Discussion:
Browser suggestion: local server
Steve Comstock
2015-11-28 13:08:16 UTC
Permalink
Hello Steve,
There are excellent, not IT motivated reasons for
using a local server, or better said locating an
(actual) interface at 127.0.0.1.
Well, I'm aware of that interface, but it is not
at all what I'm talking about; my suggestion needs
code in the browser to simulate the way a server
handles <!--#include ... --> statements.
This is not how the "Web of Things" works,
but I don't care about that.
but this is how people arrange collections of
reference documents. This is highly significant
in Emergency Management where hardware and
connectivity can be disrupted by the event itself
... but you, your laptop and trusty thumb drive
survived. There are Portable Apps ...
(http://portableapps.com/), but your trusty thumb
drive might not have its favorite laptop around.
My proposal has nothing to do with survival in an
emergency, it's far more prosaic. If I have all the
pages and files for a website on a thumb drive, then
any laptop will work because there will be some
browser on the laptop.
You can count on at least a working browser on a
working laptop, I think.
Me too.


So, if the browser supports the current standard,
and if the standard says when a browers is pointed
at a local file whose name ends in '.shtml' then
the browser should attempt to handle server side
includes in the same way a server does.
That said, the document collection should then be
XML ... because the style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not distract or
interfere with access.
Well, I don't want to step on any toes here, but
my impression is that XHTML is kinda' moribund and
that the latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).

And, it shouldn't matter: if the HTML standard were
to support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
Subject: Browser suggestion: local server
Date: Thursday, November 12, 2015, 11:08 AM
Guys,
I've been doing a lot of development using .shtml
and server side includes. Testing, however, is a
bit of a pain: I can't really test the includes
are working until I upload all the files to my
server.
It occurs to me it would be terrific if this
* If a browser (user agent) points to a local file,
and if the filename ends in '.shtml', then the
browser should endeavor to process any 'include'
statements in the file in the same way a server
would
This would also be nice because I can put a whole
website on a thumb drive then display it to a meeting
or class without having to actually connect to the
internet! Makes the site much more portable.
Is that reasonable? Desirable? How do I go about
proposing such behavior?
Kind regards,
-Steve Comstock
303-355-2752
Jonathan Garbee
2015-11-28 16:23:40 UTC
Permalink
Why not just download a copy of nginx? It can be ran from a folder directly
without any install. Or a python install (they can be portable as well) and
use its simple HTTP server module?

Getting a local server running for testing is very easy and accessible now.
I don't see why UA's should be forced to step in here.

- Garbee
Post by Steve Comstock
Hello Steve,
There are excellent, not IT motivated reasons for
using a local server, or better said locating an
(actual) interface at 127.0.0.1.
Well, I'm aware of that interface, but it is not
at all what I'm talking about; my suggestion needs
code in the browser to simulate the way a server
handles <!--#include ... --> statements.
This is not how the "Web of Things" works,
but I don't care about that.
but this is how people arrange collections of
reference documents. This is highly significant
in Emergency Management where hardware and
connectivity can be disrupted by the event itself
... but you, your laptop and trusty thumb drive
survived. There are Portable Apps ...
(http://portableapps.com/), but your trusty thumb
drive might not have its favorite laptop around.
My proposal has nothing to do with survival in an
emergency, it's far more prosaic. If I have all the
pages and files for a website on a thumb drive, then
any laptop will work because there will be some
browser on the laptop.
You can count on at least a working browser on a
working laptop, I think.
Me too.
So, if the browser supports the current standard,
and if the standard says when a browers is pointed
at a local file whose name ends in '.shtml' then
the browser should attempt to handle server side
includes in the same way a server does.
That said, the document collection should then be
XML ... because the style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not distract or
interfere with access.
Well, I don't want to step on any toes here, but
my impression is that XHTML is kinda' moribund and
that the latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).
And, it shouldn't matter: if the HTML standard were
to support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
Subject: Browser suggestion: local server
Date: Thursday, November 12, 2015, 11:08 AM
Guys,
I've been doing a lot of development using .shtml
and server side includes. Testing, however, is a
bit of a pain: I can't really test the includes
are working until I upload all the files to my
server.
It occurs to me it would be terrific if this
* If a browser (user agent) points to a local file,
and if the filename ends in '.shtml', then the
browser should endeavor to process any 'include'
statements in the file in the same way a server
would
This would also be nice because I can put a whole
website on a thumb drive then display it to a meeting
or class without having to actually connect to the
internet! Makes the site much more portable.
Is that reasonable? Desirable? How do I go about
proposing such behavior?
Kind regards,
-Steve Comstock
303-355-2752
Seth Call
2015-11-28 16:29:28 UTC
Permalink
On Firefox, (probably other browsers), there are extensions that run a web
server too...

