Did you miss the "temporary" word in saying this move is temporary?

The CVS server is still online, only the links and page describing it have been removed from the support site. The CVS server was not maintained for a while due to the amount of changes the template structure brings in. With this, people then thought that no progress was occuring due to not seeing CVS commits being made.

To correct this stance it was decided to no longer maintain the public CVS server at all, which lessened the stress placed on to the development team, and are now focusing on the internal development server.

It's that easy, and nothing more.

Edit:

I overlooked something in your reply. The "not yet ready to be announced" is not in relation to development, which I had cleared in my previous reply.

There is more to the project than coding.

This post has been edited by hpdl: Sep 14 2005, 07:00 PM

I think you don't understand. What you present as 'The best for the team' means the death of the credibility of the project.
There is no way any team (any size) can develop a complete redesign of a complex piece of software in a hop. Go ask Netscape, that died by failing to understand that.

That of course meaning you have some idea to make a living out of OSC.
Fine with me: so give us the code, and go on with your other business while WE go on with ours.

Angelo.

This post has been edited by aturetta: Yesterday, 12:53 AM

Harald, I am not clear how your reply addresses my questions regarding the secrecy that surrounds OSC. My question is non-technical; why has recent development of OSC been secret?

Historical server issues and the challenges presented by numerous changes to the code are interesting and plausibly reasonable excuses for momentary delays, but there is nothing understandable, in a project that is ostensibly under the GPL, for holding the team's work in unreleased form for month after month. It's looking from the community's point of view more and more like a closed code project.

I don't need to lecture you or anyone else about this. What's at stake is obvious. My question is simple - what unique advantage is there in the case of OSC for prolonged secrecy? You control the code and the copyright. You are the only one with the answers.

I don't write a lot in these forums, in fact, I'm not really the "replying" type of guy, I mostly just read on and eventually someone states the points that I would've contributed if I had pitched in my two cents.

First off, let me start by writing that I completely undertand both sides of this, however I think you're misunderstanding(or at least hope) each other, causing the both sides to go into a constant defend/attack mode.

But there's more than two sides, there's people like me, who's not declaring the project dead, but are also not repeating "It's free, they do it in their spare time, shut it or find another project, you can't make demands" like a religious mantra.

osC 2.2MS2 is a great source to build a shop on, a fantastic forum of developers and countless contributions. But that's the thing, it's a source to build on, the stock version is simply not up to the standard of what people want. And that's why people like me can make a living coding and implenting contribs. I've been doing this for a couple of years now, and have eagerly awaited MS3, and I must admit I was frustrated at one point or another by the almost cult like secrecy that MS3's development has been surrounded with the last coiple of month.

But I moved on, I put myself in Harald and the teams shoes and tried to figure out why they were doing things they way they do. And as I see it, MS3 is back to square one and fix everything that's wrong and build a strong codebase for the future. And that's quite a task, and requires a very small group of trusted developers if you wanted juuust the way you pictured it. And I can respect that. So I understand it being "closed source" until it's finished, hopefully from MS3 and on we'll see a bit more open development.

Public CVS, constantly updating progress can take time away from the project at hand, and become very stressfull, so I understand the closure of the CVS and progress updates halting a bit.

But isn't constantly defending yourself also tiresome? Isn't reading and writing the threads with this topic a hassle?

I respect this group of developers, I respect the goal they're trying to archieve, but critism should be given where due.

Here's what I suggest as a comprimise:

1) The Roadmap: Clarify what each project is about a tad more.
2) News: Put more news out, instaed of just posting it in the comments(like the result of the meeting).
3) CVS: Yes, I know it's down, but put it in the news, make a CVS section, were you explain how, what, where and why.
4) Timeline: You don't need a release date, but saying "we're 85% done" would make a lot of people happy.

Other than that, concentrate on making MS3 and other things osC, most of us would be happy with just these few extra points.

To conclude my uselesss waste of space, I would like to thank you for a wonderful piece of software, and the time and effort you put into it.

Hope my feedback is useful!

This post has been edited by RL2000: Yesterday, 05:55 AM

All of the above is fair analysis in support of cautionary development measures, but the issue at stake is the open source integrity of the project. It may be necessary to keep development in the hands of a select few, but the regular release of code, even without an invitation to the community for contributions, maintains both the team's and the community's confidence that the project is still an open source effort.

I am not asking for daily release of code. Weekly or even monthly code releases would do the trick. There is no plausible reason that I can see why unstable versions of the development code, clearly labeled as such, could not be a policy. Without code release, there develops an inevitable atmosphere of frustration and suspicion. The only way for an open source project to forestall or counter such a atmosphere is for it to behave like an open source project - release often and let thousands of eyes examine and learn from the code.

All I have been asking is why OSC hasn't been following an open model for the last several months? Why is it an exception to the philosophy of the GSL?

Wow...I didn't relize this would go into this. But I think its a good thing that people start saying stuff.

*Edit* - offtopic

Another question the Harold didn't replay is what do we do about security releases?

And isnt the point of the cvs to be updated as VERSION control, and it also says that it might be broken from time to time. ok so we are redoing the template system, cool beans, but thats no resason not to upload development. After looking at ALL the changes accross the osc website and other countries osc sites it almost appears to be masking what the real point of the project is.

