From szybalski at gmail.com  Sun Oct  3 20:17:14 2021
From: szybalski at gmail.com (Lukasz Szybalski)
Date: Sun, 3 Oct 2021 19:17:14 -0500
Subject: [moin-user] Moin no longer in debian stable?
In-Reply-To: <20210928011154.GB23973@gate.kl-ic.com>
References: <20210928011154.GB23973@gate.kl-ic.com>
Message-ID: <CAKkTUv2YWgWQgAmsnPA05nHwnJBJD7qRS+3kx9R=L+3igxAzrQ@mail.gmail.com>

Hello,
We need to pick a lane.

a) Either convert moin1.9 to python 3, and re-release stable as is.
(knowing we have to release 1.9.py3-bugfixes if any later.
or
b) Release moin 2.0 if its stable enough as new stable.

Otherwise we risk total loss of confidence in the moin project and eventual
death.

As it is right now, my own site is down and soon google will forget all my
ranking and bye bye 10 years of content work.

So which shall it be?

Who can or assist in doing a or b?

Thanks
Lucas


On Mon, Sep 27, 2021 at 8:37 PM Keith Lofstrom <keithl at kl-ic.com> wrote:

> On Sun, Sep 26, 2021 at 07:28:28PM -0500, Lukasz Szybalski wrote:
>
> >    I use moinmoin in lucasmanual.com
> >    I see the package is not in debian stable anymore?
> >
> >    Can someone tell me why not, and what needs to get done to bring it
> back?
>
> There is no debian stable moin.moin package anymore,
> but with the sage advice of gentleman and scholar Paul
> Boddie, there are two workarounds.
>
> Moin 2.0 is somewhere over the rainbow, not available
> for practical purposes.  Existing Moin 1.x deployments
> will probably need a lot of tweaking to be compatible
> with Moin 2.0 (eventually), a process I dread.
>
> Moin 1.x depends on Python 2, which is no longer being
> updated - all the Kool Kids are using Python 3.
>
> However - python2 is still available for LTS versions of
> Debian, such as the Ubuntu 20.04 LTS distro I am deploying
> on my Rimuhosting virtual server.
>
> To make moin 1,x use python2, perhaps all I need to do
> is change the first line of  /usr/local/bin/moin
>
> from #! /usr/bin/python
> to   #! /usr/bin/python2
>
> Moin will not install as a debian package, but it can still
> be built from "source".  Perhaps we can write simplified
> install procedures for our own flavors of Debian.  Perhaps
> some clever person can create a .deb package, which we can
> share privately.
>
> Gentleman Paul also works on moinlight, a python script
> that creates static html web pages out of a deployed
> moin instance.
>
> Though I use moin as my virtual web notebook, I can set
> up a moinlight based script to create a separate set of
> URLs on my server that serve up static "release" versions
> of my work, directing the web searchbots to those.  I also
> hope to create archival DVDs with static content on them.
>
> And if Moin 1.x/python 2 develops a critical security flaw,
> I can edit on a private server and deploy only static html.
>
> That's all hypothetical for now - I haven't finished
> setting up my new server.  However, if we struggle
> through this transition together, we can learn from
> each other's successes and failures.  I will contribute
> plenty of failures ... :-)
>
> Keith
>
> --
> Keith Lofstrom          keithl at keithl.com
> _______________________________________________
> moin-user mailing list
> moin-user at python.org
> https://mail.python.org/mailman/listinfo/moin-user
>


-- 
http://lucasmanual.com/ <http://lucasmanual.com/blog/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.python.org/pipermail/moin-user/attachments/20211003/f5fbd152/attachment.html>

From paul at boddie.org.uk  Mon Oct  4 19:48:21 2021
From: paul at boddie.org.uk (Paul Boddie)
Date: Tue, 05 Oct 2021 01:48:21 +0200
Subject: [moin-user] Moin no longer in debian stable?
In-Reply-To: <CAKkTUv2YWgWQgAmsnPA05nHwnJBJD7qRS+3kx9R=L+3igxAzrQ@mail.gmail.com>
References: <20210928011154.GB23973@gate.kl-ic.com>
 <CAKkTUv2YWgWQgAmsnPA05nHwnJBJD7qRS+3kx9R=L+3igxAzrQ@mail.gmail.com>
Message-ID: <46670240.NCf9ZCDhB8@jason>

On Monday, 4 October 2021 02:17:14 CEST Lukasz Szybalski wrote:
> Hello,
> We need to pick a lane.
> 
> a) Either convert moin1.9 to python 3, and re-release stable as is.
> (knowing we have to release 1.9.py3-bugfixes if any later.
> or
> b) Release moin 2.0 if its stable enough as new stable.

