gnuplot development

Hello

--- Tatsuro MATSUOKA wrote:
> In my opinion, the flag '-Wl,--subsystem,windows -mwindows' in wx-config for wxMSW is to be
> removed. 
> I have posted my opinion to the wx-develop post.  However at the moment, I cannot say that my
> opinion
> will be accepted. Therefore I propose the patch attached.

My proposal was rejected by the wxWidgets developper.

http://groups.google.co.jp/group/wx-dev/browse_thread/thread/65a3af28e9a21bca# 

However I can get useful information for disabling the the flag '-Wl,--subsystem,windows -mwindows': 
override it by a subsequent -Wl,--subsequent,console.


I will reconsider the patch.

Regards

Tatsuro

--------------------------------------
Yahoo! JAPAN - Internet Security for teenagers and parents.
http://pr.mail.yahoo.co.jp/security/

Hello 

After discussion with Vadim Zeitlin, who is one of the developer of wxWidget, in the following address

http://groups.google.co.jp/group/wx-dev/browse_thread/thread/65a3af28e9a21bca/89264cf1c66e16f3#89264cf1c66e16f3

I consider two different patch. First one, 'makefile.mgw.1.patch' is the same as that I have proposed
in the previous post.
Second one is add overrideng option to when PIPES flag is on: 'makefile.mgw.2.patch'  

First patch:
-	WX_LIBS = $(shell wx-config --libs)
+	WX_LIBS = $(shell wx-config --libs | sed -e 's/ -Wl,--subsystem,windows -mwindows//')
This removes  '-Wl,--subsystem,windows -mwindows' flag

Second patch:
 	WX_LIBS = $(shell wx-config --libs)
+	ifdef PIPES
+		WX_LIBS += -Wl,--subsystem,console
+	endif
add overrideing option for gnuplot.exe and wgnuplot_pipes.exe

I cannot figure out which is better.  Please discuss this matter and produce better patch.
Of course, more smart ideas are highly welcome.

Regards

Tatsuro
--- Tatsuro MATSUOKA  wrote:

> My proposal was rejected by the wxWidgets developper.
> 
> http://groups.google.co.jp/group/wx-dev/browse_thread/thread/65a3af28e9a21bca# 
> 
> However I can get useful information for disabling the the flag '-Wl,--subsystem,windows
> -mwindows': 
> override it by a subsequent -Wl,--subsequent,console.
> 
> 
> I will reconsider the patch.


--------------------------------------
Yahoo! JAPAN - Internet Security for teenagers and parents.
http://pr.mail.yahoo.co.jp/security/

Tatsuro MATSUOKA wrote:
> Hello
> 
> --- Tatsuro MATSUOKA wrote:
>> In my opinion, the flag '-Wl,--subsystem,windows -mwindows' in wx-config for wxMSW is to be
>> removed. 
>> I have posted my opinion to the wx-develop post.  However at the moment, I cannot say that my
>> opinion
>> will be accepted. Therefore I propose the patch attached.
> 
> My proposal was rejected by the wxWidgets developper.
> 
> http://groups.google.co.jp/group/wx-dev/browse_thread/thread/65a3af28e9a21bca# 
> 

The flag "-mwindows" has nothing to do with whether an application is 
using GDI libraries or not. It specifies the type of startup code and 
entry point of the application. So I cannot follow the wxwidgets 
developers argumentation. But it's their decision to make.

> However I can get useful information for disabling the the flag '-Wl,--subsystem,windows -mwindows': 
> override it by a subsequent -Wl,--subsequent,console.

I am sorry but this is IMO a very bad solution to take.
This depends on the order of arguments on the command line, is not 
portable and depends no binutils' actual interpretation of such a 
double-specfication.
And it is simply wrong. An application is either a GUI-subsystem 
application or a console-subsystem application. If it's not a GUI 
application, one shouldn't say so in the first place and not change 
one's mind halfway through. It's not a clean method, IMO

benjamin

Tatsuro MATSUOKA wrote:
> Hello 
> 
> After discussion with Vadim Zeitlin, who is one of the developer of wxWidget, in the following address
> 
> http://groups.google.co.jp/group/wx-dev/browse_thread/thread/65a3af28e9a21bca/89264cf1c66e16f3#89264cf1c66e16f3
> 
> I consider two different patch. First one, 'makefile.mgw.1.patch' is the same as that I have proposed
> in the previous post.
> Second one is add overrideng option to when PIPES flag is on: 'makefile.mgw.2.patch'  
> 
> First patch:
> -	WX_LIBS = $(shell wx-config --libs)
> +	WX_LIBS = $(shell wx-config --libs | sed -e 's/ -Wl,--subsystem,windows -mwindows//')
> This removes  '-Wl,--subsystem,windows -mwindows' flag

I'd propose a more robust version like
sed -e "s+-Wl,--subsystem,windows++g" -e "s+-mwindows++g"

Removing should not be dependent on spaces and the actual order of the 
erroneus commands.

> Second patch:
>  	WX_LIBS = $(shell wx-config --libs)
> +	ifdef PIPES
> +		WX_LIBS += -Wl,--subsystem,console
> +	endif
> add overrideing option for gnuplot.exe and wgnuplot_pipes.exe
> 
> I cannot figure out which is better.  Please discuss this matter and produce better patch.
> Of course, more smart ideas are highly welcome.

The former. The latter looks wrong and is highly unportable as it 
depends on the order of the flags on the command line and their 
interpreation by binutils.

benjamin

Hello Benjamin


--- Benjamin Lindner  wrote:
The flag "-mwindows" has nothing to do with whether an application is 
using GDI libraries or not. It specifies the type of startup code and 
entry point of the application. So I cannot follow the wxwidgets 
developers argumentation. But it's their decision to make.

I agree. But it seem to difficult to make their decision to change.

> However I can get useful information for disabling the the flag '-Wl,--subsystem,windows -mwindows':

> override it by a subsequent -Wl,--subsequent,console.

I am sorry but this is IMO a very bad solution to take.
This depends on the order of arguments on the command line, is not 
portable and depends no binutils' actual interpretation of such a 
double-specfication.
And it is simply wrong. An application is either a GUI-subsystem 
application or a console-subsystem application. If it's not a GUI 
application, one shouldn't say so in the first place and not change 
one's mind halfway through. It's not a clean method, IMO

OK. I under stand your statement.  Surely it should be clean.

> Tatsuro MATSUOKA wrote:
> > Hello 
> > 
> > After discussion with Vadim Zeitlin, who is one of the developer of wxWidget, in the following
> address
> > 
> >
>
http://groups.google.co.jp/group/wx-dev/browse_thread/thread/65a3af28e9a21bca/89264cf1c66e16f3#89264cf1c66e16f3
> > 
> > I consider two different patch. First one, 'makefile.mgw.1.patch' is the same as that I have
> proposed
> > in the previous post.
> > Second one is add overrideng option to when PIPES flag is on: 'makefile.mgw.2.patch'  
> > 
> > First patch:
> > -	WX_LIBS = $(shell wx-config --libs)
> > +	WX_LIBS = $(shell wx-config --libs | sed -e 's/ -Wl,--subsystem,windows -mwindows//')
> > This removes  '-Wl,--subsystem,windows -mwindows' flag
> 
> I'd propose a more robust version like
> sed -e "s+-Wl,--subsystem,windows++g" -e "s+-mwindows++g"


> Removing should not be dependent on spaces and the actual order of the 
> erroneus commands.
> 
> > Second patch:
> >  	WX_LIBS = $(shell wx-config --libs)
> > +	ifdef PIPES
> > +		WX_LIBS += -Wl,--subsystem,console
> > +	endif
> > add overrideing option for gnuplot.exe and wgnuplot_pipes.exe
> > 
> > I cannot figure out which is better.  Please discuss this matter and produce better patch.
> > Of course, more smart ideas are highly welcome.
> 
> The former. The latter looks wrong and is highly unportable as it 
> depends on the order of the flags on the command line and their 
> interpreation by binutils.

I submit a patch considering Benjamin suggestion.

Benjamin. Thank you for your comprehensive and extensive suggestions.

Regards

Tatsuro


--------------------------------------
Yahoo! JAPAN - Internet Security for teenagers and parents.
http://pr.mail.yahoo.co.jp/security/

Clark Gaylord wrote:
> This was going to be just a quick note about our 
> "pre-production/production" web content design, but I started rummaging 
> around through the archives and thought I'd go ahead a document the 
> history for reference. Sorry for the lack of brevity. --ckg
> 
> The design for managing the web content is that the sourceforge site is 
> considered "pre-production" which then gets to the "production" 
> gnuplot.info site once/day. It was originally intended that the 
> sourceforge project would be the primary site for gnuplot *development* 
> (which I established 1 February 2000; HBB joined in April 2000). The 
> idea was that we could work with the web site effectively on a 
> "pre-prod" server (now this is sourceforge), which then is occasionally 
> in a state of flux, and automatically this will get rolled into 
> production once per day. There is a very small risk that the state of 
> flux would happen at the moment of sync, and this is obviously a pretty 
> cheesy design. However, it was a very easy way to give us a "poor man's 
> CMS", and the content has never had a "must be updated in fifteen 
> minutes because CNN is running a story that is going to give us $4M 
> RIGHT NOW!" nature. More importantly, the only thing we must do to 
> perform a major rewrite of the web site is to break the mirror while we 
> work on it (or if we wanted to have a real "development" web site 
> distinct from the public facing web site).
> 
> I think we may have had http://www.gnuplot.vt.edu canonical in February 1999 
> (ftp was started in 1998), prior to the gnuplot.org domain. I picked up 
> the content from Alex Woo in late 1998, and I think Lars ran the old 
> cmpc1 ftp server. That was about when we started having this 
> staging/production environment. I think Lars Hecking had the primary 
> content shortly after we set up the gnuplot.org domain, and he operated 
> until 2004 when we migrated web content to sourceforge. John Turner had 
> the gnuplot.org domain between 1998 and 2001, when we lost it and I 
> registered gnuplot.info, both of which have always been hosted at 
> Virginia Tech. gnuplot.info came on the air 3 Jan 2002. In 2000, I 
> created the sourceforge space to support gnuplot *development* (that's 
> why the project is named "gnuplot development") and give us a CVS home. 
> Mailing lists were migrated from Dartmouth to sourceforge in 2003. We 
> started using sourceforge as our pre-production web site in late 2004, 
> when Lars had to move off of his server at ucc.ie.
> 
> The suitability of the various options is arguable. Sourceforge tends to 
> have a more "high availability" server environment (we "think" ....) 
> Though there have certainly been plenty of issues with sourceforge over 
> the last decade, this is much more true of a single server that some guy 
> (i.e. me) takes care of (actually, the server has the support from a 
> "team" of sysadmins, currently at the Virginia Tech Transportation 
> Institute http://www.vtti.vt.edu/ ). VT is better connected to the 
> research and international communities due to its involvement in 
> Internet2 and National Lambda Rail, but this is from a strictly 
> "performance" perspective not necessarily based on path redundancy, etc. 
> So, from that perspective you could argue either way. It really is about 
> having the ability for developers to work on the site without having to 
> worry so much about breaking it temporarily.
> 
> Of course if we were to lose the VT hosting option someday I would not 
> mind using sourceforge if necessary, though I would probably first look 
> for something to replicate the existing design.
> 
> Anyway, I hope that explains why we have it set up this way and it makes 
> sense.
> 
> --ckg
> 

Perhaps due to longevity as a gnuplot site or general content across VT 
domain this mirror got a google PR of 7 the last time I looked.

I've rarely seen a PR that high.

That will have a strong bearing on the google results for any search 
relating to the content which could be a simple as someone looking for a 
plotting utility.

That argument alone would seem to be a good reason for keeping it (and 
getting the content sorted before the G notices all the content it has 
cached no longer exists, returns 404 and decided to dump the PR7).

Any site with a PR7 deserves respect .

greetz.

plotter@... wrote:

> 
> Perhaps due to longevity as a gnuplot site or general content across VT 
> domain this mirror got a google PR of 7 the last time I looked.
> 
> I've rarely seen a PR that high.
> 
> That will have a strong bearing on the google results for any search 
> relating to the content which could be a simple as someone looking for a 
> plotting utility.
> 
> That argument alone would seem to be a good reason for keeping it (and 
> getting the content sorted before the G notices all the content it has 
> cached no longer exists, returns 404 and decided to dump the PR7).
> 
> Any site with a PR7 deserves respect .
> 
> greetz.
> 

He he. Too late. It's dropped to 20th today and google shows it's title 
as "It Works!"

It still gets pr7 , so hopefully next time G bots it will get back up to 
date.

Anyway, the site has some serious kudos with Google which is an asset.

/regards/

> Given that today there is SourceForge, the need to host community projects 
> on individual's accounts seems less pressing. As I said before, it would 
> make sense to me to have gnuplot.sourceforge.net be the "canonical" site 
> (and have gnuplot.info point to it), and to make Clark's site at VT a 
> clearly designated mirror.
> 
> Votes? Opinions? Discussion?

- I think http://www.gnuplot.info looks more "professional" than gnuplot.sf.net.

- The following link
        http://www.gnuplot.info/demo
  (notice the missing trailing slash) goes to:
	http://azog.bioinformatics.vt.edu:49280/demo/
  but not
	http://gnuplot.sourceforge.net/demo

- I think the VT site should have index.html with an immediate
  redirection to http://www.gnuplot.info.

