Request for feedback - Avalonia Accelerate 🚀 #16997
Replies: 39 comments 116 replies
-
|
I have a few questions:
Overall a lot of these sound fairly useful. |
Beta Was this translation helpful? Give feedback.
-
|
We're not yet ready to talk about the technical details, but to answer a few of your questions: Is the Visual Designer planned to work like the Visual Studio one for WPF Is Avalonia Edit going to be deprecated? |
Beta Was this translation helpful? Give feedback.
-
Good to hear you have avoided the biggest issue with the WPF designer. I never really understood what the actual point of it was when it failed to do layouts properly. |
Beta Was this translation helpful? Give feedback.
-
|
Several HTML page designers somehow made it better than what WPF was doing. Conceptually, it's very similar to what Avalonia needs. |
Beta Was this translation helpful? Give feedback.
-
|
Amazing! Really looking forward to some of these new features, even if I think some of these (like the Markdown Renderer in particular) could be a free offering as well. Also, I assume the "Enhanced Code Completion" in IDEs will come for free users, as well? It'd seem a bit weird to me if there were 2 separately maintained products for this. All in all, I'm really happy to see the commitment to further Avalonia development! 🚀 |
Beta Was this translation helpful? Give feedback.
-
|
I too agree that Markdown Renderer and Code Completion should be a default Avalonia features, these are two most criticized items among newcomers which I tried to invite to Avalonia framework development, given current state of other (not C#) UI design frameworks |
Beta Was this translation helpful? Give feedback.
-
|
I hope dev tool related (devtool, designer, hotreload) and complex controls can be purchased separately. |
Beta Was this translation helpful? Give feedback.
-
|
We might break apart the bundle at a later date, but right now the plan is to sell it together. |
Beta Was this translation helpful? Give feedback.
-
|
I agree an à la carte model is important when discussing the wide number of features above. Specifically, controls should each have their own licensing option -- reasons are in my comment below. |
Beta Was this translation helpful? Give feedback.
-
While we might consider individual offerings in the future, we're starting with a bundled approach. This aligns with industry standards, simplifies pricing, and allows us to include niche tools that might not be viable as standalone applications. |
Beta Was this translation helpful? Give feedback.
-
|
.app (and all the associated nonsense that goes with it) : It should support universal binaries (x64 and Apple silicon). I dit it (manually then in our CICD) but it is pain in the a$$... :) |
Beta Was this translation helpful? Give feedback.
-
|
Absolutely! Thats on the list. |
Beta Was this translation helpful? Give feedback.
-
|
When will it be launched? |
Beta Was this translation helpful? Give feedback.
-
We're hoping to have phase 1 early access by the end of the year. But it's ready when it's ready; we'd rather ship quality than ship to a deadline. |
Beta Was this translation helpful? Give feedback.
-
|
No one forgives a vendor for delivering a flawed product early or on time. Everyone prefers a working product, even if it's late. |
Beta Was this translation helpful? Give feedback.
-
|
Hello @MikeCodesDotNET I would like to ask, if there is more improvement on going for touch and animations? As you know I recently focus to work more on Shell, and sync between animations and touch gesture is missed today. And solution to control the animation timing. Thanks in advance. |
Beta Was this translation helpful? Give feedback.
-
|
Animation rework is on our v12 roadmap. |
Beta Was this translation helpful? Give feedback.
-
|
Hi, Regarding the TreeDataGrid, will there be any bug fixes before that licensing change? I currently use it in both hobby and professional codebases and I want to mention that it is in a state where I personally do not feel like it would be worth paying for. To me, it could be (easily perhaps) if 2 issues were addressed
|
Beta Was this translation helpful? Give feedback.
-
|
Also just a confirmation question: the previewer is planned to stay outside of this right? However, the current Rider extension is much more powerful (and actually very helpful) compared to the VS extension. Is there a plan to make the VS one as "rich" as the Rider one? |
Beta Was this translation helpful? Give feedback.
-
|
Good to know, I was not aware of this. As an enterprise user that does more of embedded development I am really interested in anything that makes me faster and safer. And I would love to pay for that such that the AvaloniaVS "team" can make it as best as it gets. |
Beta Was this translation helpful? Give feedback.
-
|
There is an existing external previewer. Not sure how much better it is though. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, I believe the previewer is already good. It seems a bit less stable than the Rider one but you get around that. The more advanced you get with Avalonia XAML code completion is less an issue. The XAML warnings prevent you however from doing mistakes. The GoTo from XAML bindings to VM properties do save time. If I had to pick one thing it would be the latter... Not sure if VS would allow for that. This is getting off topic... as always... |
Beta Was this translation helpful? Give feedback.
-
|
@llfab The Rider extension is just the previewer and nothing else. Rider natively understands Avalonia XAML, so the code completion and refactoring features are second to none. You'd get all of this even without the Avalonia extension installed. As for Visual Studio, we plan to create a significantly improved extension. That will form part of this bundle, but we've not decided how to ensure that even free users get some of the benefits, and we avoid maintaining two distinct versions. |
Beta Was this translation helpful? Give feedback.
-
|
This sounds very good! |
Beta Was this translation helpful? Give feedback.
-
|
Great to see all the work on new features planned above! A lot of it makes sense and I'm sure it's going to be a big boost for both developers and Avalonia growth itself. My primary concern is with license model changes. I never like to see a license change from MIT to something much more restrictive. I'm talking specifically about the TreeDataGrid control in this case (for commercial use). I know the team really debated even putting the source code out there in the beginning under MIT so some leeway has to be given here. It's a lot of unpaid work so to speak. However, several companies (some of which I work with) view licensing changes unfavorably. While this certainly isn't a bait-and-switch it is still difficult to stomach. If you can justify doing this for one control, that type of thinking can always be expanded to other areas. It puts up a caution flag for sure. A lot of companies see this type of thing and immediately deprioritize or move away from certain software modules as well. We've seen countless examples of this over the last few years: Unity, ImageSharp, etc. that has resulted in a lot of blowback and negative press. Of course that is the extreme and we are only talking about one control here. Personally, I think the problem comes down to the subscription model itself tied to Avalonia Accelerate. As discussed before, I'm not sure you will allow unlimited lifetime use of the version that was first subscribed to. That essentially means applications have to keep paying for a license even if they don't need the latest code or aren't updating the software themselves. That becomes extremely cost prohibitive over time. Keep in mind there are a lot of passion projects out there that aren't open source but also operate in the red. Having to pay for a license in perpetuity in this case just doesn't make sense -- it would be taking on a lot more losses. There are two options that would work while keeping a subscription model:
|
Beta Was this translation helpful? Give feedback.
-
|
The TreeDataGrid license has been a subject we've discussed a considerable amount. We're very aware that the change will be controversial, but as @grokys has already stated, it doesn't get much focus at the moment unless a paying customer requests support. Ultimately, every decision we make is about finding the right balance. From a commercial perspective, we need to generate revenue to continue investing in Avalonia, which is crucial for our growth and sustainability. This license change for the TDG is the most reasonable solution to enable us to invest more in that control. If organisations aren't inclined to purchase a subscription to access the Avalonia Accelerate features, that's entirely their prerogative. They'll remain free to use the older, unmaintained MIT version of TreeDataGrid. They can continue their Avalonia projects without financially supporting the project, which is part of the reality of shipping open-source software. |
Beta Was this translation helpful? Give feedback.
