Thread: Re: [DOCS] Doc fixes and improvements
Thom Brown wrote: > On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: > > > > Patch applied and backpatched to 9.0.X. ?Thanks. > > Whoa, no comments or objections then? I'm scared now. :S Well, you are improving an area that is long overdue for improvement, so we give you a wide berth. ;-) Also, if you need any javascript stuff done, just let me know. Also, we are planning to release Postgres 9.0 on September 20, and it would be great to have something in place by then to showcase our doc improvements, in both content and appearance. Is there any schedule for completion? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 10 September 2010 14:36, Bruce Momjian <bruce@momjian.us> wrote: > Thom Brown wrote: >> On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: >> > >> > Patch applied and backpatched to 9.0.X. ?Thanks. >> >> Whoa, no comments or objections then? I'm scared now. :S > > Well, you are improving an area that is long overdue for improvement, so > we give you a wide berth. ;-) > > Also, if you need any javascript stuff done, just let me know. > > Also, we are planning to release Postgres 9.0 on September 20, and it > would be great to have something in place by then to showcase our doc > improvements, in both content and appearance. Is there any schedule for > completion? Well it's mostly done now. I need to do some more testing and hopefully others can try it out on weird browsers too because I can only do so much testing myself. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On 10 September 2010 14:39, Thom Brown <thom@linux.com> wrote: > On 10 September 2010 14:36, Bruce Momjian <bruce@momjian.us> wrote: >> Thom Brown wrote: >>> On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: >>> > >>> > Patch applied and backpatched to 9.0.X. ?Thanks. >>> >>> Whoa, no comments or objections then? I'm scared now. :S >> >> Well, you are improving an area that is long overdue for improvement, so >> we give you a wide berth. ;-) >> >> Also, if you need any javascript stuff done, just let me know. >> >> Also, we are planning to release Postgres 9.0 on September 20, and it >> would be great to have something in place by then to showcase our doc >> improvements, in both content and appearance. Is there any schedule for >> completion? > > Well it's mostly done now. I need to do some more testing and > hopefully others can try it out on weird browsers too because I can > only do so much testing myself. Actually, that wasn't entirely true. I have to re-adapt it for the current site as my changes were for the new version of the site which we're not using yet. I think it's just a case of changing one line from last I checked. Not sure where to take that though as I don't have a working copy of the current site. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On Fri, Sep 10, 2010 at 15:43, Thom Brown <thom@linux.com> wrote: > On 10 September 2010 14:39, Thom Brown <thom@linux.com> wrote: >> On 10 September 2010 14:36, Bruce Momjian <bruce@momjian.us> wrote: >>> Thom Brown wrote: >>>> On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: >>>> > >>>> > Patch applied and backpatched to 9.0.X. ?Thanks. >>>> >>>> Whoa, no comments or objections then? I'm scared now. :S >>> >>> Well, you are improving an area that is long overdue for improvement, so >>> we give you a wide berth. ;-) >>> >>> Also, if you need any javascript stuff done, just let me know. >>> >>> Also, we are planning to release Postgres 9.0 on September 20, and it >>> would be great to have something in place by then to showcase our doc >>> improvements, in both content and appearance. Is there any schedule for >>> completion? >> >> Well it's mostly done now. I need to do some more testing and >> hopefully others can try it out on weird browsers too because I can >> only do so much testing myself. > > Actually, that wasn't entirely true. I have to re-adapt it for the > current site as my changes were for the new version of the site which > we're not using yet. I think it's just a case of changing one line > from last I checked. Not sure where to take that though as I don't > have a working copy of the current site. Find me on IRC sometime and we can work that out - especially if you can get to the point of having a patch ready to debug :-) -- Magnus Hagander Me: http://www.hagander.net/ Work: http://www.redpill-linpro.com/
Excerpts from Thom Brown's message of vie sep 10 09:39:50 -0400 2010: > Well it's mostly done now. I need to do some more testing and > hopefully others can try it out on weird browsers too because I can > only do so much testing myself. Is this going to be incorporated into the main stylesheet.css that's used in the tarball docs, or is it only for the website? -- Álvaro Herrera <alvherre@commandprompt.com> The PostgreSQL Company - Command Prompt, Inc. PostgreSQL Replication, Consulting, Custom Development, 24x7 support
On 10 September 2010 16:26, Alvaro Herrera <alvherre@commandprompt.com> wrote: > Excerpts from Thom Brown's message of vie sep 10 09:39:50 -0400 2010: > >> Well it's mostly done now. I need to do some more testing and >> hopefully others can try it out on weird browsers too because I can >> only do so much testing myself. > > Is this going to be incorporated into the main stylesheet.css that's > used in the tarball docs, or is it only for the website? I don't know. I haven't ever looked at the version provided in the tarball. Does it currently appear like the live site's documentation when viewing it? -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On Fri, Sep 10, 2010 at 15:48, Magnus Hagander <magnus@hagander.net> wrote: > On Fri, Sep 10, 2010 at 15:43, Thom Brown <thom@linux.com> wrote: >> On 10 September 2010 14:39, Thom Brown <thom@linux.com> wrote: >>> On 10 September 2010 14:36, Bruce Momjian <bruce@momjian.us> wrote: >>>> Thom Brown wrote: >>>>> On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: >>>>> > >>>>> > Patch applied and backpatched to 9.0.X. ?Thanks. >>>>> >>>>> Whoa, no comments or objections then? I'm scared now. :S >>>> >>>> Well, you are improving an area that is long overdue for improvement, so >>>> we give you a wide berth. ;-) >>>> >>>> Also, if you need any javascript stuff done, just let me know. >>>> >>>> Also, we are planning to release Postgres 9.0 on September 20, and it >>>> would be great to have something in place by then to showcase our doc >>>> improvements, in both content and appearance. Is there any schedule for >>>> completion? >>> >>> Well it's mostly done now. I need to do some more testing and >>> hopefully others can try it out on weird browsers too because I can >>> only do so much testing myself. >> >> Actually, that wasn't entirely true. I have to re-adapt it for the >> current site as my changes were for the new version of the site which >> we're not using yet. I think it's just a case of changing one line >> from last I checked. Not sure where to take that though as I don't >> have a working copy of the current site. > > Find me on IRC sometime and we can work that out - especially if you > can get to the point of having a patch ready to debug :-) Ok, Thom has merged these changes onto the current website code, and I've put it up on http://magnus.webdev.postgresql.org - note that you need to go back to 8.3 to actually see anything, because the db doesn't have the latest docs. It'd be good to get some more testing on that one to make sure it doean't break things - in particular, someone to test IE and Safari. It should work, but better safe than sorry... -- Magnus Hagander Me: http://www.hagander.net/ Work: http://www.redpill-linpro.com/
Excerpts from Magnus Hagander's message of vie sep 10 15:20:41 -0400 2010: > Ok, Thom has merged these changes onto the current website code, and > I've put it up on http://magnus.webdev.postgresql.org - note that you > need to go back to 8.3 to actually see anything, because the db > doesn't have the latest docs. > > It'd be good to get some more testing on that one to make sure it > doean't break things - in particular, someone to test IE and Safari. > It should work, but better safe than sorry... Just gave it a try on IE6 on Linux here, and it looks OK. No rounded corners and no shadows, but nothing's out of place either. -- Álvaro Herrera <alvherre@commandprompt.com> The PostgreSQL Company - Command Prompt, Inc. PostgreSQL Replication, Consulting, Custom Development, 24x7 support