- It don't think the mirroring works correctly -- for example,
	http://www.gnuplot.info/download.html
  still announces 4.2.5 instead of 4.2.6:
	http://gnuplot.sourceforge.net/download.html
  These problems are happening very often!

- What are proposals following from this discussion?
  A. Should http://www.gnuplot.info be an immediate redirect to sf.net?
     Would  http://www.gnuplot.info/demo/  still work correctly in that case?

  B. Could the mirroring of http://www.gnuplot.info be finally fixed so that it
     really works in 24 h intervals (or faster)?
     Would a move to another server in the world help?

  Any other possibility?

---
PM

Petr Mikulik wrote:
>> Given that today there is SourceForge, the need to host community projects 
>> on individual's accounts seems less pressing. As I said before, it would 
>> make sense to me to have gnuplot.sourceforge.net be the "canonical" site 
>> (and have gnuplot.info point to it), and to make Clark's site at VT a 
>> clearly designated mirror.
>>
>> Votes? Opinions? Discussion?
> 
> - I think http://www.gnuplot.info looks more "professional" than gnuplot.sf.net.
> 
> - The following link
>         http://www.gnuplot.info/demo
>   (notice the missing trailing slash) goes to:
> 	http://azog.bioinformatics.vt.edu:49280/demo/
>   but not
> 	http://gnuplot.sourceforge.net/demo
> 
> - I think the VT site should have index.html with an immediate
>   redirection to http://www.gnuplot.info.

That would blow out any advantage of having a well ranked site with 
google. Once it gets a whiff of the redirection it will probably remove 
.info from it's listing and only refer to the target.


> 
> - It don't think the mirroring works correctly -- for example,
> 	http://www.gnuplot.info/download.html
>   still announces 4.2.5 instead of 4.2.6:
> 	http://gnuplot.sourceforge.net/download.html
>   These problems are happening very often!

That is presumably something that can easily be corrected. Mirroring is 
not a new idea.

Clearly all efforts should be made to ensure an accurate mirroring.

regards.


> 
> - What are proposals following from this discussion?
>   A. Should http://www.gnuplot.info be an immediate redirect to sf.net?
>      Would  http://www.gnuplot.info/demo/  still work correctly in that case?
> 
>   B. Could the mirroring of http://www.gnuplot.info be finally fixed so that it
>      really works in 24 h intervals (or faster)?
>      Would a move to another server in the world help?
> 
>   Any other possibility?
> 
> ---
> PM
> 
> 
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry® Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay 
> ahead of the curve. Join us from November 9-12, 2009. Register now!
> http://p.sf.net/sfu/devconf
> _______________________________________________
> gnuplot-beta mailing list
> gnuplot-beta@...
> https://lists.sourceforge.net/lists/listinfo/gnuplot-beta
> 
>

I don't think that we need to worry too much about
Google page rank - except maybe for vanity reasons.

If somebody is searching for "gnuplot", this person 
will find either gnuplot.info or gnuplot.sf.net, among
the top 10 entries, regardless. It's a VERY specialized
search term.

Best,

		Ph.



On Tuesday 22 September 2009 07:39:01 am plotter@... wrote:
> Petr Mikulik wrote:
> >> Given that today there is SourceForge, the need to host community
> >> projects on individual's accounts seems less pressing. As I said before,
> >> it would make sense to me to have gnuplot.sourceforge.net be the
> >> "canonical" site (and have gnuplot.info point to it), and to make
> >> Clark's site at VT a clearly designated mirror.
> >>
> >> Votes? Opinions? Discussion?
> >
> > - I think http://www.gnuplot.info looks more "professional" than gnuplot.sf.net.
> >
> > - The following link
> >         http://www.gnuplot.info/demo
> >   (notice the missing trailing slash) goes to:
> > 	http://azog.bioinformatics.vt.edu:49280/demo/
> >   but not
> > 	http://gnuplot.sourceforge.net/demo
> >
> > - I think the VT site should have index.html with an immediate
> >   redirection to http://www.gnuplot.info.
>
> That would blow out any advantage of having a well ranked site with
> google. Once it gets a whiff of the redirection it will probably remove
> .info from it's listing and only refer to the target.
>
> > - It don't think the mirroring works correctly -- for example,
> > 	http://www.gnuplot.info/download.html
> >   still announces 4.2.5 instead of 4.2.6:
> > 	http://gnuplot.sourceforge.net/download.html
> >   These problems are happening very often!
>
> That is presumably something that can easily be corrected. Mirroring is
> not a new idea.
>
> Clearly all efforts should be made to ensure an accurate mirroring.
>
> regards.
>
> > - What are proposals following from this discussion?
> >   A. Should http://www.gnuplot.info be an immediate redirect to sf.net?
> >      Would  http://www.gnuplot.info/demo/  still work correctly in that case?
> >
> >   B. Could the mirroring of http://www.gnuplot.info be finally fixed so that it
> >      really works in 24 h intervals (or faster)?
> >      Would a move to another server in the world help?
> >
> >   Any other possibility?
> >
> > ---
> > PM
> >
> >
> > -------------------------------------------------------------------------
> >----- Come build with us! The BlackBerry® Developer Conference in SF,
> > CA is the only developer event you need to attend this year. Jumpstart
> > your developing skills, take BlackBerry mobile applications to market and
> > stay ahead of the curve. Join us from November 9-12, 2009. Register
> > now! http://p.sf.net/sfu/devconf
> > _______________________________________________
> > gnuplot-beta mailing list
> > gnuplot-beta@...
> > https://lists.sourceforge.net/lists/listinfo/gnuplot-beta