This post has been edited by Johnson: Today, 12:21 AM

opinions thats all this is. i am also like chris and i have been fallowing osc for the past few years, and it appears dev has just ..well died.

Hi to everybody,

it's good to see that Harald is coming up with some few lines to at least point out what's right and what's wrong in here.

Nevertheless, it does not change the fact that people are eager to see MS3 really happening and coming finally out at the end of the pipe.

Due to enormous time gaps in between different milestone versions has, in my opinion, escalated the thing forward so that there exist nowadays shopsystems, which are based on osCommerce open source code but these existing systems' been developed further in various degree.

I'm not going to drop any particular names in here, as some of the products cost, and I am not really made up my mind whether it's morally right to sort of "steal" an open source code and make money out of it. Everyone sure has an opinion on that, but I don't want to start any debate of it, just make up your mind.

So back on track: what I wanted to point out, is that some people sure have vanished from osCommerce community and been jumped on some other bandwagons (shop systems), which are showing some steady progress in their development work.

It's no doubt whether it's important or not to show that the project really is moving to some direction.

All what we know regarding osCommerce 2.2 MS3 development is that it's been in progress for quite some time (fact), but we mortal osCommerce community members are not able to prove that anymore, partly due to disclosure of CVS.

Reading news regarding current development status is OK and necessary to inform us community members, but when it comes to the point that most of us are already thinking after reading time after time these somehow couraging news, there's not much space left for trust.

In the beginning of every week when I drop by reading forums, I am hoping MS3 coming out end of the week.

Trust me, I've been going through these weeks for a long time by now.

Now, here at the end I want to give and offer good spirits for osCommerce core development team for their continuous hard work. Thumbs up everybody! I want that MS3 coming SOON to us!



chekov

I don't understand why this project have to jump from a version more than 2 years old to a next release without other intermediate release and wait for weeks and weeks and years.

Only a quick note: I don't think thousands of contributions is really a selling point. At least not how contributions are organized right now. I don't know if there is somekind of moderation in the contributions download area but it sure seems like it hasn't. I think it's very hard, impossible, for a newbie to quickly know what are the 'good' contributions that they should use and not the buggy/beta ones. So they are only thousands of contributions and not useful contributions.

A simple example: language files, searching for portuguese language files i have 14 packages. Why? Shouldn't it only be only 1, maybe 2 aproved, semi-oficial language files? Now what package should I choose? What's the better one, which ones are complete?

Besides, we see titles that only serve the purpose of "tricking" the system to be at the top spots, we see non descriptive titles, we see contributions that are very old, contribs that aren't supported by their authors, etc,etc,etc...

Again, i point the example of phpbb.com, it's not an excelent example but it sure is better than OSC right now in terms of contributions.

As i don't like being critical just for the sake of it, here are my sugestions:
a) open a "free upload" contributions area, a not moderated area that everyone can upload their contribs, which means: a contrib area as it is right now but
open a "official" approved contribs where they are moderated before being published, where they have a mandatory forum thread for support, where we can see screenshots of the mod, and of course, only mods with no bugs are approved.

This is more work to do, but there's no need for "senior developers" to evaluate each mod. We can only analyze contribs in layman terms: install in a fresh OSC, works as described? fine. If we can gather a team of "mods" that analyze contribs for good programming, security, etc, better, but if that's not possible i would be happy with a contrib area where every contrib works, has templated documentation, a descriptive title, a thread for contrib support and stuff like that.


Again, don't take it the wrong way: there are contributions that work, are good and their authors support it (and yes I'm lazy to test 100s of packages just to see what's the good one for my purpose). But I think we(you, someone) should really rethink the contributions area and organize it better.

The forum rules are quite clear that discussion of other carts is not allowed. While I've occasionally questioned some grey areas, that one is quite black and white in the rules.

ed

Ed: Do you have a link to that by chance?

Thx

http://forums.oscommerce.com/index.php?act=boardrules


Its in the first part of general.. Information Exchange

There's also a link at the top next to search...


Kind Regards,
tom

This post has been edited by Darklings: Yesterday, 11:06 PM

Harald,
Will all future releases of deveopment code for OSC 2.2 be released under the GPL?

Yes, ofcourse. And in regard to your "secret" remark, there will also only be one version available (as it has been since day 1) and not a "professional" and "community" release, where one could contain additional features than the other.

Right now the development team consists of 3 people (2, excluding myself). The decision to add new developers to the team after MS3 meant the public CVS server would no longer be maintained, to place all focus and resources of the current development team on internal development.

Although it is not the best approach for the community, it is the best approach for the team to get MS3 released sooner.

And to be honest, I am not going to reply to every single post here that are based on the "doomed if we do, doomed if we don't" mentality. I'd rather put my focus on the more constructive criticism.

Anything less is on the verge of being deleted.

Since developers are only 3 why to add 19 workboard entry to the MS3 and delay the release on MS3 and not to add less workboard (only ones that are about core development) and release before MS3?

Thank you for the clarification. It puts my mind to rest and perhaps the minds of many others.

As for the "secret" remark, there was no offense intended. Undisclosed code is secret. Disclosed code is open. The use of the term secret was natural, arising from the facts.