Household in microsoft.* hierarchy
Julien ÉLIE
[Follow up to news.admin.hierarchies.]
535 newsgroups have been removed this morning on news.microsoft.com,
which implies that only 1772 newsgroups currently remain.
I will wait during at least a week before sending any new control
articles for that hierarchy. (Just to be sure it was not a mistake.)
Incidentally, I wonder whether I should send 535 rmgroup articles
(which would be clean) followed by a checkgroups, or only one checkgroups
without any rmgroup control articles.
Any idea on that subject?
If I send 535 rmgroup articles, it will probably one every 30 seconds,
so that innfeed does not drop articles because of too many consecutive
reloads.
For your information, here is the list of removals:
microsoft.public.activex.programming.control.webwiz
microsoft.public.arabic.backoffice
microsoft.public.arabic.feedback
microsoft.public.arabic.games
microsoft.public.arabic.localization
microsoft.public.arabic.ntserver
microsoft.public.arabic.ntworkstation
microsoft.public.arabic.office2000
microsoft.public.arabic.sqlserver
microsoft.public.arabic.webdeveloper
microsoft.public.arabic.win9x
microsoft.public.ausnz.msf
microsoft.public.automap
microsoft.public.backoffice.sbs.beta
microsoft.public.backoffice.server.beta
microsoft.public.br.architecture
microsoft.public.br.dotnet.vjsharp
microsoft.public.br.retail
microsoft.public.br.sms.general
microsoft.public.br.technet
microsoft.public.br.vfp
microsoft.public.br.xml
microsoft.public.ccf
microsoft.public.cert.developer.mcts
microsoft.public.certification.winnt-9x
microsoft.public.cmserver.evaluation
microsoft.public.cn.applicationcenter
microsoft.public.cn.biztalk
microsoft.public.cn.design.gallery
microsoft.public.cn.dotnet.embedded
microsoft.public.cn.dotnet.languages.vb
microsoft.public.cn.dotnet.mobility
microsoft.public.cn.ie40
microsoft.public.cn.inetexplorer.ie5
microsoft.public.cn.inetexplorer.ie6beta
microsoft.public.cn.mcspmember.technical
microsoft.public.cn.mscs
microsoft.public.cn.office.templates.misc
microsoft.public.cn.onenote
microsoft.public.cn.siteserver
microsoft.public.cn.snaserver
microsoft.public.cn.sourcesafe
microsoft.public.cn.vbscript
microsoft.public.cn.visio
microsoft.public.cn.win9x
microsoft.public.cn.windows.server.dns
microsoft.public.cn.winme
microsoft.public.cn.xml
microsoft.public.commerceserver.businessdesk
microsoft.public.commerceserver.datawarehousing
microsoft.public.commerceserver.sdk
microsoft.public.commerceserver.solutionsites
microsoft.public.data.ado.rds
microsoft.public.ddk.win2000.multimedia
microsoft.public.ddk.win9x.printer.subsystems
microsoft.public.ddk.win9x.still_image
microsoft.public.ddk.win9x.vcomm
microsoft.public.de.biztalk
microsoft.public.de.expression.blend
microsoft.public.de.expression.design
microsoft.public.de.german.anwender.directx
microsoft.public.de.german.win2000.applications
microsoft.public.de.german.win2000.dns
microsoft.public.de.german.win2000.hardware
microsoft.public.de.german.win2000.ras_routing
microsoft.public.de.german.windowsme
microsoft.public.de.german.zertifizierung.fachwissen.developer
microsoft.public.de.windows.tabletpc
microsoft.public.digitaldashboard
microsoft.public.directx.networking
microsoft.public.dotnet.academic.assignment_manager
microsoft.public.dotnet.vsa
microsoft.public.enable.developer
microsoft.public.es.biztalk.server
microsoft.public.es.ie30
microsoft.public.es.ie4
microsoft.public.es.ie4_win31
microsoft.public.es.ie5
microsoft.public.es.ieak
microsoft.public.es.msdn.estudiantes
microsoft.public.es.netmeeting
microsoft.public.es.photodraw
microsoft.public.es.pictureit
microsoft.public.es.proxyserver
microsoft.public.es.reader
microsoft.public.es.vinterdev
microsoft.public.es.visualj
microsoft.public.es.vsnetbeta
microsoft.public.excel.datamap
microsoft.public.excel.querydao
microsoft.public.exchange2000.announcements
microsoft.public.exchange2000.application.conversion
microsoft.public.exchange2000.applications
microsoft.public.exchange2000.clustering
microsoft.public.exchange2000.documentation
microsoft.public.exchange2000.interop
microsoft.public.exchange2000.kms
microsoft.public.exchange2000.protocols
microsoft.public.exchange2000.realtime.collaboration
microsoft.public.exchange2000.setup.installation