It is a difficult choice. Moin 1.9 is what most of us know, but it isn't 
always particularly nice to work with. Then again, we are at least familiar 
with it. Moin 2.0 seems rather up in the air:

https://moinmo.in/MoinMoin2.0

I guess the actual repository might provide more accurate information about 
its state:

https://github.com/moinwiki/moin

My impression is that quite a bit of the technology is the same or similar, so 
some of the same observations might apply, but one could argue that it is more 
actively developed than Moin 1.9.

> Otherwise we risk total loss of confidence in the moin project and eventual
> death.

Well, unless there is a vibrant community on GitHub using fashionable 
communications tools to coordinate development, I would say that we might have 
seen a substantial loss of confidence, or interest, or motivation, already. 
There is practically no engagement on this list or the developer list.

> As it is right now, my own site is down and soon google will forget all my
> ranking and bye bye 10 years of content work.

Yes, we really want to avoid that kind of outcome.

> So which shall it be?
> 
> Who can or assist in doing a or b?

I am open to either "lane", in principle, although I am not going to burn 
myself out pursuing this, either. Previous experience suggests that people can 
be enthusiastic right up to the point of doing something, and then they all 
step back and leave other people to finish the potentially open-ended task.

Institutional users of Moin might also want to consider investing in the 
solution, as well. I'm all for helping Moin users have a viable solution for 
the future, but I am not in favour of shoring up the bottom line of various 
organisations and corporations because people don't want to have an awkward 
conversation with their bosses.

I don't know whether Thomas, Roger or anyone closer to Moin is willing to say 
what they think might be done to deliver a usable solution from either of the 
possible starting points, but I would welcome the engagement from them on the 
topic. Otherwise, I guess it will be up to us to determine what the situation 
is and what we can usefully do to remedy it.

Paul



From szybalski at gmail.com  Wed Oct  6 20:00:35 2021
From: szybalski at gmail.com (Lukasz Szybalski)
Date: Wed, 6 Oct 2021 19:00:35 -0500
Subject: [moin-user] Moin no longer in debian stable?
In-Reply-To: <46670240.NCf9ZCDhB8@jason>
References: <20210928011154.GB23973@gate.kl-ic.com>
 <CAKkTUv2YWgWQgAmsnPA05nHwnJBJD7qRS+3kx9R=L+3igxAzrQ@mail.gmail.com>
 <46670240.NCf9ZCDhB8@jason>
Message-ID: <CAKkTUv06kOZZit7YRJaHMwukiF_FhzwpgcpzeMYv+M9ZerU7Cw@mail.gmail.com>

Paul,
Thank you for your comments.

To explore the 2.0 upgrade, (while I'm not familiar with its planned or
completed goals); does the moin 2.0 include "upgrade" scripts to convert
existing 1.9 repo's to 2.0 repo's assuming python 2 is no longer available?

If the upgrade scripts are there and the conversion process is somewhat
seamless, then to unify development and community I would say "for long
term" it might be more beneficial. But if the 2.0 is not backward
compatible aka does not contain an upgrade process, then that might be
challenging.

