For instance there is a new Free Pascal debugger being developed, separate from GDB which will be superior to GDB when it is completed. I only got to know about because someone had a problem and one of the core Lazarus developers told him about it.And the person who asked about it is actually an experienced Lazarus user.
The purpose of blogging is to introduce such features to existing prospective and existing users alike. I come to the forum because I have specific issues in mind. There are times when I browse at leisure, covering developer blogs etc, but there is nothing for one of my preferred coding tool Lazarus.
But there is a blog article about the Pascal debugger!
http://lazarus-dev.blogspot.fi/2014/05/de-bug-wars-new-hope.html
You don't bother to read even the few blogs that are out there, you just want to find some excuse to complain.
I don't quite understand you. This would be ok from a newbie but you have been around for years and the same thing goes on. I don't remember any contribution from you (I mean none, zero) but I remember lots of similar complaints during the years.
After seeing that article I remember having read it in the past, but then again it only drives home the point I am trying to make.
It is the only blog entry for the whole of 2014. I am sure that there must be at least 4 major initiatives in addition to debugger which were started in 2014, but who knows about them? And you know what, users cannot help with testing, bug finding and feedback in this initiative, let alone the others which are in the pipeline. And why is that,
because they don't know about them in the first place.
This is particularly poignant because it was Graeme who asked about it. Graeme has been and still is major controbutor to Lazarus and if he doesn't know about something as major as a new debugger, you know there is a problem. Just like me he probably learned about it earlier and forgot it about or didn't know how much progress had been made.
Speaking of Graeme I am sure you Juha will remember that recently Graeme inquired about the state of UTF development on the mailing list, and you pointed out to him that he himself had created or edited a page in the Wiki about UTF, and had probably forgotten about it.
What I am saying is that the issues Lazarus faces are not technical or even one of commitment, but are actually cultural, and the cultural issues are set by the gatekeepers if you prefer that term, or the managers and moderators of the central Lazarus infrastructure.
Now getting back to the wiki. One of the major issues newcomers face is getting Lazarus working or a recent build from trunk working. The last few days I have started working on get Lazarus to compile, and one of the pages mentioned installing that I had to install the fpc package to compile Lazarus. The version of fpc in the packages is 2.4.4, which is about 3 years old whilst a current compile requires 2.6.4. You could argue that the wiki page is out of date. I wouldn't call it out of date, I say it is plain wrong and outright misleading, because someone would try it and fail and try again believing they have done something wrong when in fact is is the page that is wrong. But if the wiki entry was dated then it could be described as out of date, rather than plain wrong and misleading. There are pages as well which say if you want to compile Lazarus,
get the source from svn or sourceforge then type make
. There is nothing about requirements such as development libraries, utilities and whatnots, and I'm looking at and asking myself "what on earth is this is. this is plain wrong".
The problem is that wiki pages are made of multiple contributions and you can't go around deleting stuff willynilly because a lot of people are going be annoyed although they deserve an upbraiding worthy of Linus. Wiki pages also state they date they were last edited, but some of the sections go back years. So if the moderators were to lay down some rules such as "Dude, before you put anything in some section or edit it, make sure you test it and it works, ie start of with a clean machine or at least a new configuration, follow the instructions that you laid down, and make sure they work, before putting it up. Date it and state the pre requirements, and if it will disrupt or make the wiki page confusing, create a separate on the wiki or elsewhere perhaps Github and link to it. Otherwise if it winds up wasting time for very obvious shortcoming (to a more experienced person) we are going to point the finger at you and say "you done it".
This is not a matter of commitment and contributing, because people are committed and contributing, but is how they are guided and channelled into doing it right that matters. It is more a matter of people with the right depth of technical knowledge and experience and standing within Lazarus setting out the cultural style, simply saying that "round hereabouts
This Is How We Do It"
. But Lazarus is a highly technical environment, so even if someone is a good evangelist, how do they go about establishing the culture surrounding Lazarus without the necessary technical nous to be effective in that role? Lazarus is not like Drupal where
Webchick could enter into that role. You are working at the lowest level of the operating system with a wide range of equally complicated APIs. That takes time and long experience, and only core Lazarus developers can do that. All it takes is one or more of the core developers to devote a few months to setting the culture and the approach, for the rest of the masses can carry on from there.then they can go back to doing what they do best., ie hard core development. It doesn't have to be forever.
Now Lazarus is not like Drupal where people can devote a lot of time to volunteering where they know that they will be making profitable contacts and developing skills that would lead them into 60k to 120K jobs. There doesn't seem to be that much in Delphi, let alone Lazarus. It mostly small developers and hobbyists so clearly you are not going get that kind of commitment and that is why stuff must be made really easy. Drupal is a huge thing, but no where Lazarus in complexity, but between, drupal.org, groups.drupal.org and drupal.stackexchange.com you will find everything you need to solve a Drupal issue, not to mention the hundreds of blogs offering advice and tips. but you can hardly say thte same thing for FPC and Lazarus.
There is a lot of wonderful stuff going on in Lazarus, but most of it can't be seen or heard of in the official forums or mailing lists, because the developers don't find the culture here welcoming or the facilities easy to integrate with it. Perhaps the site and the forums should be broken down into a separate strictly controlled and moderated core, and a freer looser environment which enables outsiders to contribute their stuff. For instance fpgui seems to be getting more popular, but because discussion of fpgui is not allowed here, fpgui users have to go to a separate newsgroup for help. But this means that users of fpgui have to keep switching between the forum and fpgui newsgroup which is painful and disruptive of workflow. It also means people who don't use fpgui don't get to see all the things in fpgui which could probably help their own work, and wind up with a rather narrow view of all the facilities available to Lazarus and FreePascal users. And this can be extrapolated to all other wonderful Lazarus stuff out there.
Of course I don't know how much activity the forum's hardware infrastructure can accommodate, but if it can handle a few extra forums then why not. If people are creating Lazarus forums and help websites away from here with the end result of scattering the information, then it isn't really helpful.
Don't think I am not creating stuff to contribute, but I don't want to inflict half-finished stuff on an innocent, unsuspecting Lazarus users.
I could add more stuff it is already TLDR, but to sum it up they problem is not one of contribution or commitment, but ease of contributions and their inclusion, coupled with a scattering and disconnection of resources and technical info and some guidelines and standards.