> Il 07/05/20 15:52, Norman Megill ha scritto:
> > Set.mm is not a part of the programs. It is a file of data like a Word
> > or an Excel file. It should be kept in the current directory.
On Thu, 7 May 2020 21:17:16 +0200, Giovanni Mascellani
wrote:
> I tend to agree with this point of view:
On Thu, May 7, 2020 at 12:19 PM Giovanni Mascellani
wrote:
>
> Il 06/05/20 18:40, David A. Wheeler ha scritto:
> > So "\Documents and Settings" was sort-of like /home.
> > This was HIDEOUS for programs and users, because
> > "Documents and Settings" is absurdly long AND has spaces in it,
> >
Il 06/05/20 18:40, David A. Wheeler ha scritto:
> So "\Documents and Settings" was sort-of like /home.
> This was HIDEOUS for programs and users, because
> "Documents and Settings" is absurdly long AND has spaces in it,
> which is pain for Windows shell similar to the pain of Unix shell.
Not to
Hi,
Il 07/05/20 15:52, Norman Megill ha scritto:
> Set.mm is not a part of the programs. It is a file of data like a Word
> or an Excel file. It should be kept in the current directory.
I tend to agree with this point of view: set.mm and any other Metamath
database is just a file like any other
Forwarded Message
Subject: Git
Date: Thu, 7 May 2020 11:45:59 +0200 (CEST)
From: fl
To: Megill Norman
Hi Norm,
Can you post this:
providing only communication through git is not a good Idea. Git should be
reserved forp who wants to modify the programs (Metamath or mmj2).
On Tuesday, May 5, 2020 at 12:00:28 AM UTC-4, David A. Wheeler wrote:
>
>
> We could continue to post Metamath-exe with a precompiled binary
> and have people store it in C:\metamath. That's not how software
> is usually installed on Windows, and it's clunky, but it *works*.
There are many
On Wed, 6 May 2020 00:45:22 -0700 (PDT), "'B. Wilson' via Metamath"
wrote:
> There also seems to be some tension about installation expectations. Some
> people like the fact that metamath Just Works by downloading and extracting
> an archive to wherever. Others, like myself, seem to prefer
On Wed, 6 May 2020 00:45:22 -0700 (PDT), "'B. Wilson' via Metamath"
wrote:
> Would it make sense to have "search path" variables for metamath and mmj2?
It could. All Metamath tools (at least Metamath & mmj2)
would have to be modified to support it, and that's a minor pain.
But once done things
> My eyes are certainly Linux-tinted, but is there some sort of analogue to
> the above for Windows?
Sort of. Please sit back, the story is long, below.
I've lived through all this history: I've used Unix & Linux of many flavors,
and I've also used CP/M & MS-DOS 1.0 on & Windows on.
--- David
On Wed, 6 May 2020 00:45:22 -0700 (PDT), "'B. Wilson' via Metamath"
wrote:
> Would it make sense to have "search path" variables for metamath and mmj2?
>
> The FHS describes /var/lib and /var/local/lib as places for "variable data"
> associated with programs in /usr and /usr/local,
Here are some brainstorming questions and thoughts:
Would it make sense to have "search path" variables for metamath and mmj2?
The FHS describes /var/lib and /var/local/lib as places for "variable data"
associated with programs in /usr and /usr/local, respectively. Ostensibly,
the purpose of
Forwarded Message
Subject: Current problem
Date: Tue, 5 May 2020 20:19:24 +0200 (CEST)
From: fl
To: Megill Norman
Hi Norm,
Can you post this.
I don't know how you do, but my own version of mmj2 has no problems to find
set.mm in the current directory.
--
FL
On Tuesday, May
On Tue, 5 May 2020 08:00:35 -0700, David Starner wrote:
> It'd be a lot easier to handle set.mm if it were more standard to have
> set.mm as a constant and editing being done to separate .mm files.
I don't think so. Which version of set.mm would be constant?
--- David A. Wheeler
--
You
It'd be a lot easier to handle set.mm if it were more standard to have
set.mm as a constant and editing being done to separate .mm files.
--
The standard is written in English . If you have trouble understanding
a particular section, read it again and again and again . . . Sit up
straight. Eat
FL:
> For me it's not "/usr/bin" it's "/usr/local/bin"
That's true for many people. Metamath-exe actually already supports this,
just use "make install" with autoconf (as always that's the default).
If you want /usr/bin (as is common for packages installed with a
package manager), use
Forwarded Message
Subject: Directories
Date: Tue, 5 May 2020 11:36:00 +0200 (CEST)
From: fl
To: Megill Norman
Hi Norm,
Can you post this.
For me it's not "/usr/bin" it's "/usr/local/bin" and set.mm is not a
library it's a file of data. You have to
be able to browse it and
On Mon, 4 May 2020 18:41:16 -0700 (PDT), "'B. Wilson' via Metamath"
wrote:
...
Your post does *not* discuss what to do with Windows.
We need to decide how to help people install Metamath on Windows,
since that's a common platform.
We could continue to post Metamath-exe with a precompiled
On Mon, 4 May 2020 18:41:16 -0700 (PDT), "'B. Wilson' via Metamath"
wrote:
> Thank you for starting this discussion. As a package manager, I certainly
> empathsize strongly with efforts to make the current installation more
> standards-compliant. In that vein, I would recommend against
this discussion
goes!
Cheers,
2020年5月4日月曜日 2時16分07秒 UTC+9 David A. Wheeler:
>
> I want to clarify a key point in my earlier post,
> "[Metamath] Proposed installation conventions so things will be easier to
> install".
>
> To make it easier to install mmj2 and o
On Mon, 4 May 2020 08:32:59 -0700 (PDT), Norman Megill
wrote:
> While I'm not sure of the best way to go forward, let me explain why set.mm
> and the metamath program are currently in the same directory.
...
> Putting both into one directory allows an unsophisticated Windows user to
> click on
nly access to set.mm was via
the program, so it was important to make getting started as simple and
quick as possible for such users.
Norm
On Sunday, May 3, 2020 at 1:16:07 PM UTC-4, David A. Wheeler wrote:
>
> I want to clarify a key point in my earlier post,
> "[Metamath] Proposed
This looks reasonable. I'm not completely sure I follow which pieces are
recommendations which people can follow or not, and which pieces are
default places for scripts to look (which people can also not follow,
since there will be ways to override defaults). But I'm sure this will
be clear as
I want to clarify a key point in my earlier post,
"[Metamath] Proposed installation conventions so things will be easier to
install".
To make it easier to install mmj2 and other tools, I want to change the
conventions
so that "set.mm" is by default stored in its *own* di
We need to make it easier for new Metamath users to join the fun.
Typical Metamath users have multiple things installed. I think they should be
able to install things wherever they want, *but* we should have some
conventions so that things "just work" on a simple install by default.
Here are
24 matches
Mail list logo