Thank you
Lucas


On Mon, Oct 4, 2021 at 6:48 PM Paul Boddie <paul at boddie.org.uk> wrote:

> On Monday, 4 October 2021 02:17:14 CEST Lukasz Szybalski wrote:
> > Hello,
> > We need to pick a lane.
> >
> > a) Either convert moin1.9 to python 3, and re-release stable as is.
> > (knowing we have to release 1.9.py3-bugfixes if any later.
> > or
> > b) Release moin 2.0 if its stable enough as new stable.
>
> It is a difficult choice. Moin 1.9 is what most of us know, but it isn't
> always particularly nice to work with. Then again, we are at least
> familiar
> with it. Moin 2.0 seems rather up in the air:
>
> https://moinmo.in/MoinMoin2.0
>
> I guess the actual repository might provide more accurate information
> about
> its state:
>
> https://github.com/moinwiki/moin
>
> My impression is that quite a bit of the technology is the same or
> similar, so
> some of the same observations might apply, but one could argue that it is
> more
> actively developed than Moin 1.9.
>
> > Otherwise we risk total loss of confidence in the moin project and
> eventual
> > death.
>
> Well, unless there is a vibrant community on GitHub using fashionable
> communications tools to coordinate development, I would say that we might
> have
> seen a substantial loss of confidence, or interest, or motivation,
> already.
> There is practically no engagement on this list or the developer list.
>
> > As it is right now, my own site is down and soon google will forget all
> my
> > ranking and bye bye 10 years of content work.
>
> Yes, we really want to avoid that kind of outcome.
>
> > So which shall it be?
> >
> > Who can or assist in doing a or b?
>
> I am open to either "lane", in principle, although I am not going to burn
> myself out pursuing this, either. Previous experience suggests that people
> can
> be enthusiastic right up to the point of doing something, and then they
> all
> step back and leave other people to finish the potentially open-ended task.
>
> Institutional users of Moin might also want to consider investing in the
> solution, as well. I'm all for helping Moin users have a viable solution
> for
> the future, but I am not in favour of shoring up the bottom line of
> various
> organisations and corporations because people don't want to have an
> awkward
> conversation with their bosses.
>
> I don't know whether Thomas, Roger or anyone closer to Moin is willing to
> say
> what they think might be done to deliver a usable solution from either of
> the
> possible starting points, but I would welcome the engagement from them on
> the
> topic. Otherwise, I guess it will be up to us to determine what the
> situation
> is and what we can usefully do to remedy it.
>
> Paul
>
>
> _______________________________________________
> moin-user mailing list
> moin-user at python.org
> https://mail.python.org/mailman/listinfo/moin-user
>


-- 
http://lucasmanual.com/ <http://lucasmanual.com/blog/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.python.org/pipermail/moin-user/attachments/20211006/a49ec67e/attachment.html>

From crosseyedpenguin at yahoo.com  Thu Oct  7 17:45:37 2021
From: crosseyedpenguin at yahoo.com (Roger Haase)
Date: Thu, 7 Oct 2021 21:45:37 +0000 (UTC)
Subject: [moin-user] Moin no longer in debian stable?
In-Reply-To: <CAKkTUv06kOZZit7YRJaHMwukiF_FhzwpgcpzeMYv+M9ZerU7Cw@mail.gmail.com>
References: <20210928011154.GB23973@gate.kl-ic.com>
 <CAKkTUv2YWgWQgAmsnPA05nHwnJBJD7qRS+3kx9R=L+3igxAzrQ@mail.gmail.com>
 <46670240.NCf9ZCDhB8@jason>
 <CAKkTUv06kOZZit7YRJaHMwukiF_FhzwpgcpzeMYv+M9ZerU7Cw@mail.gmail.com>
Message-ID: <533729771.839247.1633643137426@mail.yahoo.com>

 The best way to install Moin2 is to clone the repo using git.?https://moin-20.readthedocs.io/en/latest/admin/install.html#installation-for-developers

When you get to "Typing ?./m? (or ?m? on Windows) will display a menu similar to:", there is a missing?
. activate?
command in the docs..?
Then display the menu using ./m; then ./m import19 <path to your 1.9 wiki/data>
Then ./m run and point your browser to?http://127.0.0.1:8080
Note the old FrontPage is now Home by default, see configs.
Easiest way to fix is to create a Home page with the content
{{FrontPage}}


If you just want to look at moin2, try:?https://moinmoin2.pythonanywhere.com/Home

You will not be able to login nor update any pages.
Roger Haase



    On Wednesday, October 6, 2021, 05:00:52 PM MST, Lukasz Szybalski <szybalski at gmail.com> wrote:  
 
