Google Summer of Code 2019 Ideas

Here you can discuss ReactOS related topics.

Moderator: Moderator Team

dizt3mp3r
Posts: 1488
Joined: Mon Jun 14, 2010 5:54 pm

Google Summer of Code 2019 Ideas

Post by dizt3mp3r » Tue Jan 08, 2019 2:47 pm

https://reactos.org/wiki/Google_Summer_ ... 2019_Ideas

My vote goes to Paravirtualisation or perhaps more importantly but less sexy, the Kernel mode test suite.
Skillset: VMS sysadmin 20 years, fault Tolerance, cluster, Vax, Alpha, ftSparc. DCL, QB45, VB6, NET, PHP, JS, CMS, Graphics, Project Manager, DOS, Windows admin from 1985. Quad Electronics. Classic cars & motorbikes. Artist watercolours. Historian.

milawynsrealm
Posts: 43
Joined: Wed Jul 20, 2011 10:48 pm

Re: Google Summer of Code 2019 Ideas

Post by milawynsrealm » Tue Jan 08, 2019 5:10 pm

For the PDB support, I found a github project put there by Microsoft (https://github.com/Microsoft/microsoft-pdb). It seems to be under the MIT license, so it should be usable by the project.

Ancient
Posts: 81
Joined: Tue Mar 27, 2018 11:32 pm

Re: Google Summer of Code 2019 Ideas

Post by Ancient » Wed Jan 09, 2019 3:57 am

USB support through USB C, USB boot, USB PnP, removable and fixed drives supporting all Linux file systems and NTFS.

Secondary target, basic HDMI through 1080P video and sound.

Third target, ANYTHING else missing to get the beta out the door.

Totally open source Chromium (no binary blobs) for React OS. As Windows is moving to Chromium, ROS could be all caught up by the end of summer.

Ask for 3-5 programmers.

knowmad
Posts: 17
Joined: Mon Oct 20, 2014 11:04 pm

Re: Google Summer of Code 2019 Ideas

Post by knowmad » Wed Jan 09, 2019 4:11 am

I think finishing USB support, NTFS. PnP and printing. It could very well be possible any of those four could be developed into a more complete and perhaps even usable state. But any progress in anything is all good and I do hope there will be more than one mentored student this time around.

ThFabba
Developer
Posts: 285
Joined: Sun Jul 11, 2010 11:39 am

Re: Google Summer of Code 2019 Ideas

Post by ThFabba » Wed Jan 09, 2019 9:30 am

Coming up with good GSoC projects for ROS usually means coming up with the smallest/simplest projects you can think of. All of the "ROS could really use this" problems are way too complicated for someone to get familiar with and solve in just a few months, especially if they have no Windows kernel development knowledge (which few professionals do, let alone students).
Most of the things that we're stuck on, we're stuck on for a reason: because they're really hard problems.

So I would advise: think small!
GSoC is more about gaining long-term contributors than solving our biggest problems. Which is actually why having a smaller number of students can be good -- it means we don't have to stretch our extremely limited mentoring capabilities out too thin, therefore the students will have a better experience and higher chances of success, which will make them more likely to stick around.

PurpleGurl
Posts: 1788
Joined: Fri Aug 07, 2009 5:11 am
Location: USA

Re: Google Summer of Code 2019 Ideas

Post by PurpleGurl » Wed Jan 09, 2019 12:14 pm

knowmad wrote:
Wed Jan 09, 2019 4:11 am
I think finishing USB support, NTFS. PnP and printing. It could very well be possible any of those four could be developed into a more complete and perhaps even usable state. But any progress in anything is all good and I do hope there will be more than one mentored student this time around.
I agree. Students already did USB and NTFS work, so we could use more. PnP is a bit tricky. Now they might be able to help with printing.

The testing and conformity tools are worth pulling for, as they'd let us research differences and help us get closer to Windows in the deeper parts of ROS that are not otherwise reachable. That can improve things for existing devs as they'd have more raw data to work with.

dizt3mp3r
Posts: 1488
Joined: Mon Jun 14, 2010 5:54 pm

Re: Google Summer of Code 2019 Ideas

Post by dizt3mp3r » Wed Jan 09, 2019 12:49 pm

Ancient wrote:
Wed Jan 09, 2019 3:57 am
Third target, ANYTHING else missing to get the beta out the door.
You need to dispense with this idea once and for all. Nothing anyone does during GSOC is going to deliver a beta, not this year, not next, nor the year after that. It is nonsensical to keep going on about it.

Stability improvements will be gained through enhancing the kernel test suite and that is important to me, that is why I mentioned it. Kernel programmers seem to be few and far between and it would be good to attract up and coming kernel programmers. As GSOC is all about attracting students I suppose our own personal wishes are largely irrelevant as the students will volunteer to do only what they are able to do.
Skillset: VMS sysadmin 20 years, fault Tolerance, cluster, Vax, Alpha, ftSparc. DCL, QB45, VB6, NET, PHP, JS, CMS, Graphics, Project Manager, DOS, Windows admin from 1985. Quad Electronics. Classic cars & motorbikes. Artist watercolours. Historian.

shunesburg
Posts: 178
Joined: Wed Feb 21, 2018 3:46 pm
Location: Somewhere in France

Re: Google Summer of Code 2019 Ideas

Post by shunesburg » Wed Jan 09, 2019 12:58 pm

We need in priority (in this order) : a generic driver for Ethernet card, a good support of USB device (USB Key, WiFi Stick, Printer...), a Print Spooler, a WiFi Manager, and a ReactOS Update.

dizt3mp3r
Posts: 1488
Joined: Mon Jun 14, 2010 5:54 pm

Re: Google Summer of Code 2019 Ideas

Post by dizt3mp3r » Wed Jan 09, 2019 2:11 pm

- but this is not about what we need. It is about what we can entice students to create.
Skillset: VMS sysadmin 20 years, fault Tolerance, cluster, Vax, Alpha, ftSparc. DCL, QB45, VB6, NET, PHP, JS, CMS, Graphics, Project Manager, DOS, Windows admin from 1985. Quad Electronics. Classic cars & motorbikes. Artist watercolours. Historian.

PurpleGurl
Posts: 1788
Joined: Fri Aug 07, 2009 5:11 am
Location: USA

Re: Google Summer of Code 2019 Ideas

Post by PurpleGurl » Wed Jan 09, 2019 2:24 pm

dizt3mp3r wrote:
Wed Jan 09, 2019 2:11 pm
- but this is not about what we need. It is about what we can entice students to create.
True. We have to think more in terms of the long game. Right now, give them the low hanging fruit, and if they want to stay to do the other stuff, they can. We don't want to assign a hard one they give up on and won't bother to finish, or a moderate one that is almost done that they won't return for. It would be better to finish a small one, get a good taste for ROS, and want to aspire for more later.

Ancient
Posts: 81
Joined: Tue Mar 27, 2018 11:32 pm

Re: Google Summer of Code 2019 Ideas

Post by Ancient » Thu Jan 10, 2019 7:42 am

dizt3mp3r wrote:
Wed Jan 09, 2019 12:49 pm
Ancient wrote:
Wed Jan 09, 2019 3:57 am
Third target, ANYTHING else missing to get the beta out the door.
You need to dispense with this idea once and for all. Nothing anyone does during GSOC is going to deliver a beta, not this year, not next, nor the year after that. It is nonsensical to keep going on about it.

Stability improvements will be gained through enhancing the kernel test suite and that is important to me, that is why I mentioned it. Kernel programmers seem to be few and far between and it would be good to attract up and coming kernel programmers. As GSOC is all about attracting students I suppose our own personal wishes are largely irrelevant as the students will volunteer to do only what they are able to do.
In 3 years if not out of beta ROS will be dead imo. You'll have no developer interest at all, attrition and forking will result in a few forks maintained by singletons, and there will exist no cogent community. If Google will help, ask for what you need. Bites at the apple will vanish shortly. Today it's 8 core, 16 thread on 32 GB of DDR6 RAM with a 16 GB video card, in 3 years it will be 128 core, 256 thread, 1TB god knows what RAM with 256 GB video cards popping up on 32K 3d no glasses needed holographic immersive floors. A circa 2003 goal in the still distant future for a single core 32 bit system with no drivers will not be taken seriously.

You will want to recruit teams of 2 or 3. Maybe 3 teams if you are lucky.

If the goal is to languish this into oblivion, well that plan is so far working very well. I wish it weren't. Ask Google for a professional software development manager to help set up teams with the goal of getting something out the door within a year. Google has resources, some at Google corporate are talented and bored or interested in strange stuff.

Ancient
Posts: 81
Joined: Tue Mar 27, 2018 11:32 pm

Re: Google Summer of Code 2019 Ideas

Post by Ancient » Thu Jan 10, 2019 7:45 am

Best of luck!

ThFabba
Developer
Posts: 285
Joined: Sun Jul 11, 2010 11:39 am

Re: Google Summer of Code 2019 Ideas

Post by ThFabba » Thu Jan 10, 2019 10:39 am

What? We don't have any special relationship with Google. The GSoC program is offered as-is. You don't get professional developers, or managers, you get students. And you don't get teams, you get individuals.

dizt3mp3r
Posts: 1488
Joined: Mon Jun 14, 2010 5:54 pm

Re: Google Summer of Code 2019 Ideas

Post by dizt3mp3r » Thu Jan 10, 2019 2:11 pm

Ancient wrote:
Thu Jan 10, 2019 7:42 am
If the goal is to languish this into oblivion, well that plan is so far working very well. I wish it weren't.
Honestly, you are just another opinionated chap that wants ReactOS now and as it isn't happening in your own personal timeframes then it is no good for you. Well, that's fine, so that means you can go now and look for a project/product that suits your own needs/timescales and leave ReactOS to itself.

If you've been on the ReactOS forums for a while as many of us have then you wouldn't be able to count the number of times that the same old argument has been proposed. People come, they moan and then they leave. You might as well do the same now and save us the time.

Reality though, is that nothing that you are going to say/request is going to change anything unless you commit cash to pay for developers. Three to five years is the current reality and if you can't face that then you simply don't understand the scale of the task.

Beta in any case is a woolly target. All you need is to find a version of ReactOS that works for you in a specific instance.

Anyway, this is now a silly thread and a regurgitation of an old discussion had many times. I refer you to those. No-one, certainly not the developers are going to listen to your demands for more speed now!
Skillset: VMS sysadmin 20 years, fault Tolerance, cluster, Vax, Alpha, ftSparc. DCL, QB45, VB6, NET, PHP, JS, CMS, Graphics, Project Manager, DOS, Windows admin from 1985. Quad Electronics. Classic cars & motorbikes. Artist watercolours. Historian.

dizt3mp3r
Posts: 1488
Joined: Mon Jun 14, 2010 5:54 pm

Re: Google Summer of Code 2019 Ideas

Post by dizt3mp3r » Thu Jan 10, 2019 2:38 pm

Hijacked thread now returning discussions to GSOC 2019 and what we can suggest as possibly interesting technical challenges for new students as well as those we'd like to push for ourselves and our own benefit...
Skillset: VMS sysadmin 20 years, fault Tolerance, cluster, Vax, Alpha, ftSparc. DCL, QB45, VB6, NET, PHP, JS, CMS, Graphics, Project Manager, DOS, Windows admin from 1985. Quad Electronics. Classic cars & motorbikes. Artist watercolours. Historian.

Post Reply

Who is online

Users browsing this forum: Ahrefs [Bot], DotBot [Crawler] and 2 guests