request for decisions for the release of LyX 2.1

[Uwe Stöhr]

Dear LyX developers,

as I don't understand what is nowadays going with LyX I trigger a discussion.

My point of view:

- LyX 2.1 is THE chance to advertise LyX since a while.

- For an unknown reason an RC had to be released before the docs were ready while I requested 2 more
weeks. But an undocumented feature is an unknown/unused and therefore untested. The goal is to have
a stable release therefore it is important that all features are tested. I am not confident with
that but accepted it.

- I requested that everybody had a look at
http://wiki.lyx.org/LyX/NewInLyX21
to list all new features and to add some images to beautify the page. But except of Jürgen, nobody
did anything. While stepping through the docs I found 3 features that where not on that page and one
feature that was removed. That is unsatisfying. This page is our main advertising place as far as I see.

- There are more people testing an RC release than a Beta release so there should be enough time
between the first RC and the final one. Less than 2 weeks is not enough. For example it will not be
enough time for the translators. If we want to keep as many people active as possible we should take
the necessary time so have their work in.

- To assure a stable as possible release it is important to get as many feedback as possible and
time to fix the things that were reported. It should also be considered to release as many RCs as
possible. That is why I sent out a mail more than 30 persons with request to test LyX 2.1RC1. It
took a while until I got the first feedback mails (this Tuesday). So obviously people need their time.

- We don't have a plan yet how to advertise LyX and when. In the past I sent out mails to
contributing editors of computer magazines, to student groups at universities etc.. What are the
plans for LyX 2.1? We should synchronize our actions. Sending a mail to press people will only make
sense right after the release. We know that many of use won't have time around Easter (the same is
for press people) so we should consider to wait until Easter, I mean LyX 2.1 is our chance and we
should not let it go!

- When a problem occurs that deserves a test build, trust the maintainer of a platform that he knows
what he is doing. Feedback is the most important point in the RC period and never harms!

- Formalism doesn't help us. The user don't care about formalism, but only if the program is stable.

- Even today I got 2 feedback mails that the float counter does not work in RC1. We should therefore
consider to release an RC2 because that would uncover further issues. Testers are happy to get a new
version test further than to continue testing with a version that has an annoying bug. And more
important, maybe we have other uninitialized variable bugs and these can cause very strange things
as we have seen.

- We should not release LyX this weekend. We need more time for user feedback and if possible an
RC2. Also the translators need more time. After 3 years of development there is no reason not to
wait e.g. another 2 weeks.

- We should discuss our strategy, time frames etc. It is not OK that we have a single release
manager who decides alone. If we are a community we should vote about the main things like release
dates, pre-releases, advertising activities.

- As nowadays we become more formalistic, well then we should vote about some rules and store them
somewhere. E.g. the German Wikipedia has a lot of rules. this is controversial but they are the
result of a fair vote and discussion and they can be changed any time by a new vote. Personally, I
don't like formalism and many rules, but obviously others do. Then we should be that consequent to
do it right to avoid the much stress we now have. It is not OK that a few say this is now a rule and
some say "+1" or so.


I would like to hear your opinions and hope that we can do some decisions.

thanks and best regards
Uwe

[Vincent van Ravesteijn]

Post by Uwe Stöhr
Dear LyX developers,
as I don't understand what is nowadays going with LyX I trigger a discussion.

To be honest, this makes me want to stop reading immediately. Also,
the remarks that you need to have a position within the project that
allows you to ignore the others don't really help. Finally, the bloat
about formalism reduces your chance further to get a proper response.

Vincent

[Vincent van Ravesteijn]

Post by Uwe Stöhr
- For an unknown reason an RC had to be released before the docs were ready
while I requested 2 more weeks. But an undocumented feature is an
unknown/unused and therefore untested. The goal is to have a stable release
therefore it is important that all features are tested. I am not confident
with that but accepted it.

If this is really your opinion, you should have updated the
documentation while we were in beta-testing phase, because that it is
the time were features are tested. Instead, you wait until we conclude
on the list that 2.1 is ready to be released, and then you start
updating the documentation. Apparently, you didn't care about testing
before, and now you start screaming you need testing.

