Thread: First thoughts about the (pre) Sailfish OS 2.0
View Single Post
javispedro's Avatar
javispedro is offline javispedro
2015-09-16 , 20:40
Posts: 2,355 | Thanked: 5,249 times | Joined on Jan 2009 @ Barcelona
#415
Originally Posted by pycage View Post
I'm always reading "open-source this, open-source that"... but would it really bring a benefit?
It surprises me that I would see doubts about this on this forum of all places..

Originally Posted by pycage View Post
Most people are consumers, not contributing to the open-source stuff they use at all. The people who develop and contribute to open-source projects are very rare.
And yet both the many users and the minority of developers benefit. Developers make patches and receive donations/reputation and users enjoy significantly increased level of customization of the platform.

Originally Posted by pycage View Post
I think this thread resembles this as well. There are only a few developers among many consumers.
I think this thread is a good example indeed. A number of people is complaining about some perceived mis-features of Sailfish 2.0. A developer appears with a number of patches (possible because Qml is "open source" in the webOS ausmt sense). Everyone wins. Zero cost to Jolla.

Originally Posted by pycage View Post
So we also don't see much progress with Nemo or the open-source parts of Sailfish, except for the work Jolla is doing.

The successful open-source projects are successful, because people employed by companies do paid work on them in the interest of their company. The Linux kernel is a prominent example of how a range of companies are working together on a common
operating system.
And if you look at Mer or Nemo, you'd notice that most contributions are coming from people paid by Jolla.
Did you expect random people would do Jolla's work for free?

Just be thankful that they're scratching their itches and gracefully embrace their free contributions. Like every other open source project.

Originally Posted by pycage View Post
And with open-sourcing the few remaining closed parts of Sailfish, this would not change.
Maybe. So what? Why is that an argument against open sourcing 100% of Sailfish?

I call this the "Netscape mentality".

You should never _expect_ people to work for you for free. If it happens, then you embrace it with your arms open. But since when the deal's been "i'll only open source this if you implement feature X for me?".

Originally Posted by pycage View Post
The community consists of consumers with a few active developers among them. Most open-source projects are one-man-shows that disappear once the developer loses interest. Almost all bigger projects are actually paid work.
No argument. But again, so what?

Originally Posted by pycage View Post
The community of developers is a utopia.
Couldn't disagree more.

I have developed a number of small utilities or hacks for desktop Linux that most people have never heard about. Bus factor=1. Yet I still receive daily emails about them; with ideas, patches, bug reports, questions and comments. Random people I have never heard about have even packaged them for some Linux distributions. Some of them large ones, some of them probably one-man shows.

Do you know what do I call this?

I call this a fscking miracle.

The number of man-hours contributed by these persons greatly exceeds the amount of time I've spent on these projects. I use them almost daily and yet hardly ever find a big enough free time slot to work on them myself. Thus, I almost always encourage any person who emails me to set up a fork. They never do it either, because they lack the time too.

And yet with the few man-minutes that everyone can offer, the projects improve, and _I_ benefit from the results.

To sum it up: I get a _daily_ reminder about this wonderful utopia of developers. And my small utilities are mostly niche garbage.

I'm sorry your impressions are different.

Originally Posted by pycage View Post
Why don't we see open devices running Mer? I think the best person to answer this question would be Aaron Seigo of the Vivaldi project.
I would ask Jolla. Why did they need to make SailfishOS? And why did they need to make it closed?

Just to repeat past Nokia mistakes in an era in which we are enjoying peak levels of open-sourceness in the competition?

I would never expect Jolla to be the developer of the most closed source mobile operating system. Yet now this scenario no longer seems as unlikely as it would have seemed on day 1. Again, ironically, many other open operating systems will be reusing the components Jolla developed!

Originally Posted by pycage View Post
It simply is not possible to find
acceptable mobile hardware running on open drivers. The Jollavdevices are as open as
it can get. And they are running Mer on top of a closed-source Android hardware adaptation layer dictated by the manufacturer.
Trade-offs. You have the 1000eur phone quite close, on this very forum, which shows you can actually do acceptable open mobile hardware. Stallman was even close to approving it; the FSF did not approve it based on what many people believe is a small technicality. That's a device that is "as open as it can get".

But despite the Android-only hardware, I find I easily like Jolla's hardware, possibly even more than the software. TOH. Schematics!. Actual access to people who know the hardware. I can ask for an unlocked bootloader and I get it. Without having to needlessly explain "why would anyone want an unlocked bootloader" for the nth time. All these things are mostly unheard on most other manufacturers. Even the manufacturers chosen by Ubuntu are crap in this regard.

One of the primary reasons I use a Jolla daily.

Originally Posted by pycage View Post
On the other end is the Silica UI that was promised to be opened (stay tuned as stskeeps said...), but actually is almost entirely BSD-licensed and open already.
Why didn't anyone reimplement the small closed-source part of Silica to make the
BSD-licensed open components run on top of Nemo, so that you have the full
Silica experience on Nemo? After more than two years of having the BSD-licensed Silica code out in the open, I really wonder. Maybe because the community of developers is a utopia...
Again, why would I work for free? I am paying Jolla to do this for me!

Now, release the source of the mail client UI, and I will quickly make a hackish patch to enable multiple identity support, since that's one of my itches*. Since I only have a few man-minutes of free time maybe I will just hardcode the list of identities to avoid having to understand how account setup works. The result would be useless as a Nemo contributon, but still useful for some, developer-y people. And maybe someone else will eventually develop the UI to configure it.

* fictitious example

Originally Posted by pycage View Post
IMHO it is NOT Jolla who is to blame for the lack of contributions to Sailfish.
But they are definitely to blame for the closeness of Sailfish.
 

The Following 18 Users Say Thank You to javispedro For This Useful Post: