Evergreen ILS Website

IRC log for #evergreen, 2014-05-28

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
04:58 pinesol_green Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html>
06:46 timf joined #evergreen
07:44 collum joined #evergreen
08:16 rjackson-isl joined #evergreen
08:20 mrpeters joined #evergreen
08:22 akilsdonk joined #evergreen
08:26 Dyrcona joined #evergreen
08:39 Dyrcona joined #evergreen
08:44 ericar joined #evergreen
08:47 yboston joined #evergreen
08:51 mrpeters is there a handy way to dump srfsh output to text instead of dumping to screen?
08:52 mrpeters maybe just creating a .srfsh file and feeding it to srfsh and > foo.log
08:52 mrpeters yes indeed...that works (for the logs)
08:55 mrpeters rjackson-isl++ jboyer-isl++
08:56 jboyer-isl You seeing folder sharing problems at other installations, or just being cautious?
08:57 mrpeters no, seeing them
08:57 mrpeters a big list of them for a newly migrated library
08:57 jboyer-isl Good times tracking them down the first time.
08:58 mrpeters oh yeah its a nightmare, thats why i was hoping you still had my nagios checks
08:58 mrpeters but i can see you've even improved upon them
08:58 jboyer-isl I do like to tinker.
09:08 RoganH joined #evergreen
09:13 kbeswick joined #evergreen
09:17 dbwells joined #evergreen
09:17 csharp so is anyone out there in Evergreen land running their staff workstations in virtualized environments?  I'm specifically interested in whether anyone's running Windows on VMWare, but I wanted to cast a wide net
09:18 RoganH I have one library that runs on the MS platform (can't remember it's name).
09:18 csharp one of our member libraries is complaining of constant freezes and needs for restarts and I was looking for a comparison point
09:19 RoganH I know a few more that have talked about it.  I'll check to see where they are in their testing.
09:19 csharp RoganH: thanks!
09:19 RoganH Actually, now that I think about it, we have another who has been virtual for a long time and they might be VMWare.  I tend to forget about them because they did it before it was cool.
09:20 RoganH They had a really top notch IT guy at the time.
09:20 csharp yeah - these guys have been on it for a few years - it was a county IT departmental decision, and it apparently works well everywhere except for the library :-/
09:20 phasefx csharp: I put together a vmplayer image once with xubuntu and an auto-started staff client, as a last ditch-workaround for the client crashing on a particular machine; not sure how that played out afterward
09:21 csharp phasefx: yeah, I would *love* to steer them towards something like Lubuntu, but that's almost certainly out of the question ;-)
09:21 * csharp reboots the Fedora buildslave
09:22 phasefx csharp: if it's all configured, they can just pretend it's a weird app :)
09:22 RoganH OK, I'm seeing a very narrow problem with searching for email.  Can I get anyone else to test it in their staff clients?
09:22 csharp heh - yeah, I'll see if I can sneak that in ;-)
09:22 RoganH What happens is if you search for an email that starts with an 'a.' will fail.
09:23 RoganH So, for example, a.smith@generic.net will generate an error.  Or A.smith@generic.net
09:23 csharp @eightball what will happen if I type 'google' into Google?
09:23 pinesol_green csharp: Yes!
09:23 RoganH But it's that narrow, b.smith@generic.net will.
09:23 bshum RoganH: On my staff client, if I key in "a" (just that) into the email field, it works fine
09:24 RoganH or aa.smith@generic.net will work.
09:24 bshum My guess is a timeout from overabundunt results
09:24 bshum I've seen that happen before on the city field
09:24 jeff bshum: what about searching for a.
09:24 bshum When using our most populated city
09:24 bshum Ah that, hmm
09:24 csharp we had a problem with email searches until we added an index on them
09:24 bshum Yes, I get hits for "a." as well
09:25 RoganH bshum: I get results for "a." as well but not "a.email@address.net"
09:25 csharp RoganH: that worked ok for me too
09:25 jeff RoganH: what is the error, and is there any corresponding error in the database logs or opensrf logs?
09:25 bshum Well I wouldn't expect any results for that, given that I have no patrons with that style of address :)
09:25 bshum For me anyways
09:26 bshum But I get the usual "no patrons found matching search criteria" popup when I try it.