Vincent

[Uwe Stöhr]

Post by Vincent van Ravesteijn

Post by Uwe Stöhr
- For an unknown reason an RC had to be released before the docs were ready
while I requested 2 more weeks. But an undocumented feature is an
unknown/unused and therefore untested. The goal is to have a stable release
therefore it is important that all features are tested. I am not confident
with that but accepted it.

If this is really your opinion, you should have updated the
documentation while we were in beta-testing phase, because that it is
the time were features are tested. Instead, you wait until we conclude
on the list that 2.1 is ready to be released, and then you start
updating the documentation. Apparently, you didn't care about testing
before, and now you start screaming you need testing.

During the last 3 release cycles it works that way that strings could be changed in the beta phase.
I learned from the 1.5 release cycle where I had to update the docs several times because strings
and menu names were changed.
I always made clear that it doesn't make sense to update the docs before the strings are almost
freezed. We once also had a string freeze for some weeks before the release.
However, for this release cycle I started to update the docs in the first week of January.

regards Uwe

[Vincent van Ravesteijn]

Post by Uwe Stöhr
- For an unknown reason an RC had to be released before the docs were ready
while I requested 2 more weeks. But an undocumented feature is an
unknown/unused and therefore untested. The goal is to have a stable release
therefore it is important that all features are tested. I am not confident
with that but accepted it.

At Feb 21 Georg asked what needed to be done. At Feb 27 you answered:
"I guess I will need at least 2 more weeks and I guess I will uncover
further bugs." Three weeks later, at Mar 20, I froze master.
Technically speaking, you had well more than the 2 more weeks that you
requested.

I'm not sure what I did wrong here.

Vincent

[Uwe Stöhr]

Post by Vincent van Ravesteijn

Post by Uwe Stöhr
- For an unknown reason an RC had to be released before the docs were ready
while I requested 2 more weeks. But an undocumented feature is an
unknown/unused and therefore untested. The goal is to have a stable release
therefore it is important that all features are tested. I am not confident
with that but accepted it.

"I guess I will need at least 2 more weeks and I guess I will uncover
further bugs." Three weeks later, at Mar 20, I froze master.
Technically speaking, you had well more than the 2 more weeks that you
requested.
I'm not sure what I did wrong here.

Hi Vincent,

I expected a string freeze like in the past release cycles. I will not blame you for this because we
all are to blame here. We should start to write down how we want to release that it becomes more
transparent and that we don't forget about that for the next release.

The time between the freeze and the release was too short. I wrote you that need more time to
update the docs and worked that time at least 2 hours a day on this. If nobody is working on the
docs. well then LyX can of course be released anyway but there were at least 3 people actively
working on the docs. I also wrote you that in my private mail parallel to this thread that you don't
get the impression that I want to blame you personally. (That attempt sadly failed.)

You saw that the updating of the docs uncover some bugs and it is therefore a good idea to freeze
LyX afterwards. Moreover as we all want a stable as possible release the docs help a lot because
testers will follow its descriptions and test the new features. And that is what the RC-phase is
about in my opinion: the program is feature-frozen and string frozen. Users get some weeks to test
it thoroughly and we fix as many bugs as possible.

regards Uwe

[Vincent van Ravesteijn]

Op 28 apr. 2014 23:17 schreef "Uwe >And that is what the RC-phase is about
in my opinion: the program is feature-frozen and string frozen. Users get
some weeks to test thoroughly and we fix as many bugs as possible.
You completely misunderstood. Bèta releases are for testing. RC are release
candidates. Bèta releases should be feature frozen, RCs should be frozen
except for major issues. I really do not know where you got another
impression from.

Vincent

[Uwe Stöhr]

Post by Vincent van Ravesteijn
Op 28 apr. 2014 23:17 schreef "Uwe >And that is what the RC-phase is about
in my opinion: the program is feature-frozen and string frozen. Users get
some weeks to test thoroughly and we fix as many bugs as possible.
You completely misunderstood. Bèta releases are for testing. RC are release
candidates. Bèta releases should be feature frozen, RCs should be frozen
except for major issues. I really do not know where you got another
impression from.