-
|
@grokys Yes, I'm talking specifically about the licensing and cost changes to the TreeDataGrid control. However, I don't think we are on the same page. I'm (we're) talking about using a Rider pricing model in this case even after you switch licenses and go to subscription. One specific example: I can't ship with the TreeDataGrid right now due to it's bugs -- even the current MIT version. That means no matter what I'm going to have to use some future version of this control tied to whatever licensing model you are switching to. I was hoping to be able to pay support and fix what I needed once the list of needs is clear. If this was Rider we would have no problem: I would just pay the $170 a year fee and keep going. Happier even as I get a better supported control with several fixes. There is little risk and no commitment. With Avalonia Accelerate this is a LIFETIME commitment which amounts to likely 10s of thousands of dollars -- for one control (I think you will be asking 2 or 3 times what dotUltimate costs). Do you see the difference here? The cost/benefit equation tips dramatically as soon as you remove "the last version you paid for can be used indefinitely". What was a no-brainer is now an extremely difficult to justify cost. All the while, the decision to re-write using the TreeDataGrid was taken because it was free at the time, works better than DataGrid, and works with AOT, etc. It was unexpected for you to change that but I won't repeat my other comments about the licensing which stand above. |
Beta Was this translation helpful? Give feedback.
-
|
@robloo, we haven't finalised any pricing decisions yet, which is precisely why we've created the survey. We want to hear from the community and understand their perspectives on value for money. Regarding the dotUltimate cost, are you looking at the 'For Organizations' or 'Individual Use' price? It's worth noting that the vast majority of our users are using Avalonia professionally. |
Beta Was this translation helpful? Give feedback.
-
Good call, I mistakenly just looked at the individual use pricing. So my numbers would have to be adjusted accordingly. Nevertheless, $470 (close to the 2-3x I was predicting) and you have a lifetime license at the initial subscription is still much different than 10s of thousands of dollars required for the lifetime of the app. Still quite different numbers and commitments/risk here. Bottom line, I think a different pricing model for CONTROLS that match industry expectations, other software, different use cases and company size is needed.
It seems compromising might also be more appropriate considering the history of this control specifically. Personally, I'm happy to pay as long as it's fair pricing for each use case. |
Beta Was this translation helpful? Give feedback.
-
|
@MikeCodesDotNET I still recommend to keep TreeDataGrid license as is, and re-branding the paid version to another name, like TreeGrid , to avoid any confusion. |
Beta Was this translation helpful? Give feedback.
-
|
Are you considering a different pricing for small businesses similar to how QT works? https://www.qt.io/pricing/qt-for-small-business |
Beta Was this translation helpful? Give feedback.
-
|
Pricing is still under consideration, which is why we've launched the community survey. Our aim is to set a price point that makes this subscription an obvious choice for all Avalonia developers, balancing value and accessibility. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for all the effort. But I believe Android AOT should be free because it is foundamental to dotnet ecosystem. |
Beta Was this translation helpful? Give feedback.
-
|
Determining which features should be paid or free isn't easy. Our general approach is to consider the engineering resources required for development and ongoing support. In this instance, Native AOT for Android demands substantial time and effort from multiple developers, which is why we've decided to include it in our paid subscription. Rest assured, running your Android apps on the Mono runtime will remain available at no cost. |
Beta Was this translation helpful? Give feedback.
-
|
No official Webview for Avalonia Hybrid? |
Beta Was this translation helpful? Give feedback.
-
|
For most people, Avalonia is expected to remain free, but I understand that any business needs profits to keep operating. But in the long run, I am not sure what the direction and goal of Avalonia's operations are. I personally feel that the Avalonia team is now in a hurry to realize. I am not sure whether the operational pressure from wages, R&D costs, etc. on the Avalonia team is really great. Simply put, I am worried that Avalonia will change its subscription rules significantly in the future, just like Unity and Oracle, and even include the part that originally belonged to MIT in the subscription. Or in the future, will the part of Avalonia that belongs to MIT be changed to commercial licensing, etc.? Why are we worried about this? Because for most people, especially small and medium-sized enterprises, the occurrence of unpredictable costs may cause the business to stop operating. Even rewriting the software on hand in other programming languages is a huge workload. |
Beta Was this translation helpful? Give feedback.
-
|
@kyuranger First off, I want to put your mind at ease: Avalonia is absolutely staying free, and we have zero plans to change the license. We know that our permissive license is a big reason why so many of you choose Avalonia, and we're not about to mess with that. Changing the license would be incredibly short-sighted, and that's not how we operate. We focus on the long-term impact of decisions on our community first and business second. As for why we're moving in the direction of paid tools, it's pretty simple; we've got big plans for Avalonia! We want to keep pushing the boundaries and build the best UI framework. To do that, we need to hire incredible talent, which isn't cheap. I've always tried to be open about our plans and what we're thinking. Last year, I wrote about our strategy for balancing growth and sustainability, which mentioned unreleased products, so I hope you can see that this has been in the works for a while. To give you additional context, I wrote the initial business plan for Avalonia ~2021, and the high-level plan was always to:
That's where Avalonia Accelerate comes in. It's not about changing course but enabling us to make Avalonia even better (for everyone). I understand why seeing new things go into a paid subscription might raise eyebrows, especially given what's happened with other tech companies. But we're not Unity or Oracle, and our culture and approach to community engagement is very different. I think it's important to emphasise that there's no obligation to purchase a subscription. Our strategy has always been to add value through new features and improvements, so you can continue building with Avalonia as you have for the last decade and ignore our paid offerings if you'd like. The core of Avalonia will remain free and open-source, just as it's always been. We're in this for the long haul and committed to growing Avalonia in a way that works for everyone. I hope this clears things up, but if you have more questions or thoughts, please don't hesitate to share. I'm always here to chat! |
Beta Was this translation helpful? Give feedback.
-
|
@kyuranger: as an enterprise user but also as a private sponsor I would love to express: nothing is for free. Non of us is working "for free". If it wasn't for enterprises paying for support agreements or custom development that is on Avalonia's roadmap anyhow, the development of Avalonia would potentially be 1 or 2 guys doing some small development on a Saturday. In the long run that would not be good and fast enough to keep up with the rest of the world. Hence, all of you, really each and everybody, should be happy about the recent development and the fact that these guys are creating a sustainable business model. Even for small businesses the risk of Avalonia becoming an after-work dev project is much greater than the risk of paying a small fee. Or take the other perspective: if each and every user paid 50 USD per year, we may not have this discussion at all. It is important to understand that OSS sustainability has not been solved in the majority of projects. Most of them only work because large enterprises rely on it and have paid devs to do it (e.g. SkiaSharp). As Mike pointed out, Avalonia will remain MIT. I would like to encourage all of you to think about what you could contribute even if it is not mandatory. So maybe contribute some time and write unit tests (something easy to be approved) or become a sponsor. Everybody can afford a few coins per month... |
Beta Was this translation helpful? Give feedback.
-
Thank you for answering my question. In my opinion, Avalonia should be comparable to Unity in the future. Unity is a cross-platform game development tool, while Avalonia is a cross-platform software development tool (although Avalonia can also develop games). I hope Avalonia will have better development in the future. Best wishes. |
Beta Was this translation helpful? Give feedback.