Like others, I thank Clark Gaylord for his history of the gnuplot
domain issue, and more substantively, for all the work he has done
on behalf of gnuplot.

That said, I think sourceforge has now matured to the point where
it no longer makes sense to think of gnuplot.sf.net as a sort of
"pre-production" site.  People will expect that if a project has a
presence on sf, that is the project's canonical site.

Some projects (I think) have an sf presence for downloadable
files, version control, etc., but choose to maintain a separate
"shop window" website on the pattern foo.org.  In that case the sf
website, foo.sf.net (as opposed to the "project page"), would be a
redirect to foo.org.  That would be a possible pattern for gnuplot
to adopt (with gnuplot.info == foo.org).

However, I agree with Philipp Janert that the present situation --
where the supposedly canonical site is in fact a mirror of the
truly primary site, with some synchronization problems to boot --
is not very satisfactory.

Allin Cottrell

> > - I think the VT site should have index.html with an immediate
> >   redirection to http://www.gnuplot.info.
> 
> That would blow out any advantage of having a well ranked site with 
> google. Once it gets a whiff of the redirection it will probably remove 
> .info from it's listing and only refer to the target.

I don't see any advantage of having "www.gnuplot.vt.edu" on top of google 
nowadays. It's no longer the primary site as it was 10 years ago.

> > 
> > - It don't think the mirroring works correctly -- for example,
> > 	http://www.gnuplot.info/download.html
> >   still announces 4.2.5 instead of 4.2.6:
> > 	http://gnuplot.sourceforge.net/download.html
> >   These problems are happening very often!
> 
> That is presumably something that can easily be corrected. Mirroring is 
> not a new idea.

It seems that it cannot be easily corrected -- these mirroring problems 
occur for several years.

> Like others, I thank Clark Gaylord for his history of the gnuplot
> domain issue, and more substantively, for all the work he has done
> on behalf of gnuplot.

I agree.

Unfortunately, the mirroring does not work. Therefore:

> > - What are proposals following from this discussion?
> >   A. Should http://www.gnuplot.info be an immediate redirect to sf.net?
> >      Would  http://www.gnuplot.info/demo/  still work correctly in that case?
> > 
> >   B. Could the mirroring of http://www.gnuplot.info be finally fixed so that it
> >      really works in 24 h intervals (or faster)?
> >      Would a move to another server in the world help?


It seems that "www.gnuplot.info" should not be a redirect to gnuplot.sf.net,
but an independent site. Then there are two possibilities:

A. gnuplot.sf.net stays the main editable site, the new gnuplot.info its
   mirror.
B. new gnuplot.info sites is the main editable site, gnuplot.sf.net is its
   redict (I don't think it makes sense to have a mirror)
C. let DNS of gnuplot.info point to gnuplot.sf.net (is this feasible?)

Requirements:
A. - many people have access to gnuplot.sf.net (organized by sf accounts), 
     less people have account on new gnuplot.info
   - mirroring works reliably and fast
     (max 12 h delay or sync can be launched on demand by a special e-mail, 
     touching a special file, or whatever the current technology allows)

B. - many people have access to this account

What are your votes?