We had a lot of string changes during the beta phase and the last feature of algorithm2e was added
in January.

However, with an RC you attract more users to test a release than with a beta because people have
the feeling to test an almost ready product. Therefore an RC2 makes sense. Take for example
LibreOffice - many of the bugs are reported against RCs and they usually never release with only one
RC. In the past we also had a longer RC-phase. This might sound unpleasant but this way we can get
more feedback. Now we have e.g. again the case that we have lyx2lyx issues. 2 are uncovered by a bug
report after the freeze. Georg is doing a great job of fixing them right now but I think you can
understand that I wanted to have them fixed before LyX 2.1.0 final.

But OK, LyX is now out and from the time of the release we found (thanks to your daughter ;-) ) a
good date. Easter holidays are over, today many people returned to work or the university.
I already prepared some mails to press people but some already wrote about LyX - perfect!

regards Uwe

[Vincent van Ravesteijn]

Post by Uwe Stöhr

Post by Vincent van Ravesteijn
Op 28 apr. 2014 23:17 schreef "Uwe >And that is what the RC-phase is about
in my opinion: the program is feature-frozen and string frozen. Users get
some weeks to test thoroughly and we fix as many bugs as possible.
You completely misunderstood. Bčta releases are for testing. RC are
release
candidates. Bčta releases should be feature frozen, RCs should be frozen
except for major issues. I really do not know where you got another
impression from.

We had a lot of string changes during the beta phase and the last feature
of algorithm2e was added in January.

Well, yes, that's my fault. If it caused you extra work, it is me to blame
for it. I'm sorry about that.

Post by Uwe Stöhr
However, with an RC you attract more users to test a release than with a
beta because people have the feeling to test an almost ready product.

Yes, and with the final release you attract even more users, so let's do
multiple final releases, and fix as many bugs as possible.

Vincent

[Uwe Stöhr]

Post by Vincent van Ravesteijn

Post by Uwe Stöhr
However, with an RC you attract more users to test a release than with a
beta because people have the feeling to test an almost ready product.

Yes, and with the final release you attract even more users, so let's do
multiple final releases, and fix as many bugs as possible.

:-)

regards Uwe

[Pavel Sanda]

blame you for this because we all are to blame here. We should start to
write down how we want to release that it becomes more transparent and that
we don't forget about that for the next release.

Uwe,

I don't think this is not matter of forgetting/transparency. I agree with you
that all multiple beta/rc are better than what we had now, but it really boils
down to the fact that we did not have man power for doing that; we can feel
lucky that 2.1 is out at all ;)

What I have been thinking about is changing the model how we select release
managers for next releases.

It is consistent that devs active in the begining of the release cycle are
different from devs active in the end. 99% of work, by which I mean doing
tarballs, testing, keeping list up-to-date with mails, pushing other devs to
finish their work etc is at the end of cycle.

I remember that in all last three major releases we had the same problem that
it was actually the community, who had to scream and push things forward.

So my proposal for the next release would be that we let things to evolve in
it's usual chaotic way and in the moment there is a feeling in the community
that release is due, we try to push someone active at that time to take the
responsibility (perhaps the one who screams most ;)

What do others think about that?
Pavel

[Pavel Sanda]

Post by Pavel Sanda
I don't think this is not matter of forgetting/transparency. I agree with you

this is not matter -> this is a matter

[Richard Heck]

Post by Pavel Sanda
So my proposal for the next release would be that we let things to evolve in
it's usual chaotic way and in the moment there is a feeling in the community
that release is due, we try to push someone active at that time to take the
responsibility (perhaps the one who screams most ;)
What do others think about that?

I think that's a good idea: Figure out when it is time for the work to
be done who should do it.

Richard

[Benjamin Piwowarski]

Post by Pavel Sanda
So my proposal for the next release would be that we let things to evolve in
it's usual chaotic way and in the moment there is a feeling in the community
that release is due, we try to push someone active at that time to take the
responsibility (perhaps the one who screams most ;)
What do others think about that?

