How about a stable version of CB

[quBit]

Dear developer,

Before I write my message I must express my appreciation and enthusiasm regarding the work on Code::Blocks IDE. My point is that everyday you release a new version which is incompatible with it being a serious development tool. Wouldn't it be much more better not to release a daily build and to work on a set of well defined requirements ? My second point, is that there are A LOT of features that you seem to focus in parallel and most of them work in a manner that cannot make C::B a serious competitor to commercial IDEs. I am sure that you aim to have a release and a standard for your own work and not to remain to the stage of infinite nightly builds.

[kelo81]

Are you saying that CodeBlocks is not a serious competitor?, I finally found in Codeblocks all the features that where missing in the MSVC IDE itself, like integration with several compilers, things like "code completion" working for EVERY C++ compiler under ANY SUPPORTED OS's. So, tell me quBit, why do you think C::B is not a serious competitor, if you can see that most OpenSource projects are using it as a favorite tool for multiplatform builds... appart of the amazing Eclipse of course.

In fact, if you see that there are some unestable features: Why don't you just try to improve them instead of posting those messages in the forum?, C::B is an OpenSource project, that means everyone can take a look at the code and try the make it better... that's the most important feature an OpenSource project may have. A year ago, CodeBlocks was a smaller project, without even a half of the actual features... the next year, I think it will be really better, and the following year too... and allways for free.

MSVC has a good IDE... but they don't improve the product, each release they make is worst... Thousands of MB's and also you have to download hundreds of "security updates" every week... that's NOT a serious tool!

[cacb]

Dear developer,

Before I write my message I must express my appreciation and enthusiasm regarding the work on Code::Blocks IDE. My point is that everyday you release a new version which is incompatible with it being a serious development tool. Wouldn't it be much more better not to release a daily build and to work on a set of well defined requirements ? My second point, is that there are A LOT of features that you seem to focus in parallel and most of them work in a manner that cannot make C::B a serious competitor to commercial IDEs. I am sure that you aim to have a release and a standard for your own work and not to remain to the stage of infinite nightly builds.

I also appreciate the hard work and enthusiasm put into Code::Blocks. I am talking about it to all my colleagues using MSVC, and I am using it for a private project making a portable Windows/Linux application. It works great, although I am now way behind on the builds (4237 on Win and 4066 on Kubuntu).

I would have an easier job convincing new people of its greatness (it *is* great!) if an up to date stable version existed, since I would be able to simply tell them "download version 2.x" or whatever it would be. So my recommendation is to pause new development for a while (say 1 month), produce a nice and stable version that most people can be happy with, and link to it on the front page of http://www.codeblocks.org.  Then continue with the development.

In general, a stable release every ~6 months or so would be perfect for me. Even once per year would be fine.

But don't get me wrong: This is one of the more interesting projects I have encountered. A portable IDE is just what lots of people need, myself included. Thanks one more time for all your efforts!

[XayC]

The lack of recent stable version (release, call it like you want) of Code::Blocks, in my opinion, is damaging CB.

I was recently thinking to propose the use of CB in a open software project which supports both Windows and Linux, they use MSVC for Windows and make for Linux. But I can't propose a IDE if it has not a release or something like that.

This made me think about the fact that a CB release should be provided as soon as possible, you can't ask other people to use an IDE which comes only as a nightly build.
Also maybe developers should think about changing the release schedule to a time-based one from the current that is features-based.

It would be nice to know what developers or Mandrav (who would decide about the releases) think about this. Don't you think it would greatly help CB to spread around the world having a release? Am I missing some good reason to delay the release of a stable version?

Thanks, XayC

[mandrav]

The reasons on why it's taking us so long are clearly stated on the main site (well, plus some real-life issues that held some of us back for some periods).
Our main goal is to have a stable SDK. Only then a new release will see the light of day. People wanting to make their own plugins have to be certain that they will work as-is, at least until our next major release.
Towards that goal, a huge step has been made these days by incorporating the new logging system. A couple of things are left and I guess then we could plan for a release. Hopefully it won't take us long so stay tuned .

[XayC]

Yes, having a stable SDK is indeed a good reason to wait before making a release.
Thanks for clarifying this point.

Regards, XayC

[JGM]

Wouldn't it be much more better not to release a daily build and to work on a set of well defined requirements ?

I love seeing updates every day, is great to know this is a really active project, and not like other projects that take 3 year to release something that works while this works better every day! 

[cacb]

I love seeing updates every day, is great to know this is a really active project, and not like other projects that take 3 year to release something that works while this works better every day! 

According to the web pages, the last stable release was dated Oct 25, 2005. That is now more than 2 years ago.  :shock:

Although the nightly builds mostly work well (I have tried a few), I cannot see how to persuade an open source project, or even a company to start using it without a stable release.