https://addons.mozilla.org/en-US/firefox/addon/pow-plain-old-webserver/
Post by Jonathan Garbee
Why not just download a copy of nginx? It can be ran from a folder
directly without any install. Or a python install (they can be portable as
well) and use its simple HTTP server module?
Getting a local server running for testing is very easy and accessible
now. I don't see why UA's should be forced to step in here.
- Garbee
Post by Steve Comstock
Hello Steve,
There are excellent, not IT motivated reasons for
using a local server, or better said locating an
(actual) interface at 127.0.0.1.
Well, I'm aware of that interface, but it is not
at all what I'm talking about; my suggestion needs
code in the browser to simulate the way a server
handles <!--#include ... --> statements.
This is not how the "Web of Things" works,
but I don't care about that.
but this is how people arrange collections of
reference documents. This is highly significant
in Emergency Management where hardware and
connectivity can be disrupted by the event itself
... but you, your laptop and trusty thumb drive
survived. There are Portable Apps ...
(http://portableapps.com/), but your trusty thumb
drive might not have its favorite laptop around.
My proposal has nothing to do with survival in an
emergency, it's far more prosaic. If I have all the
pages and files for a website on a thumb drive, then
any laptop will work because there will be some
browser on the laptop.
You can count on at least a working browser on a
working laptop, I think.
Me too.
So, if the browser supports the current standard,
and if the standard says when a browers is pointed
at a local file whose name ends in '.shtml' then
the browser should attempt to handle server side
includes in the same way a server does.
That said, the document collection should then be
XML ... because the style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not distract or
interfere with access.
Well, I don't want to step on any toes here, but
my impression is that XHTML is kinda' moribund and
that the latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).
And, it shouldn't matter: if the HTML standard were
to support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
Subject: Browser suggestion: local server
Date: Thursday, November 12, 2015, 11:08 AM
Guys,
I've been doing a lot of development using .shtml
and server side includes. Testing, however, is a
bit of a pain: I can't really test the includes
are working until I upload all the files to my
server.
It occurs to me it would be terrific if this
* If a browser (user agent) points to a local file,
and if the filename ends in '.shtml', then the
browser should endeavor to process any 'include'
statements in the file in the same way a server
would
This would also be nice because I can put a whole
website on a thumb drive then display it to a meeting
or class without having to actually connect to the
internet! Makes the site much more portable.
Is that reasonable? Desirable? How do I go about
proposing such behavior?
Kind regards,
-Steve Comstock
303-355-2752
Steve Comstock
2015-11-28 16:44:39 UTC
Permalink
Post by Seth Call
On Firefox, (probably other browsers), there are extensions that run a web
server too...
https://addons.mozilla.org/en-US/firefox/addon/pow-plain-old-webserver/
Cool. I use Firefox almost exclusively.

Oh, wait ... not available for Firefox 42.0, the latest
and greatest version, which is what I run.

Of course, I would have to include Firefox and the addon
in my thumbdrive, and install them if not available on the
PC I am making a presentation from.
Post by Seth Call
Post by Jonathan Garbee
Why not just download a copy of nginx? It can be ran from a folder
directly without any install. Or a python install (they can be portable as
well) and use its simple HTTP server module?
Well see, I didn't know about these options. I'll explore them. Thanks.
Post by Seth Call
Post by Jonathan Garbee
Getting a local server running for testing is very easy and accessible
now. I don't see why UA's should be forced to step in here.
OK, well I was just trying to make it de rigeur for all current
browsers so there's nothing to install. Just another lazy developer
I guess. :-)

Thanks again.