I think that's a good idea: Figure out when it is time for the work to
be done who should do it.

Seems good. Also, maybe it would be a good idea for the next freezes to leave the master branch as the development (no change) and use e.g. a stable-2.2 branch for the stable code. I think this led to some confusion. 

Benjamin 

[Vincent van Ravesteijn]

Post by Benjamin Piwowarski

Post by Pavel Sanda

Post by Pavel Sanda
So my proposal for the next release would be that we let things to

evolve in

Post by Pavel Sanda
it's usual chaotic way and in the moment there is a feeling in the

community

Post by Pavel Sanda
that release is due, we try to push someone active at that time to

take the

Post by Pavel Sanda
responsibility (perhaps the one who screams most ;)
What do others think about that?

I think that's a good idea: Figure out when it is time for the work to
be done who should do it.

Seems good. Also, maybe it would be a good idea for the next freezes
to leave the master branch as the development (no change) and use e.g.
a stable-2.2 branch for the stable code. I think this led to some
confusion.
Benjamin

People don't want that either. They appear to be used to freezing master
and any change is confusing.

Vincent

[Jürgen Spitzmüller]

Post by Uwe Stöhr
- We should not release LyX this weekend. We need more time for user
feedback and if possible an RC2. Also the translators need more time. After
3 years of development there is no reason not to wait e.g. another 2 weeks.

I see no reason to delay the release any further. LyX 2.1 actually should
have been released long ago.
All (potential) problems you list can be fixed in the maintenance phase.
There is only one really urgent bug (#9049), but we have way too less
information to debug it.

Also, I think it is the release manager's job to fix the release date, and
I fully support Vincent's decisions.

I have nothing more to say. I do not see the need to raise a fundamental
discussion about LyX 2.1 release at this point.

Jürgen

[Uwe Stöhr]

I do not see the need to raise a fundamental discussion about LyX 2.1 release at this point.

Well, we only release a major version ever 2 - 3 years. So we should take the time to bring it to a
shape that we can advertise it. For example if there is e.g. an issue with the installation, press
people don't give a program a good review. The same is with incomplete translations.

I started this thread for various reasons, one of the main reason is that I miss a concept for LyX
and the 2.1 release. What is our plan for advertisement, when do we think it is a good time to
release, who of the press people do we ant to inform, what do we think is our market
scientific-only, schools, business usage, .... According to our decision we can develop a small
campaign.
Look e.g at LibreOffice. I like their campaigns and this is how I found their program. Another good
example are the Calligra people. Because of their press campaign I noticed Krita, which I now use
instead of Gimp. They got reviews in many popular computer webpages.

Besides the strategy I find it more than fair to give the translators time to get their things in
before the release. Sure one can do this also for 2.1.1 but imagine how do they would feel. And
waiting some weeks for user feedback is never an issue. Our file format is now fix but what is if a
user encounters a bug in this regard. In the last release we had some lyx2lyx bugs that were first
reported after the release of 2.0.0 as far as I remember. That should be avoided. We tested lyx2lyx
as best as we can but getting more feedback - and if it is that no lyxl2yx bug occurs - the better
it is.

However, I have seen that the release was already postponed.

Also, I think it is the release manager's job to fix the release date

That is too simple. The release date matters. For example in Germany the semester just has begun so
if we advertise LyX for students this is a bad time - the mail boxes of student organizations are
full and from my time at the University the beginning of a semester was the most stressy time. In
these days one does not try out a new LyX. What about school organizations? - Easter holidays just
have begun. What about press? - it's Easter time. What about companies? - it's Easter time. I mean
every country has its own specialties and with a strategy who we want to address and how we will
find a suitable release date.

regards Uwe

[Jean-Marc Lasgouttes]

Post by Uwe Stöhr
I started this thread for various reasons, one of the main reason is
that I miss a concept for LyX and the 2.1 release. What is our plan
for advertisement, when do we think it is a good time to release, who
of the press people do we ant to inform, what do we think is our
market scientific-only, schools, business usage, .... According to
our decision we can develop a small campaign.