-
No, I am not complaining about paying. At least Avalonia is not forcing users to pay now. What I am worried about is that Oracle and Unity will suddenly change the subscription rules, which will cause a sudden increase in corporate usage costs. For example, like JAVA, many companies have changed from Oracle JDK to Open JDK because Oracle keeps changing the subscription rules. This is a huge project. This also puts a lot of pressure on the company's cash flow and has affected the normal operation of the company. As the person in charge of the company, I must consider this and choose the right technology stack. |
Beta Was this translation helpful? Give feedback.
-
|
Interesting. Lot's of good stuff. Would it be possible for the developers behind this initiative to simply form a new company that markets and sells these products? Perhaps XPF as well? Would that not be less confusing to explain? |
Beta Was this translation helpful? Give feedback.
-
|
To answer your questions: Would it be possible for the developers behind this initiative to simply form a new company that markets and sells these products? Perhaps XPF as well? We already have. It's called AvaloniaUI OU, and it's based in Estonia. It was formed by and employs the core Avalonia team. Would that not be less confusing to explain? I don't see the confusion. Avalonia, the OSS UI framework, is owned by AvaloniaUI, the company. It's similar to Qt, the OSS UI framework being owned by Qt, the company. Would that not be less intimidating to any competitors? Not sure I follow. Would that not be more fair to the open source nature of Avalonia? Again, I'm not sure I follow. But, my view is that OSS at the kind of scale we're operating at inherently creates challenges when it comes to fairness. |
Beta Was this translation helpful? Give feedback.
-
|
Lot's of very useful stuff, this will accelerate app development for sure! The only downside for me is the removal of TreeDataGrid, as I am currently using it in a few of my demo apps. However, since those apps are not official projects of my company, which prefers more widely used technologies, it would be difficult for me to justify a license. Would it be possible to keep the current license for the existing version of TreeDataGrid, and switch to the new license for future versions? |
Beta Was this translation helpful? Give feedback.
-
The MIT license legally cannot be revoked. It contains no provisions to do so. As such you are free to keep using the MIT licensed version forever. It will however likely stop working at some point if the base Avalonia makes enough changes unless someone forks and maintains it. Any fork would also be available under MIT. |
Beta Was this translation helpful? Give feedback.
-
|
@thevortexcloud is correct. If the community wishes to fork and maintain the TreeDataGrid with an MIT license, they are welcome to do that. |
Beta Was this translation helpful? Give feedback.
-
|
Ok, thank you, that sound good! |
Beta Was this translation helpful? Give feedback.
-
|
Company I’m curious about your company “AvaloniaUI OU”. Do you plan to market products and services under that name, or is it the same entity that receives donations for the open-source initiative? Clarifying this would help us understand your approach better. Clarification Sometimes it’s a bit challenging to distinguish when you’re speaking as “Mike the businessman” versus “Mike the open-source steward”. Providing clearer distinctions could help avoid any potential confusion or uncertainty. Inclusivity I appreciate the complexity of wearing multiple hats, and I understand it can be tricky to balance different roles. Some community members might be concerned that the open-source project could be adjusted to favor a single business, possibly limiting other options. For example, with features like Hot Reload, one could imagine it might be implemented in a way that excludes both free and paid alternatives. Perhaps having a separate company that markets and sells these products under a different name could help navigate this delicate balance. |
Beta Was this translation helpful? Give feedback.
-
They already do. It's just not obvious unless you read the small print on the bottom of the page. |
Beta Was this translation helpful? Give feedback.
-
|
Company Clarification Inclusivity |
Beta Was this translation helpful? Give feedback.
-
|
In our technical stack selection for a large project, I previously chose Avalonia from among MAUI, UNO, and Avalonia. Today, we suddenly discovered Avalonia Accelerate, and now the team is re-evaluating our choice. What I want to say is that the change in licensing has indeed raised some concerns for us. |
Beta Was this translation helpful? Give feedback.
-
|
I get the feeling that some people in this thread are now having general licensing in mind. That has very well been excluded many times but also in this discussion. So we are really only talking about stuff that pretty much currently does not exist at all or already exists on a mediocre level whereas that stat will be left untouched with respect to licensing. @chatterzhao: I do not see a world where decision making is as simple as just looking at license in black or white. The size of license fee (whether moderate or high) will have an impact and don't forget the extra cost of pain using something that is less practical. I agree that Avalonia has not yet reached the point of a Qt situation, where orgs would pay insanee license fees and even fees per sold item. However, for Desktop and Embedded development there is clearly much less pain on a daily bases compared to MAUI and UNO with browser and mobile closing in. I value your feedback but I get the feeling that if some devs would not use Avalonia just because some add-on features are licensed, the rest of the community can live with that. |
Beta Was this translation helpful? Give feedback.
-
|
@llfab I think some are extrapolating based on directional changes. If TreeDataGrid is going closed source or paid now. What's next in a few years? Things like WebView, DataGrid and FlowDocument are pretty fundamental to the framework historically as well (WPF). Otherwise, I agree with you. |
Beta Was this translation helpful? Give feedback.
-
|
I guess, TDG being fundamental is one perspective. The other one is the actual reality that making it right apparently requires a ton of work. It is hard to match these two perspectives. Again it depends on the numbers but as a user I would prefer to pay a fair fee instead of waiting many more years to get it right. Gout feeling is that leaving the current TDG more or less as is (free to the community to improve it) unlicensed and creating a clone that is brought forward fast but must be paid is not such a bad option as a potential trade-off for these perspectives. Still these elements are not as fundamental as Grid, Panel, ListView right. Also, (but again that is my selfish kind of view) many Avalonia customers are doing more kind of embedded UI development. Still Windows or Mac or Linux full desktop but UI's that are not really based on such kind of controls. And the perspective of those developers is, that they would not want some man-years of the core team development spent on such a more detailed things. I'm really not in these details but I remember a decade back when WPF also did not have such great things and my former enterprise had to pay DevExpress to get something nice... Again it really depends on how much work some of these controls are and how basic they are for a large part of the users. Conceptually, I would understand though to make very detailed controls that are heavy work at the same time "paid". I would be less worried about: what is next. It's not like Qt just raising prices on everything, right. It is based on the estimation that it would require a lot of work and may not be so essential. It's handling limited resources where not everybody can be happy. Leaving everything free will also make some unhappy. Like me. Because for my enterprise the highest risk is that the entire Avalonia work is slowed down or canceled because working on in doesn't get the core team devs paid. If I did not know better, my suggestion would be: lets but a number on TDG (say 100K) with the majority of Avalonia users needing this thing which would be at least 10K users. It would then just need everybody to give 10 bucks to get it done. But that's not going to happen either, right... |
Beta Was this translation helpful? Give feedback.
-
Agreed. Keep in mind though the current TDG is actively being held back -- the community is not allowed to submit PRs for it as-is due to time constraints of the core team.
That's far too narrow of a view. A significant number of Avalonia users expect it to be a next-gen WPF.
WPF not having a DataGrid for many years was a fundamental roadblock that stopped its adoption. It was always a mistake for it not to have been there from the beginning. A single WPF dev for the most part fixed this but too late.