Magnus Hagander wrote: > On Fri, Sep 10, 2010 at 15:48, Magnus Hagander <magnus@hagander.net> wrote: > > On Fri, Sep 10, 2010 at 15:43, Thom Brown <thom@linux.com> wrote: > >> On 10 September 2010 14:39, Thom Brown <thom@linux.com> wrote: > >>> On 10 September 2010 14:36, Bruce Momjian <bruce@momjian.us> wrote: > >>>> Thom Brown wrote: > >>>>> On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: > >>>>> > > >>>>> > Patch applied and backpatched to 9.0.X. ?Thanks. > >>>>> > >>>>> Whoa, no comments or objections then? ?I'm scared now. :S > >>>> > >>>> Well, you are improving an area that is long overdue for improvement, so > >>>> we give you a wide berth. ?;-) > >>>> > >>>> Also, if you need any javascript stuff done, just let me know. > >>>> > >>>> Also, we are planning to release Postgres 9.0 on September 20, and it > >>>> would be great to have something in place by then to showcase our doc > >>>> improvements, in both content and appearance. ?Is there any schedule for > >>>> completion? > >>> > >>> Well it's mostly done now. ?I need to do some more testing and > >>> hopefully others can try it out on weird browsers too because I can > >>> only do so much testing myself. > >> > >> Actually, that wasn't entirely true. ?I have to re-adapt it for the > >> current site as my changes were for the new version of the site which > >> we're not using yet. ?I think it's just a case of changing one line > >> from last I checked. ?Not sure where to take that though as I don't > >> have a working copy of the current site. > > > > Find me on IRC sometime and we can work that out - especially if you > > can get to the point of having a patch ready to debug :-) > > Ok, Thom has merged these changes onto the current website code, and > I've put it up on http://magnus.webdev.postgresql.org - note that you > need to go back to 8.3 to actually see anything, because the db > doesn't have the latest docs. I don't see the fix for javascript font size for Firefox fixed-width fonts: http://magnus.webdev.postgresql.org/docs/8.3/static/sql-syntax-lexical.html Is that installed? I assmume not. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On Fri, 2010-09-10 at 15:32 -0400, Alvaro Herrera wrote: > Excerpts from Magnus Hagander's message of vie sep 10 15:20:41 -0400 2010: > > > Ok, Thom has merged these changes onto the current website code, and > > I've put it up on http://magnus.webdev.postgresql.org - note that you > > need to go back to 8.3 to actually see anything, because the db > > doesn't have the latest docs. > > > > It'd be good to get some more testing on that one to make sure it > > doean't break things - in particular, someone to test IE and Safari. > > It should work, but better safe than sorry... > > Just gave it a try on IE6 on Linux here, and it looks OK. No rounded > corners and no shadows, but nothing's out of place either. IE8 on Win7 looks good. No rounded corners but that is as expected. JD > > -- > Ãlvaro Herrera <alvherre@commandprompt.com> > The PostgreSQL Company - Command Prompt, Inc. > PostgreSQL Replication, Consulting, Custom Development, 24x7 support > -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On 10 September 2010 20:36, Bruce Momjian <bruce@momjian.us> wrote: > Magnus Hagander wrote: >> On Fri, Sep 10, 2010 at 15:48, Magnus Hagander <magnus@hagander.net> wrote: >> > On Fri, Sep 10, 2010 at 15:43, Thom Brown <thom@linux.com> wrote: >> >> On 10 September 2010 14:39, Thom Brown <thom@linux.com> wrote: >> >>> On 10 September 2010 14:36, Bruce Momjian <bruce@momjian.us> wrote: >> >>>> Thom Brown wrote: >> >>>>> On 9 September 2010 01:48, Bruce Momjian <bruce@momjian.us> wrote: >> >>>>> > >> >>>>> > Patch applied and backpatched to 9.0.X. ?Thanks. >> >>>>> >> >>>>> Whoa, no comments or objections then? ?I'm scared now. :S >> >>>> >> >>>> Well, you are improving an area that is long overdue for improvement, so >> >>>> we give you a wide berth. ?;-) >> >>>> >> >>>> Also, if you need any javascript stuff done, just let me know. >> >>>> >> >>>> Also, we are planning to release Postgres 9.0 on September 20, and it >> >>>> would be great to have something in place by then to showcase our doc >> >>>> improvements, in both content and appearance. ?Is there any schedule for >> >>>> completion? >> >>> >> >>> Well it's mostly done now. ?I need to do some more testing and >> >>> hopefully others can try it out on weird browsers too because I can >> >>> only do so much testing myself. >> >> >> >> Actually, that wasn't entirely true. ?I have to re-adapt it for the >> >> current site as my changes were for the new version of the site which >> >> we're not using yet. ?I think it's just a case of changing one line >> >> from last I checked. ?Not sure where to take that though as I don't >> >> have a working copy of the current site. >> > >> > Find me on IRC sometime and we can work that out - especially if you >> > can get to the point of having a patch ready to debug :-) >> >> Ok, Thom has merged these changes onto the current website code, and >> I've put it up on http://magnus.webdev.postgresql.org - note that you >> need to go back to 8.3 to actually see anything, because the db >> doesn't have the latest docs. > > I don't see the fix for javascript font size for Firefox fixed-width > fonts: > > http://magnus.webdev.postgresql.org/docs/8.3/static/sql-syntax-lexical.html > > Is that installed? I assmume not. I just checked and the existing geckostyle.js is present, as well as the geckofixes.css it's supposed to introduce. I also disabled javascript in Firefox to see the effect, and the monospace fonts are very small (previously looked fine), so the script appears to be working. Remember that the script for the live site (geckostyle.js) has a different name to the upcoming site (monospacefix.js) which I was hosting before. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > >> >> Actually, that wasn't entirely true. ?I have to re-adapt it for the > >> >> current site as my changes were for the new version of the site which > >> >> we're not using yet. ?I think it's just a case of changing one line > >> >> from last I checked. ?Not sure where to take that though as I don't > >> >> have a working copy of the current site. > >> > > >> > Find me on IRC sometime and we can work that out - especially if you > >> > can get to the point of having a patch ready to debug :-) > >> > >> Ok, Thom has merged these changes onto the current website code, and > >> I've put it up on http://magnus.webdev.postgresql.org - note that you > >> need to go back to 8.3 to actually see anything, because the db > >> doesn't have the latest docs. > > > > I don't see the fix for javascript font size for Firefox fixed-width > > fonts: > > > > ? ? ? ?http://magnus.webdev.postgresql.org/docs/8.3/static/sql-syntax-lexical.html > > > > Is that installed? ?I assmume not. > > I just checked and the existing geckostyle.js is present, as well as > the geckofixes.css it's supposed to introduce. I also disabled > javascript in Firefox to see the effect, and the monospace fonts are > very small (previously looked fine), so the script appears to be > working. > > Remember that the script for the live site (geckostyle.js) has a > different name to the upcoming site (monospacefix.js) which I was > hosting before. OK, here is what I see: http://momjian.us/tmp/pg_manual.png I have font size of 14, with fixed size of 12 in my Firefox preferences. If I turn off javascript is looks more consistent: http://momjian.us/tmp/pg_manual2.png -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 10 September 2010 20:50, Bruce Momjian <bruce@momjian.us> wrote: > Thom Brown wrote: >> >> >> Actually, that wasn't entirely true. ?I have to re-adapt it for the >> >> >> current site as my changes were for the new version of the site which >> >> >> we're not using yet. ?I think it's just a case of changing one line >> >> >> from last I checked. ?Not sure where to take that though as I don't >> >> >> have a working copy of the current site. >> >> > >> >> > Find me on IRC sometime and we can work that out - especially if you >> >> > can get to the point of having a patch ready to debug :-) >> >> >> >> Ok, Thom has merged these changes onto the current website code, and >> >> I've put it up on http://magnus.webdev.postgresql.org - note that you >> >> need to go back to 8.3 to actually see anything, because the db >> >> doesn't have the latest docs. >> > >> > I don't see the fix for javascript font size for Firefox fixed-width >> > fonts: >> > >> > ? ? ? ?http://magnus.webdev.postgresql.org/docs/8.3/static/sql-syntax-lexical.html >> > >> > Is that installed? ?I assmume not. >> >> I just checked and the existing geckostyle.js is present, as well as >> the geckofixes.css it's supposed to introduce. I also disabled >> javascript in Firefox to see the effect, and the monospace fonts are >> very small (previously looked fine), so the script appears to be >> working. >> >> Remember that the script for the live site (geckostyle.js) has a >> different name to the upcoming site (monospacefix.js) which I was >> hosting before. > > OK, here is what I see: > > http://momjian.us/tmp/pg_manual.png > > I have font size of 14, with fixed size of 12 in my Firefox preferences. > > If I turn off javascript is looks more consistent: > > http://momjian.us/tmp/pg_manual2.png The font size shown in your 2nd screenshot looks smaller than the existing docs. I did a comparison of Magnus' version and the live version. The same monospace example in both renders the font as precisely a calculated 12.7667px. In fact it looks more like the proportional font is too small. My value for that is still at the default value of 16. Does yours differ? -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > > OK, here is what I see: > > > > ? ? ? ?http://momjian.us/tmp/pg_manual.png > > > > I have font size of 14, with fixed size of 12 in my Firefox preferences. > > > > If I turn off javascript is looks more consistent: > > > > ? ? ? ?http://momjian.us/tmp/pg_manual2.png > > The font size shown in your 2nd screenshot looks smaller than the > existing docs. I did a comparison of Magnus' version and the live > version. The same monospace example in both renders the font as > precisely a calculated 12.7667px. > > In fact it looks more like the proportional font is too small. My > value for that is still at the default value of 16. Does yours > differ? Here is what I think you want to see. This is my screen with proportional and fixed set to 14, with javascript turned off: http://momjian.us/tmp/pg_manual3.png That looks right. This is Ubuntu with Firefox 3.6.9. And this is the display of my font defaults: http://momjian.us/tmp/fonts.png Does that help? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
