How should Qt be developed in the future? Report from the Qt Contributor Summit.
-
Dear Qt Contributors!
One of the naggings questions the attendes had at their minds here at the Qt Contributors Summit in Berlin was that of what the future of Qt development could look like, now that Nokia has decided to invest into Qt only "in the near term". The worries were both with Qt the technology, but also with the Qt development team that we are friends and colleagues with and need to continue to work on Qt 5.0 and beyond. There was a discussion in the halls about a potential Qt foundation that could take the lead. In the light of "this previous discussion":https://qt-project.org/forums/viewthread/18055 a session was held about this idea. This is a write-up of the results of this meeting.
Present were roughly a hundred Qt contributors, representing Nokia, individiual developers and a range of Qt partners, all with a vital interest in Qt. It was extremely surprising and at the same time encouraging to see how strong the consensus was about the following results:
- Qt should not end up in the hand of any one single company, should Nokia decide to sell it.
- Instead, it should be developed cooperatively.
- There should be openness and a level playing field amongst all involved contributors and companies.
- If the current Qt ecosystem cannot be maintained, there is a danger of Qt being fragmented or forked based on the GPL and LGPL versions.
- The Open Governance process is tremendously important.
The agreement was that a setup that implements these ideas would provide for the most stable and continuous development and growth for Qt, possibly at the expense of a temporary slowdown. This consensus was communicated to Nokia during the summit. We will continue working on this in the hope to find a solution for the future development of Qt that provides the necessary long term stability.
All the best,
"Mirko":/member/2481
-
Thanks Mirko
I for one, would like to see Qt-project.org continue, and think that if it needs some legal rights from Nokia to do that, I hope that they will help.
The core thing to remember, is that Qt is very strong. I see interesting projects every day, so I know that these companies are going to do what is in their best interest which is to help Qt stay strong.
Regardless of what Nokia does not does not do, the community is moving. In the past few days, I've been in conversations about things like independent code repository, a Qt Partner Alliance, and many other things.
We are in a transition, but the future of Qt is bright!
-
There is an sdk called Mosync which is crossplatform c++ / html 5 / js api for mobile development. I would like Qt to move to a similar path in future, i.e a high performance cross platform c++ api for desktop, mobile and embedded and I don't mind paying for such a toolkit. This is better than Qt belonging to any particular OEM and tied down to a single platform. There is a similar effort behind phonegap. several companies are contributing to it. I wish the same thing happen for Qt. I think that once Qt is completely free from Nokia, several big companies will be willing to join.
Nokia could have ported Qt to WP8 since it support native code. But Nokia now seems like an unofficial microsoft center.
-
What Jayakrishnan.. said - LGPL is nice, but what is even nicer is a more sensible commercial licensing scheme, there are many people who won't pay a four digit price for a commercial license, but will gladly pay a more reasonably priced license for indie developers, including me. IMO Qt has the capacity to sustain itself and be 100% independent.
There is always the problem of consensus when it comes to cooperation. Cooperation can only be successful if there is coherency across the cooperators, without that there are different directions of different parties and things get harder to move. Alas, people have problems with being objective, and seem to always involve personal or corporate interests and yield to the bias those produce.
-
utcenter:
please take in mind, a commercial license includes support.
Support costs money. This is an easy calculation. I don't know what price an hour is for Digia, but multiply it by the time needed for a typical support case and multiply it by the average number of cases per person per year. The costs must be more then that.If you are thinking on commercial licenses, why do you need them? We (in our company) need them for serious support in urgent cases. But we also create valuable income by using Qt.
If I would make that calculation above with our internal costs per hour, there might not be more then 30 hours per license per year (and that includes no gain of money). And they additionally do other things that have to be calculated too (like managers, packaging, ...)
-
Well, it depends on what do you mean by "support"
Support doesn't seem to cover my personal N1 annoyance with Qt - holes of missing functionality in the APIs - support doesn't deal with API design limitations
Bugs - by now the common practice should be companies paying users to find bugs, not users paying companies to solve bugs
Those are the only two cases I ever needed support with a commercial product, either a case of missing or broken feature. I know many people reach for support in often absurd and trivial scenarios, but not me.
I don't really need a commercial license for the sake of support, but for the sake of static linkage combined with API functionality holes that I fill, i.e. so I can make the changes I need. So, it comes at no surprise that I'd be happy with a 200 $ commercial license WITHOUT support rather than a XXXX $ commercial license for support I won't be using. That's what I mean by flexible licensing, addressing the different needs of different users.
And BTW, there are plenty of frameworks that are way cheaper than Qt and still offer customer support - MoSync, Marmalade, JUCE... etc..., so I don't see a valid reason for Digia's time to be that much more precious.
-
I have no idea how the support of MoSync etc are, but I have for Qt support:
We had many cases where we needed support. Bugs (where we got interem fixes), problems which we got solved, even once a case where the graphics were too slow and which was fixed by a huge amount of time.
This is support as commercial companies want it to have.
Shall the digia people say:
You are a prio1 licensee, you get fixes, you are only prio 2, you don't get anything?
-
With a non BSD-style license (OR WHATEVER) someone must always work
for the corporation that attached the license by declaring what petty or
non petty changes were made for the benefit of the corporation.The problem is about consumerism. With a BSD-style license "someone can
make a transaction"-- by not sacrificing labor or time for the corporation
that attached the license.Would YOU be happier as a consumer with a BSD-style license!?
-
@Gerolf - I personally won't have any trouble with not receiving the support I don't want and don't want to pay for. I am not a commercial company, I am an indie developer, and the whole point was that Qt has no licensing scheme fit for anything other than commercial companies. E.g. a wide range of market for smaller commercial customers is not being addressed, cuz I am sure there are plenty of other people who'd want Qt commercial for purposes other than commercial support.
-
utcenter: Do you actually know the prices? I talked to digia once about the price scheme and I found it very reasonable, especially for a one-developer company.
-
And what fits within your idea of "very reasonable"? "This post":http://stackoverflow.com/questions/352896/qt-commercial-licenses says 5500 Euro for 3 platforms plus annual fee, but it is from 4 years ago.
In my experience, when a dealer has good prices, he puts them on display, which doesn't seem to be the case with Digia :)
-
Never trust the internet;-) Ask Digia directly.