If it takes another month to make a stable SDK, then I think it is good. If takes much longer than the remainder of 2007 to make a stable release of C::B, I would suggest the plans be reconsidered. I support the idea of a time based release schedule.

The reason for saying this is that I would love to see C::B becoming a success soon, also in commercial use....! In any case, I will stay tuned  :wink:

[gd_on]

I completely understand all the reasons why there is not still a stable release. Myself, for my own projects, I use the nightlies, almost every one, and I'm very happy with them. But I have not been able to convince my company to install C::B at work as the default IDE. Their reason : no stable release (previous one too old, outdated)  , no full self install (with MinGW for windows for example) ....
If you compare with Thunderbird or Firefox (these two softs are for official use in my company), they release from time to time a new stable version (for different reasons, and sometimes they say it's a security release ( , oups, as Micro$oft). With TB or FF, if you wish, you can also work with nightlies. Myself, I work with both type of version : the stable one at work, and the nightlies at home (where I'm free to do what I want)  because I maintain a Thunderbird theme and want it to be ready each time there is a new official release.
So in a few words : at home, no problems to work with C::B.
At work, practically forbidden to use C::B. 

gd_on

[quBit]

    Well people, my point seems to be understood. C::B seems or promises to be the single viable solution for a free IDE on the Windows platform. DevCPP is old and not maintained any more. wxDevCPP is almost DevCPP and is too tightly coupled with wxWidgets which, although still with a large user-base, is an obsolete framework (new lightweight toolkits which target strictly GUI, such as SmartWin or Win32-Generics which make use of modern C++ techniques, are preferred). There are many more examples and I don't even want to think about the VS Express.
    Good luck to developers and I hope they find the time to finalize the work in a decent period of time. I think a  version which is stable has some quality or corectness guarantees (at least implicit) for its features would drive much more attention than maintaining a user base by keeping them occupied with nightly builds.
    It would be a shame not to have such a promising project as C::B on the open-source "market", laughing at the commercial IDEs for which developers pay hundreds or thousands of dollars (or euros in my case) and they are not even what you wished for.

Best regards, quBit.

[Morphius Faydal]

    Well people, my point seems to be understood. C::B seems or promises to be the single viable solution for a free IDE on the Windows platform. DevCPP is old and not maintained any more. wxDevCPP is almost DevCPP and is too tightly coupled with wxWidgets which, although still with a large user-base, is an obsolete framework (new lightweight toolkits which target strictly GUI, such as SmartWin or Win32-Generics which make use of modern C++ techniques, are preferred).

You do realize That Code::Blocks is a wxWidgets project, right?  And what's wrong with wxWidgets?

[thomas]

And what's wrong with wxWidgets?

Don't get me started 

[MortenMacFly]

And what's wrong with wxWidgets?

Don't get me started 

Please someone: Block any further comments of Thomas concerning wxWidgets... faaaast!!! ;-)

[LeRoi]

I totally agree with "quBit", "XayC" and "cacb"

Code::blocks is a great IDE, with great features, and native execution on multiple plateforms

but, trying to get familiar with CB, I found my self obliged to mention some points about code::blocks :
 * the latest stable release of CB goes back to Oct 25, 2005, wich -as a personal point of view- is a long time, espacially when considering all the new features that have been implemented since that date

 

I love seeing updates every day, is great to know this is a really active project

* well, that's true, but no one will update it's CB svn build every day !!, maybe to get rid of a corrected bug, but than again, he may get new ones, and you all know that the last thing that a programmer wants to take car of, is the instability of the IDE(I'm not saying CB nighty builds are instables)
when I tried to install a nightly build the first time, I got some real problems, with CB, the wxWidgets and the compiler(s)
Poeple are often waiting for an all-in-one package. There are so many steps, and time to make a helloworld wxwidget application that it can be frustrating to any new comer to CB. compiling wxwidgets with the custom settings, installing and configuring the compiler, setting up the debugger(a nightmare in somecases) to finally get CB to debug correctly an application.
when compared to MS VS or Delphi, ..., you write the code, you type F9 and bingo ....

Im not writing this just to criticize CB and it's authors, I like CB, I really like it for many many reasons, and that's why I'm posting here, so after saying my points, I will make some suggestions  :
* I'm not against daily updates, but strongly against 2years+ stable updates ! 6 mounths for stable releases seems to be a great cycle .
* why not to put more effort on the wiki, it's very important for new developpers, and sometimes, it lacks for informations, and the user have to go through the forum ...

well nothing to says more, you are doing a great job, keep working, and I'm hoping to see a a stable version soon  :lol:

[dje]

but no one will update it's CB svn build every day !!

I do !
Less than 2 mn to be up to date, it is worth !

Dje