I just noticed that the developer docs on puredata.org were moved to the "developer" section. Please put them back where they were, that is where they belong. The "developer" section is not for documentation, the documentation section is. That has been long since established.
Also, there are many, many links out there that point to the old location for the developer docs.
Lastly, please consult this list before making such drastic changes to the website. I've put a lot of work into that section, so you could have at least asked.
.hc
------------------------------------------------------------------------
¡El pueblo unido jamás será vencido!
hi
Hans-Christoph Steiner wrote:
I just noticed that the developer docs on puredata.org were moved to the "developer" section. Please put them back where they were, that is where they belong. The "developer" section is not for documentation, the documentation section is. That has been long since established.
Also, there are many, many links out there that point to the old location for the developer docs.
Lastly, please consult this list before making such drastic changes to the website. I've put a lot of work into that section, so you could have at least asked.
i admit it was really bad practice, esp. not to ask.
but before i put this back, could we clarify: why is "documentation" the place where it belongs? or: what makes the "developer" section less "the right place"?
the thing i currently see is, that there are 2 branches /dev and /doc/dev which both hold documentation about development.
information is surely to be in the directory where you are currently not looking, since it is unclear, where it should be. (and i am not the only one who feels like this).
after all, the /dev section IS "documentation about development", and not "development" itself.
so in theory(!) i see no reason to have 2 separate branches with the same topic but different content.
however, you are completely right: in practice there are external links pointing to the old & deprecated location, which must be handled. the good thing is, that this could be handled very easily: i can set up a rewrite rule on the webserver, so that http://puredata.org/doc/dev is automatically redirected to http://puredata.org/dev which should make all the links valid again.
and finally: please use relative linking whenever possible. there is really no point in linking from http://puredata.org/bi/ba to "https://www.puredata.info/bi/borg" when "borg" would have done the same.
mfga.dsr IOhannes
why i didn't ask? because we are really bad at decision-making. most discussions stop at the point where "everybody" (this is: you, me, probably a single 3rd person) has clarified their poisition.
On Tue, 26 Sep 2006, IOhannes m zmoelnig wrote:
so in theory(!) i see no reason to have 2 separate branches with the same topic but different content.
Do it like DMOZ and the Yahoo index do. To overcome the fact that categories in the index are arborescent but categories in your head are not, pick one of the two possibilities as your main one and use a hyperlink to the other one.
and finally: please use relative linking whenever possible. there is really no point in linking from http://puredata.org/bi/ba to "https://www.puredata.info/bi/borg" when "borg" would have done the same.
isn't it the only way to switch from http to https by using a link? (is https:borg a valid relative URL?) is there any reason to make an explicit switch from http to https in this case?
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Mathieu Bouchard wrote:
On Tue, 26 Sep 2006, IOhannes m zmoelnig wrote:
so in theory(!) i see no reason to have 2 separate branches with the same topic but different content.
Do it like DMOZ and the Yahoo index do. To overcome the fact that categories in the index are arborescent but categories in your head are not, pick one of the two possibilities as your main one and use a hyperlink to the other one.
and finally: please use relative linking whenever possible. there is really no point in linking from http://puredata.org/bi/ba to "https://www.puredata.info/bi/borg" when "borg" would have done the same.
isn't it the only way to switch from http to https by using a link? (is https:borg a valid relative URL?) is there any reason to make an explicit switch from http to https in this case?
that was the point: people are automatically redirected to https://<whatever> when (before) they log into the portal, in order to not transmit passwords in plain-text. when they now edit their page, they often just copy the link from the linkbar (which now is on https). anonymous (http) visitors of the website, now get loads of https-links (and the poor server's load jumps up)
mfg.asdr IOhannes
Hallo, IOhannes m zmoelnig hat gesagt: // IOhannes m zmoelnig wrote:
that was the point: people are automatically redirected to https://<whatever> when (before) they log into the portal, in order to not transmit passwords in plain-text. when they now edit their page, they often just copy the link from the linkbar (which now is on https).
Maybe it's possible to redirect people back to plain http after they're authenticated?
Ciao
On Wed, 27 Sep 2006, IOhannes m zmoelnig wrote:
that was the point: people are automatically redirected to https://<whatever> when (before) they log into the portal, in order to not transmit passwords in plain-text. when they now edit their page, they often just copy the link from the linkbar (which now is on https). anonymous (http) visitors of the website, now get loads of https-links (and the poor server's load jumps up)
Why not use challenge-response authentication over plain http? That's secure enough, yes?
