Is there a separate mailing list for Lazarus developers?

This mailing list appears to be mainly for Lazarus users rather than the developers.

Is there a separate mailing list for that, or does it take place in Bugzilla or some other place?

vfclists . wrote

Is there a separate mailing list for Lazarus developers?

Yes there is.
Can be joined by invitation only. Most of the members also have a commit access to SVN repo.

You will be invited, too, if you became a Lazarus developer. :)

Juha

A lot of Lazarus developments and improvements, whether fully implemented or planned go unnoticed and I wonder if this is the reason why.

Is it possible to view their archives or have some kind of read only access to it?

vfclists . wrote

A lot of Lazarus developments and improvements, whether fully implemented or planned go unnoticed and I wonder if this is the reason why.

What might those unnoticed developments and improvements be? I believe this is a misunderstanding or maybe hopeful thinking from your side. The discussion is about technical details of some piece of code, release schedules and maybe organizing transportation for a meeting. You would be bored.

There is no secret agenda about the future of Lazarus that its developers try to hide from you.
The development process is as open as it gets.
The missing features are well known: Project groups, Android and other targets, docking, installing packages from a repository, Unicode, etc ...
The Mantis bug tracker has loads of feature requests, too. Everybody agrees those features are needed but somebody actually has to do them. Just talking and writing about them would be useless.

Is it possible to view their archives or have some kind of read only access to it?

No.
Besides, I think your time is better used if you learn the Lazarus code base and then provide patches.
And, if you still really REALLY want to read the developers mailing list, you must provide so many quality patches that you finally get a commit access and get invited to the list.
I can recommend this route.

Juha

I have always had this nagging feeling that something wasn't quite right about Lazarus and this explains it. I have been using Lazarus on and off for about 4 years now. From the start I don't think I used more the 10% of what Lazarus offered and in the succeeding 4 years my usage of its features increased by 10%.

The additional features that I can think of are using the Object Tree to navigate the components, changing the parent to move objects around, using Control-Shift-Down Arrow to jump to the definition and Control Click with the mouse to jump to the definition of a procedure or a type, but I suspect that there have been more improvements since then.

I think the developers fear that if they open up the list even for viewing alone there will be lots of acrimonious debates and disagreements with non developers over how Lazarus should evolve, such as the everlasting Unicode debates, and also criticism about its internal operation.

It may be a well founded fear as seen in other open source projects but it ought to be reconsidered, with the understanding that whatever things non-developers/participants disagree with should be raised politely and in a tempered manner that doesn't offend or dishearten existing developers.

As for being able to participate by submitting patches or fixing bugs first many users may consider such things outside their field of competence, or may only be interested in areas that affect them directly or are deeply knowledgeable about, such the issue with transparency that Kostas recently submitted a patch for. For me it was always a non-starter because compiles were breaking due to missing units and I was always afraid of breaking my installations to try new things even for my own use. It is only now that I am becoming more confident and even then I am still wary.

I think for testing stock VM images should be used so that contributors can just load, test and report their findings without the fear of messing up their working installations. Testing is one area that can result in more participants if involves just being click monkeys.

I think that opening the list may allow people who are highly skilled in specialized areas to see where they can contribute and offer their help. Going through Bugzilla to select bugs for fixing is okay for young developers or students who want to cut their teeth or find a way to enhance their skills and acquire some experience, but some of us are too long in the tooth for that. I also suspect that some of those bugs wouldn't creep into the code in the first place if the development process was open enough for more eyes to spot them in the first place.

There a lot of projects which are more open but I suspect that using dynamic languages helps with rapid iterations and enable to them to manage debates and issues without slowing development, whereas a compiled language doesn't afford that time and flexibility.

I think being able to see what is coming and follow their development will encourage more participants and more bug fixing as well.

In any case if Lazarus developers feel that this approach has proved to be the best over the years then they are entitled to continue as is.

vfclists . wrote

I have always had this nagging feeling that something wasn't quite right about Lazarus and this explains it.

