ReactOS Forum

Well, Jeromê Gardou is going from ROS. He has done great work, we appreciate all of them. But, it's big loss for ROS! Okay, we are out of money... what will we do then?

1) This is reaction on https://reactos.org/node/898 if anybody wondered
2) The contract is ended at the end of the month, so 10 more days of awesomeness
3) Jeromê is just switching his attention to what he finds interesting instead of what the community finds interesting - he works on ReactOS since 2010 and he has done cca 1000 commits over that time - see here http://code.reactos.org/user/zefklop

Kind of great actually

Maybe he will continue work on related things. I think at least already started work will be finished even if the contract ends.

Don't shock me again so much. I got feared when I read it.

You people really need to stop overreacting and taking things out of context. Jerome has been with the project for years and he spent much of that period working on ReactOS in his spare time. The reason we initiated a contract with him was that so he could commit more time over the course of a month to specific tasks because we needed attention paid to some specific things. So of course he is not leaving.

Perhaps his contract can be extended? Because the memory manager is so crucial it would help the OS to run more stable with less problems and more applications would run. I would be willing to donate towards that goal for any developer to continue work on it.

Would this slow development down?

Amine is working on it. It's not so much money as again German red tape.

ctasan wrote:Well, Jeromê Gardou is going from ROS. He has done great work, we appreciate all of them. But, it's big loss for ROS! Okay, we are out of money... what will we do then?

No he is not. He just won't be paid for a specific assignment. The devs are mostly volunteers. The reason for paying certain ones is to help them clear their schedules to work on specific code. They all have a proven track record working on ROS already. Nor is the project out of funds. It is just that he has completed or is completing the task the money was earmarked for. He is still free to work on ROS - as is anyone who is able, not in a conflict of interest, and hasn't seen Windows source code.

What I understand was that the news was intended to be a celebration, not considered a loss.

Exactly, Jeromê's work delivered in this month is exciting and I hope he and the foundation will be able to close another contract in a near future! It's of course compliments to the whole team, but the focus made possible by this contract is impressive.

ctasan has so far a track record of using words such as "sad" or "inadequate" when talking about ROS. But let's hope he will stay with the project long enough to get positive and appreciate every small improvement, every bug fixed and every new app running

Yeah, this is just a "Release note" that should have been released one month ago really however it kept in our ToDo list. This is more like a "Ey guys!We hired thanks to your donations a Dev and look how wonderful changes are happening" (he is working in Java, MM, and AFD issues among others).
Hope we can convince him to keep working in such wonderful areas, because his work (and his teaming with the rest of the Devs) have shown a great ReactOS push.

Let's see if we can convince him a little bit more!!

Well, I have this records, Black_Fox. But one day, I will say only "legend" about ReactOS. I also wait some of major improvements: for example support for setups that have 16-bit stubs, proper DX support and my list goes on..m

Well, I'm writing here good news: Jeromê is doing a legend work!

He has created a new branch for reimplementing networking stuff, which is a blocker for ReactOS now. Branch name is tcpip_revolution. Absolutely this is a revolution.

I'm glad to see that! You?

I know too little about the current state of networking in ReactOS regarding the lwIP integration. From reading the commit messages I can guess that the stack wasn't good enough and so that lwIP is going to be used much more thoroughly and handling only minor number of cases will have to be written. I'm interested in watching the further development, especially since it seems like a full rewrite of TCP/IP stack.

P.S. I took a look at wiki and actually found that the correct way of spelling zefklop's name is "Jérôme", we had it a bit wrong.

Yes, there seems to be 2 main issues there. First is totally gutting the old code to get LWIP to doing more, and the other is getting rid of race conditions that cause crashes. It seems it might be easier to gut out all the pre-LWIP networking components and then build new supporting code that is compatible with it and any changes in ROS since then.