-Steve
Post by Seth Call
Post by Jonathan Garbee
- Garbee
Post by Steve Comstock
Hello Steve,
There are excellent, not IT motivated reasons for
using a local server, or better said locating an
(actual) interface at 127.0.0.1.
Well, I'm aware of that interface, but it is not
at all what I'm talking about; my suggestion needs
code in the browser to simulate the way a server
handles <!--#include ... --> statements.
This is not how the "Web of Things" works,
but I don't care about that.
but this is how people arrange collections of
reference documents. This is highly significant
in Emergency Management where hardware and
connectivity can be disrupted by the event itself
... but you, your laptop and trusty thumb drive
survived. There are Portable Apps ...
(http://portableapps.com/), but your trusty thumb
drive might not have its favorite laptop around.
My proposal has nothing to do with survival in an
emergency, it's far more prosaic. If I have all the
pages and files for a website on a thumb drive, then
any laptop will work because there will be some
browser on the laptop.
You can count on at least a working browser on a
working laptop, I think.
Me too.
So, if the browser supports the current standard,
and if the standard says when a browers is pointed
at a local file whose name ends in '.shtml' then
the browser should attempt to handle server side
includes in the same way a server does.
That said, the document collection should then be
XML ... because the style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not distract or
interfere with access.
Well, I don't want to step on any toes here, but
my impression is that XHTML is kinda' moribund and
that the latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).
And, it shouldn't matter: if the HTML standard were
to support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
Subject: Browser suggestion: local server
Date: Thursday, November 12, 2015, 11:08 AM
Guys,
I've been doing a lot of development using .shtml
and server side includes. Testing, however, is a
bit of a pain: I can't really test the includes
are working until I upload all the files to my
server.
It occurs to me it would be terrific if this
* If a browser (user agent) points to a local file,
and if the filename ends in '.shtml', then the
browser should endeavor to process any 'include'
statements in the file in the same way a server
would
This would also be nice because I can put a whole
website on a thumb drive then display it to a meeting
or class without having to actually connect to the
internet! Makes the site much more portable.
Is that reasonable? Desirable? How do I go about
proposing such behavior?
Kind regards,
-Steve Comstock
303-355-2752
Seth Call
2015-11-28 16:50:51 UTC
Permalink
I feel forced to recommend this too:

In Python 2.x, you can do:

*python -m SimpleHTTPServer*

Or in Python 3.x



*python3 -m http.server*
Then, the current folder that you ran the command will have it's files
served from port 8000.
Post by Steve Comstock
Post by Seth Call
On Firefox, (probably other browsers), there are extensions that run a web
server too...
https://addons.mozilla.org/en-US/firefox/addon/pow-plain-old-webserver/
Cool. I use Firefox almost exclusively.
Oh, wait ... not available for Firefox 42.0, the latest
and greatest version, which is what I run.
Of course, I would have to include Firefox and the addon
in my thumbdrive, and install them if not available on the
PC I am making a presentation from.
Post by Seth Call
Why not just download a copy of nginx? It can be ran from a folder
Post by Jonathan Garbee
directly without any install. Or a python install (they can be portable as
well) and use its simple HTTP server module?
Well see, I didn't know about these options. I'll explore them. Thanks.
Post by Seth Call
Post by Jonathan Garbee
Getting a local server running for testing is very easy and accessible
now. I don't see why UA's should be forced to step in here.
OK, well I was just trying to make it de rigeur for all current
browsers so there's nothing to install. Just another lazy developer
I guess. :-)
Thanks again.
-Steve
Post by Seth Call
Post by Jonathan Garbee
- Garbee
On Sat, Nov 28, 2015 at 8:08 AM, Steve Comstock <
Hello Steve,
There are excellent, not IT motivated reasons for
using a local server, or better said locating an
(actual) interface at 127.0.0.1.
Well, I'm aware of that interface, but it is not
at all what I'm talking about; my suggestion needs
code in the browser to simulate the way a server
handles <!--#include ... --> statements.
This is not how the "Web of Things" works,
but I don't care about that.
but this is how people arrange collections of
reference documents. This is highly significant
in Emergency Management where hardware and
connectivity can be disrupted by the event itself
... but you, your laptop and trusty thumb drive
survived. There are Portable Apps ...
(http://portableapps.com/), but your trusty thumb
drive might not have its favorite laptop around.
My proposal has nothing to do with survival in an
emergency, it's far more prosaic. If I have all the
pages and files for a website on a thumb drive, then
any laptop will work because there will be some
browser on the laptop.
You can count on at least a working browser on a
working laptop, I think.
Me too.
So, if the browser supports the current standard,
and if the standard says when a browers is pointed
at a local file whose name ends in '.shtml' then
the browser should attempt to handle server side
includes in the same way a server does.
That said, the document collection should then be
XML ... because the style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not distract or
interfere with access.
Well, I don't want to step on any toes here, but
my impression is that XHTML is kinda' moribund and
that the latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).
And, it shouldn't matter: if the HTML standard were
to support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
Subject: Browser suggestion: local server
Date: Thursday, November 12, 2015, 11:08 AM
Guys,
I've been doing a lot of development using .shtml
and server side includes. Testing, however, is a
bit of a pain: I can't really test the includes
are working until I upload all the files to my
server.
It occurs to me it would be terrific if this
* If a browser (user agent) points to a local file,
and if the filename ends in '.shtml', then the
browser should endeavor to process any 'include'
statements in the file in the same way a server
would
This would also be nice because I can put a whole
website on a thumb drive then display it to a meeting
or class without having to actually connect to the
internet! Makes the site much more portable.
Is that reasonable? Desirable? How do I go about
proposing such behavior?
Kind regards,
-Steve Comstock
303-355-2752
Steve Comstock
2015-11-29 13:18:48 UTC
Permalink
Well, lots of suggestions on how to do this without
changing the standards for UAs. And I researched
and even tested several of these.