Magnus Hagander <magnus@hagander.net> writes: > Ok, Thom has merged these changes onto the current website code, and > I've put it up on http://magnus.webdev.postgresql.org - note that you > need to go back to 8.3 to actually see anything, because the db > doesn't have the latest docs. You know, I hate to say it, but this is just way too busy for my taste. The typography is screaming "look at me" instead of staying out of the way. I can't imagine wanting to read any large amount of docs in this presentation. Maybe I'm in the minority. regards, tom lane
On 10 September 2010 21:03, Bruce Momjian <bruce@momjian.us> wrote: > Thom Brown wrote: >> > OK, here is what I see: >> > >> > ? ? ? ?http://momjian.us/tmp/pg_manual.png >> > >> > I have font size of 14, with fixed size of 12 in my Firefox preferences. >> > >> > If I turn off javascript is looks more consistent: >> > >> > ? ? ? ?http://momjian.us/tmp/pg_manual2.png >> >> The font size shown in your 2nd screenshot looks smaller than the >> existing docs. I did a comparison of Magnus' version and the live >> version. The same monospace example in both renders the font as >> precisely a calculated 12.7667px. >> >> In fact it looks more like the proportional font is too small. My >> value for that is still at the default value of 16. Does yours >> differ? > > Here is what I think you want to see. This is my screen with > proportional and fixed set to 14, with javascript turned off: > > http://momjian.us/tmp/pg_manual3.png > > That looks right. This is Ubuntu with Firefox 3.6.9. > > And this is the display of my font defaults: > > http://momjian.us/tmp/fonts.png > > Does that help? Yes, according to these, the proportional font defaults to 16px: http://nst.sourceforge.net/nst/docs/tech/ch02s04.html http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > Magnus Hagander <magnus@hagander.net> writes: >> Ok, Thom has merged these changes onto the current website code, and >> I've put it up on http://magnus.webdev.postgresql.org - note that you >> need to go back to 8.3 to actually see anything, because the db >> doesn't have the latest docs. > > You know, I hate to say it, but this is just way too busy for my taste. > The typography is screaming "look at me" instead of staying out of the > way. I can't imagine wanting to read any large amount of docs in this > presentation. > > Maybe I'm in the minority. Do you have any specific recommendations? What would tone it down? -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On Fri, 2010-09-10 at 21:11 +0100, Thom Brown wrote: > On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > Magnus Hagander <magnus@hagander.net> writes: > >> Ok, Thom has merged these changes onto the current website code, and > >> I've put it up on http://magnus.webdev.postgresql.org - note that you > >> need to go back to 8.3 to actually see anything, because the db > >> doesn't have the latest docs. > > > > You know, I hate to say it, but this is just way too busy for my taste. > > The typography is screaming "look at me" instead of staying out of the > > way. I can't imagine wanting to read any large amount of docs in this > > presentation. > > > > Maybe I'm in the minority. > > Do you have any specific recommendations? What would tone it down? I find it very pleasant. It reminds me of the python docs which are nice to read as well. Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
Thom Brown <thom@linux.com> writes: > On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: >> You know, I hate to say it, but this is just way too busy for my taste. >> The typography is screaming "look at me" instead of staying out of the >> way. �I can't imagine wanting to read any large amount of docs in this >> presentation. >> >> Maybe I'm in the minority. > Do you have any specific recommendations? What would tone it down? I still think the colored backgrounds are overused, and there's way too much bold text. AFAICS all the fixed-width-font text is now bold too, to no improvement in readability. regards, tom lane
Thom Brown wrote: > On 10 September 2010 21:03, Bruce Momjian <bruce@momjian.us> wrote: > > Thom Brown wrote: > >> > OK, here is what I see: > >> > > >> > ? ? ? ?http://momjian.us/tmp/pg_manual.png > >> > > >> > I have font size of 14, with fixed size of 12 in my Firefox preferences. > >> > > >> > If I turn off javascript is looks more consistent: > >> > > >> > ? ? ? ?http://momjian.us/tmp/pg_manual2.png > >> > >> The font size shown in your 2nd screenshot looks smaller than the > >> existing docs. ?I did a comparison of Magnus' version and the live > >> version. ?The same monospace example in both renders the font as > >> precisely a calculated 12.7667px. > >> > >> In fact it looks more like the proportional font is too small. ?My > >> value for that is still at the default value of 16. ?Does yours > >> differ? > > > > Here is what I think you want to see. ?This is my screen with > > proportional and fixed set to 14, with javascript turned off: > > > > ? ? ? ?http://momjian.us/tmp/pg_manual3.png > > > > That looks right. ?This is Ubuntu with Firefox 3.6.9. > > > > And this is the display of my font defaults: > > > > ? ? ? ?http://momjian.us/tmp/fonts.png > > > > Does that help? > > Yes, according to these, the proportional font defaults to 16px: > http://nst.sourceforge.net/nst/docs/tech/ch02s04.html > http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK OK, so why are we not just setting the fixed-width font to be the same size as the proportional? We are mixing the two types in the same paragraph, so why not just set them the same? Using the javascript I sent, you can probe for the preference settings. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On Fri, 2010-09-10 at 16:42 -0400, Bruce Momjian wrote: > Thom Brown wrote: > > Yes, according to these, the proportional font defaults to 16px: > > http://nst.sourceforge.net/nst/docs/tech/ch02s04.html > > http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK > > OK, so why are we not just setting the fixed-width font to be the same > size as the proportional? We are mixing the two types in the same > paragraph, so why not just set them the same? Using the javascript I > sent, you can probe for the preference settings. Generally monospace is kept smaller than proportional (if it is larger, that is wrong). Mono is easy to read all the way down to 8pt (not that I am recommending that). Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On 10 September 2010 21:25, Tom Lane <tgl@sss.pgh.pa.us> wrote: > Thom Brown <thom@linux.com> writes: >> On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: >>> You know, I hate to say it, but this is just way too busy for my taste. >>> The typography is screaming "look at me" instead of staying out of the >>> way. I can't imagine wanting to read any large amount of docs in this >>> presentation. >>> >>> Maybe I'm in the minority. > >> Do you have any specific recommendations? What would tone it down? > > I still think the colored backgrounds are overused, and there's way too > much bold text. AFAICS all the fixed-width-font text is now bold too, > to no improvement in readability. The fixed-width-font text in examples shouldn't appear bold as I've applied no such style to those. How about this: http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html Lighter example boxes, slightly smaller monospace and no bold monospace text within paragraphs on the page. The intention of making a clear visual distinction between normal paragraph sections and examples/output was to make it easier to scan. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > On 10 September 2010 21:25, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > Thom Brown <thom@linux.com> writes: > >> On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > >>> You know, I hate to say it, but this is just way too busy for my taste. > >>> The typography is screaming "look at me" instead of staying out of the > >>> way. ?I can't imagine wanting to read any large amount of docs in this > >>> presentation. > >>> > >>> Maybe I'm in the minority. > > > >> Do you have any specific recommendations? ?What would tone it down? > > > > I still think the colored backgrounds are overused, and there's way too > > much bold text. ?AFAICS all the fixed-width-font text is now bold too, > > to no improvement in readability. > > The fixed-width-font text in examples shouldn't appear bold as I've > applied no such style to those. > > How about this: > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html > > Lighter example boxes, slightly smaller monospace and no bold > monospace text within paragraphs on the page. > > The intention of making a clear visual distinction between normal > paragraph sections and examples/output was to make it easier to scan. Looks much better. Here is "normal": http://momjian.us/tmp/pg_manual4.png and here is "Large": http://momjian.us/tmp/pg_manual5.png -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
