Version 1.7 delays

Question

Given that there is currently little to no activity regarding a release of 1.7 Is anybody working on the bugs that plague the current version, or is what we've got the best we can hope for until Soon comes along?

If this project is stagnating now. How does/will that impact on plugin developement, if the core of 1.6.3 is to remain bugged?

Comments

COMIC SAAAAAANS!

---

All, is there any update on release date of version 1.7 ?
I want to start my site but want to do with version 1.7 so am waiting for it.
any tentative date ?

---

See Scott's reply to this post...

http://www.question2answer.org/qa/37570/what-about-1-7-when-will-it-released

NOTE: It will be a beta (so possibly unstable? version)

---

Thanks Funrunna.
So it seems that version 1.7 (beta) will be released before 12th September 2014.
Even 1.7 beta should be good for my purpose.

Answer 1

Regarding your first question, in short, you have to wait until an official release of 1.7. There isn't much anybody can do about it (except for Scott and Gideon who are the ones responsible for finishing and releasing the new version). Having said that, I've been sending some pull requests (code enhancements proposals which can include bug fixes or new features, for the ones who don't know much about GitHub) to the GitHub project for the last months.

You can see the pull requests here: https://github.com/q2a/question2answer/issues?state=open and you can apply those changes to any 1.7 version you are using. But again, whether they will or will not become part of the core is Scott's decision so even if you test it and like it, if Scott decides not to merge them then they won't be part of 1.7. By the way, I don't mean Scott is the bad guy here, he's just applying some additional criteria to filter the changes, for example, if the change results in dropping backward compatibility then, usually, those changes are discarded as plugins might depend on that compatibility.

Now, I've been proposing fixes to the bugs I've found (latest one https://github.com/q2a/question2answer/pull/58) and some to bugs other people found (latest one https://github.com/q2a/question2answer/pull/57). I wonder what other bugs are there. I've seen many issues posted here but 99% seem to be configuration-related issues so I don't even get involved in them. However, you claim there are many other bugs out there. Can you provide a list of them? I guess creating a new question here with all of them would help to measure their impact and probably estimate their complexity and maybe, if I'm in the mood, maybe I can provide a few fixes .

Regarding your second question, release 1.7 does not provide radical changes that would enhance plugin development. After taking a look at the code and giving 1.7 a try, I see there are enhancements in private messaging. That can result in some related plugins but, apart for that, I don't thing the new release would have any considerably positive impact on plugins.

My personal opinion about this matter is that although there is a need for developers adding code to the core, there is also a need for people testing that new code. So far I haven't seen anyone providing a pull request to the core so Scott is really alone out there. Asking for a new release implies the responsibility of testing it. So far nobody raised a single bug of 1.7 which means nobody is testing it. The best way to get that 1.7 version sooner is testing what there is right now and posting any bug found. Of course, if it is possible to post the bug with the fix, that is even better . So as you can see I do think there is plenty of work to do but I don't think it is all in Scott's hands.

Comments

I already saw that but nothing major has been done since then.

---

It seems the development focus after that post was published and up to the last commit in GitHub has been on item 'Improve private messaging'. So you can take a look it. Anyway, there had been some noticeable changes such as using mysqli, upgrading CK editor and PHP Mailer that might require some testing. For instance, there is currently a bug with CK editor on Mac OSX. I didn't manage to reproduce it but it might mean the new 1.7 release won't have the newer CK editor version (this is just a guess as it is not my call).

Giving those features a try and providing feedback is essential to both: help in the testing process and motivate the developers to carry on the work, showing them they're not alone :)

---

Thanks pupi for you comprehensive response to my question!

I will follow the links you provided (thanks again) and if I may, I'll add any comments later?

Re; Version 1.6.3 bugs... Having used/implemented Q2A on an active site since v 1.4.1. and with an established commuunity of some 40 active 'regulars' and just over 200 registered users. Those figures changed dramatically because of various frustrations (bugs?) in the software.

I would accept that the definition of what is a 'bug' and what is not. Is quite different depending on one's understanding of a literal definition of the meaning (coder vs user) and concentrating on the latter for a moment. I had many 'complaints' about the very badly implemented 'User Wall' facility, which came along after v 1.5.

The User Wall was a great idea (badly needed) but functionally, it was a disaster. I do appreciate the improvements to date eg. unrestricted deletion of posts and it was brought somewhat nearer to the surface (easier to find and access etc.)

Other regular 'user' issues came about because CKeditor. As we all know. Is rather less than user friendly and has a bit of a learning curve. Even allowing for its idiosyncrasies, adding pictures, links etc. Is too much of a challenge for normal users.

Just for clarification of my own standpoint. I mention the above as examples of what I would say are are not strictly 'bugs' within Q2A core with the caveat, that the latter (CKE) has a few within it but that's another issue. and the former, (user wall) just lacked said functionality and was rather buried away. Of course by definition, if a function doesn't exist then there isn't a bug... To conclude, these things matter to 'ordinary' users who after all, just want to add some content or other or post quick messages to their friends etc. And when things don't just 'click'n'go' They desert a websites in droves!

On the issue of what I would suggest are 'real' bugs?

Q2A is supposed to integrate with 'wordpress' out of the box and to a degree, it pretty well does. I had things set up and running smoothly for 2 out of 7 months of integration. That would suggest that things were 'configured' right at an 'admin' level? However, in 'real world' usage. many 'cracks' appeared to the point that I just installed it (again losing users) I regularly looked on here for info on resolving issues and others frustratingly posted similar problems to my own experiences but I over time I noted they were seemingly ignored and saw little point in repeating problems/points already raised? The same could be said of plug-ins such as the FaceBook module. I found its integration/implementation horrendously frustrating. It too would seem to work for a week or two and then suddenly keel over.

I accept that 'plug-ins' are effective bolt ons and as such, there is bound to be some scope for bugs to manifest. Whether they are bugs in the core or a given module, is impossible for me to call and only you guys who develop the latter can say for sure. Either way, doesn't that suggest that there are plenty of things to smooth out in 1.6.3 and surely the improvements made leading to .4, .5, - .9 etc. would at least show progress rather than have a leap from 1.6.3 to a promised 1.7 which is for whatever (and I'm sure valid) reason, seemingly on hold? and it looks like indefinitely.

As to testing and feeding back... I am also prepared to do that and it makes sense to encourage developers etc. But I call your attention to Jay's comments earlier. I would also add, that plugin developers have offered to share/lighten Scott's load and I don't think that has been taken up? Among others, you yourself have admirably made what look like practical and certainly 'active' improvements to what could and arguably should be part of the core. So it would seem there is quite capable help on hand and yet little movement?

I understand that for developers in particular. Upgrades to HTML5 and mysql are welcome and needed and so, the dev version of 1.7 probably has something to offer you guys. I see you mentioned a potential problem ahead regarding the proposed CKed upgrade and I note the point.

I personally, haven't had the software online for a while now. (since March) Because I was hoping for improvements in things like Wall Posting and when Scott announced in April that this was at least on the 'agenda' then it seemed to make sense to wait for 1.7 upward.

Finally and an observation really...

Q2A is fundamentally a good and functional piece of software although I did notice that 'spamming' had become a bit of an issue on last install. But essentially, It is a very good basic Q&A platform. Any problems are seemingly due to the implementation of additional functionality. So Kudos to GidGreen.

Perhaps from one perspective. A vanilla version minus 'wordpress' and other geegaws but with improved anti-spam control. Is an option... Delivery and maintenance of that might lessen the 'burden' on Scott. I'm assuming there is a framework for plugin dev's to work in for taking care of ancillary services?

I personally (and my own target userbase) want/need wallposting and blogging capabilities (it is a social networking age) and it is a pity that at least one of these has been dismissed as a core function. However, It is good to see there is hope outside of core developer philosophy.

Thank you again for the reply Pupi.

Regards,

---

Ok, that was quite a detailed comment :)

Firstly, your point about "bug" meaning is important: the most appropriate meaning of a bug is a feature or result that is unexpected or does not match the original definition of the feature. So missing or uncomfortable features are not bugs by definition.

Secondly, as you said, the uncomfortable Wall feature is not itself a bug so changing that would mean a new analysis and development process and it wouldn't be just a fix as in the case of bugs.

Regarding WordPress I have to confess it is not one of my strengths so I haven't tested Q2A under that platform.

Regarding Facebook integration I don't know exactly what issues you had. However, I remember discussing something related to how multiple logins for a single user are handled. Long and boring discussion with the developer of the Open Login plugin here: http://www.question2answer.org/qa/31438 . Honestly, I don't remember right now what features the core provides for handling multiple logins in detail but I remember that my conclusion after deeply analyzing the core, the DB schema and the login plugin module was: "Do not ever use any multiple login plugin"

You mentioned 1.6.4. That is a HUGE keyword. I've been suggesting and resuggesting creating that version with bugfixes that have already been added to 1.7 (and probably some pull requests that might be added to 1.7 in the future). So far thumbs down to that suggestion. Maybe I can make a bag of bugfixes that can be applied to 1.6.3 and create a branch in GitHub for users to download, which would, in other words, mean creating an unofficial 1.6.4 Q2A release. Anyway, that would mean adding some work on my shoulders that won't have much of an impact in terms of development because they would only be bugfixes rather than new features... not to mention I would be swimming against the tide. I believe if 1.6.4 is ever going to exist then it should be released by Scott or Gideon.

Regarding this comment "there is quite capable help on hand and yet little movement?" I'm not so sure about "capable" but I certainly can confirm the word "help" from my side. I won't make any comment on "little movement" :)

My conclusion is that for now, I won't change my course of action: if I find a bug I'll try to fix it, if I think about a code improvement I'll try to propose it, if someone else posts something here that seems to be a bug (and not a configuration issue!) I'll first research it and then try to fix it too, if I provide a fix I'll implicitly or explicitly resuggest a 1.6.4 release. What will happen with Q2A nor what I will do in the future I don't really know... but it seems no one actually does :)

Answer 2

Three mistakes of Q2A that I think.

 

Mistake1: Github as development environment

 

The old users of Q2A like me know that originally Q2A did not use Github. There was discussion to use Github for before. By the then argument, only merits of Github were discussed, and demerits were not discussed at all. The development style is different by an engineer. And I thought that the development style of Gid did not match Github (Me too). Therefore, I opposed using Github. If a state of the development was not published in Github, there should have been few users with uneasiness. If Q2A continues using Github in future, development leader should understand merits and demerits of Github.

 

Mistake2: Succession from Gid

 

Gid is the great developer whom I respect. He has the special skill that ordinary man does not have. When Gid leaved from development, it is a mistake to have chosen an specific (individual) user as a successor. There was not the developer with technique and management capability equal to Gid at all in those days. The hole of Gid might not be filled up in one developer. At least Gid and I knew it. The hole of Gid might not be filled up with this method either, but it was necessary to make the development system by the team.

 

Mistake3: True work of the development leader

 

Work of the leader of the big project is not development. The main work of the leader ...

• Summarize the whole project
• Decide aim of project
• Make schedule, and manage it
• Announce things of project to users

Gid went all things include development alone. Because he had great ability, he was able to do it alone. The opinion of pupi is right for generalization, but is not for real scene. Without the strong core development system, the cooperator will not readily appear. We should discuss these things again. Q2A will decline when we do not change the development system from the origin.

 

This is my opinion.

Comments

I think you've raised some important points sama...

1. A discussion that is only one sided, is not a discussion... It is a dictatorship!

2. I don't mean this in the least bit sarcastically. But 'Superman' is superman and of course, cannot be replaced in terms of one clear vision/goal and implementation etc.

3. "Work of the leader of the big project is not development. The main work of the leader ...

Summarize the whole project
Decide aim of project
Make schedule, and manage it
Announce things of project to users"

I agree with the role of a 'leader/manager' and it is the understanding of that set of principles. Which defines good management. However, Managers are paid good money because they know who can do what the most effectively and allocate time and resources to achieve desired/known objectives which are clearly defined... Because most people are 'not' management material. They tackle things differently and therefore earn less. Irrespective of how technically competent they are in a given area of expertise!

At the risk of playing devil's advocate here. If your suggestion were in fact the way things were done in the future. (and in principle I accept the common sense of it) Then there is also the propensity for further 'issues' to arise... For example conflict/s of interest?

There are plugin and other developers who provide their time, energy and final 'products' for free. Whilst yet others have a 'commercially' vested interest even if they also 'donate' to a given project. The latter, will inevitably what to give some things freely and want to place a financial or other value on their efforts for others... I am not condoning one, nor condemning the other, merely pointing out. That the or a manager 'role' has to consider many things not necessarily obvious to all parties potentially or otherwise involved in a project.

Natural leaders are not followed because they carry the title leader/manager/boss etc. Real leaders are those who can command 'mutual' respect from everyone around them. And always, because they are unbiased and thus 'trusted' to meet the needs of all for common good. They are also followed even when something goes against the grain of an individual within a group. Because they can and will willingly demonstrate why a path must be walked and so they 'earn' respect.

These are things I'm sure must be in the minds of GidGreen and latterly Scott too. So it is not always clear cut, as to why one path is walked, in preference to another. Even if a majority potentially prefers the alternative!

regards.

---

I agrees with you sama . 1 vote up .

Your opinion about github is correct . But some times very helpful for developers like us . We can track everything line by line changes . We can see in which commit what are the files changes . Is it not nice thing . ??

---

@sama55 I agree with your points and plan is required for the community.

I also respect Gideon's decision to appoint Scott as lead because of his Q2A development knowledge .

I feel Scott should announce probable release date (like Gid used to do), announce features/bugs planned and work with other developers to make it happen.

At same time we all should do rigorous testing of Q2A so that it will be always be stable.