 Paul,Thank you for your comments.
To explore the 2.0 upgrade, (while I'm not familiar with its planned or completed goals); does the moin 2.0 include "upgrade" scripts to convert existing 1.9 repo's to 2.0 repo's assuming python 2 is no longer?available?
If the upgrade scripts are there and the conversion process is somewhat seamless, then to unify development and community I would say "for long term" it might be more beneficial. But if the 2.0 is not backward compatible aka does not contain an upgrade process, then that might be challenging.
Thank youLucas

On Mon, Oct 4, 2021 at 6:48 PM Paul Boddie <paul at boddie.org.uk> wrote:

On Monday, 4 October 2021 02:17:14 CEST Lukasz Szybalski wrote:
> Hello,
> We need to pick a lane.
> 
> a) Either convert moin1.9 to python 3, and re-release stable as is.
> (knowing we have to release 1.9.py3-bugfixes if any later.
> or
> b) Release moin 2.0 if its stable enough as new stable.

It is a difficult choice. Moin 1.9 is what most of us know, but it isn't 
always particularly nice to work with. Then again, we are at least familiar 
with it. Moin 2.0 seems rather up in the air:

https://moinmo.in/MoinMoin2.0

I guess the actual repository might provide more accurate information about 
its state:

https://github.com/moinwiki/moin

My impression is that quite a bit of the technology is the same or similar, so 
some of the same observations might apply, but one could argue that it is more 
actively developed than Moin 1.9.

> Otherwise we risk total loss of confidence in the moin project and eventual
> death.

Well, unless there is a vibrant community on GitHub using fashionable 
communications tools to coordinate development, I would say that we might have 
seen a substantial loss of confidence, or interest, or motivation, already. 
There is practically no engagement on this list or the developer list.

> As it is right now, my own site is down and soon google will forget all my
> ranking and bye bye 10 years of content work.

Yes, we really want to avoid that kind of outcome.

> So which shall it be?
> 
> Who can or assist in doing a or b?

I am open to either "lane", in principle, although I am not going to burn 
myself out pursuing this, either. Previous experience suggests that people can 
be enthusiastic right up to the point of doing something, and then they all 
step back and leave other people to finish the potentially open-ended task.

Institutional users of Moin might also want to consider investing in the 
solution, as well. I'm all for helping Moin users have a viable solution for 
the future, but I am not in favour of shoring up the bottom line of various 
organisations and corporations because people don't want to have an awkward 
conversation with their bosses.

I don't know whether Thomas, Roger or anyone closer to Moin is willing to say 
what they think might be done to deliver a usable solution from either of the 
possible starting points, but I would welcome the engagement from them on the 
topic. Otherwise, I guess it will be up to us to determine what the situation 
is and what we can usefully do to remedy it.

Paul


_______________________________________________
moin-user mailing list
moin-user at python.org
https://mail.python.org/mailman/listinfo/moin-user



-- 
http://lucasmanual.com/
_______________________________________________
moin-user mailing list
moin-user at python.org
https://mail.python.org/mailman/listinfo/moin-user
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.python.org/pipermail/moin-user/attachments/20211007/e7aaead0/attachment.html>

From paul at boddie.org.uk  Thu Oct  7 20:08:46 2021
From: paul at boddie.org.uk (Paul Boddie)
Date: Fri, 08 Oct 2021 02:08:46 +0200
Subject: [moin-user] Moin no longer in debian stable?
In-Reply-To: <CAKkTUv06kOZZit7YRJaHMwukiF_FhzwpgcpzeMYv+M9ZerU7Cw@mail.gmail.com>
References: <20210928011154.GB23973@gate.kl-ic.com> <46670240.NCf9ZCDhB8@jason>
 <CAKkTUv06kOZZit7YRJaHMwukiF_FhzwpgcpzeMYv+M9ZerU7Cw@mail.gmail.com>
Message-ID: <1898774.HanNXZ5boA@jason>

On Thursday, 7 October 2021 02:00:35 CEST Lukasz Szybalski wrote:
> Paul,
> Thank you for your comments.

No problem! Note that I am just another observer with regard to Moin 2, so 
this is as much a voyage of discovery for me as it is for you.