Thom Brown <thom@linux.com> writes: > On 10 September 2010 21:25, Tom Lane <tgl@sss.pgh.pa.us> wrote: >> I still think the colored backgrounds are overused, and there's way too >> much bold text. �AFAICS all the fixed-width-font text is now bold too, >> to no improvement in readability. > The fixed-width-font text in examples shouldn't appear bold as I've > applied no such style to those. Hmm. I'm seeing nothing but boldface on Magnus' pages, using Safari. > How about this: > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html Curiously, that *doesn't* show as bold. > The intention of making a clear visual distinction between normal > paragraph sections and examples/output was to make it easier to scan. Agreed, but a little goes a long way. If you overdo it, it'll grate on the reader after twenty or thirty pages. The example blocks on this page are suitably subtle, maybe even a bit too far --- I find that depending on what angle I look at my LCD display, the background color can be practically indistinguishable from the rest of the page. But at other angles there's a clear distinction. regards, tom lane
On 10 September 2010 21:42, Bruce Momjian <bruce@momjian.us> wrote: > Thom Brown wrote: >> On 10 September 2010 21:03, Bruce Momjian <bruce@momjian.us> wrote: >> > Thom Brown wrote: >> >> > OK, here is what I see: >> >> > >> >> > ? ? ? ?http://momjian.us/tmp/pg_manual.png >> >> > >> >> > I have font size of 14, with fixed size of 12 in my Firefox preferences. >> >> > >> >> > If I turn off javascript is looks more consistent: >> >> > >> >> > ? ? ? ?http://momjian.us/tmp/pg_manual2.png >> >> >> >> The font size shown in your 2nd screenshot looks smaller than the >> >> existing docs. ?I did a comparison of Magnus' version and the live >> >> version. ?The same monospace example in both renders the font as >> >> precisely a calculated 12.7667px. >> >> >> >> In fact it looks more like the proportional font is too small. ?My >> >> value for that is still at the default value of 16. ?Does yours >> >> differ? >> > >> > Here is what I think you want to see. ?This is my screen with >> > proportional and fixed set to 14, with javascript turned off: >> > >> > ? ? ? ?http://momjian.us/tmp/pg_manual3.png >> > >> > That looks right. ?This is Ubuntu with Firefox 3.6.9. >> > >> > And this is the display of my font defaults: >> > >> > ? ? ? ?http://momjian.us/tmp/fonts.png >> > >> > Does that help? >> >> Yes, according to these, the proportional font defaults to 16px: >> http://nst.sourceforge.net/nst/docs/tech/ch02s04.html >> http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK > > OK, so why are we not just setting the fixed-width font to be the same > size as the proportional? We are mixing the two types in the same > paragraph, so why not just set them the same? Using the javascript I > sent, you can probe for the preference settings. That javascript checked whether the values were the same, and if they weren't, it introduced a stylesheet which made an adjustment, which is pretty much what the current site does, except we don't bother checking the rendering engine, just the default font settings. It seems that I'm a lot further off than I thought I was. :/ Back to the drawing board. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > >> > Here is what I think you want to see. ?This is my screen with > >> > proportional and fixed set to 14, with javascript turned off: > >> > > >> > ? ? ? ?http://momjian.us/tmp/pg_manual3.png > >> > > >> > That looks right. ?This is Ubuntu with Firefox 3.6.9. > >> > > >> > And this is the display of my font defaults: > >> > > >> > ? ? ? ?http://momjian.us/tmp/fonts.png > >> > > >> > Does that help? > >> > >> Yes, according to these, the proportional font defaults to 16px: > >> http://nst.sourceforge.net/nst/docs/tech/ch02s04.html > >> http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK > > > > OK, so why are we not just setting the fixed-width font to be the same > > size as the proportional? ?We are mixing the two types in the same > > paragraph, so why not just set them the same? ?Using the javascript I > > sent, you can probe for the preference settings. > > That javascript checked whether the values were the same, and if they > weren't, it introduced a stylesheet which made an adjustment, which is > pretty much what the current site does, except we don't bother > checking the rendering engine, just the default font settings. > > It seems that I'm a lot further off than I thought I was. :/ Back to > the drawing board. I think the original adjustment was flawed because it just set a percentage change. I think we can just assign them to match, rather than trying to figure out what percentage to use. Where is the javascript file? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
Tom Lane wrote: > Thom Brown <thom@linux.com> writes: > > On 10 September 2010 21:25, Tom Lane <tgl@sss.pgh.pa.us> wrote: > >> I still think the colored backgrounds are overused, and there's way too > >> much bold text. �AFAICS all the fixed-width-font text is now bold too, > >> to no improvement in readability. > > > The fixed-width-font text in examples shouldn't appear bold as I've > > applied no such style to those. > > Hmm. I'm seeing nothing but boldface on Magnus' pages, using Safari. > > > How about this: > > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html > > Curiously, that *doesn't* show as bold. > > > The intention of making a clear visual distinction between normal > > paragraph sections and examples/output was to make it easier to scan. > > Agreed, but a little goes a long way. If you overdo it, it'll grate on > the reader after twenty or thirty pages. The example blocks on this > page are suitably subtle, maybe even a bit too far --- I find that > depending on what angle I look at my LCD display, the background color > can be practically indistinguishable from the rest of the page. But > at other angles there's a clear distinction. Yeah, I think you want the blocks to appear distinct, but not chop up the text that is above/below it --- not an easy task. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 10 September 2010 21:59, Bruce Momjian <bruce@momjian.us> wrote: > Thom Brown wrote: >> >> > Here is what I think you want to see. ?This is my screen with >> >> > proportional and fixed set to 14, with javascript turned off: >> >> > >> >> > ? ? ? ?http://momjian.us/tmp/pg_manual3.png >> >> > >> >> > That looks right. ?This is Ubuntu with Firefox 3.6.9. >> >> > >> >> > And this is the display of my font defaults: >> >> > >> >> > ? ? ? ?http://momjian.us/tmp/fonts.png >> >> > >> >> > Does that help? >> >> >> >> Yes, according to these, the proportional font defaults to 16px: >> >> http://nst.sourceforge.net/nst/docs/tech/ch02s04.html >> >> http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK >> > >> > OK, so why are we not just setting the fixed-width font to be the same >> > size as the proportional? ?We are mixing the two types in the same >> > paragraph, so why not just set them the same? ?Using the javascript I >> > sent, you can probe for the preference settings. >> >> That javascript checked whether the values were the same, and if they >> weren't, it introduced a stylesheet which made an adjustment, which is >> pretty much what the current site does, except we don't bother >> checking the rendering engine, just the default font settings. >> >> It seems that I'm a lot further off than I thought I was. :/ Back to >> the drawing board. > > I think the original adjustment was flawed because it just set a > percentage change. I think we can just assign them to match, rather > than trying to figure out what percentage to use. Where is the > javascript file? It wasn't introduced for the existing site, but you can grab it from here: http://pgweb.darkixion.com:8081/media/js/monospacefix.js http://pgweb.darkixion.com:8081/media/css/monospacefixes.css -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > >> That javascript checked whether the values were the same, and if they > >> weren't, it introduced a stylesheet which made an adjustment, which is > >> pretty much what the current site does, except we don't bother > >> checking the rendering engine, just the default font settings. > >> > >> It seems that I'm a lot further off than I thought I was. :/ ?Back to > >> the drawing board. > > > > I think the original adjustment was flawed because it just set a > > percentage change. ?I think we can just assign them to match, rather > > than trying to figure out what percentage to use. ?Where is the > > javascript file? > > It wasn't introduced for the existing site, but you can grab it from here: > > http://pgweb.darkixion.com:8081/media/js/monospacefix.js > http://pgweb.darkixion.com:8081/media/css/monospacefixes.css OK, that helps, and I see the issue now. You basically have no way to push the calculation up into the style sheet. There are several options: 1. Use document.styleSheet[] to push changes into the style sheet 2. Use javascript to set a computed "style" string for every item with a specific class. 3. Use a single monospacefixes.css file 4. Use one of several monospacefixes.css files based on a computed value #1 and #2 are overkill, I think. #3 is what we have now, and is obviously deficient. #4 seems good to me; using http://pgweb.darkixion.com:8081/media/js/monospacefix.js I would add:if (propSize / monoSize >= 1.8) document.write('<styletype="text/css" media="screen">@import "/media/css/monospacefixes1_8.css";</style>\n');else if (propSize/ monoSize >= 1.4) document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_4.css";</style>\n');elseif (propSize / monoSize >= 1.2) document.write('<style type="text/css"media="screen">@import "/media/css/monospacefixes1_2.css";</style>\n');else if (propSize / monoSize >= 1.1) document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_1.css";</style>\n'); That requires the creation of four small style sheets, and should cover most of the common ratios. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