09:26 RoganH Well, what prompted this is a valid email address that is in a record.
09:26 RoganH But, made up ones cause it too.
09:26 jeff RoganH: that was my next question. thanks.
09:27 RoganH It is like it's timing out but that doesn't make sense since I can give it far more intensive searches and it responds fine.
09:27 jeff RoganH: that seems to rule out the likelihood of it being a malformed patron record tripping up the client when it tries to display one of the results matching your email address criteria.
09:27 jeff searching in 2.5.1 for email a.smith@example.com i get the usual "no patrons found" message, no errors.
09:27 RoganH We've been playing around to narrow down the criteria.  It's odd.
09:28 RoganH Specifically I get a 500 network error.
09:28 jeff and server side, what is the error?
09:28 RoganH I tested this last night at 11 pm when there was no server load.
09:29 RoganH I haven't gone that far yet.  I was looking to see if others knew of anything similar but it sounds like a unique to us thing now.
09:30 RoganH The weird ones are kind of fun so long as they don't affect too many situations.  :)
09:30 jeff RoganH: not that I think that method has changed much recently, but what version are you on?
09:30 RoganH jeff: 2.5.3
09:30 jeff thanks.
09:32 bshum Well, I modded one of my test accounts to include a style of email like a.test@somewhere.com
09:32 bshum And while it took awhile to search, it did retrieve the patron in search
09:33 bshum Might just be a timing issue of some sort.  Where the DB takes just a little too long, and the service stops paying attention?
09:33 jeff server logs are going to be the next best clue.
09:33 bshum Indeed.
09:34 RoganH bshum: that's what I'm thinking but yeah, just odd since it coughs up responses to just "a" or "a." with no problem but we've all seen weirder
09:35 jeff can you tell me the exact string that you were searching on that does fail?
09:35 jeff i think you did above, but i just wanted to be sure.
09:35 RoganH There are days I would accept that it's because nibiru is too close in orbit or somehting.
09:52 Dyrcona RoganH: It's the polar shift.
09:52 RoganH Dyrcona++
10:02 jwoodard joined #evergreen
10:11 jeff RoganH: if you can get more detailed client logs or server logs, i'm intrigued by your a.something@example.com issue.
10:11 RoganH jeff:  I am too.  I'm slammed this morning with more critical things but I'll let you know what I find out.
10:14 * jeff nods
10:22 ldw joined #evergreen
10:24 afterl joined #evergreen
10:31 jeff another interesting challenge with linked/shared/cloned addresses: what happens when you receive returned mail for one patron?
10:31 jeff you could retrieve the patron that owns the address and uncheck the valid checkbox.
10:31 jeff (which is likely what i'll do here)
10:31 jeff but there isn't an easy way to end up with no address on the subordinate patron account.
10:32 jeff a "copy this shared address to this user" option would probably help.
10:33 csharp and possibly an ON DELETE trigger that does it automatically if the "master" address is deleted?
10:35 jeff or prevent deletion at that low level, and have something higher up handle it.
10:35 jeff the idea is neat, and could probably use some interface TLC
11:19 ldw joined #evergreen
11:21 Bmagic who is using keepalived ?
11:44 csharp bshum: can you catch me up on the status of 14.04 and Evergreen?
11:49 Dyrcona csharp: I can tell you that on the vms and other computers where we at MVLC have tried upgrading from 12.04 to 14.04, we've had horrible luck.
11:49 csharp yeah, I expected that from my early tests
11:50 csharp right now I'm going for clean install - I just didn't want to duplicate effort
11:50 Dyrcona Even going from 13.10 to 14.04 on my laptop left it in a state that it wouldn't boot.
11:50 csharp my "desktop" ubuntu upgrades went fine
11:50 csharp but we don't really upgrade
11:51 csharp we install the base OS and install debs from there
11:51 csharp though, right now I'm just working on building from source on 14.04
11:53 * csharp wants to start messing with upgrading dojo (coming from a very naive perspective, hoping to learn as I go ;-) )
11:53 Dyrcona csharp: Good luck with that.
11:55 * Dyrcona realizes that his lunch salad would be better if he had added apple slices.