But they miss the mark, in my opinion.


The advantage in requiring UAs to handle SSI include
statements if it is pointing to a local file is that
I can have a portable site with no need to connect
to the internet and no need to install supplemental
software or write code that might not work on some
other platform.

If UAs did this, I can show my portable website no
matter if the browser is IE, Chrome, Firefox, Opera,
Safari, ... and no matter if the platform is Windows,
Mac, Linux, ...

So this enhancement would support: useability, flexibilty,
portability.


In the absence, I'll continue testing by loading up all
my files to the server and then pointing to the starting
page on the server.


Thanks anyway folks.


-Steve
Post by Steve Comstock
Hello Steve,
There are excellent, not IT motivated reasons for
using a local server, or better said locating an
(actual) interface at 127.0.0.1.
Well, I'm aware of that interface, but it is not
at all what I'm talking about; my suggestion needs
code in the browser to simulate the way a server
handles <!--#include ... --> statements.
This is not how the "Web of Things" works,
but I don't care about that.
but this is how people arrange collections of
reference documents. This is highly significant
in Emergency Management where hardware and
connectivity can be disrupted by the event itself
... but you, your laptop and trusty thumb drive
survived. There are Portable Apps ...
(http://portableapps.com/), but your trusty thumb
drive might not have its favorite laptop around.
My proposal has nothing to do with survival in an
emergency, it's far more prosaic. If I have all the
pages and files for a website on a thumb drive, then
any laptop will work because there will be some
browser on the laptop.
You can count on at least a working browser on a
working laptop, I think.
Me too.
So, if the browser supports the current standard,
and if the standard says when a browers is pointed
at a local file whose name ends in '.shtml' then
the browser should attempt to handle server side
includes in the same way a server does.
That said, the document collection should then be
XML ... because the style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not distract or
interfere with access.
Well, I don't want to step on any toes here, but
my impression is that XHTML is kinda' moribund and
that the latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).
And, it shouldn't matter: if the HTML standard were
to support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
Subject: Browser suggestion: local server
Date: Thursday, November 12, 2015, 11:08 AM
Guys,
I've been doing a lot of development using .shtml
and server side includes. Testing, however, is a
bit of a pain: I can't really test the includes
are working until I upload all the files to my
server.
It occurs to me it would be terrific if this
* If a browser (user agent) points to a local file,
and if the filename ends in '.shtml', then the
browser should endeavor to process any 'include'
statements in the file in the same way a server
would
This would also be nice because I can put a whole
website on a thumb drive then display it to a meeting
or class without having to actually connect to the
internet! Makes the site much more portable.
Is that reasonable? Desirable? How do I go about
proposing such behavior?
Kind regards,
-Steve Comstock
303-355-2752
Gannon Dick
2015-11-30 18:37:15 UTC
Permalink
Steve,

What you need is a WAMP (Windows) or LAMP (*nix) Stack. These include a Server, Scripting and a Data Base integrated to the Local Host at IP 127.0.0.1. They come in executable zip files.

Apache2 (can) will process Server Side Includes (*.shtml) by itself as an alternative to the Scripting (Perl, PHP, etc.). This is a nice "feature" since using the script engine has a cost. The scripting is much more versatile than Server Side includes, and the reason UA vendors don't handle these is because one quickly needs more sophisticated tools. Server Side includes are not used much because the code cannot be reused. Of course it can be copied, and even inserted automatically, but write once, run often is not all that is meant by the term.