microsoft.public.expression.encoder
microsoft.public.faculty_members
microsoft.public.faculty_public
microsoft.public.fdtnservices.hailstrm.sdk
microsoft.public.fdtnservices.netmyservices.sdk
microsoft.public.fdtnservices.passport.sdk
microsoft.public.fi.design.gallery
microsoft.public.fi.dotnet.languages.vb
microsoft.public.fortran
microsoft.public.fox.books
microsoft.public.fox.fox2x
microsoft.public.fox.fox2x.setup.environment
microsoft.public.fox.fox2x.xplat
microsoft.public.fox.vfp.language-menus
microsoft.public.fox.vfp.lck-api
microsoft.public.fox.vfp.xplat
microsoft.public.fr.access.runtime
microsoft.public.fr.administration_centrales
microsoft.public.fr.antigen
microsoft.public.fr.axapta
microsoft.public.fr.ccs
microsoft.public.fr.cmserver
microsoft.public.fr.collectivites_territoriales
microsoft.public.fr.commerceserver
microsoft.public.fr.dotnet.olymars
microsoft.public.fr.dotnet.vc
microsoft.public.fr.dotnet.visualstudio
microsoft.public.fr.dotnet.vjsharp
microsoft.public.fr.dungeonsiege
microsoft.public.fr.encarta
microsoft.public.fr.exchange.applications
microsoft.public.fr.forefront
microsoft.public.fr.hiserver
microsoft.public.fr.hpc
microsoft.public.fr.msn.desktop_search
microsoft.public.fr.netmeeting
microsoft.public.fr.office.templates.misc
microsoft.public.fr.onenote
microsoft.public.fr.photodraw
microsoft.public.fr.proxy
microsoft.public.fr.sante
microsoft.public.fr.siteserver
microsoft.public.fr.sqlserver.analysisservices
microsoft.public.fr.sqlserver.notificationsvcs
microsoft.public.fr.teamsystem
microsoft.public.fr.vinterdev
microsoft.public.fr.windows95
microsoft.public.fr.windowsnt
microsoft.public.fr.xml
microsoft.public.fr.zootycoon
microsoft.public.games.ageofkings
microsoft.public.games.zone.allegiance
microsoft.public.games.zone.asherons_call
microsoft.public.games.zone.beta
microsoft.public.games.zone.board
microsoft.public.games.zone.bridgeclub
microsoft.public.games.zone.card
microsoft.public.games.zone.fighterace
microsoft.public.games.zone.fighterace.flightschool
microsoft.public.games.zone.links2001
microsoft.public.games.zone.lstour
microsoft.public.games.zone.lstour.tech
microsoft.public.games.zone.puzzles
microsoft.public.games.zone.tanarus
microsoft.public.games.zone.zonelan
microsoft.public.greetings.workshop
microsoft.public.hk.systemsmanagementserver
microsoft.public.hk.visualstudioanddevelopment
microsoft.public.hostedapps.dev
microsoft.public.hu.games
microsoft.public.hu.project
microsoft.public.hu.windows.server.general
microsoft.public.il.hebrew.access
microsoft.public.il.hebrew.architects
microsoft.public.il.hebrew.asp
microsoft.public.il.hebrew.drushim
microsoft.public.il.hebrew.excel
microsoft.public.il.hebrew.feedback
microsoft.public.il.hebrew.fun
microsoft.public.il.hebrew.groupware
microsoft.public.il.hebrew.ie3
microsoft.public.il.hebrew.iis
microsoft.public.il.hebrew.junior
microsoft.public.il.hebrew.mcse
microsoft.public.il.hebrew.ntworkstation
microsoft.public.il.hebrew.officedev
microsoft.public.il.hebrew.olap.datawarehouse
microsoft.public.il.hebrew.powerpoint
microsoft.public.il.hebrew.vc
microsoft.public.il.hebrew.vfoxpro
microsoft.public.il.hebrew.webdevelopers
microsoft.public.il.hebrew.win95
microsoft.public.il.hebrew.win98
microsoft.public.il.hebrew.windows2000
microsoft.public.il.israeli.sitebuilder
microsoft.public.in.aspdotnet
microsoft.public.inetexplorer.ie5.outlookexpress.stationery
microsoft.public.inetsdk.programming.comctl32
microsoft.public.inetsdk.programming.components.design_time
microsoft.public.inetsdk.programming.components.packaging
microsoft.public.inetsdk.programming.components.usage
microsoft.public.inetsdk.programming.docobjects
microsoft.public.inetsdk.programming.mstask
microsoft.public.inetsdk.programming.shell_objmodel
microsoft.public.inetsdk.sdk_setup
microsoft.public.inetserver.dbweb
microsoft.public.inetserver.webtool
microsoft.public.internet.activex.conferencing.sdk
microsoft.public.internetexplorer.rights_management
microsoft.public.internet.netmeeting.beta
microsoft.public.it.biztalk
microsoft.public.it.learning.faculty_connection
microsoft.public.ja.office.templates.misc
microsoft.public.java.afc
microsoft.public.java.cab
microsoft.public.java.jdirect
microsoft.public.java.macvm
microsoft.public.java.visualj++.activex