> To explore the 2.0 upgrade, (while I'm not familiar with its planned or
> completed goals); does the moin 2.0 include "upgrade" scripts to convert
> existing 1.9 repo's to 2.0 repo's assuming python 2 is no longer available?

It looks like it. Conveniently, looking at the current history on GitHub, the 
following appears:

import19 should offer choice of output converters, fixes #1122

Which suggests that there are indeed converters. And I missed that Moin 2 
seems to support Python 3, which is not particularly prominent from the wiki 
page. (It says "master branch: Python 3.5+" three quarters of the way down the 
page.)

> If the upgrade scripts are there and the conversion process is somewhat
> seamless, then to unify development and community I would say "for long
> term" it might be more beneficial. But if the 2.0 is not backward
> compatible aka does not contain an upgrade process, then that might be
> challenging.

I am starting to think that with Python 3 compatibility, Moin 2 should be the 
emphasis of any real effort, given that converting Moin 1.9 to Python 3 would 
be the tedious chore that it was everywhere else in the Python ecosystem 
(thanks to the general mismanagement of the language roadmap).

So, I looked at Moin 2 again, and found a few things out...

Firstly, getting started could be better documented. The wiki page is not 
helpful here, nor is the README.rst file on GitHub. But there is the Sphinx 
documentation [1]:

https://moin-20.readthedocs.io/en/latest/

Specifically, this:

https://moin-20.readthedocs.io/en/latest/admin/install.html#installation-for-developers

Somewhere, the requirements for going any further will probably mention the 
need for venv and pip, although not prominently (you need to look at the 
quickinstall.py program's docstring), so given the Debian context, let me note 
this command to be somewhat helpful:

sudo apt-get install python3-venv python3-pip

Now, the quickinstall.py program will fail without venv and pip, not even 
showing any help text (which should appear if you put "help" after the program 
name on the command line), but it will also keep failing if it failed the 
first time...

"""
Traceback (most recent call last):
  File "./quickinstall.py", line 794, in <module>
    choice = 'cmd_%s' % args[1]
IndexError: list index out of range
"""

So, do this to try and remedy that:

rm activate m

I nevertheless found that the program will complain about an absence of "pip" 
in the log. [2] If we're pursuing a Python 3 strategy, we need pip3 on Debian. 
The quickinstall.py program is explicitly a Python 3 program, so I went 
through the program and parameterised it to take a common executable name for 
pip, introducing "pip3" as the value of this parameter. [3]

Cleaning up and running the program again got me lots of complaints about 
packages not building:

"""
  Running setup.py bdist_wheel for Flask-Script: started
  Running setup.py bdist_wheel for Flask-Script: finished with status 'error'
  Complete output from command /home/paulb/Software/Python/moin-2.0/moin-venv-
python3/bin/python3 -u -c "import setuptools, tokenize;__file__='/tmp/pip-
install-inu3e504/Flask-Script/setup.py';f=getattr(tokenize, 'open', open)
(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, 
__file__, 'exec'))" bdist_wheel -d /tmp/pip-wheel-ik38pfch --python-tag cp37:
  usage: -c [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...]
     or: -c --help [cmd1 cmd2 ...]
     or: -c --help-commands
     or: -c cmd --help

  error: invalid command 'bdist_wheel'

  ----------------------------------------
  Failed building wheel for Flask-Script
  Running setup.py clean for Flask-Script
"""

And so on, ending with warnings about mismatching setuptools versions:

"""
ERROR: setuptools==40.8.0 is used in combination with setuptools_scm>=6.x

Your build configuration is incomplete and previously worked by accident!
"""

The problem with this kind of thing is that you probably have to be the person 
who did all this Python packaging to understand what you did wrong. [4] I 
didn't really find a working remedy for the "bdist_wheel" issue: random people 
on the Internet suggest all sorts of "smash all the buttons" remedies 
involving the wheel package, setup.py, upgrading pip, and so on.

What initially "worked" for me was this:

pip3 install --upgrade pip

But because this puts a new version in my ~/.local directory hierarchy, I then 
needed to make sure that the full path to this separate pip3 program was given 
as the parameter I added to the quickinstall.py program. Still, the setuptools 
error remained, and I also got a complaint about "compile_catalog" not being 
supported.