That sounds a little paranoid to me.

I think the developers fear that if they open up the list even for viewing alone there will be lots of acrimonious debates and disagreements with non developers over how Lazarus should evolve, such as the everlasting Unicode debates, and also criticism about its internal operation.

Such debates are seen already on the forum and this mailinglist. We don't need more of them.

And that is exactly how I started out.
At first I was just a user of Lazarus, and I only used the release versions. And of course I hit bugs, and I reported them in bugtracker. When some of those bugs were not picked up soon enough (I'm a little impatient), I tried to see if I could fix them.
At that time I never even wrote a single component of myself, so I really had to study the LCL and Lazarus code very hard. I then discussed these issues in the bugtracker with several of the devels and got very usefull feedback.
This encouraged me to submit patches and some of them got accepted.

I think for testing stock VM images should be used so that contributors can just load, test and report their findings without the fear of messing up their working installations

I was afraid of that too when I started fixing bugs (in my area of interest, just because I wanted the to be fixed). At that time I was using WinMe, which as you might know is an OS that tends to get screwed up very easily when installing software.
The beauty of Lazarus is that you can very easily have a "stable" release version and trunk up and running.
(See wiki, or use the fpcup tool from Reinier)

If you have SVN (or GIT) than it is virtually impossible to screw up. If you mess up the code so much you cannot get back, just do svn revert (or the GIT equivalent) and you're back.
No need for VM images (which one could not provide for Windows at all).

I think that opening the list may allow people who are highly skilled in specialized areas to see where they can contribute and offer their help.

They already can, and they do not need access to the devel list for that. Just submit patches in bugtracker.
If it doesn't get picked up, it is perfectly OK to ask on the mailinglist if one of the devels will look at it.

I also suspect that some of those bugs wouldn't creep into the code in the first place if the development process was open enough for more eyes to spot them in the first place.

That makes no of all you can see all commits in the svn repository. Second, most commits are not discussed on the devel list either.

And yes, on the devel list their are discussions on the technical aspects of the development of Lazarus as well, but we mostly use it to discuss release schedules, ask other devels feedback or advice on problems we are working on, or use it as a quick method to notify that some revision caused a problem that needs quick fixing.

It is not secret, but there is also no need for non developers to listen in.

I think being able to see what is coming and follow their development will encourage more participants and more bug fixing as well.

Well, before I got invited to the devel list I never even knew it existed. And I never missed it.

I have always felt that I could discuss anything about Lazarus on the mailinglist or the forum.
I have had plenty of discussion with many of the devels, long before I ever wrote a patch, and they have always been willing to answer any question I had.
And exactly that willingness to answer my questions and give me feedback about proposed patches has encouraged me to keep sending in more patches over the years.

Bottom line is: if you want to have influence in the way Lazarus evolves, you have to submit code.

Bart

I am with Juha on the topic. I think you expect to much from that list.

Try the wiki
http://wiki.lazarus.freepascal.org/Lazarus_IDE_Tools http://wiki.lazarus.freepascal.org/New_IDE_features_since http://wiki.lazarus.freepascal.org/Lazarus_1.2.0_release_notes#Previous_release_notes http://wiki.lazarus.freepascal.org/Lazarus_1.4.0_release_notes#Previous_release_notes

As for the unquoted rest of your mail, I noted your opinion, but do not share it. (based on what I actually know the content of that list to be)

As for images for VM: manpower would be needed to create them. Nice idea.

To give you an idea, here are the subjects (and begin of msg, / cut by thunderbird when copy many) of the last mail-threads, I started. (no special selection, the last 15 threads I started, first mail of each thread)

1 or 2 of them might have had content, that might have been of interest to others. e.g. the 2nd msg, where I was seeking help and that could happen to anyone....

Most of them are setting dates, making sure stuff is merged, .... Some point to descriptions of potential bugs, on the forum, or other web-pages. So the maintainer of the related code in Lazarus could check them out. I could have send them to the one person directly...