In my mind this is pretty simple. If the feature/control is in WPF or WinUI we should have that in Avalonia "core" itself. I've ported many simpler controls over myself and plan a few more if things allow for it. Whether the control exists in WPF/WinUI seems like a very reasonable place to draw the line. It keeps things competitive, aligns with what control vendors already expect, and keeps the community happy. Of course as you said some of these controls are just so expensive to develop -- but we have all of WPF and WinUI open sourced now which we didn't when most of these controls were first attempted.
That's your viewpoint but also keep in mind the majority of companies try to maximize profit long-term. It's a very rare company that is willing to sacrifice some profit for the greater good and pushing the industry forward. That does conflict with the open-source mindset as I've stated earlier. Avalonia UI the company is still fairly new and the community has to get used to the new dynamics. Also keep in mind your budget for development is quite a bit higher than a hobby dev. But let me put this in different terms you can relate with. What if Avalonia started charger a per-install fee like Unity? Every time you distribute your app you have to pay a royalty so to speak. I bet that would raise some concerns from your viewpoint. Well, it's the same for open-source hobbyist developers right now that were using Avalonia entirely free -- including the TreeDataGrid. Now that is going to change and they have to find a budget for it or switch controls.
That is a valid viewpoint I understand. The core team is fundamentally trying to work on the framework full time and that's the reason driving everything. Completely fair.
Note that ALL of Blender was open source for 100K.... that said, 2 developer man-years of developers doesn't seem out of the ordinary for such a control. Keep in mind porting the WPF DataGrid is also a cheaper option at this point we didn't have in the beginning. |
Beta Was this translation helpful? Give feedback.
-
|
Well said! |
Beta Was this translation helpful? Give feedback.
-
|
One thing more that hasn't been discussed is how to handle conflicts of interest. Avalonia is careful to discuss with control vendors to make sure they don't step on any toes. However, what about conflicts of interest with the community itself? I'll give some examples:
Finally, keep in mind what happened with C# hot reload / edit and continue itself when Microsoft attempted to take it fully closed source and block other IDEs who relied on it. Open core is a great principle but all have different ideas of what "core" is. How will you handle things when the community disagrees with the direction of the company? In cases like these everyone is going to think their viewpoint is the right one and it can get difficult to reconcile. I'm voicing these concerns here ahead to time to try to smooth things over long-term. The more direction/viewpoints are understood by everyone the easier things are to manage without major disagreement. |
Beta Was this translation helpful? Give feedback.
-
Mike responded to a bit along the same lines: |
Beta Was this translation helpful? Give feedback.
-
Avalonia's text backend already supports virtualization it is just not used by TextBox and TextBlock |
Beta Was this translation helpful? Give feedback.
-
|
Agree. It is clear that everybody need Button, ListView, Panel kind of things. But these detailed controls are quite more specific. And: there are 3rd party vendors who build these and may through some relevant dev resource at it. I like this kind of line. It's relatively simple to draw. |
Beta Was this translation helpful? Give feedback.
-
Wait, really? Now I am curious why nobody has hooked either of those controls up to it... |
Beta Was this translation helpful? Give feedback.
-
|
I disagree that only text virtualization is needed. In my mind the goal should be support WPF/WinUI parity with controls. The line has already been drawn there by Microsoft. So it seems the clearest way to make everyone happy. It's what 3rd party control libraries have designed for and it keeps the community happy. Personally, I think WebView and DataGrid should be part of the framework itself. FlowDocument and rich text editing aren't needed though as long as the core framework doesn't exclude important things (sounds like it doesnt). Now if we need paid controls to fill gaps immediately that totally makes sense. But I also wouldnt expect Avalonia to block those controls from coming in to the framework by contributors either. As I said WebView was planned by myself eventually. If I ever had time between projects again I would port WPF's DataGrid as well. |
Beta Was this translation helpful? Give feedback.
-
|
Given the pushback against making TDG a paid component, perhaps the best course of action would be to archive the repository and encourage forks (though not under the Avalonia namespace, as then we'd be responsible for its maintenance). It's unlikely that we'd make a ton of money off it anyway, and it'd be a lot of work to get it up to a state where it has all expected features. |
Beta Was this translation helpful? Give feedback.
-
I think you're underestimating how much time it takes to review a PR. In some cases we receive a well-defined PR with a good description, unit tests and clear fix: in that case reviewing is simple. In many other cases, it takes more time to review the PR than to just do the work ourselves. If the PR is a feature, then we must maintain that feature in perpetuity, after the original contributor has gone (which they will - they're not getting paid to maintain it after all). Lets put it this way: community PRs in general are a cost, not a saving. As other have said offering money for PRs will make this much worse.
Not necessarily, one can have a license that is acceptable for OSS projects and hobbyists but not for commercial products, such as AGPL-3; which is what we were considering for TDG. |
Beta Was this translation helpful? Give feedback.
-
To me, it sounded like the proposed changes for TDG is just open source with a different licence. However I interpreted the comment I replied to as only giving the source code to those that pay for it. Typically with source available stuff you need to sign an NDA as well as some other contracts before the provider will even tell you where the code is hosted. Some companies also charge an additional fee for access. For open source, you need a FOSS license and for public access to the code. |
Beta Was this translation helpful? Give feedback.
-
No, I'm not. I spent several hours yesterday just responding to your counter PR of our PR! 😆
A community PR that fixes a non-trivial defect costs the contributor far more to produce than the PR costs the maintainer to review. Producing a community PR is also very expensive, and is a cost, not a saving. The point I was trying to make is that no one will work for free. If you want something, it has to be paid for. Even OSS is paid for, by someone. I mean, an unemployed, starving, homeless person, who can't support his family who is also starving, isn't writing OSS and publishing it online in pure generosity... |
Beta Was this translation helpful? Give feedback.
-
|
I don't think that arguing about PRs in this way gets us anywhere. All code that is taken into the framework brings maintenance cost. Each large PR that is created without consulting first is extra cost even if done in good will. It doesn't matter. No great product is created without the vision and competence of a core group. We better strengthen the core group by making it a sustainable and fun thing than arguing about what rights the community shall have. Funding a core group with cash ist just way more efficient than PRs even if they may be helpful. I find the brilliance and effort of this group isn't valued enough. As isn't the transparency. Try get this from MSFT who instead have invented yet another disappointing UI framework and then discontinued them without asking. Avalonia showed that making WPF plat wasn't so far out of reach. |
Beta Was this translation helpful? Give feedback.
-
Agreed. Well said. |
Beta Was this translation helpful? Give feedback.
-
|
Reading the comments yesterday has been incredibly frustrating. There is an undertone of entitlement from a minority of users. I think it's worth being clear: if you're not a paying customer, we have zero obligation to provide you with anything not available today. While we want to improve Avalonia and see the user base grow, we do this by ensuring the project is sustainable. I don't have the time nor the energy to answer everything said yesterday, but I'll respond to a few comments:
We've always been clear that Avalonia will stay open-source. Period. Your comments suggesting we can't be trusted aren't just off-base; they're disrespectful. We've consistently shown our commitment to transparency and openness. We've said Avalonia will remain OSS, and we stand by that. If you don't believe us, that's your call, but remember that words have an impact. Hearing we can't be trusted is pretty hurtful, especially considering how hard we've worked for the entire community.