(I then got into all sorts of other problems with regard to the .local 
directory connected to the idiocy of various Linux desktop developers, but let 
me save the rants for the footnotes.)

And so, I will leave things there for today, given that I also want to spend 
some of my spare time on other things, and also that this starts to resemble 
less attractive things about my day job. I have probably overlooked something 
here, but not being immersed in this project, I feel that things could be a 
bit more obvious.

But I hope that my moaning has indicated some areas where obstacles might be 
removed so that more people can investigate the presumably good stuff beyond 
those obstacles.

Paul

Footnote rants:

[1] Sphinx?! Still, unlike many of the projects whose documentation I find on 
Read The Docs these days, at least there is quite a bit of it here. But I 
thought that wikis were good for documentation and that is why we are using 
Moin, along with the nicer syntax than the overly cute but frustrating 
reStructuredText syntax (double dots, double commas, leading and trailing 
underscores, precise numbers of underlining symbols, and so on).

[2] Why does this need to go to a log? It is a critical error. I suppose the 
output of the program does summarise the log fairly nicely, and this is useful 
when pip is repeatedly failing, but then the program probably shouldn't finish 
with a message vaguely suggesting success ("Type ". activate" to activate 
venv, then "./m" for menu" or "Successfully created or updated venv").

[3] Yes, I know that the Python core developers want everyone to pretend that 
Python didn't exist before version 3, but in the real world, people still need 
to distinguish between legendary Python and Python 3. (I work in a domain 
where there are still Python 2 programs being actively used, which shows how 
much the need for constructive outreach to actual users was overlooked, as 
opposed to cheerleading amongst advocates and pundits, combined with guilt-
shaming and making needless work for the rest of the community.)

[4] Back when everyone decided that Python should have a packaging system 
"like CPAN", I guess no-one advocating that position ever had to deal with 
random Perl packages failing to build or install, leaving them in a situation 
where there is no obvious remedy. This is why we have software distributions 
like Debian where processes are in place to build and reliably deploy 
packages.



From keithl at kl-ic.com  Thu Oct  7 21:34:10 2021
From: keithl at kl-ic.com (Keith Lofstrom)
Date: Thu, 7 Oct 2021 18:34:10 -0700
Subject: [moin-user] Moin 2 raw text?
In-Reply-To: <mailman.17726.1633651748.4167.moin-user@python.org>
References: <mailman.17726.1633651748.4167.moin-user@python.org>
Message-ID: <20211008013410.GA18543@gate.kl-ic.com>

On Thu, 7 Oct 2021 21:45:37 +0000 (UTC) Roger Haase <crosseyedpenguin at yahoo.com> wrote:

> If you just want to look at moin2, try:?https://moinmoin2.pythonanywhere.com/Home

Unfamiliar interface ... "different" or "inferior?". 
10 minutes of frobbing did not reveal the equivalent of Moin 1.9:

More Actions:
   Raw Text

I hope to learn what is typed in, to make what comes out.  I'll git the
github version, and frob some more.

One really important advantage (for me) of Moin 1.9 is that I can compose
raw text with other tools - building raw text for tables with C programs,
for example, then cut and paste.  Will Moin 2 permit that?

Slightly disturbing is that the banner area is a little taller.  As
screens get wider and wider (and simultaneously runty-er and runty-er),
tall banners are a step backwards.  OTOH, if the whole banner can be
collapsed with a not-yet-discovered button click, I can learn to love it.

----

Anyway, if there's a vote for converting Moin1.9 to Python 3, or
releasing Moin2 *AS THAT DEMO SEEMS TO BEHAVE NOW* as "Stable" and
going with that ...  I'm in the Moin1.9 migration camp.  FORK BACK!

Not sure what I can contribute to converting Moin1.9, besides testing
beta code and making revealing mistakes and documenting what I did to
make them (I have daily backups, I won't lose my work).  But then,
I can't make hello.c compile the same way twice running, so if the
retrograde renegade team needs a crash test dummy, I'm your dummy.

As far as difficult discussions with my boss, my wife lets me think
that's me ... until I make a mistake, as defined by her.  She won't
take kindly to rewriting her wiki content and learning a different
computer tool, so I would rather kick in some money to the Moin1.9
migration, instead of paying a divorce attorney and moving to a
fleabag hotel.

