Tech Insider					     Technology and Trends


		       Geeklog Mailing List Archives

From tony at tonybibbs.com  Fri Aug 20 10:53:34 2004
From: tony at tonybibbs.com (Tony Bibbs)
Date: Fri, 20 Aug 2004 09:53:34 -0500
Subject: [geeklog-devel] Input From PJ of Groklaw
Message-ID: <4126106E.9040505@tonybibbs.com>

I recently was made aware of some issues Groklaw was having.  Seems as 
though Stories and commennts come up missing sporadically.  I contacted 
PJ and she says she's been working with Rakaz (she calls him MathFox). 
on the issues. If anybody sees Rakaz in IRC they should bring this up 
and see how we might help.  Also, any changes related to performance 
that we have made we should share with Groklaw immediate so they might 
modify their existing code (vinny I think I saw CVS commits from you in 
particular related to performance and comments).

Aside from that PJ had a number of feature requests so I'm putting them 
here so that they will be considered in either of the branches:

1) PJ likes to have a couple versions of the same page loaded.  In one 
she uses to view the article, in the othe she edits it.  She'd like away 
to do this in one page.
2) Comment moderation has been mentioned numerous times.  However, her 
request is that admins would be able to moderate the comments.  I think 
that is a bit more realistic (and useful).
3) She would like to mark bad comments as hidden (basic trolling).  
Naturally they would show up for the admin but not the regular users.
4) She would like quotations, blocks of text in an article, to 
optionally have a slightly darker shading than the rest of the page, so 
it would be clearly a quotation, without having to use quotation marks. 
Blaine can we borrow the forum code for this?
5) She'd like pagination of commments.  This is a good idea as it would 
help performance.
6) She'd like users to edit their own comments and even delete them if 
no one has responded to it yet.
7) Similar to #1 she wants to be able to edit and review a static page 
in one sweep.
8) Add the ability to convert a PDF to plain text (great idea to 
compliment our new PDF generation)
9) This one is hard to rearticulate so I'm quoting her exactly: "It 
would be nice if there could be a work area, where everyone could work 
on a document together, which I could let certain groups have access 
to.  It's the single hardest part of Groklaw, the difficulty of working 
together fast when we're all over the world.  I know I could do it now 
if I give everyone the power, but for security reasons, it has to be a 
walled off area, so they can't wander about.  My readers know how to do 
everything, and sometimes we get volunteers with an agenda, so I need to 
be careful.  When we do audio transcripts, for example, we typically 
break it up into mp3s and then everyone gets sent one. It would be 
pleasant to be able to have folks just come and grab what they want and 
have a place to input the result on Groklaw, instead of it all being 
funneled through me."

Obviously some of these we've seen before.  Some we haven't.  I think 
quite a few are low hanging fruit and others would take quite a bit of 
work.  Nonetheless, I think we should consider these.  I know in a truly 
democratic system (which this isn't necessarily) we give all users the 
same say but I think we would be wise to at least address a few of these 
ideas as I'd hate to see Groklaw change platforms.  To be clear, 
changing platforms wasn't mentioned in the message she sent me but I 
can't imagine it wasn't thought off considering they have content 
disappearing.

Also, this may be a good time to try recruiting more help off of 
geeklog.net.  I was surprised at the help I got for Auth_Enterprise by 
nagging people there.  I think if we create a request for help and list 
some of these request (along with any prexisting issues on the project 
site) and drop Groklaw's name we might get more help because....well, 
nobody really wants to see SCO win this thing and helping, even if in 
this indirect way, would probably motivate people.

Thoughts, questions or concerns?

--Tony

From: tony at tonybibbs.com (Tony Bibbs)
Date: Fri, 20 Aug 2004 10:36:46 -0500
Subject: [geeklog-devel] [Fwd: Re: Geeklog and Groklaw]
Message-ID: <41261A8E.4030906@tonybibbs.com>

Some clarification to the list I sent.