Bruce Momjian wrote: > > It wasn't introduced for the existing site, but you can grab it from here: > > > > http://pgweb.darkixion.com:8081/media/js/monospacefix.js > > http://pgweb.darkixion.com:8081/media/css/monospacefixes.css > > OK, that helps, and I see the issue now. You basically have no way to > push the calculation up into the style sheet. There are several > options: > > 1. Use document.styleSheet[] to push changes into the style sheet > 2. Use javascript to set a computed "style" string for every item with > a specific class. > 3. Use a single monospacefixes.css file > 4. Use one of several monospacefixes.css files based on a computed value > > #1 and #2 are overkill, I think. #3 is what we have now, and is > obviously deficient. #4 seems good to me; using > http://pgweb.darkixion.com:8081/media/js/monospacefix.js I would add: > > if (propSize / monoSize >= 1.8) > document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_8.css";</style>\n'); > else if (propSize / monoSize >= 1.4) > document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_4.css";</style>\n'); > else if (propSize / monoSize >= 1.2) > document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_2.css";</style>\n'); > else if (propSize / monoSize >= 1.1) > document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_1.css";</style>\n'); Duh, I don't know what I was thinking. This is the way to do it --- create the CSS inline rather than link in a file: if (monoSize != propSize){ document.write(' <style type="text/css"> #docContainer tt, #docContainer pre, #docContainer code { font-size: ' + (propSize / monoSize) + 'em; } </style> ');} -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On Fri, 2010-09-10 at 15:32 -0400, Alvaro Herrera wrote: > Excerpts from Magnus Hagander's message of vie sep 10 15:20:41 -0400 2010: > > > Ok, Thom has merged these changes onto the current website code, and > > I've put it up on http://magnus.webdev.postgresql.org - note that you > > need to go back to 8.3 to actually see anything, because the db > > doesn't have the latest docs. > > > > It'd be good to get some more testing on that one to make sure it > > doean't break things - in particular, someone to test IE and Safari. > > It should work, but better safe than sorry... > > Just gave it a try on IE6 on Linux here, and it looks OK. No rounded > corners and no shadows, but nothing's out of place either. IE8 on Win7 looks good. No rounded corners but that is as expected. JD > > -- > Álvaro Herrera <alvherre@commandprompt.com> > The PostgreSQL Company - Command Prompt, Inc. > PostgreSQL Replication, Consulting, Custom Development, 24x7 support > -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On Fri, 2010-09-10 at 21:11 +0100, Thom Brown wrote: > On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > Magnus Hagander <magnus@hagander.net> writes: > >> Ok, Thom has merged these changes onto the current website code, and > >> I've put it up on http://magnus.webdev.postgresql.org - note that you > >> need to go back to 8.3 to actually see anything, because the db > >> doesn't have the latest docs. > > > > You know, I hate to say it, but this is just way too busy for my taste. > > The typography is screaming "look at me" instead of staying out of the > > way. I can't imagine wanting to read any large amount of docs in this > > presentation. > > > > Maybe I'm in the minority. > > Do you have any specific recommendations? What would tone it down? I find it very pleasant. It reminds me of the python docs which are nice to read as well. Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On Fri, 2010-09-10 at 16:42 -0400, Bruce Momjian wrote: > Thom Brown wrote: > > Yes, according to these, the proportional font defaults to 16px: > > http://nst.sourceforge.net/nst/docs/tech/ch02s04.html > > http://answers.yahoo.com/question/index?qid=20080211131203AA6CErK > > OK, so why are we not just setting the fixed-width font to be the same > size as the proportional? We are mixing the two types in the same > paragraph, so why not just set them the same? Using the javascript I > sent, you can probe for the preference settings. Generally monospace is kept smaller than proportional (if it is larger, that is wrong). Mono is easy to read all the way down to 8pt (not that I am recommending that). Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
Joshua D. Drake wrote: > On Fri, 2010-09-10 at 21:11 +0100, Thom Brown wrote: > > On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > > Magnus Hagander <magnus@hagander.net> writes: > > >> Ok, Thom has merged these changes onto the current website code, and > > >> I've put it up on http://magnus.webdev.postgresql.org - note that you > > >> need to go back to 8.3 to actually see anything, because the db > > >> doesn't have the latest docs. > > > > > > You know, I hate to say it, but this is just way too busy for my taste. > > > The typography is screaming "look at me" instead of staying out of the > > > way. I can't imagine wanting to read any large amount of docs in this > > > presentation. > > > > > > Maybe I'm in the minority. > > > > Do you have any specific recommendations? What would tone it down? > > I find it very pleasant. It reminds me of the python docs which are nice > to read as well. I know you like the Python docs, and I think they _look_ good too at a glance, but see this page: http://docs.python.org/c-api/memory.html#examples The many horizontal lines really break up the text blocks and make me feel like I am reading through a mine field or obstacle course. This might be part of the issue we are dealing with now in the new format. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 11 September 2010 01:34, Bruce Momjian <bruce@momjian.us> wrote: > Bruce Momjian wrote: >> > It wasn't introduced for the existing site, but you can grab it from here: >> > >> > http://pgweb.darkixion.com:8081/media/js/monospacefix.js >> > http://pgweb.darkixion.com:8081/media/css/monospacefixes.css >> >> OK, that helps, and I see the issue now. You basically have no way to >> push the calculation up into the style sheet. There are several >> options: >> >> 1. Use document.styleSheet[] to push changes into the style sheet >> 2. Use javascript to set a computed "style" string for every item with >> a specific class. >> 3. Use a single monospacefixes.css file >> 4. Use one of several monospacefixes.css files based on a computed value >> >> #1 and #2 are overkill, I think. #3 is what we have now, and is >> obviously deficient. #4 seems good to me; using >> http://pgweb.darkixion.com:8081/media/js/monospacefix.js I would add: >> >> if (propSize / monoSize >= 1.8) >> document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_8.css";</style>\n'); >> else if (propSize / monoSize >= 1.4) >> document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_4.css";</style>\n'); >> else if (propSize / monoSize >= 1.2) >> document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_2.css";</style>\n'); >> else if (propSize / monoSize >= 1.1) >> document.write('<style type="text/css" media="screen">@import "/media/css/monospacefixes1_1.css";</style>\n'); > > Duh, I don't know what I was thinking. This is the way to do it --- > create the CSS inline rather than link in a file: > > if (monoSize != propSize) > { > document.write(' > <style type="text/css"> > #docContainer tt, > #docContainer pre, > #docContainer code { > font-size: ' + (propSize / monoSize) + 'em; > } > </style> '); > } I came up with something very similar last night but I hadn't finished playing with it: http://pgweb.darkixion.com:8081/media/js/monospacefix.js But yes, I really didn't like the idea of having individual CSS files for varying differences of font type when it could just be calculated.I'll try to get time to test it properly this weekend. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > I came up with something very similar last night but I hadn't finished > playing with it: > http://pgweb.darkixion.com:8081/media/js/monospacefix.js > > But yes, I really didn't like the idea of having individual CSS files > for varying differences of font type when it could just be calculated. > I'll try to get time to test it properly this weekend. Nice. Do you need those paragraphs with "x"? Do they show up in the document? Does it work if they have no text node? If fixed with is larger then proportional, we leave it alone? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