--Gannon
--------------------------------------------
On Sun, 11/29/15, Steve Comstock <***@trainersfriend.com> wrote:

Subject: Re: Browser suggestion: local server
To: "Gannon Dick" <***@yahoo.com>, "Ian Hickson" <***@hixie.ch>, public-html-***@w3.org, ***@opera.com, ***@google.com, ***@inter-locale.com, team-***@w3.org, "Ian Jacobs" <***@w3.org>, "Ulrik Dobashi Hansen" <***@808.dk>, "Bert Bos" <***@w3.org>
Date: Sunday, November 29, 2015, 7:18 AM

Well, lots of suggestions
on how to do this without
changing the
standards for UAs. And I researched
and even
tested several of these.

But they miss the mark, in my opinion.


The advantage
in requiring UAs to handle SSI include
statements if it is pointing to a local file is
that
I can have a portable site with no need
to connect
to the internet and no need to
install supplemental
software or write code
that might not work on some
other
platform.

If UAs did this,
I can show my portable website no
matter if
the browser is IE, Chrome, Firefox, Opera,
Safari, ... and no matter if the platform is
Windows,
Mac, Linux, ...

So this enhancement would support: useability,
flexibilty,
portability.


In the absence, I'll
continue testing by loading up all
my files
to the server and then pointing to the starting
page on the server.


Thanks anyway folks.


-Steve



On
On 11/12/2015 11:36 AM, Gannon Dick
Hello Steve,
There are
excellent, not IT motivated reasons for
using a local server, or better said
locating an
(actual) interface at
127.0.0.1.
Well,
I'm aware of that interface, but it is not
at all what I'm talking about; my
suggestion needs
code in the browser to
simulate the way a server
handles
<!--#include ... --> statements.
This is not how the "Web of
Things" works,
but I don't care about that.
but this is how
people arrange collections of
reference documents.  This is highly significant
in Emergency Management where hardware
and
connectivity can be disrupted
by the event itself
... but you,
your laptop and trusty thumb drive
survived.  There are Portable Apps ...
(http://portableapps.com/), but your
trusty thumb
drive might not have
its favorite laptop around.
My proposal has nothing to do with
survival in an
emergency, it's far
more prosaic. If I have all the
pages
and files for a website on a thumb drive, then
any laptop will work because there will be
some
browser on the laptop.
You can count on at least a working
browser on a
working laptop, I
think.
Me too.
So, if the browser
supports the current standard,
and if
the standard says when a browers is pointed
at a local file whose name ends in
'.shtml' then
the browser
should attempt to handle server side
includes in the same way a server does.
That said, the document collection
should then be
XML ... because the
style, spin, persuasion,
salesmanship whatever you want to call it that
XHTML inherits from HTML should not
distract or
interfere with
access.
Well, I don't want to step on any toes
here, but
my impression is that XHTML
is kinda' moribund and
that the
latest HTML version is actually gaining
steam. Of course, I could be totally wrong (it
wouldn't be the first time).
And, it shouldn't
matter: if the HTML standard were
to
support my suggestions, presumably that would
also be supported in XHTML.
c.f.
http://Stratml.us/
http://www.rustprivacy.org/2015/stratml/cap_sml/vfsroot/
--Gannon
--------------------------------------------
On Thu, 11/12/15, Steve Comstock
   Subject: Browser
suggestion: local server
   To: "Ian
team-***@w3.org,
"Ian Jacobs"
"Ulrik Dobashi Hansen"
   Date: Thursday, November
12, 2015, 11:08 AM
   Guys,
   I've been doing a lot
of development using .shtml
   and server side includes.
Testing, however, is a
   bit of a pain: I
can't really test the includes
   are working until I
upload all the files to my
   server.
   It occurs to me it would
be terrific if this
   could be part of some
   * If a browser (user
agent) points to a local file,
 
   and if the filename ends in '.shtml',
then the
     browser
should endeavor to process any 'include'
     statements in the file
in the same way a server
 
   would
   This would also be nice
because I can put a whole
   website on a thumb drive
then display it to a meeting
   or class without having
to actually connect to the
   internet! Makes the site
much more portable.
   Is that reasonable?
Desirable? How do I go about
   proposing such
behavior?
   Kind regards,
   -Steve Comstock
   303-355-2752
Loading...