microsoft.public.java.visualj++.compiler
microsoft.public.java.visualj++.general
microsoft.public.java.visualj++.ide_addins
microsoft.public.java.visualj++.installation
microsoft.public.java.visualj++.wfc
microsoft.public.java.visualj++.wfc.applications
microsoft.public.java.visualj++.wfc.controls
microsoft.public.java.visualj++.wfc.database
microsoft.public.java.visualj++.wfc.html
microsoft.public.java.vm.migration
microsoft.public.java.win16vm
microsoft.public.jp.adsi.general
microsoft.public.jp.applicationcenter
microsoft.public.jp.cmserver.general
microsoft.public.jp.commerceserver
microsoft.public.jp.crm
microsoft.public.jp.dotnet.framework
microsoft.public.jp.dotnet.framework.adonet
microsoft.public.jp.dotnet.framework.aspnet
microsoft.public.jp.dotnet.languages.csharp
microsoft.public.jp.dotnet.languages.vc
microsoft.public.jp.dotnet.vsnet
microsoft.public.jp.games
microsoft.public.jp.games.ageofempire
microsoft.public.jp.hiserver.general
microsoft.public.jp.home
microsoft.public.jp.home.hagaki
microsoft.public.jp.inetexplorer.mac
microsoft.public.jp.inetserver.asp
microsoft.public.jp.infopath
microsoft.public.jp.interconnect
microsoft.public.jp.mediacenter
microsoft.public.jp.mom
microsoft.public.jp.msinformation
microsoft.public.jp.office.templates.misc
microsoft.public.jp.outlookexpress.mac
microsoft.public.jp.producer
microsoft.public.jp.proxy
microsoft.public.jp.publisher
microsoft.public.jp.resourcekit
microsoft.public.jp.rights_mgmt_svcs
microsoft.public.jp.sbs
microsoft.public.jp.servicesforunix
microsoft.public.jp.sharepoint.portalserver
microsoft.public.jp.softwareupdatesvcs
microsoft.public.jp.storage
microsoft.public.jp.virtualpc
microsoft.public.jp.virtualpc.mac
microsoft.public.jp.win95
microsoft.public.jp.win98
microsoft.public.jp.windows.64bit
microsoft.public.jp.windowsce
microsoft.public.jp.windows.directx
microsoft.public.jp.windowsme
microsoft.public.jp.windowsnt.ws
microsoft.public.jp.windows.server.setup
microsoft.public.jp.windows.tabletpc
microsoft.public.jp.windowsxp.compatibility
microsoft.public.ko.ie30
microsoft.public.ko.ie40
microsoft.public.kr.appcenter.general
microsoft.public.kr.asp.qna
microsoft.public.kr.biztalk.general
microsoft.public.kr.cms.general
microsoft.public.kr.commerceserver.general
microsoft.public.kr.directx
microsoft.public.kr.dotnet.vc
microsoft.public.kr.dotnet.webservices
microsoft.public.kr.dotnet.windowsforms
microsoft.public.kr.dotnet.xml
microsoft.public.kr.frontpage
microsoft.public.kr.his.general
microsoft.public.kr.ie60.qna
microsoft.public.kr.iis.qna
microsoft.public.kr.isa.general
microsoft.public.kr.koie50.qna
microsoft.public.kr.mis.general
microsoft.public.kr.mom.general
microsoft.public.kr.officexp.qna
microsoft.public.kr.outlookexpress
microsoft.public.kr.platsdk.qna
microsoft.public.kr.sms.qna
microsoft.public.kr.sps.general
microsoft.public.kr.sql
microsoft.public.kr.technet
microsoft.public.kr.vc.qna
microsoft.public.kr.visio.qna
microsoft.public.kr.win2000.qna
microsoft.public.kr.windows98.qna
microsoft.public.kr.windows.hardware
microsoft.public.kr.windows.mediacenter
microsoft.public.kr.windows.network_web
microsoft.public.kr.windows.qna
microsoft.public.kr.windows.server.clustering
microsoft.public.kr.windows.server.dns
microsoft.public.kr.windows.server.general
microsoft.public.kr.windows.server.migration
microsoft.public.kr.windows.server.networking
microsoft.public.kr.windows.server.security
microsoft.public.kr.windows.server.setup
microsoft.public.kr.winme.qna
microsoft.public.kr.winntsvr.qna
microsoft.public.kr.winntwks.qna
microsoft.public.kr.wmt.qna
microsoft.public.learning-network-manager
microsoft.public.lrn
microsoft.public.lt.developer
microsoft.public.lv.developer
microsoft.public.lv.license
microsoft.public.lv.office
microsoft.public.lv.server
microsoft.public.lv.windows
microsoft.public.mac.rdc.rdc2
microsoft.public.mcis
microsoft.public.mcis.chatserver
microsoft.public.mcis.membership
microsoft.public.mcis.newserver
microsoft.public.messaging.misc
microsoft.public.mls
microsoft.public.mobileexplorer
microsoft.public.mom.managementpack.sql
microsoft.public.mom.security
microsoft.public.mom.ui
microsoft.public.mom.workgroupedition
microsoft.public.mscom.webservices.general
microsoft.public.msf