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard - tél:+1.514.383.3801 - http://artengine.ca/matju | Freelance Digital Arts Engineer, Montréal QC Canada
Mathieu Bouchard wrote:
On Wed, 27 Sep 2006, IOhannes m zmoelnig wrote:
that was the point: people are automatically redirected to https://<whatever> when (before) they log into the portal, in order to not transmit passwords in plain-text. when they now edit their page, they often just copy the link from the linkbar (which now is on https). anonymous (http) visitors of the website, now get loads of https-links (and the poor server's load jumps up)
Why not use challenge-response authentication over plain http? That's secure enough, yes?
so how?
mf.asdr IOhannes
Hallo, IOhannes m zmoelnig hat gesagt: // IOhannes m zmoelnig wrote:
Mathieu Bouchard wrote:
Why not use challenge-response authentication over plain http? That's secure enough, yes?
so how?
I think, Matju refers to basic or digest http authentificaton, which is a step backwards from using SSL, so as we already have SSL working, I would not recommend it. But redirecting users to plain http-pages after they have logged in over SSL is a pretty common design pattern in Web applications. I found this page which addresses this and various other login stuff:
http://plone.org/documentation/how-to/secure-login-without-plain-text-passwo...
Search for "came_from".
Ciao
On Sep 26, 2006, at 6:26 AM, IOhannes m zmoelnig wrote:
hi
Hans-Christoph Steiner wrote:
I just noticed that the developer docs on puredata.org were moved to the "developer" section. Please put them back where they were, that is where they belong. The "developer" section is not for documentation, the documentation section is. That has been long since established. Also, there are many, many links out there that point to the old location for the developer docs. Lastly, please consult this list before making such drastic changes to the website. I've put a lot of work into that section, so you could have at least asked.
i admit it was really bad practice, esp. not to ask.
but before i put this back, could we clarify: why is "documentation" the place where it belongs? or: what makes the "developer" section less "the right place"?
the thing i currently see is, that there are 2 branches /dev and / doc/dev which both hold documentation about development.
information is surely to be in the directory where you are currently not looking, since it is unclear, where it should be. (and i am not the only one who feels like this).
after all, the /dev section IS "documentation about development", and not "development" itself.
By that standard then everything on the web is documentation. We are using a narrower definition which makes a lot more sense. Manuals, HOWTOs, tutorials, these are documentation. Everything in /docs/ developer was a HOWTO, manual, or tutorial
The developer section is meant for developer tools: key links, bugtrackers, wikis for project management, etc. The content in the wikis will probably ultimately end up being boiled down to docs, but they are not there yet. The bugtracker is documentation on bugs, but it is not a manual, howto, or tutorial.
Please read the archives for a more in depth discussion.
so in theory(!) i see no reason to have 2 separate branches with the same topic but different content.
however, you are completely right: in practice there are external links pointing to the old & deprecated location, which must be handled. the good thing is, that this could be handled very easily: i can set up a rewrite rule on the webserver, so that http://puredata.org/ doc/dev is automatically redirected to http://puredata.org/dev which should make all the links valid again.
We discussed this stuff at length when we set up the website (check the archives). Its been working well as far as I've seen. I see no reason to change it now.
and finally: please use relative linking whenever possible. there is really no point in linking from http://puredata.org/bi/ba to "https://www.puredata.info/bi/borg" when "borg" would have done the same.
Yes indeed, that's just a general rule in web site building.
mfga.dsr IOhannes
why i didn't ask? because we are really bad at decision-making. most discussions stop at the point where "everybody" (this is: you, me, probably a single 3rd person) has clarified their poisition.
It takes 10 seconds to send such an email. It would have saved us all a lot of time and hassle.
.hc
------------------------------------------------------------------------
Computer science is no more related to the computer than astronomy is related to the telescope. -Edsger Dykstra
hi.
Hans-Christoph Steiner wrote:
On Sep 26, 2006, at 6:26 AM, IOhannes m zmoelnig wrote:
after all, the /dev section IS "documentation about development", and not "development" itself.
By that standard then everything on the web is documentation. We are using a narrower definition which makes a lot more sense. Manuals, HOWTOs, tutorials, these are documentation. Everything in /docs/developer was a HOWTO, manual, or tutorial
like http://puredata.info/docs/developer/sourcenotes ?
The developer section is meant for developer tools: key links, bugtrackers, wikis for project management, etc. The content in the wikis will probably ultimately end up being boiled down to docs, but they are not there yet. The bugtracker is documentation on bugs, but it is not a manual, howto, or tutorial.