--Tony

-------- Original Message --------
Subject: 	Re: Geeklog and Groklaw
Date: 	Fri, 20 Aug 2004 15:21:50 +0000
From: 	Pamela Jones <pj at groklaw.com>
To: 	Tony Bibbs <tony at tonybibbs.com>
References: 	<200408182042.i7IKgpu32509 at statler.metalab.unc.edu> 
<69B65C55-F17F-11D8-9895-000D93C0C6E2 at groklaw.com> 
<412610B2.1000109 at tonybibbs.com>



Hi Tony,

Couple of things:  rakaz and mathfox are two different people, but both
work on Groklaw. But mathfox is the webmaster.

Second, on the request you list as number one:  I need to explain
better.  What mathfox has said is that if I simultaneously open up two
copies of the same page of Groklaw, it can cause problems.   I would
like to be able to do that, because of the reasons I mentioned. But on
that request (as distinct from my request re static pages), I am not
asking that it be viewable on the same page. Two open at the same time
without crashing.

Also, on public requests for help. Please remove the moderator request
regarding hiding.  I have enough diligent Utah people attacking me for
moderating as it is. I'd appreciate keeping that part more private.  

Thanks,

PJ 

On Fri, 2004-08-20 at 14:54, Tony Bibbs wrote:
> I've forwarded your input on to our developers and cc'd you both in.  
> PJ, as you think of more things feel free to send them to me.
> 
> Thanks,
> 
> --Tony
> 
> Pamela Jones wrote:
> 
> > Hi Tony,
> >
> > Thank you for Geeklog. I've enjoyed it a lot. I'm thrilled to hear you 
> > are working on making it scale, and just in time!   I think there are 
> > some issues having to do with scale that we are starting to run into.  
> > MathFox has done some  personalizing, as you probably know.  But with 
> > a thousand comments on a story yesterday, well...it's a lot.
> >
> > He thinks maybe I caused the problem with the disappearing stories, by 
> > having three different pages of Groklaw open at the same time. I don't 
> > think that is it, though, because I wasn't working on Groklaw at the 
> > time, and was doing email instead when it happened. On the other hand, 
> > he's usually right. It might be a good idea if you contact him and you 
> > two can talk with more specifics. I'm cc'ing him with this.
> >
> > Also, comments are starting to disappear sometimes.  I don't know 
> > why.  We are analyzing the logs now, so I can't really answer your 
> > question as to why yet.  We have issues with bandwidth because ibiblio 
> > rations us, and sometimes it's hard to know what is a real problem and 
> > what is just an ibiblio issue.  But when the disappearing stories went 
> > poof, all three turned up as drafts, but early versions, missing huge 
> > chunks, and with comments attached, but not necessarily attached to 
> > the right story.  When I looked at the stats, it would say 6 hits, 27 
> > comments, which is impossible, obviously, or at least inconceivable, 
> > and there wouldn't be 27 comments in any case, so while some data was 
> > saved, it was garbled.
> >
> >  I'm generally a problem.  I do like to open several pages at once and 
> > work that way, and MathFox scolds me about that. If you could make it 
> > so I could do that, I'd be thrilled. For example, I like to open an 
> > article and go down the comments and have a second window open with 
> > the page where I can edit any mistakes I've made and readers catch.
> >
> > Also, in a perfect world, I'd be able to go through comments, flag the 
> > really good ones and they'd show up in a special place at the top of 
> > the list, a kind of Best of the Comments on this Article, so really 
> > busy folks could just read those.  I'd like it so I could "anoint" a 
> > crew to handle that choosing. With me being able to overrule them.
> >
> > I'd also love it if I could hide comments and have them show up 
> > nowhere unless I made them visible again.  The troll dance.  That way 
> > moderators could hide a comment, and I could later go through and 
> > either confirm by deleting or restore, if they go too far.  But they'd 
> > have to disappear from search and every form of view.
> >
> > Another thing I'd love would be if we could have it so quotations, 
> > blocks of text in an article, could optionally have a slightly darker 
> > shading than the rest of the page, so it would be clearly a quotation, 
> > without having to use quotation marks.
> >
> > It would be nice if there could be a work area, where everyone could 
> > work on a document together, which I could let certain groups have 
> > access to.  It's the single hardest part of Groklaw, the difficulty of 
> > working together fast when we're all over the world.  I know I could 
> > do it now if I give everyone the power, but for security reasons, it 
> > has to be a walled off area, so they can't wander about.  My readers 
> > know how to do everything, and sometimes we get volunteers with an 
> > agenda, so I need to be careful.  When we do audio transcripts, for 
> > example, we typically break it up into mp3s and then everyone gets 
> > sent one. It would be pleasant to be able to have folks just come and 
> > grab what they want and have a place to input the result on Groklaw, 
> > instead of it all being funneled through me.
> >
> > I'd love it if there was a way to have comments up to a certain number 
> > and then they'd click to read the rest.  When we get 400 plus per 
> > story, and sometimes more than twice that, it slows things down quite 
> > a lot.  If the only way to do it is a la Slashdot, though, I'd rather 
> > stick to what we have already.  I hate the Slashdot method, because it 
> > repeats way too many comments on the next page.
> >
> > Sometimes readers ask for the ability to edit their own comments or 
> > even delete them.
> >
> > It would be very nice if the static pages could be viewed in edit. For 
> > example, if I want to edit the IBM timeline page, I have to open two 
> > windows to be able to see what I'm doing, because in edit, you see 
> > only the html, not the finished look, and you really want both.
> >
> > Someone is trying to work out a  way to write a script to quickly 
> > change a pdf to plain text.  If it were built in, my whole life would 
> > get better.
> >
> > Sorry you asked yet?
> >
> > : )
> >
> > When we figure out what happened with the disappearing story, I'll 
> > write to you again and let you know.
> >
> > I love the idea of working together, because there will surely be more 
> > Groklaws in the future.  Mathfox  has, I believe, donated some stuff 
> > back. But some things we kind of keep under wraps for the moment, 
> > mainly because I'd rather the astroturfers not know all our tricks.
> >
> > PJ
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > On Aug 18, 2004, at 4:42 PM, tony_bibbs wrote:
> >
> >> PJ,
> >>
> >> I'm Tony Bibbs, one of the maintainers for Geeklog.  We have watched 
> >> Groklaw grow and have been excited to see it has, for the most part, 
> >> grown with the site.  I noticed you have run into a problem with lost 
> >> stories and I'd like to get some insight into the cause so we might 
> >> address this should it prove to be an issue with the Geeklog code or 
> >> the Geeklog database.
> >>
> >> I've read many of the posts related to this and everybody seemed to 
> >> have an opinion on how to improve things.  Mine is quite simply, let 
> >> us work with you to address any issues you might have...it only helps 
> >> us make Geeklog better and you get, I hope, the reassurance we value 
> >> you and your site.
> >>
> >> So that you know, I'm responsible for the next version of Geeklog 
> >> (version 2) and it is going to be built with large scale deployments 
> >> in mind and will include an upgrade path from 1.3.x.  To that end if 
> >> you have any suggestions on how we might improve Geeklog we would 
> >> love any feedback you might have.
> >>
> >> Thanks,
> >>
> >> --Tony
> >>
> 
> 

From: vfuria at gmail.com (Vincent Furia)
Date: Fri, 20 Aug 2004 11:30:42 -0400
Subject: [geeklog-devel] Input From PJ of Groklaw
In-Reply-To: <4126106E.9040505@tonybibbs.com>
References: <4126106E.9040505@tonybibbs.com>
Message-ID: <8319e2d604082008307b8a49bd@mail.gmail.com>

5)  Comment Pagination: Already checked into CVS.  And yes, big
performance improvement on stories with LOTS of comments.

I haven't seen Rakaz around in a while, but next time I see him I'll
let him know the details of all the performance improvements I've
added.

I can't imagine what would cause stories and comments to diappear. 
There a couple places where, using my imagination, I could see
multiple Admins editing or adding an item at the same time could cause
problems, but that is more likely just some corruption rather than
"disappearance".

-Vinny

On Fri, 20 Aug 2004 09:53:34 -0500, Tony Bibbs <tony at tonybibbs.com> wrote:
> I recently was made aware of some issues Groklaw was having.  Seems as
> though Stories and commennts come up missing sporadically.  I contacted
> PJ and she says she's been working with Rakaz (she calls him MathFox).
> on the issues. If anybody sees Rakaz in IRC they should bring this up
> and see how we might help.  Also, any changes related to performance
> that we have made we should share with Groklaw immediate so they might
> modify their existing code (vinny I think I saw CVS commits from you in
> particular related to performance and comments).
> 
> Aside from that PJ had a number of feature requests so I'm putting them
> here so that they will be considered in either of the branches:
> 
> 1) PJ likes to have a couple versions of the same page loaded.  In one
> she uses to view the article, in the othe she edits it.  She'd like away
> to do this in one page.
> 2) Comment moderation has been mentioned numerous times.  However, her
> request is that admins would be able to moderate the comments.  I think
> that is a bit more realistic (and useful).
> 3) She would like to mark bad comments as hidden (basic trolling).
> Naturally they would show up for the admin but not the regular users.
> 4) She would like quotations, blocks of text in an article, to
> optionally have a slightly darker shading than the rest of the page, so
> it would be clearly a quotation, without having to use quotation marks.
> Blaine can we borrow the forum code for this?
> 5) She'd like pagination of commments.  This is a good idea as it would
> help performance.
> 6) She'd like users to edit their own comments and even delete them if
> no one has responded to it yet.
> 7) Similar to #1 she wants to be able to edit and review a static page
> in one sweep.
> 8) Add the ability to convert a PDF to plain text (great idea to
> compliment our new PDF generation)
> 9) This one is hard to rearticulate so I'm quoting her exactly: "It
> would be nice if there could be a work area, where everyone could work
> on a document together, which I could let certain groups have access
> to.  It's the single hardest part of Groklaw, the difficulty of working
> together fast when we're all over the world.  I know I could do it now
> if I give everyone the power, but for security reasons, it has to be a
> walled off area, so they can't wander about.  My readers know how to do
> everything, and sometimes we get volunteers with an agenda, so I need to
> be careful.  When we do audio transcripts, for example, we typically
> break it up into mp3s and then everyone gets sent one. It would be
> pleasant to be able to have folks just come and grab what they want and
> have a place to input the result on Groklaw, instead of it all being
> funneled through me."
> 
> Obviously some of these we've seen before.  Some we haven't.  I think
> quite a few are low hanging fruit and others would take quite a bit of
> work.  Nonetheless, I think we should consider these.  I know in a truly
> democratic system (which this isn't necessarily) we give all users the
> same say but I think we would be wise to at least address a few of these
> ideas as I'd hate to see Groklaw change platforms.  To be clear,
> changing platforms wasn't mentioned in the message she sent me but I
> can't imagine it wasn't thought off considering they have content
> disappearing.
> 
> Also, this may be a good time to try recruiting more help off of
> geeklog.net.  I was surprised at the help I got for Auth_Enterprise by
> nagging people there.  I think if we create a request for help and list
> some of these request (along with any prexisting issues on the project
> site) and drop Groklaw's name we might get more help because....well,
> nobody really wants to see SCO win this thing and helping, even if in
> this indirect way, would probably motivate people.
> 
> Thoughts, questions or concerns?
> 
> --Tony
> _______________________________________________
> geeklog-devel mailing list
> geeklog-devel at lists.geeklog.net
> http://lists.geeklog.net/listinfo/geeklog-devel
>

From: mathfox at xs4all.nl (Peter Roozemaal)
Date: Fri, 20 Aug 2004 17:46:41 +0200
Subject: [geeklog-devel] Re: Input From PJ of Groklaw
In-Reply-To: < 4126106E.9040505@tonybibbs.com>
References: < 4126106E.9040505@tonybibbs.com>
Message-ID: < 41261CE1.7060806@xs4all.nl>

Hello all,

Tony Bibbs wrote:
> I recently was made aware of some issues Groklaw was having.  Seems as 
> though Stories and commennts come up missing sporadically.  I contacted 
> PJ and she says she's been working with Rakaz (she calls him MathFox).

Rakaz (Niels) is another person than me.

> on the issues. If anybody sees Rakaz in IRC they should bring this up 
> and see how we might help.  Also, any changes related to performance 
> that we have made we should share with Groklaw immediate so they might 
> modify their existing code (vinny I think I saw CVS commits from you in 
> particular related to performance and comments).

The database is the most serious performance bottleneck. We have 7?
webservers talking to a single database and Geeklog does a lot of
queries per page. We have a few stories with over 1000 comments!

Another issue is that regular expressions can take huge amounts of CPU
time when stories go large. PJ writes stories that don't allways fit in
64k, so I enlarged the bodytext field to mediumtext (16M). The regular
expression match in COM_extractLinks caused a time-out in stories that
had more that 50 links. (footnotes and back). BTW, we dropped the
"what's related" box.

> Aside from that PJ had a number of feature requests so I'm putting them 
> here so that they will be considered in either of the branches:
> 
[...]
> 
> Obviously some of these we've seen before.  Some we haven't.  I think 
> quite a few are low hanging fruit and others would take quite a bit of 
> work.  Nonetheless, I think we should consider these.  I know in a truly 
> democratic system (which this isn't necessarily) we give all users the 
> same say but I think we would be wise to at least address a few of these 
> ideas as I'd hate to see Groklaw change platforms.  To be clear, 
> changing platforms wasn't mentioned in the message she sent me but I 
> can't imagine it wasn't thought off considering they have content 
> disappearing.

I have been looking for a suitable platform, but couldn't find another
one that
  1. was open source
  2. had a decent security record
  3. could be installed easily
  4. had a good management interface
and I'ld hate to convert databases to a new environment.

> Also, this may be a good time to try recruiting more help off of 
> geeklog.net.  I was surprised at the help I got for Auth_Enterprise by 
> nagging people there.  I think if we create a request for help and list 
> some of these request (along with any prexisting issues on the project 
> site) and drop Groklaw's name we might get more help because....well, 
> nobody really wants to see SCO win this thing and helping, even if in 
> this indirect way, would probably motivate people.

Greetings,
	Peter.

From: tony at tonybibbs.com (Tony Bibbs)
Date: Fri, 20 Aug 2004 10:58:37 -0500
Subject: [geeklog-devel] Re: Input From PJ of Groklaw
In-Reply-To: <41261CE1.7060806@xs4all.nl>
References: <4126106E.9040505@tonybibbs.com> <41261CE1.7060806@xs4all.nl>
Message-ID: <41261FAD.7070400@tonybibbs.com>

> Rakaz (Niels) is another person than me.

Sorry about that.  Us techies get caught up in IRC/system names instead 
of real names it's not hard to do.  Apologies to you and Niels.

> The database is the most serious performance bottleneck. We have 7?
> webservers talking to a single database and Geeklog does a lot of
> queries per page. We have a few stories with over 1000 comments!

Yeah, this is clearly a problem.  We have recently implemented use of 
PHP sessions.  We didn't do much with it other than store important 
things like $_USER in it but now that we have that we need to 
investigate using the single query made to populate $_SESSION form the 
database in ways that will prevent the lots of little queries that have 
cluttered the code.  Given that we need to start reviewing the code again.

Also, have you considered adding another database server and load 
balancing them?  I think MySQL's replication has reached a point were 
you could do this reliably.  Obvious it makes things more complex from 
the administration side but it might be worth investigating.

>
> Another issue is that regular expressions can take huge amounts of CPU
> time when stories go large. PJ writes stories that don't allways fit in
> 64k, so I enlarged the bodytext field to mediumtext (16M). The regular
> expression match in COM_extractLinks caused a time-out in stories that
> had more that 50 links. (footnotes and back). BTW, we dropped the
> "what's related" box.

Hrm, good idea. Dirk, it might be worth upgrading that field.  Yeah, the 
regex's would be a killer on larger stories.  Thanks, we'll review it 
and see how we can fix this.

> I have been looking for a suitable platform, but couldn't find another
> one that
>  1. was open source
>  2. had a decent security record
>  3. could be installed easily
>  4. had a good management interface
> and I'ld hate to convert databases to a new environment.

Well, like I said, we want to be sure that Geeklog grows as the needs of 
the communities grow.  Groklaw is an outlier in terms of size but our 
goal is to support sites that become this successful.  I hope that by 
working together we can get to a point where Geeklog is sleeker, meaner 
and cleaner.  All this reminds me of where Mozilla is with Thunderbird 
and Firefox.  I think Geeklog is to a point where we need to bust out 
with code analyzing an realling streamlining things.  Any chance we 
might be able to get a copy of the Groklaw DB?  I know that is asking a 
lot but it would help us a lot.

--Tony

From: tony at tonybibbs.com (Tony Bibbs)
Date: Fri, 20 Aug 2004 10:58:37 -0500
Subject: [geeklog-devel] Re: Input From PJ of Groklaw
In-Reply-To: <41261CE1.7060806@xs4all.nl>
References: <4126106E.9040505@tonybibbs.com> <41261CE1.7060806@xs4all.nl>
Message-ID: <41261FAD.7070400@tonybibbs.com>

> Rakaz (Niels) is another person than me.

Sorry about that.  Us techies get caught up in IRC/system names instead 
of real names it's not hard to do.  Apologies to you and Niels.

> The database is the most serious performance bottleneck. We have 7?
> webservers talking to a single database and Geeklog does a lot of
> queries per page. We have a few stories with over 1000 comments!

Yeah, this is clearly a problem.  We have recently implemented use of 
PHP sessions.  We didn't do much with it other than store important 
things like $_USER in it but now that we have that we need to 
investigate using the single query made to populate $_SESSION form the 
database in ways that will prevent the lots of little queries that have 
cluttered the code.  Given that we need to start reviewing the code again.

Also, have you considered adding another database server and load 
balancing them?  I think MySQL's replication has reached a point were 
you could do this reliably.  Obvious it makes things more complex from 
the administration side but it might be worth investigating.

>
> Another issue is that regular expressions can take huge amounts of CPU
> time when stories go large. PJ writes stories that don't allways fit in
> 64k, so I enlarged the bodytext field to mediumtext (16M). The regular
> expression match in COM_extractLinks caused a time-out in stories that
> had more that 50 links. (footnotes and back). BTW, we dropped the
> "what's related" box.

Hrm, good idea. Dirk, it might be worth upgrading that field.  Yeah, the 
regex's would be a killer on larger stories.  Thanks, we'll review it 
and see how we can fix this.

> I have been looking for a suitable platform, but couldn't find another
> one that
>  1. was open source
>  2. had a decent security record
>  3. could be installed easily
>  4. had a good management interface
> and I'ld hate to convert databases to a new environment.

Well, like I said, we want to be sure that Geeklog grows as the needs of 
the communities grow.  Groklaw is an outlier in terms of size but our 
goal is to support sites that become this successful.  I hope that by 
working together we can get to a point where Geeklog is sleeker, meaner 
and cleaner.  All this reminds me of where Mozilla is with Thunderbird 
and Firefox.  I think Geeklog is to a point where we need to bust out 
with code analyzing an realling streamlining things.  Any chance we 
might be able to get a copy of the Groklaw DB?  I know that is asking a 
lot but it would help us a lot.

--Tony

From: mathfox at xs4all.nl (Peter Roozemaal)
Date: Fri, 20 Aug 2004 18:29:36 +0200
Subject: [geeklog-devel] Re: Input From PJ of Groklaw
In-Reply-To: <41261FAD.7070400@tonybibbs.com>
References: <4126106E.9040505@tonybibbs.com> <41261CE1.7060806@xs4all.nl> 
<41261FAD.7070400@tonybibbs.com>
Message-ID: <412626F0.6020304@xs4all.nl>

Tony Bibbs wrote:

>> The database is the most serious performance bottleneck. We have 7?
>> webservers talking to a single database and Geeklog does a lot of
>> queries per page. We have a few stories with over 1000 comments!
> 
> Yeah, this is clearly a problem.  We have recently implemented use of 
> PHP sessions.  We didn't do much with it other than store important 
> things like $_USER in it but now that we have that we need to 
> investigate using the single query made to populate $_SESSION form the 
> database in ways that will prevent the lots of little queries that have 
> cluttered the code.  Given that we need to start reviewing the code again.

PHP session variables create a load on the central file server. We can
not store them locally on the webserver as the loadbalancer will move
sessions from machine to machine. (rethinks) It might be possible if the
session variable only serves as cache for database data.