[Lazarusdev] date for 1.2 RC2
Shall we set the weekend 13/14 Jan 2014 for the RC2 ?

[Lazarusdev] codetool / designer and threads or ?
Is there any background work (maybe in a thread) that is performed when the IDE is started? And during which certain functions will not work? If I start the IDE (no last project loaded / opening with a new empty project), then double clicks on the form are just ignored for a few seconds (up to ?

[Lazarusdev] candidates for merging? blikblum, dmitry, bart, jesus Just had a quick look through unmerged revisions. Some of those may be rather feature than bug-fix. Some mention crashes, if those crashes can happen in 1.2RC then merging should be considered. If the authors/committers could check: - it was no oversight, that it is not merged. - the risk of bre?

[Lazarusdev] education-laz may be broken
Just wanted to investigate if this bug (from typhon forum) applies to our IDE too:
http://www.pilotlogic.com/sitejoom/index.php/forum/general-purpose/2741-implementing-pkg-educationlaz-corrupts-options But it does not even compile:
B:\lazarus_latest_2\components\education\educomppalette.pas(175,6?

[Lazarusdev] 1.2RC2 When? And What? (any important pending issue?) We need to decide on the course of RC2. I have not kept track of the issues raised since RC1 (and if any of them were regressions). If anyone is aware of any unfixed/not-merged regression, please bring them up. I mailed the issue/regression about hints not closing on minimize http://forum.laz?

[Lazarusdev] Conflict : r43462 IDE: fixed compilation fpc 2.7.1 The target file does not exist in the fixes branch. Does the code need to be fixed in another file, or is the revision listed in error?

[Lazarusdev] RC1: issue linking on Ubuntu 13.10 from forum: crti.o http://forum.lazarus.freepascal.org/index.php/topic,22558.msg134798.html#msg134798

[Lazarusdev] RC1: Regression, hint not hiding
http://forum.lazarus.freepascal.org/index.php/topic,22558.msg134665.html#msg134665 COnfirmed. The tooltip for showing procedure parameters, uses a different hint window now. This does not hide on Minimize.

[Lazarusdev] RC1: Maybe issue with icons/glyph an GTK2 ? See forum http://forum.lazarus.freepascal.org/index.php/topic,22558.msg134636.html#msg134636

[Lazarusdev] 0025357: [Patch] 1.0.16 merge: TSQLDBLibraryLoader, TPQEventMonitor, TFBEventMonitor and TFBAdmin are actually available in FPC 2.6.2
If someone reviews/test/... this, I can apply it to the branch, and mark the relevant revisions as merged (well those that are listed). I can also just mark the revisions, if someone already commits this to the branch. But I do not work with the sql components, so I will not review or test it m?

[Lazarusdev] codetool exception on forum 1.2rc1
Mattias, maybe worth a look? (I believe you do not read the forum normally?)
http://forum.lazarus.freepascal.org/index.php/topic,22558.msg134082.html#msg134082 I could not reproduce, and he says 1 in 10 times.

[Lazarusdev] 1.0.14 / ready to tag
Mattias, please tag when you have time.

[Lazarusdev] problem with 'compilertest.pas' in
CodeToolBoss.FPCDefinesCache.TestFilename:=CreateCompilerTestPascalFilename; main.pp line 9967 // create a test unit needed to get from the compiler all macros and search paths
CodeToolBoss.FPCDefinesCache.TestFilename:=CreateCompilerTestPascalFilename; MainBuildBoss.UpdateEnglishErrorMsgFilename; LazConf.pp function CreateCompilerTestPascalFilename: string; ...?