microsoft.public.msmq.interop
microsoft.public.mts.administration.security
microsoft.public.multimedia.directx.danimation.controls
microsoft.public.multimedia.directx.danimation.programming
microsoft.public.multimedia.directx.dtransform
microsoft.public.nl.iis
microsoft.public.nl.sms
microsoft.public.nl.windows.2k
microsoft.public.nl.windows98
microsoft.public.no.ie4
microsoft.public.nordic.ie30
microsoft.public.office.dataanalyzer
microsoft.public.office.developer.clipboard.dde
microsoft.public.office.developer.hosting.controls
microsoft.public.office.intranets
microsoft.public.outlook.teamfolders
microsoft.public.ph.philippines.certified.events
microsoft.public.pk.general
microsoft.public.pk.office2000
microsoft.public.pk.windows2000
microsoft.public.platformsdk.active.directory
microsoft.public.platformsdk.component_svcs
microsoft.public.platformsdk.database
microsoft.public.platformsdk.directx
microsoft.public.platformsdk.dist_svcs
microsoft.public.platformsdk.graphics_mm
microsoft.public.platformsdk.graphics_mm.directx
microsoft.public.platformsdk.internet.server
microsoft.public.platformsdk.localization
microsoft.public.platformsdk.messaging
microsoft.public.platformsdk.mslayerforunicode
microsoft.public.platformsdk.multimedia
microsoft.public.platformsdk.telephony.tapi_2
microsoft.public.platformsdk.telephony.tsp
microsoft.public.platformsdk.telephony.wte
microsoft.public.platformsdk.win_base_svcs
microsoft.public.pl.ie5
microsoft.public.pocketpc.binaries
microsoft.public.pocketpc.marketplace
microsoft.public.reader.developer
microsoft.public.reference_arch.commerce
microsoft.public.research.netscan.discussion
microsoft.public.research.wwmx
microsoft.public.ru.axapta
microsoft.public.ru.msdn
microsoft.public.ru.navision
microsoft.public.ru.russian.programming
microsoft.public.ru.russian.windows2000
microsoft.public.ru.russian.wishlist
microsoft.public.ru.vb
microsoft.public.schedule+
microsoft.public.security.hfnetchk
microsoft.public.security.scw
microsoft.public.siteserver.commerce
microsoft.public.siteserver.knowledgemgr
microsoft.public.siteserver.per-mbr
microsoft.public.site-server.postingacceptr
microsoft.public.siteserver.publishing
microsoft.public.siteserver.sdk
microsoft.public.siteserver.search
microsoft.public.site-server.site-mgmt
microsoft.public.site-server.webpost
microsoft.public.sms.sitecomm
microsoft.public.sqlxml.viewmapper
microsoft.public.technet.magazine
microsoft.public.th.da
microsoft.public.th.design.gallery
microsoft.public.th.office
microsoft.public.th.ts
microsoft.public.tr.biztalk
microsoft.public.tr.channel.ntw2000
microsoft.public.tr.commerceserver
microsoft.public.tr.sna
microsoft.public.tw.win2000
microsoft.public.ua.general
microsoft.public.ua.licensing
microsoft.public.ua.office
microsoft.public.ua.win
microsoft.public.ua.winsrv
microsoft.public.uddi.specification
microsoft.public.usageanalyst
microsoft.public.vb.3rdparty
microsoft.public.vb.database.rdo
microsoft.public.vb.ole.cdk
microsoft.public.vb.ole.servers
microsoft.public.vb.ownersarea
microsoft.public.vb.setupwiz
microsoft.public.vb.webclasses
microsoft.public.vb.yellowpages
microsoft.public.vc.etk
microsoft.public.vc.events
microsoft.public.vchat
microsoft.public.vc.mfc.macintosh
microsoft.public.vc.res_editing
microsoft.public.vc.yellowpages
microsoft.public.vi.dtc
microsoft.public.vsnet.jlca
microsoft.public.webservices.gxa
microsoft.public.webservices.wsdk
microsoft.public.webstoragesystem.development
microsoft.public.webstoragesystem.sdk.eng
microsoft.public.win16.programmer.gdi
microsoft.public.win16.programmer.networks
microsoft.public.win16.programmer.pen
microsoft.public.win16.programmer.tools
microsoft.public.win16.programmer.ui
microsoft.public.win2000.netware
microsoft.public.win2000.new_user
microsoft.public.win32.programmer.3rdparty
microsoft.public.win32.programmer.pen
microsoft.public.win32.programmer.yellowpages
microsoft.public.win95.commtelephony
microsoft.public.win95.dialupnetwork
microsoft.public.win95.exchangefax
microsoft.public.win95.multimedia
microsoft.public.win95.networking
microsoft.public.win95.printfontvideo
microsoft.public.win95.shellui
microsoft.public.win95.win95applets
microsoft.public.win98.comm.dun
microsoft.public.win98.comm.modem
microsoft.public.win98.display.general