---
PM

Philipp K. Janert wrote:
 > I don't think that we need to worry too much about
 > Google page rank - except maybe for vanity reasons.
 >
 > If somebody is searching for "gnuplot", this person will find either 
gnuplot.info or gnuplot.sf.net, among
 > the top 10 entries, regardless. It's a VERY specialized
 > search term.
 >
 > Best,
 >
 >         Ph.
 >
 >

Hi Philipp,

you misread what I posted. Sure, anyone who searches for "gnuplot" will 
find it.

My point is about people finding gnuplot in the first place when needing 
plotting tool. When I needed such a tool a few years back I looked and 
did not find anything suitable. I did not find gnuplot.

This brings up another point: that gnuplot site content is not coming up 
in searches anyway. I was looking for such a tool for a long time and 
only found gnuplot by someone recommending it to me on a forum somewhere.

I certainly don't see PR as a "vanity" commodity but a PR7 is an asset 
that many would pay thousands to have. It seems foolish to dismiss it. 
It may make more sense to exploit it.

regards.



 >
 > On Tuesday 22 September 2009 07:39:01 am plotter@... wrote:
 >> Petr Mikulik wrote:
 >>>> Given that today there is SourceForge, the need to host community
 >>>> projects on individual's accounts seems less pressing. As I said 
before,
 >>>> it would make sense to me to have gnuplot.sourceforge.net be the
 >>>> "canonical" site (and have gnuplot.info point to it), and to make
 >>>> Clark's site at VT a clearly designated mirror.
 >>>>
 >>>> Votes? Opinions? Discussion?
 >>> - I think http://www.gnuplot.info looks more "professional" than 
gnuplot.sf.net.
 >>>
 >>> - The following link
 >>>         http://www.gnuplot.info/demo
 >>>   (notice the missing trailing slash) goes to:
 >>>     http://azog.bioinformatics.vt.edu:49280/demo/
 >>>   but not
 >>>     http://gnuplot.sourceforge.net/demo
 >>>
 >>> - I think the VT site should have index.html with an immediate
 >>>   redirection to http://www.gnuplot.info.
 >> That would blow out any advantage of having a well ranked site with
 >> google. Once it gets a whiff of the redirection it will probably remove
 >> .info from it's listing and only refer to the target.
 >>
 >>> - It don't think the mirroring works correctly -- for example,
 >>>     http://www.gnuplot.info/download.html
 >>>   still announces 4.2.5 instead of 4.2.6:
 >>>     http://gnuplot.sourceforge.net/download.html
 >>>   These problems are happening very often!
 >> That is presumably something that can easily be corrected. Mirroring is
 >> not a new idea.
 >>
 >> Clearly all efforts should be made to ensure an accurate mirroring.
 >>
 >> regards.
 >>
 >>> - What are proposals following from this discussion?
 >>>   A. Should http://www.gnuplot.info be an immediate redirect to sf.net?
 >>>      Would  http://www.gnuplot.info/demo/  still work correctly in that case?
 >>>
 >>>   B. Could the mirroring of http://www.gnuplot.info be finally fixed so 
that it
 >>>      really works in 24 h intervals (or faster)?
 >>>      Would a move to another server in the world help?
 >>>
 >>>   Any other possibility?
 >>>
 >>> ---
 >>> PM
 >>>
 >>>
 >

I have listened to this discussion on this point, and here's my
take on it.  As is see it, some people are thinking in terms of
"what makes most logical sense" (with regard to domain names)
while others are thinking along the lines of, "certain sites have
high google rankings" (regardless of what makes logical sense?)
and we should aim to capitalize on that.

I certainly do not claim to have any privileged information on
what makes for a high google ranking, but I do agree that "we"
(advocates for gnuplot) want to "come up" on google near the top
when people are searching (without necessarily ever having heard
of gnuplot) for free/open-source graphing/plotting programs.

So here's a very basic survey, as of 2009-09-24.  (I'm assuming
that if you don't come up in the first 2 pages of google hits for
the search string the user has entered then you're more or less
invisible, though of course some people are willing to persevere
further than that.)

<survey>

(1) Google search: "free plotting program"
 http://www.gnuplot.info = #2
gnuplot sf page not on first 2 google pages

(2) Google search: "open source plotting program"
 http://www.gnuplot.info = #3
gnuplot sf page not on first 2 google pages

(3) Google search: "open-source plotting program"
Same as (2)

(4) Google search: "free graphing program"
gnuplot (in any form) not on first 2 google pages

(5) Google search: "open source graphing program"
 http://www.gnuplot.info = #2 on second google page
gnuplot sf page not on first 2 google pages

(5) Google search: "open-source graphing program"
 http://www.gnuplot.info = #3 on second google page
gnuplot sf page not on first 2 google pages

</survey>

One could vary the search strings to taste, but it seems to me two
main points emerge:

1. Clearly, http://www.gnuplot.info has much more google-recognition than
gnuplot.sf.net.

2. Gnuplot (even in the form http://www.gnuplot.info) is not showing up
as it "should" for some relevant google searches.