11:57 csharp Dyrcona: :-D thanks
11:58 csharp if nothing else, I hope *I* will get a better understanding of the dojo pieces ;-)
11:58 Dyrcona csharp: Aren't the Dojo pieces going away with the web-based staff client?
11:58 csharp Dyrcona: no, as far as I understand, they are coming along with
11:59 * csharp invites berick to correct him, though
12:01 tsbere csharp: Given some of the headaches with previous attempts of my own to upgrade dojo I would recommend we remove it, not update it. >_>
12:02 * tsbere thinks he had found that we are using features that they did not continue to include in later versions
12:05 csharp ah - well if going forward with dojo isn't certain, I'll not spend too much time on it
12:06 csharp I'm just reacting to some UI glitch-type frustrations discovered when testing acq, and was hoping to ameliorate things
12:10 ericar joined #evergreen
12:12 csharp no mention of dojo in the agendas/minutes of dev meetings over the last year
12:12 csharp I guess with angular.js in the picture, that would be the preferred toolset, though, right?
12:13 csharp @ana TOO MANY TOOLSETS
12:13 pinesol_green csharp: Tamest loony to so
12:14 dbs angular is not a toolset the way that dojo is (at least, not in terms of widgets)
12:14 dbs It tries to be minimalist
12:14 bshum csharp: So, I've been a bit distracted as of late from getting Evergreen working with 14.04.
12:14 csharp bshum: that's okay - I finally have some time to devote, so I was going to give it a go
12:14 bshum At the very least, I'm happy enough that my postgresql database is moving along with 14.04 though.
12:14 csharp awesome ;-)
12:15 bshum We changed up the core, but I didn't trust Evergreen and the bricks quite yet
12:15 csharp bshum: is this on the server with SSDs?
12:15 bshum csharp: Yes.
12:15 csharp is it awesome?
12:15 bshum I'm fairly pleased so far.
12:15 csharp rock and roll
12:16 * csharp looks forward to receiving ours in the next couple/few weeks
12:17 bshum I had enough room to change all my GIST to GIN for metabib indexes.
12:17 bshum Not sure if it matters yet, but I like it :)
12:18 csharp GIN++
12:18 csharp we saw a marked difference after that switch
12:20 ericar joined #evergreen
12:20 bshum csharp: fwiw, I think we committed the initial changes for 14.04 to OpenSRF
12:21 bshum So there is that
12:21 mrpeters so no go on 14.04 at this time?
12:21 mrpeters was considering upgrading to 14.04 when we upgrade someone to 2.6
12:22 csharp mrpeters: I'm working on that right now
12:22 csharp bshum: did you have an EG branch you were working on? or was my branch the only one?
12:22 mrpeters yeah, cool.  i need to set up their testing server within the next week or two so i may have to just roll with 12.04 for that, and then 14.04 for production
12:23 bshum csharp: I think yours is the only one I know about.
12:23 mrpeters which i dont love, but we can always rebuild test server at a later date
12:23 csharp bshum: okay - cool - thanks!
12:23 berick csharp: i would love to replace the dojo UIs with angular versions, but there is no specific plan in place to do that yet
12:23 csharp berick: do you think it's best then to stay with old dojo for the foreseeable future?
12:25 berick csharp: well, if you have the time to update dojo, i don't think anyone will stop you.  my guess is your time could be better spent, though.
12:26 csharp berick: thanks - exactly what I was thinking about
12:28 zerick joined #evergreen
13:12 jihpringle joined #evergreen
13:13 pinesol_green [evergreen|Mike Rylander] LP#1310751: Reorder query-building wrapper for QueryParser - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=c3327de>
13:15 bshum marc--
13:17 csharp @karma marc
13:17 pinesol_green csharp: Karma for "marc" has been increased 0 times and decreased 18 times for a total karma of -18.
13:17 csharp marc--
13:18 Dyrcona marc-- # for an even -20
13:21 bshum A handful (less than a dozen I'm sure) bibs didn't export because they were too large to be handled with USMARC I guess.
13:21 bshum So now I have to re-run it as MARCXML to see if it'll nab them this time around.