For some releases, we had a small "PR" team. That was good, but it is
not something that was decided by a committee at release time. This work
is begun when one feels that the release is looming. This time, we don't
have it.

PR is just another feature of a release. Just as we cannot decide in
advance who is going to develop what feature, we cannot force anybody to
do PR. And it would be too late anyway for big plans.

Post by Uwe Stöhr
I mean every country has its own specialties and with a strategy who
we want to address and how we will find a suitable release date.

Hmm, I see it coming... A different release date for different
countries. And then we could have region codes like DVDs to prevent
people to use LyX in a region of the world when we do not want it to be
used? :)

This is what happens when we think too much. Let's just stop
procrastinating and release the thing.

JMarc

[Uwe Stöhr]

Post by Jean-Marc Lasgouttes

Post by Uwe Stöhr
I started this thread for various reasons, one of the main reason is
that I miss a concept for LyX and the 2.1 release. What is our plan
for advertisement, when do we think it is a good time to release, who
of the press people do we ant to inform, what do we think is our
market scientific-only, schools, business usage, .... According to
our decision we can develop a small campaign.

For some releases, we had a small "PR" team. That was good, but it is
not something that was decided by a committee at release time. This work
is begun when one feels that the release is looming. This time, we don't
have it.

Hi JMarc,

the problem here is again bad communication. I am pretty sure that we could have got some LyX users
to help us here but we never asked them.
We don't sell LyX, sure, but it is nevertheless a product. Developing a product is just one part
marketing is as important as support and development. What does it help us when we add the next
feature for a very small user group when most of the people don't use LyX. I wrote you that LyX is
forbidden in many companies and even universities. Also at my university time I _HAD TO_ transform
my scientific papers to MS Word to be able to submit them to some chemistry scientific journals.
They explicitly denied LaTeX.

Post by Jean-Marc Lasgouttes
PR is just another feature of a release. Just as we cannot decide in
advance who is going to develop what feature, we cannot force anybody to do PR.

You mention an important point: We don't have a development plan. What do many customers need and
what does only please us developers? We never did a user survey on our webpages and the lists.
For example I use math ad chemistry equations almost daily and it would be fun to add support for
all \xarrow commands in math. But who will use this feature except of math experts?

From the feedback I got it seems that for average users the most useful feature of LyX 2.1 is the
table row/column shifting feature. So instead of implementing support for \xarrow it would make more
sense to work on the table feature so that it for example also works on a cell base (not only for
complete rows and columns). The import of table from Word and webpages also seems to be a useful
feature for average users.

You see, with a development plan we can focus our work on what many needs instead of adding this and
that which might only please us extremely experienced users. I will open a new thread for this.

Post by Jean-Marc Lasgouttes

Post by Uwe Stöhr
I mean every country has its own specialties and with a strategy who
we want to address and how we will find a suitable release date.

Hmm, I see it coming... A different release date for different countries.

No, definitely not. maybe you laugh at me but you have never launched a product, right? Why do you
think companies spend as much money for the release marketing as for the whole development of a
product? There are many people involved and th release date is a very important if not the most
important issue. If your main market is Europe, you will for example not launch a product in August
where about 50% of the customers are in vacation.

However, as we released LyX 2.1 after Easter it was possible to get mentioned in Germany's largest
computer magazine:
http://www.heise.de/newsticker/meldung/Freie-LaTeX-Umgebung-LyX-2-1-erschienen-2178581.html

(they write that LyX is also running user OS/2. I haven't told them this and already wrote them to
correct this if possible.)

regards Uwe

[Jean-Marc Lasgouttes]

Post by Uwe Stöhr
You mention an important point: We don't have a development plan. What
do many customers need and what does only please us developers? We never
did a user survey on our webpages and the lists.
For example I use math ad chemistry equations almost daily and it would
be fun to add support for all \xarrow commands in math. But who will use
this feature except of math experts?

This is how open source software works. Developers do what they want. I
do not mean that they do stuff that nobody want. But if some one wants
some "boring" feature to be implemented, they have to pay someone to do
it. It is as simple as that.