What follows?

(1) I would say, if we're interested in proselytizing (though, as
we all know, it makes none of us any money!), we should amend the
gnuplot home page (on whatever site[s] it subsists) to explicitly
mention the facts that gnuplot is free (as in free beer, though
not exactly in the Richard Stallman sense) and open source. And we
should also include the term "graphing".

(2) This is more speculative, but IMO google page rank is quite
dynamic, not necessarily strongly bound by history.  That is, I
think that _if_ the gnuplot developers (and I don't count myself
as one, I'm just an interested spectator) were to choose to make
gnuplot.sf.net the canonical site, then the ranking would fairly
quickly adjust to favour gnuplot.sf.net.

Allin Cottrell

Here's one further thought, after which I'll shut up (at least for
a while!).

IF you think there's some validity in the analysis I have offered
of the google ranking issue AND you're sympathetic to the points
Philipp Janert has made (as I am) BUT you doubt how easy and quick
it would be to substitute gnuplot.sf.net for http://www.gnuplot.info in
the google rankings, THEN you might be willing to consider the
option I mooted a few postings back, but which I'll now state a
bit more explicitly:

* Scrap the web page at gnuplot.sf.net; make this a redirect to
 http://www.gnuplot.info (so Ethan has only one website to maintain).

* Aside from the "shop window" website, everything else is at
sf.net; so, for example, any download links on the page at
 http://www.gnuplot.info would point to the appropriate place on sf.net.

-- 
Allin Cottrell
Department of Economics
Wake Forest University

Thanks to Allin for pulling some information 
together. I think this is really useful. Some
comments below (after a snip).


[snip]
>
> 1. Clearly, http://www.gnuplot.info has much more google-recognition than
> gnuplot.sf.net.
>
> 2. Gnuplot (even in the form http://www.gnuplot.info) is not showing up
> as it "should" for some relevant google searches.
>
> What follows?
>
> (1) I would say, if we're interested in proselytizing (though, as
> we all know, it makes none of us any money!), we should amend the
> gnuplot home page (on whatever site[s] it subsists) to explicitly
> mention the facts that gnuplot is free (as in free beer, though
> not exactly in the Richard Stallman sense) and open source. And we
> should also include the term "graphing".

And: "data analysis" and "visualization" and ...?

>
> (2) This is more speculative, but IMO google page rank is quite
> dynamic, not necessarily strongly bound by history.  That is, I
> think that _if_ the gnuplot developers (and I don't count myself
> as one, I'm just an interested spectator) were to choose to make
> gnuplot.sf.net the canonical site, then the ranking would fairly
> quickly adjust to favour gnuplot.sf.net.

I think this is true. One thing that seems to affect
Google is the way a page changes. I have seen
static pages drop from way on high to nowhere at
all. On the other hand, I have noticed that pages
that have changed tend to come up higher (and
Google tends to pick up changes w/in a few days).

On Thursday 24 September 2009 07:05:10 pm Allin Cottrell wrote:
> Here's one further thought, after which I'll shut up (at least for
> a while!).
>
> IF you think there's some validity in the analysis I have offered
> of the google ranking issue AND you're sympathetic to the points
> Philipp Janert has made (as I am) BUT you doubt how easy and quick
> it would be to substitute gnuplot.sf.net for http://www.gnuplot.info in
> the google rankings, THEN you might be willing to consider the
> option I mooted a few postings back, but which I'll now state a
> bit more explicitly:

I think by all means we should retain the gnuplot.info
domain name. If nothing else, it is much prettier than
gnuplot.sf.net. Also, it is more likely to be taken seriously.
(I am always a bit suspicious of projects which have only
their sf site, but don't even own their own domain.)

>
> * Scrap the web page at gnuplot.sf.net; make this a redirect to
> http://www.gnuplot.info (so Ethan has only one website to maintain).

The other way around: http://www.gnuplot.info should redirect
to sf.net or be an alias for it.

Also, I don't think Ethan maintains the website, the dev
team does. (Which is not to downgrade Ethan's contributions
in any way, but we all share responsibility for all aspects of
the project.)