[Lazarusdev] RC1 problem on Mac
http://forum.lazarus.freepascal.org/index.php/topic,22558.msg133711.html#msg133711 Maybe someone can have a look? I am not familiar with Mac

[Lazarusdev] All: How to use "Fixed in" field in Mantis (proposed new guideline)
I want to suggest the following guidelines for mantis usage. + New guideline: * Field "fixed in": (used by developer) => affects "Changelog" Set to released or will be release versions only. e.g. 1.2 / 1.4 / 1.2.2 Never to be set to SVN versions (that is what we currently do) ?

"Bart" wrote

> As for being able to participate by submitting patches or fixing bugs

first

> many users may consider such things outside their field of competence,

or

> may only be interested in areas that affect them directly or are deeply > knowledgeable about, [snip] For me it was always a non-starter because

compiles

> were breaking due to missing units and I was always afraid of breaking

my

> installations to try new things even for my own use. It is only now

that I

> am becoming more confident and even then I am still wary. And that is exactly how I started out. At first I was just a user of Lazarus, and I only used the release

versions.

I thinl that's how most devs started out. It was the same for me and FPC :)

No need for VM images (which one could not provide for Windows at all).

Note: Lazarus works more or less on ReactOS. So one could test how good it really works and then provide a ReactOS VM with Lazarus preinstalled.

Regards,
Sven
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus.freepascal.org/pipermail/lazarus/attachments/20140105/b37a4469/attachment-0001.html>

vfclists . wrote

I also suspect that some of those bugs wouldn't creep into the code in the first place if the development process was open enough for more eyes to spot them in the first place.

There are the commit logs and they are much more precise about what's happening than any mailing list and allow real review.

I think being able to see what is coming and follow their development will encourage more participants and more bug fixing as well.

Just follow the commit logs, on windows CommitMonitor works very fine for this.

vfclists . wrote

I have always had this nagging feeling that something wasn't quite right about Lazarus and this explains it.

Now I have a strong nagging feeling about you!

You just refuse to understand what people explain to you. All development is open. New features are here:
http://wiki.lazarus.freepascal.org/Lazarus_1.2.0_release_notes All bug reports are public.
This is the right list if you have _technical_ ideas or questions. Now you are just trolling and you should stop it!
Without a single code contribution you are writing empty whining and nagging in Forum (vfclists) and in this list.
Do you really feel that you should be allowed to do the same thing in developers' private list?

[...] such the issue with transparency that Kostas recently submitted a patch for.

If Kostas continues with his quality patches, he will become a Lazarus developer.
With you it is a different story.
BTW, notice how Kostas was interested in fixing code instead of peeking developers' private mails.
You could learn something from him.

For me it was always a non-starter because compiles were breaking due to missing units and I was always afraid of breaking my installations to try new things even for my own use. It is only now that I am becoming more confident and even then I am still wary.

That's what I mean.

Do you mean you feel too important person to fix bugs? Besides, we use Mantis instead of Bugzilla.

I also suspect that some of those bugs wouldn't creep into the code in the first place if the development process was open enough for more eyes to spot them in the first place.

Oh my God! Please go away troll!
Do you seriously think new bugs are first introduced in developer mailing list before they are committed?

[... and so on ...] In any case if Lazarus developers feel that this approach has proved to be the best over the years then they are entitled to continue as is.

Yes.
To everybody: He is wasting everybody's time. Don't feed him.

Juha

When people ask questions like this it is because they are making comparisons with other projects which are more open, or in the case of Lazarus easier to follow.

I follow a few groups such as pharo-dev and squeak-dev regularly and others such as web2py occasionally. In all instances once I log on to my email I can quickly get a good idea of how things are moving, ie you know how features are progressing and what bugs are being fixed.

I couldn't understand why I coudn't get an idea of developments in Lazarus* just by following the mailing lists* and wondered if there was another mailing list. The mailing list you are referring to is not listed at http://lists.lazarus.freepascal.org/mailman/listinfo and it is clearly not related to development of features as is generally understood.

The wiki only tells you about features after they have been implemented. It doesn't give an idea of how and why they evolved the way they did. Although they enable code displays to be better formatted once you have to follow more than a few different wikis and forums they are less convenient. This is an issue which has been well debated and I have no intention of raising it here.