Please read the archives for a more in depth discussion.
i did so. the discussion has taken place in 2003 when the site was set up. i do believe that several things in this design did not work out.
so in theory(!) i see no reason to have 2 separate branches with the same topic but different content.
i still don't. even if there is some artificial segregation line between the 2 branches, which i could memorize. if i want to know what is going on in "development" (e.g. what's on in pddp, or how the exact cvs download command was) i will just click on the "development" tab (and not just me as a "noob" who just wants to browse the information, but me as somebody who has worked for years with the website's current structure).
however, you are completely right: in practice there are external links pointing to the old & deprecated location, which must be handled. the good thing is, that this could be handled very easily: i can set up a rewrite rule on the webserver, so that http://puredata.org/doc/dev is automatically redirected to http://puredata.org/dev which should make all the links valid again.
We discussed this stuff at length when we set up the website (check the archives). Its been working well as far as I've seen. I see no reason to change it now.
the discussion has taken place in 2003 when the site was set up. i don't think that it has been working so well.
why i didn't ask? because we are really bad at decision-making. most discussions stop at the point where "everybody" (this is: you, me, probably a single 3rd person) has clarified their poisition.
It takes 10 seconds to send such an email. It would have saved us all a lot of time and hassle.
how much time have you lost through this hazzle? (esp. when writing an email takes you just 10 secs)
if the docs/development/ section is really just meant as a selection of how-to's, tutorials and manuals, then why don't we just use some autocollector, that grabs all tutorials,... from /dev/ (and wherever).
just to reiterate why i do this: (overlapping) information is kept in 2 separate places. information therefore gets duplicated (cloned). maintainment of the cloned information usually only effects one branch, resulting in either divergent information, outdated information (in one branch). information that does not get duplicated is missing in one branch.
mfg.asdr. IOhannes
PS: your old links should now at least work.
Put the site back the way it was. You have broken days and days of my work without even the courtesy of asking. Once the site is back to the original, then we can talk about new ideas.
So this is how you repay me? I spent many, many hours getting Gem working on the auto-builders and you fuck up my work? That is really, really low. I was holding on to the pure-data.info because I feared something like this. I thought you had learned so I signed over the domain name to you. And you do this to me. And you don't even have the courage to admit you were wrong and set things right. You're still fighting. This is a ridiculous and massive waste of time and energy.
.hc
On Sep 27, 2006, at 4:20 AM, IOhannes m zmoelnig wrote:
hi.
Hans-Christoph Steiner wrote:
On Sep 26, 2006, at 6:26 AM, IOhannes m zmoelnig wrote:
after all, the /dev section IS "documentation about development", and not "development" itself.
By that standard then everything on the web is documentation. We are using a narrower definition which makes a lot more sense. Manuals, HOWTOs, tutorials, these are documentation. Everything in /docs/developer was a HOWTO, manual, or tutorial
like http://puredata.info/docs/developer/sourcenotes ?
The developer section is meant for developer tools: key links, bugtrackers, wikis for project management, etc. The content in the wikis will probably ultimately end up being boiled down to docs, but they are not there yet. The bugtracker is documentation on bugs, but it is not a manual, howto, or tutorial. Please read the archives for a more in depth discussion.
i did so. the discussion has taken place in 2003 when the site was set up. i do believe that several things in this design did not work out.
so in theory(!) i see no reason to have 2 separate branches with the same topic but different content.
i still don't. even if there is some artificial segregation line between the 2 branches, which i could memorize. if i want to know what is going on in "development" (e.g. what's on in pddp, or how the exact cvs download command was) i will just click on the "development" tab (and not just me as a "noob" who just wants to browse the information, but me as somebody who has worked for years with the website's current structure).
however, you are completely right: in practice there are external links pointing to the old & deprecated location, which must be handled. the good thing is, that this could be handled very easily: i can set up a rewrite rule on the webserver, so that http:// puredata.org/doc/dev is automatically redirected to http:// puredata.org/dev which should make all the links valid again.
We discussed this stuff at length when we set up the website (check the archives). Its been working well as far as I've seen. I see no reason to change it now.
the discussion has taken place in 2003 when the site was set up. i don't think that it has been working so well.
why i didn't ask? because we are really bad at decision-making. most discussions stop at the point where "everybody" (this is: you, me, probably a single 3rd person) has clarified their poisition.