13:22 csharp yeah - that happens here too
13:23 csharp I'm going to investigate eeevil's suggestion to identify those records and export them piecemeal to stay under the size limit
13:23 csharp identifying the records will be the hardest part of that
13:26 egbuilder build #594 of evergreen-master-fedora-18 is complete: Failure [failed test]  Build details are at http://testing.evergreen-ils.org/buildbot/bu​ilders/evergreen-master-fedora-18/builds/594  blamelist: Mike Rylander <mrylander@gmail.com>
13:26 Dyrcona I read that as "is complete failure" though guess it amounts to the same.
13:29 csharp that's after upgrading to Fedora 20, FYI
13:30 csharp looks like there are some deps that got tore up
13:30 Dyrcona Perl 5.18?
13:30 dbwells csharp: I'm double-checking the commit I just pushed.  Are you saying that probably isn't the problem?
13:30 Dyrcona My understanding is some tests and some CPAN modules's tests fail under Perl 5.18.
13:30 bshum Yeah there's some bugs for that.
13:30 csharp dbwells: I'm saying that it's likely an OS problem
13:31 csharp I'm looking now
13:31 * Dyrcona has a laugh at "modules's" expense. :)
13:32 csharp Dyrcona: perl 5.18, yes
13:33 bmills joined #evergreen
13:35 dbwells csharp: Do you know when the Ubuntu build slave will be online again?
13:36 pastebot "mrpeters" at 64.57.241.14 pasted "2.6.0 open-ils.circ.circ_modifier.retreieve_all 404" (6 lines) at http://paste.evergreen-ils.org/50
13:36 mrpeters anyone bump up against that when installing 2.6
13:36 csharp dbwells: oops - just a sec
13:36 mrpeters we're trying to sort our debs for 2.6 at archive.georgialibraries.org...i figured Circ.pm was the most likely culprit but the deb installs the same version as the tarball
13:37 dbwells csharp: (or Evergreen Master Debian, for that matter)
13:37 mrpeters and open-ils.circ is definetly running with listener/drone, etc.
13:37 csharp dbwells: I don't have control over the Debian VM
13:38 csharp however, with mundungus up and running, creating one under everyone's control is easy ;-)
13:38 dbwells csharp: that's fine, I'd just like to see one successful buildbot test before I package 2.6.1 today :)
13:38 Dyrcona mrpeters: Not seen that one.
13:38 csharp dbwells: ok, ubuntu 12.04 buildslave is back up
13:39 dbwells csharp: sweet, thank you
13:39 mrpeters isn't there an "introspect" command or something like that I can run in srfsh to see all of the api's for open-ils.circ that exist?
13:40 Dyrcona mrpeters: Yes. introspect open-ils.circ
13:41 mrpeters awesome, just wanted to make sure introspect was the right command
13:41 mrpeters definetly not there
13:42 mrpeters fwiw, i was still using the osrf_ctl.sh so perhaps that is the problem?
13:42 vlewis joined #evergreen
13:42 mrpeters we haven't programmed osrf_control into the debs yet
13:42 Dyrcona mrpeters: I don't think so. You're probably missing something under Circ/
13:42 Dyrcona or, maybe action/publisher.
13:43 mrpeters when you say under circ...do you mean the Circ.pm?
13:43 mrpeters root@ubuntu:/usr/local/share/perl​/5.14.2/OpenILS/Application# rgrep "open-ils.circ.circ_modifier.retrieve.all" *
13:43 mrpeters Circ.pm:    api_name => 'open-ils.circ.circ_modifier.retrieve.all');
13:43 Dyrcona Nope. I mean under OpenILS/Application/Circ/
13:43 gsams joined #evergreen
13:43 Dyrcona Well, there you go.
13:44 mrpeters aha, awesome.  i'll start diffing those
13:44 mrpeters is there any way that one api would die, but the rest of open-ils.circ would still be running?
13:45 jcamins Hey, does anyone happen to know why the EG project selected Dojo rather than jQuery?
13:45 mrpeters it came up first in a google search, jcamins :P
13:45 jcamins I'm having some trouble figuring out when to recommend Dojo instead of jQuery, and I thought someone here might have an answer.
13:45 jcamins mrpeters: fair enough.
13:45 Dyrcona mrpeters: I don't know. I've not seen that before.