FWIW I have just spent over an hour exploring Mantis and it looks like the place where a lot of Lazarus development related discussions occur. Obviously it is a complex multiplatform system which is rather mature in internet years and has gathered quite an amount of cruft and I can understand why its developers prefer to focus on fixing existing issues rather than spending/wasting time debating new features.

@Matthias and Bart - I think it helps to open up the more technical discussions to outside viewing. There is always smart lurker out there who can help with some abstruse issue which escapes you or you don't have the time or resources for. Kostas issue is an example of what I am taking about. Not that many people would be familiar with that area, be involved with it and to recognize that there was a bug there let alone have knowledge of the library or API internals to fix it. Bugs for most people mean bugs in programming logic not in interfaces to specialized libraries.

@Juha - Perhaps you could follow the example of the older wiser heads such as Florian and Matthias who have responded to this thread with examples and suggestions because they had a better understanding of why I asked this question. Just because someone's manners or style rubs you the wrong way doesn't mean they are trolling - which is generally understood to be creating drama and controversy for its own sake.

A belated Happy New Year to Lazarus Developers and Congratulations for what is a good development tool given their resource constraints. ( I tend to be stingy with praise!!)

vfclists . wrote

When people ask questions like this it is because they are making comparisons with other projects which are more open, or in the case of Lazarus easier to follow.

How, unless you are part of the project, do you know, if a project is "more" open. How do you know they do not have a private list? Just because we answered truthfully, other projects may (if asked) deny it.

Also even if you knew of the absence of a private list, how do you know that developers do not use personal mail instead (also private).

In my experience any reasonable sized project (not just open source) need some mean of "in group only" discussion. Not because the content is "top secret" but because "too many cooks spoil the broth", and on public issues (even read only) you always get additional "cooks".

Well, now you ssay what you are missing. But that info is not on the dev list either.

The lazarus project is organized differently. We simple do not have a central place for this info. (that is anything on the topic, that is on top of mantis and svn logs)

Unless I can spot something on the svn log, most of the time (greater 95% [1]), I do not know what others work on.
[1] This figure is after eliminating knowledge I have from looking at svn logs. Since I do look at them, I actually have some idea.

This is true for features and bugs.

New features are provided by volunteers. Let me give an example.

I started with work on a new debugger about 3 or 4 month back. My original plan (or rather hope) was, that I get to that sometime next summer. Then I just decided from one minute to the next. A few days later, first commits, and any body looking at svn would have known (And some did notice, as some did ask on the forum)

The wiki only tells you about features after they have been implemented. It doesn't give an idea of how and why they evolved the way they did.

And neither does the developers list. Most features go the way, that the original author had in mind. Sometimes there is something (feedback, additional idea) on the forum, or normal mailing list. Sometime the idea gets picked up sometimes not.
In a few cases such a mail (even though it could, and probably should be public) is on the dev list. But just because a mail goes on the wrong list, does not mean the list is wrong. Same as good a private mail good have been sent. This is not a reason to open the list, but to pay more attention on which list we post.
When I started to contribute, and was not yet on the list, I did exchange some private mails with some of the developers. So list or not, private communication will and does happen.

FWIW I have just spent over an hour exploring Mantis and it looks like the place where a lot of Lazarus development related discussions occur.

Actually we try not to discuss on mantis. Mantis is mainly to identify what is wrong (and that may take a lot of question and answer). In a few cases though, it may be that it includes discussions (rather than notes) how best to fix it

Obviously it is a complex multiplatform system which is rather mature in internet years and has gathered quite an amount of cruft and I can understand why its developers prefer to focus on fixing existing issues rather than spending/wasting time debating new features.