It takes 10 seconds to send such an email. It would have saved us all a lot of time and hassle.
how much time have you lost through this hazzle? (esp. when writing an email takes you just 10 secs)
if the docs/development/ section is really just meant as a selection of how-to's, tutorials and manuals, then why don't we just use some autocollector, that grabs all tutorials,... from / dev/ (and wherever).
just to reiterate why i do this: (overlapping) information is kept in 2 separate places. information therefore gets duplicated (cloned). maintainment of the cloned information usually only effects one branch, resulting in either divergent information, outdated information (in one branch). information that does not get duplicated is missing in one branch.
mfg.asdr. IOhannes
PS: your old links should now at least work.
Hans-Christoph Steiner wrote:
Put the site back the way it was. You have broken days and days of my work without even the courtesy of asking. Once the site is back to the original, then we can talk about new ideas.
So this is how you repay me? I spent many, many hours getting Gem
no this is not how i _repay_ you. this is not connected anyhow to your very valuable work.
working on the auto-builders and you fuck up my work? That is really, really low. I was holding on to the pure-data.info because I feared
i don't understand what this would have changed in the short term (of course the domains could have split after some time of discrepancy)
courage to admit you were wrong and set things right. You're still fighting. This is a ridiculous and massive waste of time and energy.
currently it works as before (all the links are still valid, though redirected to /dev) i will change it today.
mfga.sdr IOhannes
On Sep 28, 2006, at 3:57 AM, IOhannes m zmoelnig wrote:
Hans-Christoph Steiner wrote:
Put the site back the way it was. You have broken days and days of my work without even the courtesy of asking. Once the site is back to the original, then we can talk about new ideas. So this is how you repay me? I spent many, many hours getting Gem
no this is not how i _repay_ you. this is not connected anyhow to your very valuable work.
working on the auto-builders and you fuck up my work? That is really, really low. I was holding on to the pure-data.info because I feared
i don't understand what this would have changed in the short term (of course the domains could have split after some time of discrepancy)
courage to admit you were wrong and set things right. You're still fighting. This is a ridiculous and massive waste of time and energy.
currently it works as before (all the links are still valid, though redirected to /dev) i will change it today.
Thank you for putting it back, but its not all back yet. There are now copies of all the pages in /dev also:
https://puredata.org/dev/folder_contents
That is quite dangerous as people might edit both, then we will lose work. Also, I had to put back the original index page on /docs/ developer
Sorry for freaking out, but I have put so much time in that stuff, and its not fun work. I am just getting to the point where I can wrap things up and get on to some interesting work, then this all happened. I needed to get releases out for Koray's upcoming workshop, and some other people are counting on a new release, but instead I spent way too long dealing with all this.
.hc
On Sep 28, 2006, at 3:53 PM, Hans-Christoph Steiner wrote:
On Sep 28, 2006, at 3:57 AM, IOhannes m zmoelnig wrote:
Hans-Christoph Steiner wrote:
Put the site back the way it was. You have broken days and days of my work without even the courtesy of asking. Once the site is back to the original, then we can talk about new ideas. So this is how you repay me? I spent many, many hours getting Gem
no this is not how i _repay_ you. this is not connected anyhow to your very valuable work.
working on the auto-builders and you fuck up my work? That is really, really low. I was holding on to the pure-data.info because I feared
i don't understand what this would have changed in the short term (of course the domains could have split after some time of discrepancy)
courage to admit you were wrong and set things right. You're still fighting. This is a ridiculous and massive waste of time and energy.
currently it works as before (all the links are still valid, though redirected to /dev) i will change it today.
Thank you for putting it back, but its not all back yet. There are now copies of all the pages in /dev also:
https://puredata.org/dev/folder_contents
That is quite dangerous as people might edit both, then we will lose work. Also, I had to put back the original index page on / docs/developer
Sorry for freaking out, but I have put so much time in that stuff, and its not fun work. I am just getting to the point where I can wrap things up and get on to some interesting work, then this all happened. I needed to get releases out for Koray's upcoming workshop, and some other people are counting on a new release, but instead I spent way too long dealing with all this.
I forgot to add that I am not opposed to the website being changed, and I certainly always like contributors and want to encourage that. But I just don't want the rug yanked out from under me, especially when I need it most.
I think the architecture of the site could be improved, but I think its totally useable the way it is now, and I really don't want to put much effort into it. I think to do it right, we'd have to look at the content we have as a whole. Personally, I think the Pd community would be best of with a wiki site, for the docs at least, with some other additions.
.hc
-
Frank Barknecht -
Hans-Christoph Steiner -
IOhannes m zmoelnig -
Mathieu Bouchard