Bruce Momjian wrote: > Thom Brown wrote: > > I came up with something very similar last night but I hadn't finished > > playing with it: > > http://pgweb.darkixion.com:8081/media/js/monospacefix.js > > > > But yes, I really didn't like the idea of having individual CSS files > > for varying differences of font type when it could just be calculated. > > I'll try to get time to test it properly this weekend. > > Nice. Do you need those paragraphs with "x"? Do they show up in the > document? Does it work if they have no text node? If fixed with is > larger then proportional, we leave it alone? Oh, and I checked your website and the text looks perfect: http://momjian.us/tmp/pg_manual6.png -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 11 September 2010 14:41, Bruce Momjian <bruce@momjian.us> wrote: > Thom Brown wrote: >> I came up with something very similar last night but I hadn't finished >> playing with it: >> http://pgweb.darkixion.com:8081/media/js/monospacefix.js >> >> But yes, I really didn't like the idea of having individual CSS files >> for varying differences of font type when it could just be calculated. >> I'll try to get time to test it properly this weekend. > > Nice. Do you need those paragraphs with "x"? I could probably get away with non-breaking spaces, but the purpose is to prevent some browsers from throwing the element away as an empty one, thereby preventing us having a sample to test against. IE tends to do this in some scenarios. > Do they show up in the document? The x's would if the user had javascript enabled and CSS disabled. > Does it work if they have no text node? Hmm... I don't know what this means. Could you explain? > If fixed with is larger then proportional, we leave it alone? Yes, this needs to be worked on too which is why I hadn't finished with it. The script doesn't yield text sizes in Opera, so the calculation won't fall into the "if" statement anyway, but yes, in theory someone might actually have their monospace default set higher than proportional... although that would be weird. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > On 11 September 2010 14:41, Bruce Momjian <bruce@momjian.us> wrote: > > Thom Brown wrote: > >> I came up with something very similar last night but I hadn't finished > >> playing with it: > >> http://pgweb.darkixion.com:8081/media/js/monospacefix.js > >> > >> But yes, I really didn't like the idea of having individual CSS files > >> for varying differences of font type when it could just be calculated. > >> ?I'll try to get time to test it properly this weekend. > > > > Nice. ?Do you need those paragraphs with "x"? > > I could probably get away with non-breaking spaces, but the purpose is > to prevent some browsers from throwing the element away as an empty > one, thereby preventing us having a sample to test against. IE tends > to do this in some scenarios. OK, I never tested if this would work (no text): document.write('<pre id="monotest" style="display: none;"></pre>');document.write('<p id="paratest" style="display: none;"></p>'); Ah, I see your 'display: none' now that prevents it from appearing --- good idea. > > Do they show up in the document? > > The x's would if the user had javascript enabled and CSS disabled. Seems they would have even worse problem than a few X's if they turned off CSS, so it seems fine. > > Does it work if they have no text node? > > Hmm... I don't know what this means. Could you explain? See above, and you already thought of that. > > If fixed with is larger then proportional, we leave it alone? > > Yes, this needs to be worked on too which is why I hadn't finished > with it. The script doesn't yield text sizes in Opera, so the > calculation won't fall into the "if" statement anyway, but yes, in > theory someone might actually have their monospace default set higher > than proportional... although that would be weird. Yep, odd, but seems worth fixing if it is just a simple != change. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On Fri, Sep 10, 2010 at 4:25 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > Thom Brown <thom@linux.com> writes: >> On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: >>> You know, I hate to say it, but this is just way too busy for my taste. >>> The typography is screaming "look at me" instead of staying out of the >>> way. I can't imagine wanting to read any large amount of docs in this >>> presentation. >>> >>> Maybe I'm in the minority. > >> Do you have any specific recommendations? What would tone it down? > > I still think the colored backgrounds are overused, and there's way too > much bold text. AFAICS all the fixed-width-font text is now bold too, > to no improvement in readability. It's way too much boldface for my taste, too. I am not sure I agree that the colored backgrounds are overused. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise Postgres Company
Robert Haas wrote: > On Fri, Sep 10, 2010 at 4:25 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: > > Thom Brown <thom@linux.com> writes: > >> On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: > >>> You know, I hate to say it, but this is just way too busy for my taste. > >>> The typography is screaming "look at me" instead of staying out of the > >>> way. ?I can't imagine wanting to read any large amount of docs in this > >>> presentation. > >>> > >>> Maybe I'm in the minority. > > > >> Do you have any specific recommendations? ?What would tone it down? > > > > I still think the colored backgrounds are overused, and there's way too > > much bold text. ?AFAICS all the fixed-width-font text is now bold too, > > to no improvement in readability. > > It's way too much boldface for my taste, too. I am not sure I agree > that the colored backgrounds are overused. I think you should look at Thom's version, which is more current: http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On Sat, Sep 11, 2010 at 5:10 PM, Bruce Momjian <bruce@momjian.us> wrote: >> It's way too much boldface for my taste, too. I am not sure I agree >> that the colored backgrounds are overused. > > I think you should look at Thom's version, which is more current: > > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html Oh, pardon me. That version looks good to me in general, but this page does not: http://pgweb.darkixion.com:8081/docs/8.4/static/sql-dropdatabase.html -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise Postgres Company
On 11 September 2010 22:10, Bruce Momjian <bruce@momjian.us> wrote: > Robert Haas wrote: >> On Fri, Sep 10, 2010 at 4:25 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote: >> > Thom Brown <thom@linux.com> writes: >> >> On 10 September 2010 21:05, Tom Lane <tgl@sss.pgh.pa.us> wrote: >> >>> You know, I hate to say it, but this is just way too busy for my taste. >> >>> The typography is screaming "look at me" instead of staying out of the >> >>> way. ?I can't imagine wanting to read any large amount of docs in this >> >>> presentation. >> >>> >> >>> Maybe I'm in the minority. >> > >> >> Do you have any specific recommendations? ?What would tone it down? >> > >> > I still think the colored backgrounds are overused, and there's way too >> > much bold text. ?AFAICS all the fixed-width-font text is now bold too, >> > to no improvement in readability. >> >> It's way too much boldface for my taste, too. I am not sure I agree >> that the colored backgrounds are overused. > > I think you should look at Thom's version, which is more current: > > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html Right, I've had a moment to made some changes. I've tweaked the javascript and went with the != comparison and changes. Also I think I *was* mistaken with not introducing bold for a set of keywords. Obviously I sneaked that in at some point to see how it would look but didn't recall doing so. So the latest changes are on my version (like your link above) again. Have I missed anyone's existing recommendations out? How's it look in Konqueror? I think it does look less brash than before; I think Tom was right that it was too in-your-face. Does it need toning down more? Other suggestions? -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On 11 September 2010 22:41, Robert Haas <robertmhaas@gmail.com> wrote: > On Sat, Sep 11, 2010 at 5:10 PM, Bruce Momjian <bruce@momjian.us> wrote: >>> It's way too much boldface for my taste, too. I am not sure I agree >>> that the colored backgrounds are overused. >> >> I think you should look at Thom's version, which is more current: >> >> http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html > > Oh, pardon me. That version looks good to me in general, but this > page does not: > > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-dropdatabase.html Fixed. Thanks for the spot. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
Thom Brown wrote: > > ? ? ? ?http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html > > Right, I've had a moment to made some changes. I've tweaked the > javascript and went with the != comparison and changes. Also I > think I *was* mistaken with not introducing bold for a set of > keywords. Obviously I sneaked that in at some point to see how it > would look but didn't recall doing so. > > So the latest changes are on my version (like your link above) again. > Have I missed anyone's existing recommendations out? How's it look in > Konqueror? I think it does look less brash than before; I think Tom > was right that it was too in-your-face. Does it need toning down > more? Other suggestions? Looks great to me! -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On fre, 2010-09-10 at 16:05 -0400, Tom Lane wrote: > You know, I hate to say it, but this is just way too busy for my > taste. The typography is screaming "look at me" instead of staying > out of the way. I can't imagine wanting to read any large amount of > docs in this presentation. I agree. It looks like angry fruit salad.
On fre, 2010-09-10 at 21:11 +0100, Thom Brown wrote: > Do you have any specific recommendations? What would tone it down? There are just too many things going on. Reduce the number of colors, reduce the number of different border designs (round, not round), make the borders less busy (remove the line), remove the shadows, reduce the number of different levels of indentation. It's a great CSS show case, and perhaps it's a good design for a web site (or perhaps not), but lengthy documentation text should probably be kept very simple.