Every developper decides for themself when to fix a bug [2] or do a feature. And each one has his own reasons.
[2] if you break in, you fix it, is a rule we all follow.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lazarus.freepascal.org/pipermail/lazarus/attachments/20140105/70120751/attachment-0001.html>

Considering the svn log, squeak developed (80 commits in 2013) approx. 50 times (!) slower than lazarus (almost 4000 commits in 2013) so no wonder you cannot follow lazarus comparable easily.

Nothing wrong with holding that discussion on this list. Go for it, I'd say. It is *very* telling that Kostas did not ask for or need access to anything else than he already had (this mailing list, the bug tracker etc).

Well, just because somebody's reply rubs you the wrong way doesn't mean the impression you create is the one you might necessarily want to give.

As said by (I think) everybody, it all starts with code contribution. Yes, in patches. Yes, on the bugtracker.
If you give excuses about not being able to contribute code but are willing to correct other people's code (which is what your posts seem to imply to me)... that's not much use, is it? Because you cannot even test your own suggestions.

The impression you're giving me is somebody who is used to a set practice of having well laid-out development goals, central steering etc, and you are unwilling to just jump in and get started. I may very well be wrong about it, but we're talking about impressions here.

That said, I hope you'll be able to contribute patches/code - it can only improve Lazarus.

A belated Happy New Year to Lazarus Developers and Congratulations for what is a good development tool given their resource constraints. <snip>

And best wishes for a good 2014 to you, too.

vfclists . wrote

In all instances once I log on to my email I can quickly get a good idea of how things are moving, ie you know how features are progressing and what bugs are being fixed.

You where looking for an announcement channel, but asking more general for any private list. Then you did (or so it seemed to me) set them equal (which they are not). It took me a few of your mails to figure that.

The idea for more insight is very old:
http://lazarus-dev.blogspot.co.uk/2007/08/lets-start-blogging.html

But it never took of. Seems we rather write code than blogs. We do not intend to keep it secret. We are just not bothered to do the work and write something about it.

vfclists . wrote

In all instances once I log on to my email I can quickly get a good idea of how things are moving, ie you know how features are progressing and what bugs are being fixed.

http://svn.freepascal.org/cgi-bin/viewvc.cgi/?root=lazarus&view=log

How much easier must it get for you to track what's going on in trunk?

FWIW I have just spent over an hour exploring Mantis and it looks like the place where a lot of Lazarus development related discussions occur.

Isn't that a logical place?
You report what you think is a bug and then others can put their information in as well.
And yes, you can use it to discuss how to fix, or if it needs fixing at all.

@Matthias and Bart - I think it helps to open up the more technical discussions to outside viewing.

I disagree.

Bart

Florian Kl?mpfl wrote

Considering the svn log, squeak developed (80 commits in 2013) approx. 50 times (!) slower than lazarus (almost 4000 commits in 2013) so no wonder you cannot follow lazarus comparable easily.

:-)

Juha

Squeak is a Smalltalk system and all the changes are versioned automatically in the live image and shared via the Monticello repository. If they use svn it is probably some ancillary systems related to the project. Between 1st and 5 Jan their repository system has recorded 25 commits to about 10 different packages.

Anyway I am not comparing a dynamic, interpreted system using a live image with a static compiled language as they are quite different paradigms and the development styles are quite different.

vfclists . wrote

[...] I can understand why its developers prefer to focus on fixing existing issues rather than spending/wasting time debating new features.

True, new features are not being debated much but they are being developed surely. This is not only about fixing bugs. This is possible because people mostly agree that those features are needed. At least currently developers are not pulling to different directions.

I try to be more constructive. The fact still is that most features are in developers' personal ToDo--lists and not discussed actively by developers.
However there are many ideas which get often discussed in forum and mailing list by many people.
So there is no lack of discussion. There is only lack of people implementing the ideas.

You forgot the most important means of communication which is the code commit history.
We are like aliens using an alien language, Pascal in this case. It is the most powerful way to communicate SW, a human language cannot do it as well.
Good program code explains to people what we want the computer to do.