microsoft.public.win98.display.multi_monitor
microsoft.public.win98.fat32
microsoft.public.win98.internet.netmeeting
microsoft.public.win98.msinfo32
microsoft.public.win98.multimedia
microsoft.public.win98.multimedia.directx5
microsoft.public.win98.power_mgmt
microsoft.public.win98.pptp
microsoft.public.win98.pws_4
microsoft.public.win98.scanreg
microsoft.public.win98.setup.win31
microsoft.public.win98.sys_file_check
microsoft.public.win98.taskscheduler
microsoft.public.win98.webtv
microsoft.public.windna.components
microsoft.public.windna.deployment
microsoft.public.windna.general
microsoft.public.windna.security
microsoft.public.windows95.oemdsp.preinstall
microsoft.public.windows.activedirectory.dsml
microsoft.public.windowscard
microsoft.public.windows.developer.winfx.aero
microsoft.public.windows.developer.winfx.announcements
microsoft.public.windows.developer.winfx.collaboration
microsoft.public.windows.developer.winfx.general
microsoft.public.windows.developer.winfx.indigo
microsoft.public.windows.developer.winfx.sdk
microsoft.public.windows.inetexplorer.ie55.outlookexpress.stationery
microsoft.public.windows.live.callforfree.generaldiscussion
microsoft.public.windows.live.onecare.antispyware
microsoft.public.windows.live.onecare.backup
microsoft.public.windows.live.onecare.firewall
microsoft.public.windows.live.onecare.install
microsoft.public.windowsmedia.beta.encoder
microsoft.public.windows.mediacenter.portable
microsoft.public.windowsmedia.encoder.studioedition
microsoft.public.windowsme.hardware
microsoft.public.windowsme.new-user
microsoft.public.windowsme.powermgmt
microsoft.public.windowsnt.dns
microsoft.public.windowsnt.dsmnfpnw
microsoft.public.windowsnt.gsnw-csnw
microsoft.public.windowsnt.mac
microsoft.public.windowsnt.oemdsp.preinstall
microsoft.public.windowsnt.personalfax
microsoft.public.windowsnt.protocol.ipx
microsoft.public.windowsnt.protocol.misc
microsoft.public.windowsnt.wntsee
microsoft.public.windows.server.prerelease
microsoft.public.windowsxp.embedded.techpreview
microsoft.public.word.internet.assistant
microsoft.public.word.macword5
microsoft.public.word.word97vba
microsoft.public.works.dos
microsoft.public.xml.xmlsqlwebrelease
microsoft.public.za.biztalk
microsoft.public.za.cmserver
microsoft.public.za.commerceserver
microsoft.public.za.dotnet
microsoft.public.za.exchange
microsoft.public.za.foxpro
microsoft.public.za.officexp
microsoft.public.za.outlook
microsoft.public.za.sql
microsoft.public.za.vsnet
microsoft.public.za.vsnet.setup
microsoft.public.za.win2kpro
microsoft.public.za.windowsxp.networking
microsoft.public.za.windowsxp.setup
--
Julien ï¿œLIE
ￜ -- Comment s'appelle cette ville ?
-- Divodurum.
-- N'essaie pas de m'amadouer ! Non, je ne veux pas de rhum ! ￜ (Astￜrix)
Adam H. Kerman
>Hi all,
>535 newsgroups have been removed this morning on news.microsoft.com,
>which implies that only 1772 newsgroups currently remain.
>I will wait during at least a week before sending any new control
>articles for that hierarchy. (Just to be sure it was not a mistake.)
>Incidentally, I wonder whether I should send 535 rmgroup articles
>(which would be clean) followed by a checkgroups, or only one checkgroups
>without any rmgroup control articles.
>Any idea on that subject?
>If I send 535 rmgroup articles, it will probably one every 30 seconds,
>so that innfeed does not drop articles because of too many consecutive
>reloads.
This is my suggestion:
I agree with waiting a week to see what else they are going to do, such as
creating Windows 7 groups.
Set up a Web page with the list of the 535 removed groups.
In each rmgroup article, include the Web page's URL, so those who process
your rmgroup messages understand what's going on and will know to expect a
vast number of them.
Send the messages over several days.
I suggest waiting for several weeks or a month before issuing checkgroups,
to see if the Windows 7 groups are going to be created.
Finally, a hint that somebody has resumed administering that News server...
Julien ÉLIE
> Send the [rmgroup] messages over several days.
>
> I suggest waiting for several weeks or a month before issuing checkgroups,
> to see if the Windows 7 groups are going to be created.
Why couldn't I send a checkgroups after having processed the removals?