On 13 September 2010 18:47, Peter Eisentraut <peter_e@gmx.net> wrote: > On fre, 2010-09-10 at 21:11 +0100, Thom Brown wrote: >> Do you have any specific recommendations? What would tone it down? > > There are just too many things going on. Reduce the number of colors, > reduce the number of different border designs (round, not round), make > the borders less busy (remove the line), remove the shadows, reduce the > number of different levels of indentation. > > It's a great CSS show case, and perhaps it's a good design for a web > site (or perhaps not), but lengthy documentation text should probably be > kept very simple. It sounds like you're recommending starting from scratch. I think maybe someone else should give it a try instead then. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On Mon, Sep 13, 2010 at 1:51 PM, Thom Brown <thom@linux.com> wrote: > On 13 September 2010 18:47, Peter Eisentraut <peter_e@gmx.net> wrote: >> On fre, 2010-09-10 at 21:11 +0100, Thom Brown wrote: >>> Do you have any specific recommendations? What would tone it down? >> >> There are just too many things going on. Reduce the number of colors, >> reduce the number of different border designs (round, not round), make >> the borders less busy (remove the line), remove the shadows, reduce the >> number of different levels of indentation. >> >> It's a great CSS show case, and perhaps it's a good design for a web >> site (or perhaps not), but lengthy documentation text should probably be >> kept very simple. > > It sounds like you're recommending starting from scratch. I think > maybe someone else should give it a try instead then. Are we sure Peter is looking at the latest version? Because the latest version is hardly a crazy mess of business. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise Postgres Company
On Mon, 2010-09-13 at 18:51 +0100, Thom Brown wrote: > On 13 September 2010 18:47, Peter Eisentraut <peter_e@gmx.net> wrote: > > On fre, 2010-09-10 at 21:11 +0100, Thom Brown wrote: > >> Do you have any specific recommendations? What would tone it down? > > > > There are just too many things going on. Reduce the number of colors, > > reduce the number of different border designs (round, not round), make > > the borders less busy (remove the line), remove the shadows, reduce the > > number of different levels of indentation. > > > > It's a great CSS show case, and perhaps it's a good design for a web > > site (or perhaps not), but lengthy documentation text should probably be > > kept very simple. > > It sounds like you're recommending starting from scratch. I think > maybe someone else should give it a try instead then. I think Thom has gone down the right path here and that with respect, the curmudgeons need to go back to their fire and tea. Our current docs are pathetic visual representations of a quality product. The "only" change I could see at this point in the path that "might" improve the work that he is done, is to drop the shadow. The colors allow the information to be give priority to the information that is supposed to have priority. The indentation is directly applicable to the indentation of the content itself. Everything is rounded (that I saw) except tables. What *exactly* is the problem here? Oh and: Thanks to Bruce for actually working *with* the process and helping Thom based on his feedback. Sincerely, Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On mån, 2010-09-13 at 14:04 -0400, Robert Haas wrote: > Are we sure Peter is looking at the latest version? Because the > latest version is hardly a crazy mess of business. Well, I'm catching up with a couple of days of emails and are clicking on the various links that are being posted, and I suspect the content at those locations is being updated as things get changed.
Peter Eisentraut <peter_e@gmx.net> wrote: > There are just too many things going on. Reduce the number of > colors, reduce the number of different border designs (round, not > round), make the borders less busy (remove the line), remove the > shadows, reduce the number of different levels of indentation. Personally, I would be disappointed if we backed off from these changes much. The current docs look like something I would have seen coming off the bidirectional print-head of a Teletype(R) machine a few decades ago. I've gone back and forth between that and Thom's latest draft trying to pick out useful information on various pages, and for me the visual cues help considerably in finding what I want. On top of that, switching to the new look is a refreshing experience each time; when I switch back to our current docs, I can actually feel my eyes tighten up as they prepare to pick out content from a mass of uniform text. I know people process visual information differently, but I really have a hard time seeing anything offensive about where Thom has taken it, and that was based on a lot of feedback from several different people. -Kevin
On mån, 2010-09-13 at 18:51 +0100, Thom Brown wrote: > It sounds like you're recommending starting from scratch. Let's say I recommend doing a lot less.
Excerpts from Kevin Grittner's message of lun sep 13 14:17:08 -0400 2010: > Peter Eisentraut <peter_e@gmx.net> wrote: > > > There are just too many things going on. Reduce the number of > > colors, reduce the number of different border designs (round, not > > round), make the borders less busy (remove the line), remove the > > shadows, reduce the number of different levels of indentation. > > Personally, I would be disappointed if we backed off from these > changes much. The current docs look like something I would have > seen coming off the bidirectional print-head of a Teletype(R) > machine a few decades ago. +1000 -- Álvaro Herrera <alvherre@commandprompt.com> The PostgreSQL Company - Command Prompt, Inc. PostgreSQL Replication, Consulting, Custom Development, 24x7 support
Kevin Grittner wrote: > Peter Eisentraut <peter_e@gmx.net> wrote: > > > There are just too many things going on. Reduce the number of > > colors, reduce the number of different border designs (round, not > > round), make the borders less busy (remove the line), remove the > > shadows, reduce the number of different levels of indentation. > > Personally, I would be disappointed if we backed off from these > changes much. The current docs look like something I would have > seen coming off the bidirectional print-head of a Teletype(R) > machine a few decades ago. I've gone back and forth between that Agreed. > and Thom's latest draft trying to pick out useful information on > various pages, and for me the visual cues help considerably in > finding what I want. On top of that, switching to the new look is a > refreshing experience each time; when I switch back to our current > docs, I can actually feel my eyes tighten up as they prepare to pick > out content from a mass of uniform text. > > I know people process visual information differently, but I really > have a hard time seeing anything offensive about where Thom has > taken it, and that was based on a lot of feedback from several > different people. Agreed. FYI, everyone should be looking at the same version: http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html If we make these changes, there is nothing in stone preventing us from tweaking it later, but I think there is general agreement that the new format is an improvement. We have to accept we will never make everyone happy. The only constructive suggestion I can make to perhaps make more people happy would be to go with only vertical lines to highlight the code blocks, e.g. instead of: +-------------+| text |+-------------+ we do| text | but I am not sure if I even would prefer that. FYI, I would like to see this rolled into our production web server in the next few days so it is ready for the 9.0 release. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On Mon, 2010-09-13 at 17:32 -0400, Bruce Momjian wrote: > FYI, I would like to see this rolled into our production web server in > the next few days so it is ready for the 9.0 release. +1 Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
All, I really like the callout and table formatting in the example format. Looks very nice and modern. -- -- Josh Berkus PostgreSQL Experts Inc. http://www.pgexperts.com
On Mon, 2010-09-13 at 18:51 +0100, Thom Brown wrote: > On 13 September 2010 18:47, Peter Eisentraut <peter_e@gmx.net> wrote: > > On fre, 2010-09-10 at 21:11 +0100, Thom Brown wrote: > >> Do you have any specific recommendations? What would tone it down? > > > > There are just too many things going on. Reduce the number of colors, > > reduce the number of different border designs (round, not round), make > > the borders less busy (remove the line), remove the shadows, reduce the > > number of different levels of indentation. > > > > It's a great CSS show case, and perhaps it's a good design for a web > > site (or perhaps not), but lengthy documentation text should probably be > > kept very simple. > > It sounds like you're recommending starting from scratch. I think > maybe someone else should give it a try instead then. I think Thom has gone down the right path here and that with respect, the curmudgeons need to go back to their fire and tea. Our current docs are pathetic visual representations of a quality product. The "only" change I could see at this point in the path that "might" improve the work that he is done, is to drop the shadow. The colors allow the information to be give priority to the information that is supposed to have priority. The indentation is directly applicable to the indentation of the content itself. Everything is rounded (that I saw) except tables. What *exactly* is the problem here? Oh and: Thanks to Bruce for actually working *with* the process and helping Thom based on his feedback. Sincerely, Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On Mon, 2010-09-13 at 17:32 -0400, Bruce Momjian wrote: > FYI, I would like to see this rolled into our production web server in > the next few days so it is ready for the 9.0 release. +1 Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On 13/09/2010 23:23, Josh Berkus wrote: > All, > > I really like the callout and table formatting in the example format. > Looks very nice and modern. +1 to all the positive comments - I think it's a great improvement. In fact, I'd take it further and make the sub-section headings (4.1.1, etc) a more prominent colour, so that they stand out more when scanning a page at high speed. I like the rollover effect on tables also - it'll make large tables easier to scan. Good one Thom et al! Ray. -- Raymond O'Donnell :: Galway :: Ireland rod@iol.ie