Juha

On 05/01/14 14:07, vfclists . wrote:

[ cut tl;irit]

There is another example of a mailing list where technical discussion, commit messages, bug reports and all kinds (trolls, phishing posts, spam) happens.

it's LKML, fwiw. And yes, there is general one, and there are ones dedicated to arm, to vm, mm, subsystems, you want it you name it it probably is there. There is even kernelnewbies mailing list too. But they have an organization with full time admins hired to keep it readable.

The difference is Laz/FPC vs Linux kernel is workflow AND manpower...

I can perfectly understand why Laz/FPC don't want to mix it all together, it is at the stage and in the world where you need a team of ml admins with knowledge to keep a ml from being spammed to death.

FPC/Laz devs have enough work as is, so they need this 'fence' around their playground, even if it seems a bit rigid to us outside it.

-- Frank Church

-L.

The workflow may not be fundamentally different but the size and activity is different.
Linux kernel is huge and requires more admin work for sure. Otherwise _this_ Lazarus list is similar with LKML. Both are open for everybody.
Both have admins and can be moderated if really needed. And ... neither of them is suitable for developers' personal communication.

Uhhh!
It may come as a shock for you but also kernel developers communicate privately. You make it sound like it was something weird that only Lazarus developers are "guilty" of.
No, some communication is private because it is not relevant or interesting for anybody else.
Developers mailing list is a little less private because all developers can see it. That's why I send there only messages which are of interest for other developers.
If I have something that is of interest for also Lazarus users, I send it here. It is as simple as that.
There is no conspiracy or scam behind your back.
So, it is not "a bit rigid", instead it makes things more flexible.

You and Frank Church miss the point completely. In your opinion private communication is a bad thing.
You want to open the developers mailing list.
Then, do you want me to publish also private mails for other developers? Then, do you want me to publish also other private mails for my personal friends?
Where do you draw the line? Do you also want to know how often I go to toilet and how I wipe my ass?

The question is about a person's right for privacy! Is this now better understood or should I write more?

Juha

Ergo, manpower.

But even stable review patches are being posted there publicly, and s-o-b requested. Because that's minimum the developers are ever required/expected to read, of all things. I understand this happens on devel list(s) in FPC/Laz too.

Well apparently it is open r/o so sorry for the 'rigid' word. I very much only meant that r/w-by-invitation to devel list is perfectly justified, IN MY VIEW. Speaking for me only.

Uhhh!

[...]

The question is about a person's right for privacy! Is this now better understood or should I write more?

/I/ understand your reaction, sorry for my choice of words, I did not mean to offend anybody, really.

Juha

-L.

Lukasz Sokol wrote

But even stable review patches are being posted there publicly, and s-o-b requested. Because that's minimum the developers are ever required/expected to read, of all things. I understand this happens on devel list(s) in FPC/Laz too.

Ok, I didn't know the details of their review system. It is organized differently.
Our bug-tracker is used for reviewing patches. The SVN repository is used for reviewing and testing other commits.
There is no formal way to review anything in the developer mailing list. Instead many people follow and test the SVN commits which is kind of a way to communicate the changes.
I think it all works well together with mailing lists and forum. Only PR is almost missing. Linux kernel and other major commercially backed projects get much more publicity.
Any mailing lists don't help with PR though.

/I/ understand your reaction, sorry for my choice of words, I did not mean to offend anybody, really.

Don't worry. I write in a provocative way sometimes, can't help it. :)

Juha

Juha Manninen wrote

On Mon, Jan 6, 2014 at 7:17 PM, Lukasz Sokol wrote:

[...]

I think it all works well together with mailing lists and forum.

OK I agree;

Only PR is almost missing. Linux kernel and other major commercially backed projects get much more publicity.

And here...

Any mailing lists don't help with PR though.

...well here I disagree - go see [briefly] https://lwn.net - (one of) their editors (semi)regularly posts interesting topics dug out on lkml. That's 1000s% more press coverage than I've ever seen on FPC/Laz. PR gain is obvious here.