In case Windows 7 groups were created after the checkgroups, I would send
newgroup articles for these groups and, again, another checkgroups.
--
Julien �LIE
� Quand on retire tout ce qu'on a dit, il reste tout ce qu'on n'a pas dit. �
(Philippe Geluck)
Adam H. Kerman
>Hi Adam,
>>Send the [rmgroup] messages over several days.
>>I suggest waiting for several weeks or a month before issuing checkgroups,
>>to see if the Windows 7 groups are going to be created.
>Why couldn't I send a checkgroups after having processed the removals?
>In case Windows 7 groups were created after the checkgroups, I would send
>newgroup articles for these groups and, again, another checkgroups.
Can you anticipate that those who take your checkgroups would process them
that frequently? I don't think it's a good idea to send different versions
of checkgroups within 30 days of each other, quite frankly.
News Guy
> 535 newsgroups have been removed this morning on
> news.microsoft.com,
Please explain why that has any relavence to usenet in general.
> I will wait during at least a week before sending any new
> control articles for that hierarchy.
I'm not trying to be difficult, or a smart-ass.
Please explain why the rest of usenet must follow what microsoft does
with it's server and the groups it chooses to carry or remove.
I'm assuming that because you're going to issue control messages for
those groups, that some news servers actually obey your messages.
Please explain how this situation evolved or came to be.
Where was it decided that you would have some special relationship to
the set of groups that start with "microsoft", and that your decisions
would be be taken to be authoritative.
Again, I'm not trying to be argumentative. I just want to understand
how or why you have this special relationship with the microsoft groups
and why microsoft gets to have it's decisions about how they configure
their server propagate to the rest of usenet.
If microsoft removed some high-traffic XP groups from their server,
would you also issue control messages to usenet to delete those groups
across the rest of usenet as well?
News Guy
> I agree with waiting a week to see what else they are going to
> do, such as creating Windows 7 groups.
If it's appropriate, useful or logical to create a Windows 7 group under
the Microsoft hierarchy - then why not simply do it?
Where is it written that only Microsoft can create new groups in the
microsoft.* hierarchy of newsgroups?
Russ Allbery
> Please explain why the rest of usenet must follow what microsoft does
> with it's server and the groups it chooses to carry or remove.
It doesn't.
> I'm assuming that because you're going to issue control messages for
> those groups, that some news servers actually obey your messages.
> Please explain how this situation evolved or came to be.
Microsoft kept saying they were going to provide control messages for
their hierarchy and then never got around to setting it up. Julien
volunteered to provide, as a service, control messages that synchronize
group lists with the groups that are active on Microsoft's servers.
That's the defined purpose and action of the control messages he issues.
This was discussed at quite some length in this newsgroup some time back
and documented in the information he posts about his stream of control
messages.
Each individual news administrator has a choice. If they want to track
Microsoft's view of the microsoft.* hierarchy and treat the hierarchy as
managed by Microsoft, they can honor those control messages. If they want
to manage the hierarchy by some other criteria, they can ignore those
control messages. Having the control messages available is strictly
better than not having them available since it makes that choice easily
available.
The newsgroup list on isc.org was already trying (via a largely
unsuccessful manual process) to track the groups as present on the
Microsoft servers, so it uses Julien's control messages since that saved
me a bunch of time and actually works.
Obviously any news administrator can create whatever group they want under
whatever name they want, but given that having a set of newgroups that
track the Microsoft groups as managed by Microsoft is useful, I don't see
much point in people doing that under the microsoft.* name. There are
lots of other Usenet hierarchies in which one can create groups obviously
independent of what Microsoft carries on their servers. But of course if
you disagree, knock yourself out.
--
Russ Allbery (r...@stanford.edu) <http://www.eyrie.org/~eagle/>
Russ Allbery
> If it's appropriate, useful or logical to create a Windows 7 group under
> the Microsoft hierarchy - then why not simply do it?
> Where is it written that only Microsoft can create new groups in the
> microsoft.* hierarchy of newsgroups?
What's the point of having a separate hierarchy otherwise? Why not just
create comp.os.ms-windows.windows7 or alt.os.windows7 if you're going to
create a group independent of the groups Microsoft carries on their public
servers?
News Guy
> > If it's appropriate, useful or logical to create a Windows 7
> > group under the Microsoft hierarchy - then why not simply do it?
>
> > Where is it written that only Microsoft can create new groups
> > in the microsoft.* hierarchy of newsgroups?
>
> What's the point of having a separate hierarchy otherwise?
> Why not just create comp.os.ms-windows.windows7 or
> alt.os.windows7 if you're going to create a group independent
> of the groups Microsoft carries on their public servers?
Because Microsoft may decide that it does not want to host a
usenet-compatible server for some, most, or all of it's previous,
current, or future products.
It appears that Microsoft is winding down it's corporate participation
with usenet by not hosting it's own local set of Windows-7 groups on
it's NNTP server.