Keith

-- 
Keith Lofstrom          keithl at keithl.com

From crosseyedpenguin at yahoo.com  Fri Oct  8 17:44:28 2021
From: crosseyedpenguin at yahoo.com (Roger Haase)
Date: Fri, 8 Oct 2021 21:44:28 +0000 (UTC)
Subject: [moin-user] Moin 2 raw text?
In-Reply-To: <20211008013410.GA18543@gate.kl-ic.com>
References: <mailman.17726.1633651748.4167.moin-user@python.org>
 <20211008013410.GA18543@gate.kl-ic.com>
Message-ID: <34739206.1209607.1633729468573@mail.yahoo.com>

 There are several name changes, "Raw" is now "Highlight" (or "</>" on tiny displays), "Edit" is now "Modify"...
Cut and paste still works.
There is not a button link to hide the banner.
Roger Haase
    On Thursday, October 7, 2021, 07:34:30 PM MST, Keith Lofstrom <keithl at kl-ic.com> wrote:  
 
 On Thu, 7 Oct 2021 21:45:37 +0000 (UTC) Roger Haase <crosseyedpenguin at yahoo.com> wrote:

> If you just want to look at moin2, try:?https://moinmoin2.pythonanywhere.com/Home

Unfamiliar interface ... "different" or "inferior?". 
10 minutes of frobbing did not reveal the equivalent of Moin 1.9:

More Actions:
? Raw Text

I hope to learn what is typed in, to make what comes out.? I'll git the
github version, and frob some more.

One really important advantage (for me) of Moin 1.9 is that I can compose
raw text with other tools - building raw text for tables with C programs,
for example, then cut and paste.? Will Moin 2 permit that?

Slightly disturbing is that the banner area is a little taller.? As
screens get wider and wider (and simultaneously runty-er and runty-er),
tall banners are a step backwards.? OTOH, if the whole banner can be
collapsed with a not-yet-discovered button click, I can learn to love it.

----

Anyway, if there's a vote for converting Moin1.9 to Python 3, or
releasing Moin2 *AS THAT DEMO SEEMS TO BEHAVE NOW* as "Stable" and
going with that ...? I'm in the Moin1.9 migration camp.? FORK BACK!

Not sure what I can contribute to converting Moin1.9, besides testing
beta code and making revealing mistakes and documenting what I did to
make them (I have daily backups, I won't lose my work).? But then,
I can't make hello.c compile the same way twice running, so if the
retrograde renegade team needs a crash test dummy, I'm your dummy.

As far as difficult discussions with my boss, my wife lets me think
that's me ... until I make a mistake, as defined by her.? She won't
take kindly to rewriting her wiki content and learning a different
computer tool, so I would rather kick in some money to the Moin1.9
migration, instead of paying a divorce attorney and moving to a
fleabag hotel.

Keith

-- 
Keith Lofstrom? ? ? ? ? keithl at keithl.com
_______________________________________________
moin-user mailing list
moin-user at python.org
https://mail.python.org/mailman/listinfo/moin-user
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.python.org/pipermail/moin-user/attachments/20211008/0d38747b/attachment.html>

From paul at boddie.org.uk  Sat Oct  9 18:02:44 2021
From: paul at boddie.org.uk (Paul Boddie)
Date: Sun, 10 Oct 2021 00:02:44 +0200
Subject: [moin-user] Moin 2 raw text?
In-Reply-To: <34739206.1209607.1633729468573@mail.yahoo.com>
References: <mailman.17726.1633651748.4167.moin-user@python.org>
 <20211008013410.GA18543@gate.kl-ic.com>
 <34739206.1209607.1633729468573@mail.yahoo.com>
Message-ID: <2825843.xRtlKuTbTa@jason>

On Friday, 8 October 2021 23:44:28 CEST Roger Haase via moin-user wrote:
>  There are several name changes, "Raw" is now "Highlight" (or "</>" on tiny
> displays), "Edit" is now "Modify"... Cut and paste still works.

Surely the closest equivalent to the raw action is the "Download" link, 
although that forces a download as opposed to letting the browser display the 
page content as plain text.

