[Interest] the path forward

Benjamin TERRIER b.terrier at gmail.com
Fri Apr 2 13:27:53 CEST 2021

On Thu, 1 Apr 2021 at 18:56, Jason H <jhihn at gmx.com> wrote:

> I've called Qt "top secret rockstar tech" multiple times and I plan on
> calling it that for a bit more. I've had 5 companies buy licenses since
> 2005.  Qt has delivered on getting my products to market in record time,
> even if that was a process frought with far more native code than it should
> have been. At the very least having a cross-platform UI saved some effort,
> and provided consistency, rather than having to target Android's stacks AND
> AVFoundation seperately.
> What I don't like is the erosion of the LGPL that Nokia set in place. I
> think it's shortsighted, and contributes to the decline of Qt overall. I
> feel that while Qt stock has done well, volume is thin, and something as
> technical as having BSD or LGPL could be beneficial for the volume. Qt is a
> project of declining hobbiest popularity, even when there are big companies
> buying into it. Compare it with say, ReactJS/Electron. Everyone knows about
> that, Qt not so much... Maybe the big player interest will circle back
> around as it comes up on job postings... I'd personally launch a head on
> WebGL offensive if I could I think there's a lot of untapped potential
> there!
> Also, I think Qt should be on a US stock market, but I am biased. ;-)
> There's some pink sheets version of Qt stock which is even thinner volume.

I am completely in phase with Jason.

Qt is a very good piece of technology, for many reasons.
Being a cross-platform UI framework is just the tip of the iceberg. It has
a great documentation, API are consistent and clear (isEmpty() vs empty())
and it covers far more than just UI: it has networks, filesystem, serial
port, webengine, etc.

And I also dislike the erosion of the LGPL and the Open-Source Qt offering.
For the past years, I've seen the popularity of Qt decline: I see less and
less Qt job offers and it is harder and harder to get Qt developers.
For me, this is one of many symptoms that Qt, as an open-source project, is
declining. I am not talking about quality here. I am talking about
visibility, user base, governance, etc.

For the visibility, well it's quite easy to spot. For example,
qt-project.org is just a redirect to https://www.qt.io/developers and from
there you need to go through 6 different pages before being offering the
link to get the download installer


And of course https://www.qt.io/download-open-source has 4 pages of content
trying to steer the user toward a commercial license before seeing the
button to continue.
And then you still have to create an account and login during the install
Of course this web page labyrinth to download the installer is changing
every couple of months.
I've never seen an open-source project trying this hard to hide itself.

It's such a pain to get Qt through the Qt website that for the past 5+
years I just go to download.qt.io and browse the directory structure
directly: way more easy and convenient.

The decline of the user base is mostly based on my point of view, but there
are other people on this mailing list who draw the same conclusion.
And it is only logical to have less and less users, when you switch from
LGPLv2 to LGPLv3, do not release new modules under LGPL, force Qt account
login (while having said that you understood why it's a bad idea a couple
of years earlier), stop updating Qt 5.15 when Qt 6 is barely usable, ...

For the governance. Well I might not be fully aware of everything. I try to
follow the mailing lists, but I don't read everything so I might have
missed something.
When I started to work with Qt 10+ years ago, I spent time learning about
the Open Governance model.
My feeling is that the Qt Project does not follow the open governance model
anymore. It seems to me that the Qt Company is making more and more one
sided decisions.
The mailing list is just the place where the Qt Company announces the
decisions that were already taken.
I know that TQC is the 1st contributor to Qt and provides hardware and
infrastructure support, which make its voice stronger.
But how come the TQC can single handedly decide that the Qt Project won't
support Qt 5 while Qt 6 gets actually usable?

I mean either TQC does support the Qt Project and it should support the
decision made by the Qt Project. That's what the Open Governance implies
(at least it's my understanding).
And if that's the case the decision to stop 5.15 support should have been
made in public.

Or TQC does not support the Qt Project. TQC only supports Commercial Qt.
In which case the support provided to the Qt Project is just a side effect
of TQC needing to support its commercial version of Qt, i.e. the Qt Project
only gets the crumbs TQC is willing to let go.

My feeling is that in 10 years Qt transitioned from the 1st case to the
Maybe I am wrong, maybe it never was the 1st case.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/interest/attachments/20210402/85f97f4b/attachment-0001.html>

More information about the Interest mailing list