We can only expect attrition and further retrenchment from them in this
regard. It would not be healthy or useful for usenet to continue to
mirror microsoft's corporate decision to withdraw or remove microsoft.*
groups in this manner.
Microsoft could choose tommorrow to end it's participation or peering
with usenet if that peering does not meet corporate objectives or it
requires too much oversight/censorship on their part. What would Julien
�LIE do in that case? Would he issue control messages to delete the
entire microsoft.* hierarchy from usenet? What purpose would that
serve?
I fail to see why formal, official or administrative usenet policy
should reflect Microsoft's policies and decisions with regard to the
microsoft.* set of groups - even if individual server operators can do
as they please.
Russ Allbery
> Russ Allbery wrote:
>> What's the point of having a separate hierarchy otherwise? Why not
>> just create comp.os.ms-windows.windows7 or alt.os.windows7 if you're
>> going to create a group independent of the groups Microsoft carries on
>> their public servers?
> Because Microsoft may decide that it does not want to host a
> usenet-compatible server for some, most, or all of it's previous,
> current, or future products.
So create newsgroups for what you want to talk about. We have active
comp.os.* hierarchies and alt.os.* hierarchies for that purpose that have
existed long before Microsoft had any participation in Usenet.
The whole point of the microsoft.* hierarchy when it was originally
created was to reflect the groups that Microsoft themselves was hosting
and maintaining. It wasn't to be the only place on Usenet to talk about
Microsoft products; there were existing hierarchies that were serving that
purpose just fine. Repurposing microsoft.* to be a general hierarchy for
independently-maintained groups about Microsoft products is pointless. We
have perfectly functional hierarchies already serving that role. The only
point in having a microsoft.* hierarchy at all is to provide groups
synchronized with the official Microsoft ones for people who want to use
those groups instead of or in addition to the other existing ones
independent of Microsoft.
> Microsoft could choose tommorrow to end it's participation or peering
> with usenet if that peering does not meet corporate objectives or it
> requires too much oversight/censorship on their part. What would Julien
> ÉLIE do in that case? Would he issue control messages to delete the
> entire microsoft.* hierarchy from usenet?
I don't know -- you'd have to ask him. If I were him, yes, I'd be tempted
to do that.
> What purpose would that serve?
Why retain the hierarchy when its entire purpose has gone away? Everyone
can go use the existing independent groups in the regular Usenet
hierarchies and create more as needed, using the existing management
infrastructure of those hierarchies.
> I fail to see why formal, official or administrative usenet policy
> should reflect Microsoft's policies and decisions with regard to the
> microsoft.* set of groups - even if individual server operators can do
> as they please.
Hopefully it's now clearer.
News Guy
> > What purpose would that serve?
>
> Why retain the hierarchy when its entire purpose has gone away?
By rights you, or anyone else, can't judge the purpose of a newgroup
other than to ask if it's being used by people to communicate
information or ideas.
Some of those 535 newsgroups are in active use. It does not serve those
participants to remove the groups they use simply because Microsoft
doesn't want to commit hard drive space to them.
Many people don't access those groups through microsoft's servers and
have no idea that microsoft has removed any groups. Why should
administrative decisions made by Microsoft for their own server affect
those people?
> Everyone can go use the existing independent groups in the
> regular Usenet hierarchies and create more as needed
You know damn well that isin't going to happen.
And I don't see why it needs to happen.
Again, the microsoft hierarchy has been around for a long time.
Microsoft has no practical, authoritative, or administrative role in the
functionality or useage of those groups. I don't see why now,
all-of-a-sudden, their corporate decisions need to be mirrored by the
usenet community.
The microsoft hierarchy exists. People know where it is, they
participate in them, they know what groups it contains - and it's a
logical hierarchy. And you believe that it can be replaced in a
complete and timely manner, and that the confusion and frustration
caused by a transition period is warranted?
> > I fail to see why formal, official or administrative
> > usenet policy should reflect Microsoft's policies and
> > decisions with regard to the microsoft.* set of groups
> > - even if individual server operators can do as they
> > please.
>
> Hopefully it's now clearer.
Obviously there are many linux/unix advocates that want to see the
microsoft hierarcy removed from usenet. That much is clear.
Russ Allbery
since I have no interest in debating this in detail or convincing you of
anything (or time to do either), I'll bow out. I'm entirely happy for you
to believe whatever you wish about what should be done about the
hierarchy, and will simply indicate my disagreement and move on.
John F. Morse
> Obviously there are many linux/unix advocates that want to see the
> microsoft hierarcy removed from usenet. That much is clear.
>
What is clear to me is there are likely more Win-droids that want to
keep reading the crap than "Linux and Unix advocates" who want it removed.
Some of us just don't carry it, period, and might suggest to Julien to
just "pull the plug."
Hey, if you want to carry it on your Windows 98 box, then do so. Who am
I or any other "Linux advocate" to tell you that you can't carry it?