> There is not a button link to hide the banner.

I guess that theming could probably introduce such flexibility, though.

Previously...

On Thursday, October 7, 2021, 07:34:30 PM MST, Keith Lofstrom <keithl at kl-
ic.com> wrote:
>
> Unfamiliar interface ... "different" or "inferior?".
> 10 minutes of frobbing did not reveal the equivalent of Moin 1.9:
> 
> More Actions:
>   Raw Text
> 
> I hope to learn what is typed in, to make what comes out.  I'll git the
> github version, and frob some more.

I have always found Moin 2 to be rather unfamiliar despite the familiar 
general styling. Some kind of Moin 1.9 theme with familiar navigation elements 
would be helpful, perhaps.

> One really important advantage (for me) of Moin 1.9 is that I can compose
> raw text with other tools - building raw text for tables with C programs,
> for example, then cut and paste.  Will Moin 2 permit that?

I imagine that editing still provides an opportunity for the pasting of raw 
text from other sources, but when I was still using Moin 1.9 and updating the 
pages separately, I was using Moin's import mechanism which was a bit 
cumbersome. I don't know whether Moin 2 reproduces that closely or has 
something else.

> Slightly disturbing is that the banner area is a little taller.  As
> screens get wider and wider (and simultaneously runty-er and runty-er),
> tall banners are a step backwards.  OTOH, if the whole banner can be
> collapsed with a not-yet-discovered button click, I can learn to love it.

Like with Moin 1.9, if theming is extensive enough then it should be possible 
to have something closer to what you are used to.

> Anyway, if there's a vote for converting Moin1.9 to Python 3, or
> releasing Moin2 *AS THAT DEMO SEEMS TO BEHAVE NOW* as "Stable" and
> going with that ...  I'm in the Moin1.9 migration camp.  FORK BACK!
> 
> Not sure what I can contribute to converting Moin1.9, besides testing
> beta code and making revealing mistakes and documenting what I did to
> make them (I have daily backups, I won't lose my work).  But then,
> I can't make hello.c compile the same way twice running, so if the
> retrograde renegade team needs a crash test dummy, I'm your dummy.

I think you (and maybe quite a few others) probably need to see a theme closer 
to Moin 1.9, a good mechanism for displaying raw text (probably using an 
action for that purpose if it doesn't exist already), another good mechanism 
for importing pages, and it would also be nice to have a decent mechanism for 
exporting wikis as static pages, which is something "moin export dump" rather 
failed to do adequately, in my opinion.

I also think that the deployment situation needs to be a lot better. Relying 
on Python packaging tools to work is not really sustainable. After getting 
myself into a mess with them the other evening, I then encountered them doing 
absurd things during my work day on Friday, which turned out to be some 
special Debian customisations to supposedly prevent users from doing "bad 
things", as if there were any further need to compound the general frustration 
involved in using them.

Of course, if Moin 2 were to be packaged for Debian, the various dependencies 
would also be packaged, but this is tedious work (technically and socially) 
that would need to be encouraged, supported and have people within the Debian 
scene committed to a successful outcome. This in contrast to the way that 
Python 2 was thrown overboard.

(And yes, I know that the Python core developers don't support Python 2 any 
more: I wrote probably several blog posts about such matters quite some time 
ago, maybe even mentioning how the core developers were actively frustrating 
and sabotaging people's attempts to maintain Python 2 independently, notably 
forbidding them from using the Python name. Parroting the rhetoric of those 
core developers and going along with various juvenile coercion techniques like 
the "Python clock" was not the respectful approach one might have expected 
from a Free Software distribution.)

> As far as difficult discussions with my boss, my wife lets me think
> that's me ... until I make a mistake, as defined by her.  She won't
> take kindly to rewriting her wiki content and learning a different
> computer tool, so I would rather kick in some money to the Moin1.9
> migration, instead of paying a divorce attorney and moving to a
> fleabag hotel.

By those criteria, an investment in Moin might be a good investment, but I 
would like to hear those closer to Moin core development suggest how existing 
Moin users might migrate, perhaps based on their own experiences. I hope that 
the demo site isn't the only Moin 2 site out there.

Paul