13:46 mrpeters im totally kidding jcamins
13:46 Dyrcona mrpeters: Usually, what I see is I've made a typo in some code and the whole of open-ils.circ fails to load.
13:46 jcamins mrpeters: oh. Well... that's still a perfectly legitimate reason to choose a JS framework.
13:47 mrpeters Dyrcona: yeah, and i assume plenty have installed 2.6 from tarballs, so we just need to find our needle in the haystack :)
13:48 Dyrcona mrpeters: Well, I don't make that assumption, but I've installed it from git.
13:48 hbrennan joined #evergreen
13:48 csharp dbwells: fedora buildslave was updated and is ready for testing... I don't know how to trigger a rebuild
13:49 mrpeters yeah, ill fire up a new vm and install the old fashioned slow way :P
13:49 Dyrcona I typically only see problems like that when I'm fooling around in development and have edited code.
13:50 Dyrcona mrpeters: Try this: perl -c /usr/local/share/perl/5.14.2​/OpenILS/Application/Circ.pm
13:50 Dyrcona That'll report any syntax errors.
13:51 mrpeters oh snap
13:52 mrpeters http://pastie.org/9231545
13:52 mrpeters same on version of Circ.pm from tarball
13:53 mrpeters i think i better do an install from the tarball and find out if there is actually a bug
13:55 Dyrcona You're missing a dependency, mrpeters.
13:55 mrpeters even better, thats an easy fix
13:55 Dyrcona JSON::XS probably needs to be installed or one of those.
13:55 mrpeters Stripe.pm?
13:56 Dyrcona Well, make sure Stripe is installed, but it's complaining about JSON.pm.
13:56 mrpeters does that come from cpan, or would libjson-perl take care of it?
13:56 Dyrcona Ubuntu?
13:56 mrpeters yeah, 12.04
13:56 jeff if you have an evergreen system and JSON::XS is not installed, you've probably missed a step. :P
13:57 mrpeters oh, im sure we did.  if its a new dependency we probably missed adding it to the debs
13:57 Dyrcona I have libjson-xs-perl
13:57 mrpeters we compiled Stripe, but i think we're missing  libjson-xs-perl then
13:57 Dyrcona libjson-xs-perl isn't a new requirement.
13:58 mrpeters no?
13:58 Dyrcona Probably any of the libjson-perl modules would work.
13:58 Dyrcona No, not new.
13:58 mrpeters interesting, wonder why we are just bumping into this
13:58 mrpeters of course, we only built 2.5.2 debs and then jumped to 2.6.0
13:59 mrpeters weird thing is i have it already
13:59 mrpeters opensrf@ubuntu:~$ dpkg -l | grep json
13:59 mrpeters ii  libjson-xs-perl                                   2.320-1build1                module for manipulating JSON-formatted data (C/XS-accelerated)
13:59 mrpeters ii  python-simplejson                                 2.3.2-1                      simple, fast, extensible JSON encoder/decoder for Python
14:00 Dyrcona Stripe is requiring JSON.
14:01 jeff Evergreen typically does a "use JSON::XS;" and Business::Stripe does "use JSON;"
14:01 Dyrcona Looks like I have a JSON-2.90 directory in .cpan/build
14:01 mrpeters root@ubuntu:/etc/ld.so.conf.d# dpkg -l|grep stripe
14:01 mrpeters ii  libbusiness-onlinepayment-stripe-perl             0.04-1                       Interface for Stripe payment system.
14:02 Dyrcona mrpeters: I installed Stripe via CPAN.
14:02 mrpeters yeah, we built it into a deb
14:02 mrpeters thats probably our problem
14:02 dbwells csharp: I'm about to push in another commit, so that should trigger the build (I think).  Either way, I'm not too worried about it.  Thanks again.
14:03 jeff when JSON > 2.mumble is installed, "use JSON;" actually uses JSON::XS -- but if you don't have JSON installed, and only have JSON::XS installed, "use JSON;" will fail like that.
14:03 mrpeters Business::Stripe, correct?
14:04 jeff mrpeters: if you built it into a deb and didn't set the deps right, then you'll end up with what you have now.