I volunteer(ed) to at least fwd the press releases of FPC/Lazarus to LWN, as they have not just kernel topics in there, AFAICU they cover F(L)OSS in general. I perceive this as least-i-can-do
- even if it doesn't land on main page (yet).

And while there probably exists net-mag(s) for Pascal topics, I don't think we should limit ourselves to announcing just there.

Juha

-L.

Bart wrote

such as the everlasting Unicode debates,

I feel that main causes of these debates are (1) that the definitions in DXE are not nice (at least completely ignoring any portability issues), but fpc/Lazarus is forced to follow rather closely whatever Delphi does. and (2) that Lazarus only can use what fpc offers and thus Unicode awareness was implemented before fpc provided Unicode-Delphi compatible Strings, resulting in incompatibility in multiple directions.

Thus nobody to be blamed here.

-Michael

Michael Schnell wrote

[...] that Lazarus only can use what fpc offers and thus Unicode awareness was implemented before fpc provided Unicode-Delphi compatible Strings, resulting in incompatibility in multiple directions.

Don't hijack this thread for Unicode please. It is filling too many thread already.
This thread has NOTHING to do with Unicode.

Juha

....

I volunteer(ed) to at least fwd the press releases of FPC/Lazarus to LWN,

Well that is what started the whole topic, it was not the private list per se. It was what was (wrongly) assumed to be see: Project progress (past, but also future plans), and technical aspects.

We actually have http://lazarus-dev.blogspot.co.uk/ Which started for that very reason
http://lazarus-dev.blogspot.co.uk/2007/08/lets-start-blogging.html

But we (the core developers) do not seem to be very active bloggers. So there is very infrequent content.

So what is needed are authors for that kind of stuff.

Someone who looks at the next release page, and picks interesting features, and writes about. If added info is needed, I am sure it will be available if questions are asked.
Or particular interesting bugs.

There just was one, that probably has the characteristics needed. And it was discussed on the public list: "Alpha transparency on GTK2 widgetset" The discussion points out the root issues which are in gtk2 itself (difference between drawable and bixpuf)

I admit, occasionally such a mail (which is of interest to all), gets hidden on the dev private list. (but it makes a small fraction of the lists traffic only). Speaking for me, I will try to put attention to my own posting behaviour, and should I post something of the like, I will consider which list more carefully.

Lukasz Sokol wrote

...well here I disagree - go see [briefly] https://lwn.net - (one of) their editors (semi)regularly posts interesting topics dug out on lkml. That's 1000s% more press coverage than I've ever seen on FPC/Laz. PR gain is obvious here.

Agreed, that would be important and part of the PR. The key word however was "dug". Those people must do lots of digging and editorial work.

I volunteer(ed) to at least fwd the press releases of FPC/Lazarus to LWN, as they have not just kernel topics in there, AFAICU they cover F(L)OSS in general. I perceive this as least-i-can-do - even if it doesn't land on main page (yet).

In addition to what Martin wrote:
We would need a person who digs and digests information and then publishes it. The info is in many places: in mailing lists, in wiki pages for new features and announcement, in SVN commit history, ... and in developers' heads.
I would say the most abundant source of info is the SVN commit history. Developers are following it. Sometimes a new feature found there is commented by others, sometimes not, but everybody follows it more or less.
That info could be collected, digested and published by someone.

If you want to try, please go ahead.
Like in all contributions, if developers see this work has substance, they start to support it for sure.
It means you could ask missing pieces directly from developers and get answers. Even access to devel mailing list could be given, but as I wrote earlier it does not help much.

Regards,
Juha

Well they do. No doubt about that.

[...]

Eventually, this /may/ happen.

However due to time (and skill) constraints I guess it is more practical to try to feed them with 'press releases' first, when they appear. For now. As I progress, who knows what will happen.

Regards, Juha

Cheers,
I guess I can call EOT :)

Lukasz