First thoughts about the (pre) Sailfish OS 2.0 - Page 37 - maemo.org

ste-phan | # 361 | 2015-09-15, 20:59 | Report

Originally Posted by marxian View Post

I suspect they are not usable for everyday use, because all those people that are claiming that they would be willing to contribute to Sailfish, if only it were FOSS, have not bothered to do so in the case of Nemo. This naturally raises the question of whether they really would contribute, or whether it's simply a load of gumflap. In this context, terms like 'we' or 'the community' usually mean 'someone else'.

To answer that question, I want to contribute to Sailfish, but since I am not a developer and Jolla doesn't let me support them apart from purchasing a tablet which I did not desire let alone need, and which in retrospect puts Sailfish on a side track with cool features like peek, cover gestures, pulley favorites menu, favorites application menu, OOM ,affected for the worse I would love to financially support freelance people willing to put effort in Sailfish OS to make it their gem.

I have stated many times also on other websites that I would pay 100 Euro for a Sailfish License fully supported on a few solid hardware (Fairphone, Yotaphone, Samsung if it must, or better Nokia ex-Windows Phone).

Would it not be nice and sufficient if Jolla would open up everything pre-Sailfish 2.0 for the community (and let Sailfish 2.0 become the bloatware Android clone alternative for big telecom giant's budget phone)? I suppose Jolla 1.x mid 2015 would be a nice enough base?

The Following 3 Users Say Thank You to ste-phan For This Useful Post:
Amboss, juiceme, oenone

Feathers McGraw | # 362 | 2015-09-15, 21:01 | Report

Originally Posted by pichlo View Post

the things I always found the most demotivational could be summed up in just two bullet points:[list][*]A lack of clear direction. The management may have an idea where the project is going but this is not being communicated to us little pawns. This leads to micromanagement: they tell you to do X, without telling you how it fits in the big picture or even who works on Y that directly interacts with your X. The impression is that even the management has no idea where they are going.

I think your post is quite insightful, but I don't think this point applies to Jolla based on what I've read about how they work. Don't they have those regular meetings/group discussions where everyone talks about what they're working on and how it fits into the big picture?

The second point fits Jolla very well though!

The Following 2 Users Say Thank You to Feathers McGraw For This Useful Post:
Amboss, pichlo

ajalkane | # 363 | 2015-09-15, 21:04 | Report

It seems to me there's an abundance of complaining about Sailfish 2.0, so I just want to say that I love the changes to the UI. Not every one of the changes, but on the whole I feel it's a good move forward. The main OS feels more fluid and it feels more exciting to use - which might seem as totally superfluous, but I'm all for being excited while fiddling my gadgets :-D.

Of course I have some annoyances with the changes... but many of them are fixable with community's fixes, and some of those I feel Jolla would well heed to take a close look to at least implement them as optional setting.

My biggest worry is the OOM, but that has nothing really to do with Sailfish 2.0. My uneducated gut-feeling is that it's mainly because QML is a memory hog, and that could be improved - but it's probably outside of Jolla's scope and requires improvements from upstream.

The Following 17 Users Say Thank You to ajalkane For This Useful Post:
Amboss, anig, Bundyo, Copernicus, Feathers McGraw, Jordi, juiceme, JulmaHerra, late88, Manatus, minimos, mosen, P@t, pichlo, vistaus, vitaminj, zamorph

MartinK | # 364 | 2015-09-15, 21:17 | Report

Originally Posted by Copernicus View Post

Which leads me to the obvious (at least to my mind) question: why in the world would you want to open-source Sailfish?

Why would I want it ? Well, that's simple - I'm working full time on the Fedora and RHEL installer.

Both Fedora and RHEL all completely open source, so Sailfish OS being half-proprietary feels awfully backwards. Especially if you can compare it with fully open distros on a daily basis and see all the problems the half open/half nature of Sailfish OS is causing.

Also so far I don't think anybody has been able to present any benefit the proprietary bits are providing to Sailfish OS - after more than 2 years after launch the only two devices officially running Sailfish OS are both made by Jolla. Even though there are some unclear shreds of information about the Intex device, I'm not sure a single obscure OEM from India after two years of trying is worth all the problems the closed bits are causing.

Not to mention that there are more ways to go about it than just close source stuff left and right, such as:

licensing your brand
licensing your artwork/themes
providing support, (security)fixes and feature development to your licensees

That way you will still get money without pissing of all your contributors who are already used to contributing to much more open projects.

And as for general reasons why you would want a fully open Sailfish OS:

security - unless everything possible is open the OS can't be really trusted due to missing peer review
community QA - community can help test stuff continually, rather than when it is actually too late to fix the the bugs as release is almost ready (you paid testers will never find all the bugs)
open development - you can watch, contribute and influence the development as it happens rather than once it is actually done and hard to change
actually being able to fix and improve stuff - without needing a change in a closed component that Jolla will never do due to being overworked as hell
much less pressure on Jolla developers from the community as people can do many things themselves and "patches or GTFO!" is now a valid response to many demands
makes a community distro (without your licensed artwork and trademarks) much more viable - you don't need to rewrite the whole UI and Sailfish OS can actually make use of the stuff you do
future proofing - if a meteorite obliterated Finland tomorrow we will be in similar situation we have been at least twice before (N900 & N9) - a half/open half closed system that is hard for the community to maintain no to say extend and develop further

The Following 10 Users Say Thank You to MartinK For This Useful Post:
Amboss, Copernicus, Feathers McGraw, Jordi, juiceme, Manatus, minimos, nodevel, salyavin

MartinK | # 365 | 2015-09-15, 21:20 | Report

Originally Posted by ajalkane View Post

My biggest worry is the OOM, but that has nothing really to do with Sailfish 2.0. My uneducated gut-feeling is that it's mainly because QML is a memory hog, and that could be improved - but it's probably outside of Jolla's scope and requires improvements from upstream.

Check if you are not another victim of [bug][1.1.9.28] tracker-miner-fs stuck in a loop , this alone can result in tracker alone eating 20-50% of memory and 100% of CPU for long periods.

The Following 2 Users Say Thank You to MartinK For This Useful Post:
juiceme, minimos

pichlo | # 366 | 2015-09-15, 21:20 | Report

Originally Posted by itdoesntmatt View Post

If they opened completely the ui, what an hardware company would have pay Jolla for? they could take os and use it as it is.. or i am missing some logic step?

I never understood it either but some people do quite well selling free software

The Following 3 Users Say Thank You to pichlo For This Useful Post:
Amboss, Copernicus, juiceme

MartinK | # 367 | 2015-09-15, 21:23 | Report

Originally Posted by pichlo View Post

I never understood it either but some people do quite well selling free software

I'll just leave this here.

The Following 8 Users Say Thank You to MartinK For This Useful Post:
Amboss, Copernicus, juiceme, Manatus, mattaustin, minimos, pichlo, salyavin

MartinK | # 368 | 2015-09-15, 21:30 | Report

Originally Posted by smoku View Post

What about folks that just want to fix annoying bugs in already existing piece of code, instead of wasting time on developing it from scratch?

That's pretty much nails it - even if Nemo Mobile might want to eventually go a different way (Glacier components instead of silica, etc.) forcing them to wasting time on reimplementing half the OS as the first step has resulted in the unfortunate lack of progress we are seeing...

The Following 2 Users Say Thank You to MartinK For This Useful Post:
juiceme, smoku

pichlo | # 369 | 2015-09-15, 21:38 | Report

Originally Posted by MartinK View Post

Also so far I don't think anybody has been able to present any benefit the proprietary bits are providing to Sailfish OS - after more than 2 years after launch the only two devices officially running Sailfish OS are both made by Jolla.

I think you may be confusing things a bit here. The above is not about persuading anyone about the benefits of closed components. It is about persuading anyone about the benefits of Sailfish.

As an OEM, you would not give a flying duck whether it is open or closed. All that would matter to you is, "how many units does it help me to sell?" And please, put your hand on your heart and tell me, why would you as an OEM take the risk with an obscure OS, with a very limited user base, backed by a small company with a proven track record of not delivering or delivering hopelessly late, in preference to a mature and popular OS backed by a multi-billion dollar company?

I know that if I were such an OEM, I would expect Jolla to pay me for taking the risk. I would definitely not pay a dime. Esoteric concepts like open or closed would play no role in my decision.

The Following 5 Users Say Thank You to pichlo For This Useful Post:
Amboss, Copernicus, Jordi, minimos

MartinK | # 370 | 2015-09-15, 21:42 | Report

Originally Posted by MisterMaster View Post

For example aren't the documents and browser apps open-source? How many have contributed to those? 1(well maybe more?) guy for the documents and a few for the browser? Well I don't know how hard it is to contribute to those so that might be one reason if it is difficult.

There could be also network effect at play - even though the apps themselves might be open source, that does not mean you won't get blocked by issues in closed components you can't fix.

Take multi user chat (MUC) as an example - the backend (Telepathy) is open as is (IIRC) the accounts framework - it might both actually already support multi user chat or could be added by the community due to the respective components being open. But the chat application UI is closed source, so even if the backend supports it or community adds it to the UI the only one who can change the UI is Jolla...

This is probably why the Cyanide Tox client and the various whatswhatever cleints are standalone apps rather than nicely integrated parts of the built-in chat system.

And the situation around Sip support - community has done it's part and has been waiting for Jolla to make the necessary changes in the closed parts for months, citing:

Originally Posted by

As the community has pretty much exhausted their possibilities and done things in the open source parts of Sailfish, it's now up to Jolla to eventually hold their part of the bargain.

The Following 5 Users Say Thank You to MartinK For This Useful Post:
Amboss, javispedro, madry72, mattaustin, max_power