14:04 Dyrcona mrpeters: installing libjson-perl might solve it for you.
14:04 mrpeters yeah, i think we've got it narrowed down
14:04 jeff I'd think that dh-make-perl should have handled that for you.
14:04 jeff Or didn't you use dh-make-perl for some reason?
14:04 mrpeters would have to defer to Andy on that...he built up the deb
14:05 bshum joined #evergreen
14:05 mceraso joined #evergreen
14:08 mrpeters Dyrcona++ jeff++ much appreciated
14:08 mrpeters you got us started down the right road now, we'll get this sorted
14:08 mrpeters apt-get install libjson-perl  did the trick!
14:10 pinesol_green [evergreen|Mike Rylander] LP#1321017: Order constituent records by quality - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4897a05>
14:15 egbuilder build #595 of evergreen-master-fedora-18 is complete: Success [build successful]  Build details are at http://testing.evergreen-ils.org/buildbot/bu​ilders/evergreen-master-fedora-18/builds/595
14:33 csharp yay!
14:46 Dyrcona \o/
14:51 mmorgan1 joined #evergreen
15:27 ericar joined #evergreen
15:35 mmorgan1 joined #evergreen
15:43 dbwells Last call for 2.6.1 and 2.5.5!  Thanks.
15:55 * jeff feels off by a day still
16:02 jeff dbwells: do i have some time to sneak in bug 1296937?
16:02 pinesol_green Launchpad bug 1296937 in Evergreen 2.5 "SIP2 Patron Information times out on too many checkouts/holds" (affected: 4, heat: 20) [Medium,Confirmed] https://launchpad.net/bugs/1296937
16:02 dbwells jeff: sure, just let me know when you are done.
16:02 jeff thanks!
16:17 edoceo joined #evergreen
16:20 jboyer-isl mrpeters: just a quick note on that stripe deb, following the usual perl lib naming conventions it would be called libbusiness-stripe-perl, there's no stripe lib that works as a Business::OnlinePayment module.
16:21 jboyer-isl It's not a big deal in either case though.
16:28 pinesol_green [evergreen|Galen Charlton] LP#1296937: move the $force_bc parameter of ->charged_items() to an implementation method - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b5ddc81>
16:31 jeff gmcharlt, tsbere: I seem to lack the ability to push to SIPServer's master branch. I'd like to push SIPServer working branch user/jeff/revert_charged_ite​ms_argument_position_change to finish off bug 1296937
16:31 pinesol_green Launchpad bug 1296937 in Evergreen 2.5 "SIP2 Patron Information times out on too many checkouts/holds" (affected: 4, heat: 20) [Medium,Confirmed] https://launchpad.net/bugs/1296937 - Assigned to Jeff Godin (jgodin)
16:32 jeff oh. i might have access.
16:32 tsbere jeff: You are on the access list
16:33 jeff and there we go.
16:33 jeff tsbere++ thanks for checking
16:33 tsbere jeff: Figured I could either check, or change modes to "figure out why I can't get SIPServer to run on my dev machine right now". Checking was faster.
16:34 jeff dbwells: I'm done. Thanks for your patience!
16:34 dbwells jeff: thanks
16:45 kbeswick joined #evergreen
16:56 jwoodard joined #evergreen
17:31 pinesol_green Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html>
17:56 ktomita_ joined #evergreen
18:41 timf joined #evergreen
19:19 pastebot "gsams" at 64.57.241.14 pasted "How I solved the duplication of "Lost Materials", "Overdues", "Lost Materials Processing Fee" in that SQL Query" (61 lines) at http://paste.evergreen-ils.org/51
19:19 gsams Thought I'd share how I fixed the report I was asking about yesterday
19:20 gsams I used a WITH statement and changed the original report to SELECT DISTINCT ON money.payment_view.id
19:21 gsams Then basically select all from that cte and sort it like it originally was
19:21 gsams then tack on the original union with the total row to that
19:21 gsams bam, no more duplicated bills
19:42 hbrennan gsams++
19:45 gsams joined #evergreen
20:16 sseng joined #evergreen
22:31 mmorgan1 joined #evergreen
23:23 mmorgan1 joined #evergreen
23:52 mmorgan1 joined #evergreen

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat