Page 1 of 2

Sinking Into Oblivion

Posted: Thu Nov 18, 2010 12:13 am
by Ubuntu
Hello,

I am quite concerned to see the lack of interest in my Google searches about ReactOS. Here I find out that version 0.3.12 has been released, and there is hardly any comment on the web about this. OSNews and Phoronix have tiny little comments about it and there is not much more to be found I am afraid.

Has anyone else noticed that ReactOS seemed to receive less attention than in the past?

I hate Ubuntu's 6 month release cycle because I think it happens too often and that once a year would be better. But I do agree with previous threads in this forum that a 6 month release cycle would be a huge benefit to ReactOS popularity, as it has done for Ubuntu. Giving people something to look forward to - as many do with Ubuntu's release dates - would be a major boost for ReactOS. But I know the devs don't like the idea.

Re: Sinking Into Oblivion

Posted: Mon Jan 10, 2011 11:59 pm
by Corbin052198
I think maybe one release a year would be good, if the developers can manage it.

I also agree about knowledge. ReactOS has far less publicity than Linux or even Wine. ReactOS has a future, if it can keep it's community. :)

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 12:34 am
by greenie
A release should happen for major changes really. Ubuntu can give 6 month release dates because they can make many changes in that time that users will benefit from. Though they still have targets and milestones for each release. They just schedule for it.(and don't forget they have paid developers which they can expect to meet milestones)

Though most changes most people who give reactos a try may not notice. and that is with the current system of only doing a release when goals are achieved. It's true that more awareness would be awesome. Though I don't think just chucking out a release every 6 months regardless of how little is done will improve the profile of reactos. In fact it kind of sounds like it will hurt the profile of reactos. The way this would differ is if devs decided to work a bit harder because they wanted their fix in the latest release. Which I don't think will happen.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 2:16 pm
by Aeneas
Well... I have always dissented to the "it's ready when it's ready" approach. Guess what:

- When you get sued, you have a time frame to respond at court. That time frame is in Austria four weeks. That is the same for grand corporations and for small consumers.

- When you have to pay back a credit to a bank within two years, you have to pay it back within two years.

- When your professor at university gives you two weeks to prepare for a test, you have two weeks to prepare for a test.

- When at work your boss orders you to do something TODAY, you have to do it TODAY.

- When your wife asks you to bring out the trash NOW, your time-horizon is NOW. ;)

In EVERY SPHERE OF LIFE, professional or private, time DOES matter. I see absolutely no reason why it should not matter in a community project. It is a sign of being well-organised and, if you will, disciplined. "It's ready when it's ready" I could say about a cake, too, but if you have guests at 5pm, well, guess what, it should be done by then.

If I remember correctly, there was some talk of having releases every three months. Well, let us see what will come out of this. - I know well that "this is just a hobby" and that "it is so difficult", however, positive publicity does not originate in explanations of this sort but rather in achievement - which, fairly or not, is measured usually in up-to-date-releases.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 2:23 pm
by Aeneas
Oh and one more thing:

I once worked in a (large) law firm and I was told something interesting:

"The client cannot measure success by quality. Quality differs from the point of view and the client does not have enough expertise to evaluate it. What matters is the time you respond: just make the answer good enough so that he can use it, but try to give it to him as quickly as possible."

It has always been true for me in my legal profession, and I could imagine that it would be not unlike for a computer project: only rarely will anyone truly evaluate what ground-breaking progress has been made in a release (and you all be sincere - do you?); but easily one can see, WAS there a release or NOT.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 2:26 pm
by vicmarcal
Aeneas wrote:
If I remember correctly, there was some talk of having releases every three months. Well, let us see what will come out of this. - I know well that "this is just a hobby" and that "it is so difficult", however, positive publicity does not originate in explanations of this sort but rather in achievement - which, fairly or not, is measured usually in up-to-date-releases.
-Find us 10 Dedicated Testers
-Find us 10 Widespread Devs

Then we can begin talking about up-to-date releases

-Pay us 3 fulltime devs

Then we can begin talking about roadmaps


You can be part of the solution, or just seating in the forum. You, as all of you.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 2:41 pm
by Aeneas
Sorry, vicmarcal, I dissent again:

it is NOT upon the people here, eventhough we are all having an obvious interest in ReactOS, to actually organise anything. If you do not like a tax law, you do not go to shoot the parliament personally and then become dictator; rather, you elect a party which changes things the way you want - and that party hires its own experts, for that matter. If you do not like the treatment of whales, you aid SeaShepard, but no, you do not go out on yacht armed with torpedoes. For most things, there is an institutionalised path which is to be followed. So is for ReactOS, perchance even informally - the core devs.