JMarc

[Vincent van Ravesteijn]

On Tue, Apr 29, 2014 at 10:42 AM, Jean-Marc Lasgouttes

Post by Uwe Stöhr
You mention an important point: We don't have a development plan. What

Post by Uwe Stöhr
do many customers need and what does only please us developers? We never
did a user survey on our webpages and the lists.
For example I use math ad chemistry equations almost daily and it would
be fun to add support for all \xarrow commands in math. But who will use
this feature except of math experts?

This is how open source software works. Developers do what they want. I do
not mean that they do stuff that nobody want. But if some one wants some
"boring" feature to be implemented, they have to pay someone to do it. It
is as simple as that.
JMarc

On one hand, if we would have a clear plan on what we want to have in LyX
2.2, for example, people might be indeed working on those things.

On the other hand, we have a list of bugs targeted to 2.1.x and 2.2.0, but
those bugs are there for years and we move them from 1.6.x to 2.0.x to
2.1.x .... and it doesn't seem to be a major point of interest to the
developers.

....or we should come up with a rule that no new features get into LyX if
they are not in the agreed plan, but I doubt whether there are developers
left if we do that.

Vincent

[Jürgen Spitzmüller]

Post by Uwe Stöhr
This is how open source software works. Developers do what they want. I do
not mean that they do stuff that nobody want. But if some one wants some
"boring" feature to be implemented, they have to pay someone to do it. It
is as simple as that.

Or join the team. I never would have joined the development team if there
would have been people who were willing to fulfill all my requests. And let
me remind you that I was completely coding-alliterate when I started. So "I
am no programmer" or "I do not have time" is not a legitimate excuse ;-)

If open source projects need business and PR plans, then I seem to have
failed to understand what OSS is about.

Jürgen

Post by Uwe Stöhr
JMarc

[Vincent van Ravesteijn]

Post by Jean-Marc Lasgouttes
This is how open source software works. Developers do what they
want. I do not mean that they do stuff that nobody want. But if
some one wants some "boring" feature to be implemented, they have
to pay someone to do it. It is as simple as that.
Or join the team. I never would have joined the development team if
there would have been people who were willing to fulfill all my
requests. And let me remind you that I was completely
coding-alliterate when I started. So "I am no programmer" or "I do not
have time" is not a legitimate excuse ;-)
If open source projects need business and PR plans, then I seem to
have failed to understand what OSS is about.
Jürgen

When I was at the GSOC mentor summit, I was asked numerous times what
the "business model of LyX" is. I didn't really have an answer to that,
other than that we are a bunch of enthousiasts working on it in our free
time.

OSS is not what it used to be. Most projects have ways of making money,
or are supported by companies. Git's maintainer is hired by Google,
major work on Git is done by GitHub people, KDE has a large investor
that allows them to hire developers, and so forth..

I also read last week that some large tech companies are going to
financially support OSS to prevent any new "Heartbleed" bugs. They say
this bug was caused partly because the developers of OpenSSL just didn't
have enough time to ensure the quality of the "product" (apparently they
are enthousiasts like us).

Vincent

[Jürgen Spitzmüller]

Post by Vincent van Ravesteijn
OSS is not what it used to be. Most projects have ways of making money,
or are supported by companies. Git's maintainer is hired by Google, major
work on Git is done by GitHub people, KDE has a large investor that allows
them to hire developers, and so forth..

Maybe this applies to big projects, but I fail to see how it applies to
LyX. Also, there are surely different interests amongst coders. Some
developers might have ambitions to code on a professional basis and thus be
"hired", ideally for their pet projects. I certainly don't. I get the money
to pay my rent otherwise, so I am not interested in any commercial
programming.

Jürgen

[Vincent van Ravesteijn]

Post by Vincent van Ravesteijn
OSS is not what it used to be. Most projects have ways of making
money, or are supported by companies. Git's maintainer is hired by
Google, major work on Git is done by GitHub people, KDE has a
large investor that allows them to hire developers, and so forth..
Maybe this applies to big projects, but I fail to see how it applies
to LyX.

LyX is rather big IMO ;)..