Let me set the record straight: we do accept PRs and fix issues but prioritise paying customers (though we also do merge community PRs occassionally). This isn't about holding back the control. It's a practical business decision. Steven is the only one working on this control, and his time is incredibly valuable.
It's tiring to keep hearing the 'WPF has it, so Avalonia should have it' mindset. We're not Microsoft. Comparing us to them is like comparing a local coffee shop to Starbucks. We can't just churn out free controls because they do. If the community want those controls? Great. The community is welcome to step up and build them. Or, alternatively, they could support us so we can do it. I ask that the vocal minority in the community drop the 'you owe us' attitude. We don't. We're working incredibly hard to give you a solid framework. Anything extra? That's a bonus, not a birthright.
The priority is to ensure the long-term sustainability of the project. We have always been very transparent about that. If the community wishes to replicate every feature we ship as part of Avalonia Accelerate, so be it. We can't stop people from doing that, though I'd hope that people understood how detrimental undercutting us could be to the sustainability of the project.
We'll make sure it continues to work as-is. We obviously won't be investing any significant energy in adding new features.
We're considering archiving the project and directing folks to an external fork. We've been upfront about the economics, but it seems there's pushback against paying for improvements. Fair enough. If that's the consensus, the community's welcome to take the reins. But let's be clear: it won't be under the AvaloniaUI umbrella.
Please see above. We have always been very clear that we will work with our customers on the TDG.
We've been offering this for five years already. We even tried cheaper tiers because some vocal community members insisted it would boost demand. Spoiler alert: it didn't work out as planned. Those lower tiers? They only brought in about €10k, which is nothing compared to our enterprise deals. Worse, it had the unintended effect of cutting into our revenue as larger enterprises wanted to purchase the indie tier. We learned the hard way that lower prices don't always mean more business. We're not trying to exclude anyone, but we need to be smart about pricing to keep Avalonia growing and improving for everyone. ConclusionWe aren't a big company. We reinvest every euro the business generates back into the project and have always been very open about the challenge of making Avalonia sustainable. We cannot grow the team without an increase in revenue. Not growing the team means PRs not getting timely reviews, issues being ignored and a general inability to keep up with the increase in community engagement. The hard truth is that our team is driving Avalonia forward. We've created most of the value you're using in your projects. Most advancements and improvements to Avalonia come from the team we employ to work on it, not the community. It's pretty jarring to see individuals who've contributed one or two minor PRs behaving like we should be deeply indebted to them. Let's keep things in perspective here. As we've discussed before, more often than not, a community PR costs us money, both to initially review but also to maintain going forward. Lastly, a note on donations. If every user this year had donated €50, we'd have millions sitting in the bank and not be discussing building Avalonia Accelerate. We'd be building and shipping everything as OSS. |
Beta Was this translation helpful? Give feedback.
-
|
I won't address all points as it's getting too long but I'll hit the main ones. First, I don't think my tone or intention is being properly conveyed. My goal is to make sure the core team and Avalonia the company understand how your decisions can be seen by others and from other viewpoints. I've been clear I understand what you are doing and why. But I also want to be sure you understand how some of the moves you are making affect others and how they can be seen externally. If anything, we are trying to come up with a solution that works for everyone. Given the amount of passion the Avalonia team puts into their work and the creativity involved I can understand sensitivity to these topics. But I'm giving matter-of-fact assessments that I hope eventually are understood.
I said "No one can / should expect license changes and that very much does raise a “caution flag” that we need to acknowledge....Nothing the core team can say will alleviate these concerns and it's something only time can reconcile." I'm afraid you still don't see this point. That isn't even what I think either. But you and the team in this case may not understand how the TDG license change can be viewed by others. That very much is a caution to them and I'm pointing it out as a caution to you and the team as well. That viewpoint has definitely been validated by other comments here. I'll only mention the composition rendering engine briefly but that conversation was similar in some ways. The point about a corporate influence and drive for income making decisions for the project now also stands. But it's something I and almost all other devs can relate with and accept. But I do ask you not price-out the small devs and hobbyists that picked up the project before anyone else. Will that make a ton of money for you? Definitely not. But it won't incur losses as it's money you otherwise wouldn't get (because they can't afford higher tiers) and it also greatly helps the brand image and reach of the product for developers in general. From my viewpoint it's strategically a good idea to keep everyone along for the ride. As someone who has constantly defended the team's viewpoint please understand I am trying to help you. There is nothing disrespectful in sharing how decisions by the core team can be received by other viewpoints. We need to be able to have that conversation in an impersonal, professional way. We are discussing ideas, topics and viewpoints and can keep them at that level. I should be able to summarize some of the expected pushback (for your benefit) without that being received in a negative way. Red/Blue team gaming concepts are fairly common.
Well, I haven't seen any community PRs merged for features. At least the few I was watching. The word as I understood it was don't submit anything unless you are a paying customer. This is all fine and you set the rules. We should see this more as "source available." But I still think my statement "The TDG is...being held back" is true here and that's not an attack on the way the team is handling this. I'm just saying the control was never given the conditions needed to flourish as other open source code was. The suggestion here was more the community was never allowed to take any of the burden off the core team and the control to grow organically. Yes, reviewing PRs is a big issue and time consuming. Longer term I still think we need code owners and community members to step up in places and be given power to review/merge.
I understand this from your perspective. But do you understand it from the users perspective? WPF, WinUI and Uno are FREE for the users of the framework to use as they want. This is the competition you are up against. Is it fair? Absolutely not, it's harder for you and I've stated that elsewhere. But I do stand by these statements. No one has ever used any "you owe us" wording in ANY of the comments I've read here. I think Avalonia (the open source framework) simply needs to accept controls that are historically and currently part of other Xaml frameworks. Am I asking the core team and the company to spend the time to do this? Absolutely not. No where have I said this. Keep in mind I've done a lot myself to fill the control gap so I'm putting action behind my words here at my own expense. The whole point is I hope you accept community contribution of controls to fill the feature gap. Again, there is no "you owe us" mindset.
I don't think anyone intends or will do this across the board. The vast majority of Avalonia Accelerate falls under the tooling category. That isn't even being debated here. Tooling to boost productivity is absolutely a great way to pull in extra income. What I'm referring to are CONTROLS, controls that are already part of other UI frameworks. Does the core team have the ability to merge in a WebView control to the open source project knowing it will undercut their paid control? This is a very clear conflict of interest. It's takes a very level-headed steward of the project to be able to handle this type of situation. I've also been clear in the issue discussions and elsewhere I want to personally bring a WebView at some point if no one beats me to it. I see it as foundational to the framework. Will it take a year or more? Probably. |
Beta Was this translation helpful? Give feedback.
-
Yes, it would be, but no comment I have read impugned your honesty. Please don't take our feedback quite so personally. For those of us who have been in the industry many decades, we have learned that trust and trustworthiness is not enough, and in many cases, it isn't even a factor. If a person says, "Dear Customer, I fully intend to support this product for the next 1,000 years!", we recognize immediately that even the most trustworthy individual is unable to fulfill this well intended and meaningful obligation. This reality is true not because the man can not be trusted, but because he isn't in complete control. No individual, no matter how trustworthy, has control of all the external factors they are subject to. Honesty is not a substitute for being in absolute control, and believing an honest person has full control because he is honest is not trust, it is fantasy, and irrational.
At the moment, isn't your only offering $7,500 per developer per year? Do you think that price is affordable and attractive for every developer?
A few moments later:
The latter is the way Microsoft became who they are. Everyone contributed a small amount, no one received a free lunch, the cost was evenly distributed, and this pricing model made Microsoft the near monopoly they are today. I'm sure every Avalonia user can afford €50. Why not require this donation, to gain access to the repo? For all practical purposes, €50 is the equivalent to free. If you took this approach, we would support you!
Hopefully this reaction is not in response to any of our comments, and we would be very surprised to learn that such a misunderstanding had occurred. To be short and clear, we do not hold this view.
A community PR that fixes a non-trivial defect costs the contributor far more to produce than the PR costs the maintainer to review.
Yes. That makes perfect sense. We believe you need a business model that requires everyone to contribute. The "free" aspect of OSS is a stupid concept, because no one works for free, and it is irrational to expect anyone to do so. In all examples of "free" OSS, someone is always paying the bill. Avalonia is amazing. Thank you. |
Beta Was this translation helpful? Give feedback.
-
|
@busitech Introducing a cost to access the repo would hamper adoption, especially if you mean viewing the basic source code. Avalonia, IMO, is borderline unusable for my projects (all non-commercial) if I didn't have source access and would hurt new adopters who are trying to create free, open source software. Both are important to fill gaps in docs and cover things that docs shouldn't. It's a bit different than WPF that clearly has invested significant man-years into documentation writing. Quality packaging/tooling improvements would 100% be worth a hobby-level cost as I hate fighting them. Onto a tangent about potential contribution changes. Not looking for a detailed reply. @MikeCodesDotNET @grokys With Accelerate on the horizon, there should be some new contribution guidelines enacted. Somebody will PR a feature that Accelerate offers and Avalonia will be in the awkward spot of rejecting their PR without policy in place or accepting the PR and maintaining two implementations of the same feature. I've already chilled some DevTools features I'd like to PR by my own decision since the feature set is not public yet. I'm not wanting to step on Avalonia's feet since the Pro DevTools announcement nor cause team members extra backporting work during an ongoing rework. There may be ways to reduce the impact on the team from issues and PRs. Nearly all new community bug reports should be immediately closed if the user hasn't submitted a repro (reopened after submitting). Community feature requests should have an explicit response of:
Tags are quick, but they aren't the same as a brief personal communication. Community PRs could be required to close an active issue and the Avalonia team could request a community review before their own. Including platform testing. If a PR is not popular enough for a community member (or several) to bother testing, it might not be popular enough for the team to review, merge, and maintain. The number of testers would be small, but won't grow unless the expectations get established. |
Beta Was this translation helpful? Give feedback.
-
|
I'm inclined to mark this thread as off-topic since we've drifted from Avalonia Accelerate to broader issues. That said, I'm still open to continuing this conversation and addressing questions. When discussing trust. Let our decade-long track record speak for itself. Judge us by our actions.
It's impossible to say today. While I understand your perspective regarding a potential conflict of interest, the situation is more nuanced. Our primary objective is to ensure Avalonia's long-term viability. Handling community contributions that replicate commercial features requires careful consideration before we can provide a definitive answer. However, I can outline some key factors for an OSS webview control:
From a purely theoretical standpoint, I acknowledge your view. However, our commercial reality necessitates prioritising the project's sustainability. Decisions are made with this as our guiding principle. While this may occasionally diverge from immediate community preferences, the long-term sustainability of Avalonia is ultimately the best thing for the community.
I had a look at our data on the sales of cheaper offerings. In the 8 months they were available, we sold just 5 agreements. If we hadn't offered significantly cheaper tiers, I would be inclined to agree with you, but we have the data. Cheaper tiers do not drive an increase in sales. The low sales figures and a significant increase in the risk of operating at a loss on support agreements made it impossible to continue to offer this. As a growing company, we can't afford to subsidise our users this way. It's important to note the unique value proposition we offer: when you submit a ticket through our private portal, you're not dealing with support staff. You're interacting directly with Avalonia's core engineering team. This level of expertise and direct access is a key differentiator that our customers find incredibly valuable. I understand that some businesses don't value this, but in the scheme of things, the price is excellent value for money.
Forced donations are a great way to destroy trust and adoption. We had hoped that folks would donate out of the desire to support the project. That isn't the case, even for those that are using it professionally. If I may, have you donated yet? If not, can you share what's held you back?
I must respectfully disagree based on our experience. The cost of reviewing and maintaining community PRs is often significantly higher than you might expect. When a complex PR is submitted, our core team frequently needs to pause their ongoing work to review and discuss the changes. This disruption comes with a substantial cost (both opportunity and financial). It's for this reason that we've implemented a policy of primarily accepting PRs from paying customers for the TreeDataGrid. This approach helps offset the costs associated with review, maintenance, and ongoing support. @stevemonaco I really like you're suggestion! |
Beta Was this translation helpful? Give feedback.
-
|
Is the feedback session over now? If so, just do it. Commence operation Avalonia Accelerate. There will be plenty of opportunities to adjust the course as you go. Avalonia couldn't be in better hands. |
Beta Was this translation helpful? Give feedback.
-
|
Indeed!!! |
Beta Was this translation helpful? Give feedback.
-
|
Thank you! |
Beta Was this translation helpful? Give feedback.
-
|
What about the secret projects? Any clue 🤗? Any 3D object model viewer |
Beta Was this translation helpful? Give feedback.
-
|
I couldn't possibly say as they're still a secret 🤐 |
Beta Was this translation helpful? Give feedback.
-
|
Welcome change for TreeDataGrid (for how its described in first post). What about DataGrid? Does/will it have a different purpose? Will TreeDataGrid get functionality of DataGrid and DataGrid will be de facto deprecated? Or will DataGrid will be deprecated for real? |
Beta Was this translation helpful? Give feedback.
-
|
I want to give a small update based on the feedback we've already received. TreeDataGridAfter carefully considering the community's feedback regarding our proposed dual licensing model, we have decided to suspend this change. Instead, we will archive the current TreeDataGrid project and direct users to a community-maintained fork. We believe this approach aligns better with the community's interests. Pricing (A ballpark figure)From the survey results, we think an initial price between €150 and €200 is the sweet spot. As we've previously said, we will increase the price with each phase of release. Importantly, we're committed to rewarding early adopters by grandfathering their subscription rates, ensuring they receive maximum value for their early support. Subscription vs PerpetualBased on the feedback, we will offer a perpetual fallback license to Avalonia Accelerate. We have yet to decide on the price, but it will be a multiple of the subscription price (between 3-5x). Sneek PeakThis is very early days, but here's a sneak peek of one of the new Devtools features 👀 metrics.mp4ThanksI would like to express our sincere gratitude to everyone who has taken the time to complete the survey and share their opinions with us. We truly appreciate having an open dialogue with our users. If you haven't had a chance to complete the survey yet, don't worry; it will remain open for the foreseeable future. Your input is important in shaping our decisions, so we encourage you to participate! |
Beta Was this translation helpful? Give feedback.
-
In an ideal world, we wouldn't offer a perpetual license, but from the feedback, many users value it. I think the situation we're in is a little different from SmartGit. As we change Avalonia, we increase the likelihood of perpetual license holders wanting to upgrade (to ensure compatibility). |
Beta Was this translation helpful? Give feedback.
-
|
With perpetual being defined as "infinite use" but "no updates", right? Ok. With SmartGit they even offered infinite updates but just no support which was a wrong decisions in the first place, I guess... |
Beta Was this translation helpful? Give feedback.
-
We want to offer a perpetual fallback license, which grants the user one year of updates from purchase. After expiration, they retain access to the version available at the time of purchase, effectively losing the year of updates. It's the same model used by JetBrains, and incentivises renewals for those who want the latest version. |
Beta Was this translation helpful? Give feedback.
-
@MikeCodesDotNET can you please also comment on this part? |
Beta Was this translation helpful? Give feedback.
-
It will only be build-time checks. Once an app is shipped, it will work forever. |
Beta Was this translation helpful? Give feedback.
-
|
I've created DotnetPackaging. It has everything you need to create DEB and APPIMAGE files for Linux. Just tell me if you need help using the library. |
Beta Was this translation helpful? Give feedback.
-
|
It looks good. |
Beta Was this translation helpful? Give feedback.
-
|
I think the current plan will cause significant concerns, such as how much the price will change in the future? The parts that will enter the paid list in the future are unknown and uncontrollable. Will the free parts still be actively maintained? These seemingly vague concerns may cause some new teams to abandon Avalonia, and there may also be some forks of Avalonia to create new communities, exacerbating the confusion. |
Beta Was this translation helpful? Give feedback.
-
It will go up, as we have discussed. The initial price will be between €150 and €200 euros. As we add more features, we will increase the price, but those who have purchased at a lower price, will be grandfathered in, meaning they will continue to receive future versions of a lower price.
Only new features will potentially be released as part of the Avalonia Accelerate subscription, and even then we still have a long list of new features planned for v12 which will be included for free. As a general rule, if its considered foundational to the core of the framework, its going to be included for free, in the OSS project.
Yes. Avalonia is the foundational element of our offerings. The whole point of Avalonia Accelerate is to enable us to invest more resources in maintaining the OSS project.
I don't understand why someone would fork Avalonia because we've released optional, commerial features. This line of thinking makes very little sense to me.
I don't know where to begin with this, so I'm going to ignore it. Please read everything I've ever written about creating a sustainable project.
This is planned for the Visual Designer, though it may not be a v1 feature. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for creating Avalonia and making so much of your work-product free and open. You have solved a big problem for me and I have nothing but appreciation and gratitude for the efforts you've made. I teach media coding at a university. We use Unity because it addresses so much of what our students want to do, and it runs on Windows and Mac. Avalonia lets me teach things Unity makes hard, while still using C# seamlessly across Windows and Mac. FWIW, my students will go into their professional lives with a high likelihood of suggesting Avalonia to their employers who aren't already using it. |
Beta Was this translation helpful? Give feedback.
-
|
@MikeCodesDotNET I am rather new to the open source community so I don't really have much experience. My take away from reading this thread is that:
Talk about being stuck between a rock and a hard place. I really don't know what to tell you, I totally understand and appreciate that you also have families to take care of; however, I just feel really sad for the community, the real community, the hobbyists that really did want to learn Avalonia and someday be worthy to make contributions, like me. While features like FlowDocuments, RichText, Printing and a Visual Designer control are really projects massive enough to warrant professional/corporate levels of funding and support, I really can not say that today, I personally (as a lower middle-class hobby developer) can justify $225/year+ for the rest of eternity or even the one-time fee that is about 1/24 of my yearly income. I am still trying to save up $600 to buy the blue prints for the earth bag house I want to build my family in the future because I will never afford a house built out of anything else. That said, it is more than a fair price point for a big corporation that will get far more value out of the software that they write with it than the control costs them. For me, I can not justify that kind of money per year just to maintain my own open-source Avalonia Prism demo application just on the off chance someone sees it and decides to offer me a job! It seems to me that the whole open-source experiment really doesn't work and that at some point the project simply dies out or that someone with money comes in to back it financially. I am just glad that I found out early enough that you all really do not even want my contributions before I wasted hundreds or thousands of hobby hours learning Avalonia and trying to port FlowDocuments just to find out that my contribution is too much trouble for you to even look at. Thanks again for your great work on Avalonia! I really do hope that you are blessed and that your funding efforts really do make Avalonia better for everyone! GOD Bless! |
Beta Was this translation helpful? Give feedback.
-
|
@MikeCodesDotNET I actually thought that with ReSharper you always had the option to cancel your subscription and revert to a perpetual license at whatever version you started with? Maybe I am misunderstanding you but this you can either pay a subscription forever OR you can pay pay many times the sub fee for a perpetual license really just makes it kind of scary. What I mean is that FlowDocuments are really important to me and upon further reflection would be willing to pay the $225/year for say 3 years while you complete the FlowDocuments, if I knew that once FlowDocuments were complete I could drop my subscription and have FlowDocuments to play with; however, in your proposed scheme I'd just loose access to everything I paid for the last few years and would essentially be totally wasting my time and my project would just be locked forever unless I wanted to turn around and give you another approx $1,200 to upgrade to the perpetual license. It is really too much risk for me, what if my wife gets sick and I can't afford to renew my subscription then all of a sudden my open source prism demo is no longer maintainable? On the other hand if you made the subscription option less scary and nothing catastrophic happened to my family in that 3 years, it is highly likely I would continue to go ahead and keep my sub active after I got FlowDocuments because you would surely come out with something else I would be really excited for by then and I would already be used to that hit on my pocketbook every year. Come to think about it how could one even use your new stuff in an open-source project at all? |
Beta Was this translation helpful? Give feedback.
-
|
You must have subscribed for at least a year to be eligible for a perpetual license with JetBrains products. We're not precisely replicating their approach, but we are taking inspiration from them. We've been slowly adding the low-level parts needed for the Rich Text Editor (such as FlowDocument) to the core of Avalonia, and that's all available as OSS and won't be changing. I will add that we will constantly be assessing the situation and looking to make Avalonia Accelerate more accessible. |
Beta Was this translation helpful? Give feedback.
-
|
Avalonia Accelerate Update 🚀 Development is progressing well and we're on track for Phase 1 release in early 2025. After consulting with customers on feature priorities, we're confident in our roadmap ahead. Here's a preview of our refreshed Dev Tools UI (work in progress) which showcases the level of polish we're bringing to Accelerate:
|
Beta Was this translation helpful? Give feedback.
-
|
Looking really great! I'm not sure if you are soliciting general feedback but a few things stood out for me:
|
Beta Was this translation helpful? Give feedback.
-
|
A-we-so-me 🥳 |
Beta Was this translation helpful? Give feedback.
-
|
Looks great! But why boolean options are not Checkboxes? Checkbox is much easier to read and to interact with |
Beta Was this translation helpful? Give feedback.
This comment was marked as off-topic.
This comment was marked as off-topic.
This comment was marked as off-topic.
This comment was marked as off-topic.
-
|
We have had this discussion... in length... already... many times. Nothing is for free! |
Beta Was this translation helpful? Give feedback.
and others
-
We need your help.
As many of you know, Avalonia has grown significantly over the past few years in terms of monthly active users and our team size. Seeing many developers and organisations build complex, rich applications for different platforms from a single codebase is fantastic! We're incredibly proud of what we've built together, but we've been thinking hard about how to keep this momentum going. To this end, we're developing a new offering with a working title of 'Avalonia Accelerate'.
What is Avalonia Accelerate?
Avalonia Accelerate is a collection of tools designed to enhance your development workflow and boost efficiency.
Our goal is to enable Avalonia developers to be more productive while ensuring that we continue to have the resources needed to grow and invest in improving Avalonia for everyone.
We hinted yesterday at one of the features, a visual, drag-and-drop designer, but we've currently 16 different tools either in active development or in the roadmap.
Avalonia Accelerate is about providing additional tools that can significantly enhance your productivity and streamline your development process. Our aim is to create a suite of tools that offer clear and tangible benefits to our users. We believe that the value these tools provide will be evident, offering a strong incentive to subscribe.
What features will it include?
Transparency is a core value for us. While there's been discussion about how much to share, I believe in being as open as possible about both our project roadmap and the business aspects that support Avalonia's development. We're committed to keeping you informed every step of the way.
First things first, we're taking a phased approach to releasing these features. Of the 16 planned tools, nine are currently in active development at various stages. By releasing incrementally, we aim to deliver value to you sooner and incorporate your feedback along the way. This approach allows us to be more responsive to your needs.
In no particular order.
Enhanced Dev Tools
The enhanced Dev Tools is packed with new features that make debugging your application UI much easier! A few of these features include:
App Packaging
Many of our users have little to no experience with the complexities of packaging for macOS and Linux. Our app packaging tool makes it incredibly easy to ship your Avalonia for both platforms. Simply open your Avalonia project, tick some boxes and click the package button! Its features include:
One of our stretch goals for the app packaging tool is to enable the signing of macOS binaries directly from a PC. This would be a game-changer, allowing developers to package and sign Mac applications without needing a Mac device!
Visual Designer
We love XAML, and we know many of you do, too. But we also recognise that it can be a barrier for newcomers. To grow our community and make Avalonia development more accessible, we're excited to be developing a drag-and-drop visual designer. Its key features include:
We expect to have an early preview of the designer ready in the first half of 2025. This tool represents a significant improvement in making Avalonia development more accessible without compromising the power and flexibility that advanced users love.
Hot Reload
We're excited to announce that Hot Reload is on our roadmap, slated for the final phase of our development cycle. We want to be transparent: this feature represents a significant engineering challenge, which is why it's scheduled for later in our timeline. Hot Reload will enable you to:
This feature will dramatically streamline your workflow, boosting productivity and allowing for more rapid iteration of your designs and code. While it's a complex undertaking, we're committed to delivering a robust, high-performance Hot Reload capability that lives up to your expectations
OnScreen Keyboard (for Embedded).
We've seen a huge amount of growth with Avalonia being used in embedded scenarios. As we've spoken to more and more organisations building for embedded devices, a theme has emerged. Most require an onscreen keyboard.
Web Dialog Control
A common request is the ability to render web content as part of an Avalonia application. A common scenario is supporting user authentication. There are various solutions for this in varying states of completeness. Our Web Dialog is intended for simple workloads, such as when you need to display an authentication window. Initially we only support Windows, macOS, iOS and Android.
For those who require advanced functionality or the ability to embed web content inside your Avalonia views, we recommend using DotNetBrowser, as it's the most featureful option, powered by Chromium, supporting Windows, macOS and Linux.
Media Player
Implementing high-performance video playback across multiple platforms presents significant challenges. While various open-source solutions for Avalonia exist, they often fall short of expectations. Leveraging our expertise in composition and rendering stack internals, we've developed a robust, performant solution to address these challenges.
Our media player supports the following:
TreeDataGrid
We're announcing a significant change to our TreeDataGrid control licensing model. We recognise this may impact some of our users, and we want to provide a clear explanation of our decision.
We've long aspired to make the TreeDataGrid a robust, feature-rich component for our users. However, we haven't been able to dedicate the resources necessary to improve it as quickly or comprehensively as we'd like. This has resulted in slower development and limited support, particularly for users outside our paid support services.
The crux of the issue lies in the complexity of the TreeDataGrid. It demands the attention of our most senior engineers, whose time is at a premium. Under the current licensing model, it's been challenging to justify allocating resources to its development, especially given the lack of direct return on investment. This has led to a situation where we can only review PRs and address issues from paid support customers, potentially making it expensive for some users to get the help they need.
To address these challenges and ensure the continued development and support of the TreeDataGrid, we've decided to implement a dual licensing model. The control will remain available under a AGPL-3 license for open-source projects, while commercial applications will require an Avalonia Accelerate license. This change will allow us to allocate resources to improve the control, benefiting all users through more frequent updates, better support, and faster issue resolution.
We plan to implement these licensing changes later this year, providing ample notice to ensure a smooth transition for all our users. We appreciate your understanding!
Markdown Renderer
As we work on improving our text rendering capabilities, one of the features coming soon is a control to render markdown content. This development is a key stepping stone in our broader initiative to implement a comprehensive rich-text editor. By adding Markdown support, we're laying the groundwork for more advanced text manipulation and formatting options in future releases.
Rich Text Box
The RichTextBox is one of our most requested controls, addressing a key community and customer request. This complex control supports FlowDocument content, including images, paragraphs, tables, and more.
Inspired by WPF's capabilities, our implementation represents a significant engineering effort to bring advanced text editing features to our platform. While ambitious in scope, this control will greatly enhance your ability to handle text within your Avalonia applications.
Printing
In response to growing enterprise adoption, we're pleased to say that we're working on enabling printing support with Avalonia. This feature addresses a key request from our enterprise users and enhances Avalonia's versatility across our supported platforms.
FastDroid (Android AOT)
We've previously shared our experiments with improving the performance of .NET on Android. We've two potential routes to explore further, but we're committing to providing our users with a route to shipping lightning-fast Android applications built with Avalonia. From our preliminary testing, we've seen a 68.30% improvement in startup performance.
Phased Release Strategy
We're taking a strategic approach to rolling out these features, to get tools into your hands as soon as they're ready. Here's what you can expect:
Licensing and Pricing Model
Avalonia Accelerate will be offered as a licensed product with a yearly subscription on a per-seat basis. Our initial pricing will reflect the early access nature of the offering, with prices adjusting as we add more features. Early adopters who subscribe during the early access phase will have their price locked in for at least two years as a thank-you for their support and feedback.
We're committed to accessibility across our community:
Your Feedback Matters
We need your help in determining a fair price for Avalonia Accelerate. Please complete our pricing survey - your input is crucial. Survey participants who provide their email will be added to our beta-testers list and receive a discount code when licenses become available.
Complete the survey here
Subscriptions will directly support the ongoing development of both Avalonia Accelerate and the core Avalonia framework. We're committed to balancing accessibility with sustainable development, ensuring we can continue to innovate and improve the platform for everyone.
We've shared this roadmap because your feedback is invaluable to shaping the future of Avalonia. As users of Avalonia, your thoughts will guide our development and help us ensure we're building things you genuinely need to be more productive. We're eager to hear your perspectives on these plans.
Update
While we're still accepting feedback, the preliminary results can be found here.
Beta Was this translation helpful? Give feedback.
All reactions