My contributions so far will be just semi-serious tests (FreeSSHd does not work...) and definitely some suggestions - until this project actually has working USB-support. Then I believe I may choose to deepen my interest a little more. And in the past I have made enough suggestions; I still believe my best one to have been to broadly advertise then need for people in dedicated IT-boards; even if you pick 10-15 hobbyists from all the thousands you can reach, that would be a good step forward.

But guess what: all my past suggestions went NOWHERE due to the lack of consent or action by your central institution - the core devs. THEY have to really wish to direct this into a more stringent way of progress.

Because even if someone did what you propose and, say, picked you three paid developers... then what? Would that change anything? - Image the three being good professionals and DISSENTING to some development decisions of others. - If you think it to the end, uncoordinated "help" may actually split and harm ReactOS.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 4:45 pm
by mrugiero
I think the "laws" approach is forgetting about children (could someone think about the children?! :lol:), and this project is not in its majority age. It's still an alpha, it is like a children, is far from complete. And children can't be sued (not in my country, except for murdering or something like :P), etc.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 5:11 pm
by Aeneas
No, I did not mean "laws approach" in such a negative manner and I apologise if my criticism appears too harsh. I wished to express my critical stance on the defetism that can be observed here periodically and, unfortunately, not without reason. Some of the brightest minds are working on this project, once it succeeds, it can make use of the wealth of applications already developed for windows (in fact, the by far biggest entry barrier for any OS - look at Plan 9! Or compare FreeDOS and Dosbox - the only reason THESE exist are the applications, few people are too impressed by DOS itself)... and yet it creeps along at a miserable pace (objectively, no offense to anyone intended) for no all too impressive reasons whatsoever but just because it cannot set a goal and actually reach it in a reasonable time.

We did not reach 0.3.13 yet. We shall see, I guess, even 0.3.15-0.3.16 ere it can finally get to 0.4.0. At that point, the "window of opportunity" due to XPs exceptionally long standing will begin to close.

Posted: Tue Jan 11, 2011 5:57 pm
by hto
What do you mean by "it cannot set a goal"? The goal is here, it's just a lack of manpower.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 6:35 pm
by vicmarcal
Aeneas wrote: But guess what: all my past suggestions went NOWHERE due to the lack of consent or action by your central institution - the core devs. THEY have to really wish to direct this into a more stringent way of progress.
Could you post a list of some of those suggestions?
Aeneas wrote: Because even if someone did what you propose and, say, picked you three paid developers... then what? Would that change anything?
A lot.
Each user/follower is expecting changes in a particular place of ReactOS:one is expecting movement towards USB (as seems to be your case), others towards DirectX, other towards stability, others towards networking. Others wants everything at the same time(impossible).

So there are 2 possibilities to make our users/followers happy:
-Having a lot of devs, so the statistic says they will be coding in all the ReactOS areas.

But unforntunatly it is not our case, so 2nd possibility:

-Having hired devs working in the code that noone is developing for now. So we can have a hired dev developing USB permanently, another in networking and other in stability; and if suddenly 3 free USB devs join the project, then the hired USB dev can work in another area (Directx?)but following the USB commits.


Also with hired devs we can set roadmaps as we pay them and they are fully working for us.They can create their own roadmaps with no pressure. Currently a roadmap is impossible since our devs are not just focused in ReactOS and we can not force them to do that.

So yes, money does a lot.

Aeneas wrote:We did not reach 0.3.13 yet. We shall see, I guess, even 0.3.15-0.3.16 ere it can finally get to 0.4.0.
More devs.More testers.More Designers.More PR guys.More money. More of all.

The jump to 0.4 is quite near. Imo the road to 0.3.13 is the road to polish all of our critical changes (MM rewrite, CC rewrite, Heap rewrite, MSG rewrite; they are a few and quite critical) and after that we will begin the road to 0.4 which will be a soft polish,improving our apps compatibility. Then comes road from 0.4 to 0.5, which will be short but critical.It will be quite fast, but 0.5 is our first Beta so we want to have it polished too.
Imo,0.4 should be released as pre-Beta instead an Alpha one, as right now 0.3.13 is 800% more apps compatible than 0.3.0.

Re: Sinking Into Oblivion

Posted: Tue Jan 11, 2011 7:22 pm
by EmuandCo
I fully second the pre beta idea

Re: Sinking Into Oblivion

Posted: Wed Jan 12, 2011 1:00 am
by Aeneas
@ vicmarcal

Why should my past proposals be that important? (But OK - I proposed involvment with international programmes preferrably of the EU; however, the EU wants to see results in fixed time frames if it is to provide especially financial help. I proposed getting involved with companies. I proposed getting involved with universities, but an Austrian professor expressed little interest, stating sincerely that for teaching, his colleagues use genuine Windows, and for research, a new Windows-like OS is not going to really advance anyone's scientific career.)

So let me just state the two things I stick to:

1. - Find forums where you can reach professionals and advertise there. For Austria, that would be, as proposed by the professor I contacted, "http://www.informatik-forum.at/".

2. Once USB works enough: create bootable USB-sticks with ReactOS & distribute them for free at universities; have a USB-image, too. - Yeah, I know, that needs some money. But I cannot imagine that USB sticks with, say, 2-4GB will be THAT expensive in, say, 2012-2013. You might distribute a hundred or so in IT faculties - in fact, the devs could do it among their acquaintances. - You really need to bring down the entry level barrier; not many people are going to dedicate a disk for the experiment or even to install a virtual machine - that is simply too much hassle. - Here was just very recently reading the excellent idea to make it possible to start ReactOS like a problem as Ubuntu from Windows through Wubi. So yes, maybe a variation could be a "program" that automatically installs ReactOS in a virtual machine. - Call it as you will, but it must be easy to check out ReactOS.

And I am very happy if Emuandco, who certainly has much better judgement of the matter than me, finds your pre-beta idea good, as this would be indeed encouraging regarding ReactOS' public image.

Re: Sinking Into Oblivion

Posted: Wed Jan 12, 2011 5:04 am
by greenie
Aeneas I think release dates just based on dates is not beneficial. We can have targets for a release which in the working world could give you a time frame when you work 8 hour days as your job. but to say to a contributor that you have to meet deadlines is ridiculous. I work 8 hour days with 3 hours travel I have to have 8 hours sleep and i need time to cook,clean,eat,shop, etc . If I watch a movie instead of working on reactos that is defiantly acceptable.
If you feel anyone should not do that. You can do legal work for me on my schedule for free :P.

Even in business I've seen many projects fail because someone thought the release date was the important part of project management. Rather than completing the project properly and on budget. Sometimes people just guess a deadline. Without even knowing has to be done. No project should be handled this way.

More frequent releases does not mean more progress is being made. I think some people believe this. This seems to be a reoccurring topic of making more frequent releases without doing more work.

It would be great to get more people involved. No one argues with that.

Though if you really feel enough isn't being done and there is no excuse for not doing it. feel free to jump in.

Companies do want to see results. Reactos is far away from where companies feel reactos will benefit them, and that is years away. Making more frequent releases will once again not provide results. Unless more work is being done. Who is doing all this extra work? is it you?

It's easy to say be better. faster. stronger. Though someone has to do the extra work. Reactos community could become more efficient and make developing tasks easier. Apart from that there will always need to be extra work to provide better/faster results.

Re: Sinking Into Oblivion

Posted: Wed Jan 12, 2011 7:50 am
by Z98
This discussion seems to be getting a bit confused. A major motivator for the current release 'cycle,' such as it is, is to try and make sure trunk does not regress and become a stinking pile of crap because developers were not careful in testing their commits and the like. The incorporation of new features is, at the moment, tangential. Each of the devs are working on some piece of ReactOS and genuine coordination between these efforts doesn't really happen. What we do have in terms of cooperation and coordination are when a few devs are working on the same piece of functionality or issue, and they go back and forth trying to figure out how to deal with the issue. But If Timo is working on something in win32k, he's not likely to be paying very close attention to what Aleksey is doing inside the registry, nor might he even know Aleksey is mucking around in the registry. Now the situation isn't quite that disorganized, but it does demonstrate that the idea of planning releases around 'features' isn't meaningful at this point. Until ROS stops tripping over itself with respect to memory management and the like, we're in this pattern of basically fixing broken things as we run into them instead of planning on adding 'new' stuff. Or more specifically, attempts to add new stuff keeps exposing major bugs or other missing components and so work is diverted to deal with those first.