Post by Vincent van Ravesteijn
Also, there are surely different interests amongst coders. Some
developers might have ambitions to code on a professional basis and
thus be "hired", ideally for their pet projects. I certainly don't. I
get the money to pay my rent otherwise, so I am not interested in any
commercial programming.

Yes, luckily people differ. I only wanted to say that I felt like we are
the exception, but I didn't mean to say we should do it different, or
that I want it to be different. I also said it in a proud way that we
are just "amateurs" ;) ...

It would be nice though if someone would invest such that bigger
features could be added to LyX, but I'm personally not looking forward
to work on LyX full time as a day-job for ever.

Vincent

[Richard Heck]

Post by Vincent van Ravesteijn
I also read last week that some large tech companies are going to
financially support OSS to prevent any new "Heartbleed" bugs. They say
this bug was caused partly because the developers of OpenSSL just
didn't have enough time to ensure the quality of the "product"
(apparently they are enthousiasts like us).

Yeah, there were like four of them. It's kind of amazing. But maybe it
will wake up the companies who are making money off OSS products that
they need to support them, as well.

Richard

[Georg Baum]

Post by Uwe Stöhr
Well, we only release a major version ever 2 - 3 years. So we should take
the time to bring it to a shape that we can advertise it. For example if
there is e.g. an issue with the installation, press people don't give a
program a good review. The same is with incomplete translations.

You speak about advertising a lot, and I agree that it would be nice to have
a good adverstising campaign. However, we do not sell LyX, so advertising is
not essential. IMHO we simply do not have the man power to do such
sophisticated adverstising as you'd like it to have: We cannot afford it,
and if we'd try, we would never see a 2.1.0 release. Given that
precondition, it is much better to release without any advertising.

I agree with you that several timings in the 2.1 release process were not
good in the past, and there was unneeded confusion and weeks without any
progress. However, I am very glad that Vincent found the time again to
finally do the release, and I think we all should thank him for that, work
together and help to make the release as good as possible.


Georg

PS: If all developers agree to adopt a more "continous integration" like
development style it is easy to release every year or so.

[Uwe Stöhr]

Am 11.04.2014 08:46, schrieb Vincent van Ravesteijn:

At first sorry for CCing that to the devel list, but I am again close to explode and need to come to
an end.

Please.. try to work with us, do not always stick to and repeat your
own insights and don't always give us the impression that you think
we're against each other.

We are not all against each others. I never wrote or meant that. It is incredible what you read out
of my mails.

Why don't you understand we don't _want_ a formal voting system,

(Please note that You is not We.) I also don't like formalism as I clearly stated. But if I am
blamed to violate rules, well then we need a system how these rules are set. Every club has a system
how to set up the club statues. This avoids many discussions and that is why almost all communities
have statutes.

because we are here just for fun. We all respect each other's
opinions, but certain people's opinion have more impact than other's
because of history, technical knowledge, track records, whatever...

Is that really your understanding of a democratic community? If so then we cannot work together in
future. Everybody has one voice and one opinion. It is absolutely undemocratic to say that someone's
opinion counts more. That reminds me of "Animal Farm" where some animals are more equal than others.

That's just a natural balance and we are happy with that. Don't be so
childish to complain about that over and over again.

I am childish because I have a different opinion? Thanks for the flowers. I only complained once
because of Jürgen's post. I never said that I really want that we elect an executive board or
something similar. Please re-read my posts before stating things that I did not say.

To quiet the further development maybe it is better to retire from active development. I will reply
to some posts on the list why I wrote what and to point out again what I think we should do. I am
not happy how decisions are currently made. That doesn't mean that I sabotage anything. Nevertheless
you accuse me for that now three times. I don't like to spent more spare time to a project where I
am not allowed to say my opinion. I did not insult anybody personally I only tried to explain what I
want. We obviously don't have the same goals, OK, but please stop blaming me for each and every bit
I do, every opinion I have etc. I just can't stand that anymore. You said we are here for fun. I
have no fun anymore. Please don't write my any more private mails, we can discuss everything on the
list and you can simply ignore me in future if you like.

Uwe