Which brings me back to my original point: IFF the
gnuplot.info domain is the "normative" domain for
the entire gnuplot project (and not just an individual's
gnuplot fan site), then it should point to something that
the entire dev team has access to - in case we need to,
or in case the maintainer is not available.

>
> * Aside from the "shop window" website, everything else is at
> sf.net; so, for example, any download links on the page at
> http://www.gnuplot.info would point to the appropriate place on sf.net.

Allin Cottrell wrote:
> I have listened to this discussion on this point, and here's my
> take on it.  As is see it, some people are thinking in terms of
> "what makes most logical sense" (with regard to domain names)
> while others are thinking along the lines of, "certain sites have
> high google rankings" (regardless of what makes logical sense?)
> and we should aim to capitalize on that.
> 
> I certainly do not claim to have any privileged information on
> what makes for a high google ranking, but I do agree that "we"
> (advocates for gnuplot) want to "come up" on google near the top
> when people are searching (without necessarily ever having heard
> of gnuplot) for free/open-source graphing/plotting programs.
> 
> So here's a very basic survey, as of 2009-09-24.  (I'm assuming
> that if you don't come up in the first 2 pages of google hits for
> the search string the user has entered then you're more or less
> invisible, though of course some people are willing to persevere
> further than that.)
> 
> <survey>
> 
> (1) Google search: "free plotting program"
> http://www.gnuplot.info = #2
> gnuplot sf page not on first 2 google pages
> 
> (2) Google search: "open source plotting program"
> http://www.gnuplot.info = #3
> gnuplot sf page not on first 2 google pages
> 
> (3) Google search: "open-source plotting program"
> Same as (2)
> 
> (4) Google search: "free graphing program"
> gnuplot (in any form) not on first 2 google pages
> 
> (5) Google search: "open source graphing program"
> http://www.gnuplot.info = #2 on second google page
> gnuplot sf page not on first 2 google pages
> 
> (5) Google search: "open-source graphing program"
> http://www.gnuplot.info = #3 on second google page
> gnuplot sf page not on first 2 google pages
> 
> </survey>
> 
> One could vary the search strings to taste, but it seems to me two
> main points emerge:
> 
> 1. Clearly, http://www.gnuplot.info has much more google-recognition than
> gnuplot.sf.net.
> 
> 2. Gnuplot (even in the form http://www.gnuplot.info) is not showing up
> as it "should" for some relevant google searches.
> 
> What follows?
> 
> (1) I would say, if we're interested in proselytizing (though, as
> we all know, it makes none of us any money!), we should amend the
> gnuplot home page (on whatever site[s] it subsists) to explicitly
> mention the facts that gnuplot is free (as in free beer, though
> not exactly in the Richard Stallman sense) and open source. And we
> should also include the term "graphing".
> 
> (2) This is more speculative, but IMO google page rank is quite
> dynamic, not necessarily strongly bound by history.  That is, I
> think that _if_ the gnuplot developers (and I don't count myself
> as one, I'm just an interested spectator) were to choose to make
> gnuplot.sf.net the canonical site, then the ranking would fairly
> quickly adjust to favour gnuplot.sf.net.
> 
> Allin Cottrell
> 

Thanks for the survey.

A couple of points (I have done quite a bit of work on SE optimisation 
so hopefully these comments are more than speculative): google results 
are a bit like quicksand since they are always changing the algo to keep 
ahead of people cheating, however general placements usually don't move 
that much unless your are using some cheat that Google decides to penalise.

PR is quite an important factor since it is reckoned to be one thing 
your can't really cheat. Sites like gnuplot.info that have a long 
history and probably thousands of external links can build a high PR. If 
such a site is redirected, google will generally reference the target 
and not return the site containing the redirection. This will add some 
kudos to the target but is unlikely to give it an equally high PR.

Some work could be done on the content but the objective needs to be 
clearly defined in advance. One cannot optimise a page for too many 
keywords since increasing the importance of new words will necessarily 
decrease those currently pulling a good result.

It's easy to do more harm than good if you already have good placement.

Proceed with caution and make sure the objective is clearly defined 
before poking around.

regards.
Peter.

Hello
Fine!!  Thanks a lot.

--- Benjamin Lindner  wrote:
> The wxt link flags incorrectly include the setting "-mwindows", which causes gnuplot to be a
> gui-subsystem application, no longer a console-subsystem application. Thus stdin/stdout will not
> work.
> You have to modify the flags returned by wx-config.
> 
> benjamin

Tatsuro

--------------------------------------
Thanks 10 years!  Yahoo! Shopping and Yahoo! Auctions
http://pr.mail.yahoo.co.jp/ec10years/