> Also, have you considered adding another database server and load 
> balancing them?  I think MySQL's replication has reached a point were 
> you could do this reliably.  Obvious it makes things more complex from 
> the administration side but it might be worth investigating.

We need a bit of help from your side in separating read-only database
operations from operations that modify the database. The config files
would need two database names, etc.

>> Another issue is that regular expressions can take huge amounts of CPU
>> time when stories go large. PJ writes stories that don't allways fit in
>> 64k, so I enlarged the bodytext field to mediumtext (16M). The regular
>> expression match in COM_extractLinks caused a time-out in stories that
>> had more that 50 links. (footnotes and back). BTW, we dropped the
>> "what's related" box.
> 
> Hrm, good idea. Dirk, it might be worth upgrading that field.  Yeah, the 
> regex's would be a killer on larger stories.  Thanks, we'll review it 
> and see how we can fix this.
> 
>> I have been looking for a suitable platform, but couldn't find another
>> one that
>>  1. was open source
>>  2. had a decent security record
>>  3. could be installed easily
>>  4. had a good management interface
>> and I'ld hate to convert databases to a new environment.
> 
> Well, like I said, we want to be sure that Geeklog grows as the needs of 
> the communities grow.  Groklaw is an outlier in terms of size but our 
> goal is to support sites that become this successful.  I hope that by 
> working together we can get to a point where Geeklog is sleeker, meaner 
> and cleaner.  All this reminds me of where Mozilla is with Thunderbird 
> and Firefox.  I think Geeklog is to a point where we need to bust out 
> with code analyzing an realling streamlining things.  Any chance we 
> might be able to get a copy of the Groklaw DB?  I know that is asking a 
> lot but it would help us a lot.

I can give you a sanitized version of the database if there is a place
where I can upload a 45 Mbyte of compressed data. I need a few hours to
clean it from personal data.

Greetings,
	Peter.

From: tony at tonybibbs.com (Tony Bibbs)
Date: Fri, 20 Aug 2004 12:50:44 -0500
Subject: [geeklog-devel] Re: Input From PJ of Groklaw
In-Reply-To: <412626F0.6020304@xs4all.nl>
References: <4126106E.9040505@tonybibbs.com> <41261CE1.7060806@xs4all.nl> 
<41261FAD.7070400@tonybibbs.com> <412626F0.6020304@xs4all.nl>
Message-ID: <412639F4.80809@tonybibbs.com>

Peter Roozemaal wrote:

> PHP session variables create a load on the central file server. We can
> not store them locally on the webserver as the loadbalancer will move
> sessions from machine to machine. (rethinks) It might be possible if the
> session variable only serves as cache for database data.

Using files versus a DB is the admin's decision.  In your case you'd 
clearly need the DB.

> We need a bit of help from your side in separating read-only database
> operations from operations that modify the database. The config files
> would need two database names, etc.

Hrm, haven't thought about this much.  Deserves some more thought from 
all of us.  Also, have you turned on query caching in MySQL.  I'm 
guessing you have but that would help a lot.

> I can give you a sanitized version of the database if there is a place
> where I can upload a 45 Mbyte of compressed data. I need a few hours to
> clean it from personal data.

k, Let me know when you get it ready and I'll let you SCP it to one of 
our servers.

--Tony

			        About USENET

USENET (Users’ Network) was a bulletin board shared among many computer
systems around the world. USENET was a logical network, sitting on top
of several physical networks, among them UUCP, BLICN, BERKNET, X.25, and
the ARPANET. Sites on USENET included many universities, private companies
and research organizations. See USENET Archives.

		       SCO Files Lawsuit Against IBM

March 7, 2003 - The SCO Group filed legal action against IBM in the State 
Court of Utah for trade secrets misappropriation, tortious interference, 
unfair competition and breach of contract. The complaint alleges that IBM 
made concentrated efforts to improperly destroy the economic value of 
UNIX, particularly UNIX on Intel, to benefit IBM's Linux services 
business. See SCO v IBM.

The materials and information included in this website may only be used
for purposes such as criticism, review, private study, scholarship, or
research.

Electronic mail:			       WorldWideWeb:
   tech-insider@outlook.com			  http://tech-insider.org/