every other vendor out there -- will do whatever benefits their bottom line, regardless of what the users ultimately want or need. That's the ugly underbelly of the investor-owned / corporate business paradigm. I.e., the wants of the one outweigh the needs of the many.

At an absolute minimum I'm putting an immediate halt on any efforts to further learn Affinity -- At least until it's clear it's worth the continued effort. I'm no Adobe fan but I'll go to Photoshop's superior toolset in a moment if Affinity goes subscription (because at that point it's no contest).

As I indicated before, we'll have to agree to disagree. My experience informs what I've stated and you don't agree. That's fine -- But there's no point in just tit-for-tat gainsaying each other. Whatever. I'm done here.

Oh believe me -- In more than two decades in IT I had more support nightmares than I've managed to forget. 🙂 Registry-diving was just a trivial example. It's not cross-platform nor is it advised for those not comfortable with it. I was just illustrating that it's not complex. However, I've used software that installed an icon / choice on the Start menu (or equivalent on other OSes) labeled something like "Support Information". Just run that, it asks for contact info (only if they didn't opt to supply it during install) and sends correct, detailed and sufficient information about your installation directly to the support team. That's all the user has to do - Run one menu item. A ticket is created automatically at the vendor and the user gets confirmation. The next step is the solution from a support engineer. I used a very similar process from SUN Microsystems many times. Couldn't be easier. And about install complexity from a user viewpoint -- For years KDE had a standard practice of "hiding" all optional choices behind an "Advanced" button. There was no requirement to click Advanced, but it was there for those who need / want it. I've not used KDE in a long while, so I'm not sure if they still do it that way but back in the "Mosfet vs. Havoc Pennington" days it was standard.

Hence my statement that if you want it, install it. All features could be checked by default, that's fine. Just allow some customization. That way you don't penalize some for the benefit of others. It's a win-win. Again, if they were enabled by default, Bob's yer uncle. No need to re-run anything. Just accept the default and hit "Install". What I mean is, code that's not installed doesn't have to be upgraded, nor updated, nor virus-scanned, etc. It has no security risk. It takes no space on disk and uses no memory nor CPU. Installing unused code is like filling the trunk of your sports car with cement, just because you can. That capability already exists in many (most?) installers (InstallShield, Wise, NSIS, etc., IIRC). All they have to do is use one that's up to the task. That's called documentation, which is also lacking. Serif should produce complete, comprehensive docs on every thing every app does. Users (current or prospective) should read and understand it first. It's worked that way for decades. And any good installer can handle (inter)dependencies, even in fairly complex scenarios. Ever installed Oracle on *nix? (and that's not even a super-good installer) They're developers. It's trivial to record and report on what's installed, both by machine- and human-readable methods (i.e., Help | About). From a Windows perspective you can even just look in the registry at the Uninstall key for that matter (but there are better, more portable strategies). I've used more than a few packages where the very first step in any support scenario is to run a command / utility / script that queries the system and reports everything needed to the vendor. Many of these tools will automatically email the results to the support team. Standard practice. I dunno... I mean, if a user isn't interested enough to even research and learn (about) a product, how serious can they be about using it? I promise you, developers read their documentation. Anyway, I think we need to agree to disagree and move on. 🙂

Yeah, MS broke that functionality partially in 7 and completely in 8 / 10. I'm in no way advocating a return to XP, but the way associations were exposed in userspace back then was hugely better for users than the way it's obfuscated now. Apparently even developers can't (easily?) change it now, although Serif seems to...

I'm using the MSI versions as well. That was a more general point, although I'm curious if the Affinity products could be installed to a network share for multiple users.

I'm not uninstalling / reinstalling LibreOffice to grab a screenshot, but here's a related setting that illustrates a very simlar point: This allows fine-grained control of how LO reads, converts and writes MS Office files -- Which is a very closely related concept to the OP's post.

Well, for example while I feel certain Photo isn't currently architected to support this, given its size it would be really nice to choose which personas to install. I personally use only the Photo. I've looked at the Develop persona but it doesn't provide the features I get from the standalone raw processor(s) I use. Liquify, Export and Tone Mapping do nothing I need. I've opened each once out of curiosity then never again. If others use them, great -- Select them for installation. But if not, why waste the resources to install, keep and maintain them? There's also in this context a minor -- but still non-zero -- security aspect to this, since any code addition statistically adds to the risk / compromise exposure footprint, so if it can be reduced so much the better. Other options might include which import / export filters to install, whether to install the "help" system / documentation, additional provided (or third-party) color profiles, etc. Apps like the Affinity suite are inherently targeted at experts* and should provide options accordingly, rather than being like a one-click phone app install. We're not talking about TikTok... * That is, professionals and / or those who operate in an essentially professional manner (regardless of income or not).