On 14 September 2010 10:23, Raymond O'Donnell <rod@iol.ie> wrote: > On 13/09/2010 23:23, Josh Berkus wrote: >> >> All, >> >> I really like the callout and table formatting in the example format. >> Looks very nice and modern. > > +1 to all the positive comments - I think it's a great improvement. > > In fact, I'd take it further and make the sub-section headings (4.1.1, etc) > a more prominent colour, so that they stand out more when scanning a page at > high speed. > > I like the rollover effect on tables also - it'll make large tables easier > to scan. > > Good one Thom et al! I think a consensus needs to be reached on what you guys want changing. Implementing changes isn't a problem, it's what changes should be implemented. Whatever is decided, it needs to be easy to visually scan the page and its sections, clean and consistent. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On Tue, Sep 14, 2010 at 9:58 AM, Thom Brown <thom@linux.com> wrote: > I think a consensus needs to be reached on what you guys want > changing. Implementing changes isn't a problem, it's what changes > should be implemented. Whatever is decided, it needs to be easy to > visually scan the page and its sections, clean and consistent. It seems like you're giving up just when we've just about got consensus. Please don't. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise Postgres Company
On Tue, 2010-09-14 at 14:58 +0100, Thom Brown wrote: > I think a consensus needs to be reached on what you guys want > changing. Implementing changes isn't a problem, it's what changes > should be implemented. Whatever is decided, it needs to be easy to > visually scan the page and its sections, clean and consistent. IMO. We are done for now. Call it 1.0 and release. This *can* be changed, easily. Let the wider community consume it for a while, see how we feel about it after we actually use it. Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On 14 September 2010 17:16, Robert Haas <robertmhaas@gmail.com> wrote: > On Tue, Sep 14, 2010 at 9:58 AM, Thom Brown <thom@linux.com> wrote: >> I think a consensus needs to be reached on what you guys want >> changing. Implementing changes isn't a problem, it's what changes >> should be implemented. Whatever is decided, it needs to be easy to >> visually scan the page and its sections, clean and consistent. > > It seems like you're giving up just when we've just about got > consensus. Please don't. No, not at all. :) I'm just suggesting that rather than people making a suggestion, and me going ahead and putting that change in straight away (like I have been doing), it should have the support of more than a single person first, otherwise there'll be lots of tweaking and un-tweaking without any real agreement. If you wish to go ahead with what we've currently got, I'll create another patch for Magnus to stage, otherwise what requires changing needs to be decided. -- Thom Brown Twitter: @darkixion IRC (freenode): dark_ixion Registered Linux user: #516935
On Tue, Sep 14, 2010 at 12:29 PM, Thom Brown <thom@linux.com> wrote: > On 14 September 2010 17:16, Robert Haas <robertmhaas@gmail.com> wrote: >> On Tue, Sep 14, 2010 at 9:58 AM, Thom Brown <thom@linux.com> wrote: >>> I think a consensus needs to be reached on what you guys want >>> changing. Implementing changes isn't a problem, it's what changes >>> should be implemented. Whatever is decided, it needs to be easy to >>> visually scan the page and its sections, clean and consistent. >> >> It seems like you're giving up just when we've just about got >> consensus. Please don't. > > No, not at all. :) I'm just suggesting that rather than people > making a suggestion, and me going ahead and putting that change in > straight away (like I have been doing), it should have the support of > more than a single person first, otherwise there'll be lots of > tweaking and un-tweaking without any real agreement. > > If you wish to go ahead with what we've currently got, I'll create > another patch for Magnus to stage, otherwise what requires changing > needs to be decided. Oh, OK. Well, +1 for pushing live what's currently at: http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html It's probably not perfect, but I think it's better than what we have now, and that's sufficient. Damn the torpedoes, full speed ahead. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise Postgres Company
On Tue, 2010-09-14 at 12:16 -0400, Robert Haas wrote: > On Tue, Sep 14, 2010 at 9:58 AM, Thom Brown <thom@linux.com> wrote: > > I think a consensus needs to be reached on what you guys want > > changing. Implementing changes isn't a problem, it's what changes > > should be implemented. Whatever is decided, it needs to be easy to > > visually scan the page and its sections, clean and consistent. > > It seems like you're giving up just when we've just about got > consensus. Please don't. I am at a point, assuming the patch is decent, and it is consistent with the current website (which I will be reviewing) to commit the changes myself. Thom has more than enough support for the changes he has made. Sincerely, Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
Robert Haas <robertmhaas@gmail.com> writes: > On Tue, Sep 14, 2010 at 12:29 PM, Thom Brown <thom@linux.com> wrote: >> If you wish to go ahead with what we've currently got, I'll create >> another patch for Magnus to stage, otherwise what requires changing >> needs to be decided. > Oh, OK. Well, +1 for pushing live what's currently at: > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html > It's probably not perfect, but I think it's better than what we have > now, and that's sufficient. Damn the torpedoes, full speed ahead. Yeah, let's just go with that version. I concur with the upthread comment that we should try living with the results for awhile before we engage in more tweaking. regards, tom lane
On Tue, Sep 14, 2010 at 2:58 PM, Thom Brown <thom@linux.com> wrote: > On 14 September 2010 10:23, Raymond O'Donnell <rod@iol.ie> wrote: >> On 13/09/2010 23:23, Josh Berkus wrote: >>> >>> All, >>> >>> I really like the callout and table formatting in the example format. >>> Looks very nice and modern. >> >> +1 to all the positive comments - I think it's a great improvement. >> >> In fact, I'd take it further and make the sub-section headings (4.1.1, etc) >> a more prominent colour, so that they stand out more when scanning a page at >> high speed. >> >> I like the rollover effect on tables also - it'll make large tables easier >> to scan. >> >> Good one Thom et al! > > I think a consensus needs to be reached on what you guys want > changing. Implementing changes isn't a problem, it's what changes > should be implemented. Whatever is decided, it needs to be easy to > visually scan the page and its sections, clean and consistent. Don't worry about changing headings. Their style is dictated by the rest of the site anyway - we're not going to make them look different in the docs. -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise Postgres Company
On Tue, Sep 14, 2010 at 9:37 AM, Robert Haas <robertmhaas@gmail.com> wrote: > On Tue, Sep 14, 2010 at 12:29 PM, Thom Brown <thom@linux.com> wrote: >> On 14 September 2010 17:16, Robert Haas <robertmhaas@gmail.com> wrote: >>> On Tue, Sep 14, 2010 at 9:58 AM, Thom Brown <thom@linux.com> wrote: >>>> I think a consensus needs to be reached on what you guys want >>>> changing. Implementing changes isn't a problem, it's what changes >>>> should be implemented. Whatever is decided, it needs to be easy to >>>> visually scan the page and its sections, clean and consistent. >>> >>> It seems like you're giving up just when we've just about got >>> consensus. Please don't. >> >> No, not at all. :) I'm just suggesting that rather than people >> making a suggestion, and me going ahead and putting that change in >> straight away (like I have been doing), it should have the support of >> more than a single person first, otherwise there'll be lots of >> tweaking and un-tweaking without any real agreement. >> >> If you wish to go ahead with what we've currently got, I'll create >> another patch for Magnus to stage, otherwise what requires changing >> needs to be decided. > > Oh, OK. Well, +1 for pushing live what's currently at: > > http://pgweb.darkixion.com:8081/docs/8.4/static/sql-syntax-lexical.html > > It's probably not perfect, but I think it's better than what we have > now, and that's sufficient. Damn the torpedoes, full speed ahead. +1 Thanks, Thom, for putting the time and energy into this. It's lovely to see. I've been following along, and I think what we have now is a very nice improvement. Once it is pushed live, I recommend publishing a blog post and soliciting feedback about the changes. Not necessarily to change anything right away, but just to collect feedback. -selena -- http://chesnok.com/daily - me
On Tue, 2010-09-14 at 14:58 +0100, Thom Brown wrote: > I think a consensus needs to be reached on what you guys want > changing. Implementing changes isn't a problem, it's what changes > should be implemented. Whatever is decided, it needs to be easy to > visually scan the page and its sections, clean and consistent. IMO. We are done for now. Call it 1.0 and release. This *can* be changed, easily. Let the wider community consume it for a while, see how we feel about it after we actually use it. Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt
On Tue, 2010-09-14 at 12:16 -0400, Robert Haas wrote: > On Tue, Sep 14, 2010 at 9:58 AM, Thom Brown <thom@linux.com> wrote: > > I think a consensus needs to be reached on what you guys want > > changing. Implementing changes isn't a problem, it's what changes > > should be implemented. Whatever is decided, it needs to be easy to > > visually scan the page and its sections, clean and consistent. > > It seems like you're giving up just when we've just about got > consensus. Please don't. I am at a point, assuming the patch is decent, and it is consistent with the current website (which I will be reviewing) to commit the changes myself. Thom has more than enough support for the changes he has made. Sincerely, Joshua D. Drake -- PostgreSQL.org Major Contributor Command Prompt, Inc: http://www.commandprompt.com/ - 509.416.6579 Consulting, Training, Support, Custom Development, Engineering http://twitter.com/cmdpromptinc | http://identi.ca/commandprompt