--
John
News Guy
> Who am I or any other "Linux advocate" to tell you that you can't
> carry it?
Who are you to tell others *not* to carry it?
Don't you understand the difference?
John F. Morse
Isn't that what I just said?
I see nothing different between the two statements above.
However, you obviously do not understand that it is up to the NSP owner
whether or not to honor rmgroup and/or checkgroup messages.
--
John
News Guy
> I'm entirely happy for you to believe whatever you wish about
> what should be done about the hierarchy,
You haven't put forward a cogent or logical argument as to why a usenet
newsgroup(s) should be removed just because that same group(s) is being
removed by one specific server somewhere out there.
If Usenet had the capacity to undertake a contract or agreement or
understanding with Microsoft that would stipulate that Microsoft would
peer with usenet so long as usenet would exactly mirror Microsoft's
groups, and that any removal by microsoft would be duplicated (to the
best of it's effort) across the rest of usenet, well then that I can
understand and not have an argument with.
In the absense of such an agreement, I fail to see the logical argument
of issuing control messages to remove these groups.
It flys in the very face of what usenet is supposed to be to remove
groups that are actively used by people to exchange ideas and
information, especially if it's done for bogus administrative reasons
justified with a thin veneer of rationality.
News Guy
> However, you obviously do not understand that it is up to the
> NSP owner whether or not to honor rmgroup and/or checkgroup
> messages.
This is not about what individual operators choose to do.
This is about the reasoning (or even the authority) for issuing the
rmgroup messages in the first place.
John F. Morse
> "John F. Morse" wrote:
>
>> However, you obviously do not understand that it is up to the
>> NSP owner whether or not to honor rmgroup and/or checkgroup
>> messages.
>>
>
> This is not about what individual operators choose to do.
>
Yes it is!
It is about libre "free" to run your own server as you wish, not as News
Guy thinks it should be ran.
Cf: http://en.wikipedia.org/wiki/Communism
> This is about the reasoning (or even the authority) for issuing the
> rmgroup messages in the first place.
>
You've already been given the reasoning, but you disagree. That is your
libre freedom, but your personal feelings, based upon misunderstanding,
are not likely to make any changes.
You, me, or anyone else has the "authority" but it also depends on
whether other peers will allow that authority to cause any action. That
is their libre freedom to choose.
--
John
News Guy
> > This is not about what individual operators choose to do.
>
> Yes it is!
No it isin't.
> > This is about the reasoning (or even the authority) for
> > issuing the rmgroup messages in the first place.
>
> You've already been given the reasoning, but you disagree.
It has not been explained why or how Julien �LIE has been given the
authority to make this decision and to send out the control messages in
question.
> You, me, or anyone else has the "authority" but it also depends
> on whether other peers will allow that authority to cause any
> action. That is their libre freedom to choose.
Is it known (or knowable) which news servers are configured to
automatically honor these rmgroup messages as issued by Julien �LIE?
Adam H. Kerman
Asked and answered and asked and answered and asked and answered...
Obviously it's a rhetorical question as you've never accepted any reasonable
answer anyone has ever given you.
Why don't you set up your own News server and create whatever group you
like in any top-level hierarchy, as you don't approve of the way anyone
else administers his server?
Adam H. Kerman
>"John F. Morse" wrote:
>>>This is not about what individual operators choose to do.
>>Yes it is!
>No it isin't.
>>>This is about the reasoning (or even the authority) for
>>>issuing the rmgroup messages in the first place.
>>You've already been given the reasoning, but you disagree.
>It has not been explained why or how Julien ALIE has been given the
>authority to make this decision and to send out the control messages in
>question.
Julien isn't Rob Maxwell. He's claiming no authority.
It's Usenet. Anybody can issue a control message (unless prevented by
his News administrator).
News Guy
> It's Usenet. Anybody can issue a control message (unless prevented
> by his News administrator).
You forgot to answer this question:
John F. Morse
Yes. Every news server administrator knows.
Considering security, some -- or most -- do not want to give out this
information.
Why do you need this information?
--
John
Adam H. Kerman
Dude, every one of your questions has been rhetorical. No one should
answer those.
News Guy
> > Is it known (or knowable) which news servers are configured to
> > automatically honor these rmgroup messages as issued by
> > Julien �LIE?
>
> Yes. Every news server administrator knows.
Are you even sure about that?
When was the last time that Julien �LIE issued rmgroup messages?
> Considering security, some -- or most -- do not want to give
> out this information.
Presumably it would be known the instant that the rmgroup messages are
sent and servers of interest are queried to see if they honored them.
So in that case I fail to see how could it could be a security issue
before they are sent.
> Why do you need this information?
To get an idea as to how the participation, distribution or exposure of
those groups would be reduced, as a result of knowing which servers
would, and would not be carrying those groups after the rmgroup messages
are sent.
News Guy
> > Is it known (or knowable) which news servers are configured to
> > automatically honor these rmgroup messages as issued by
> > Julien �LIE?
>
> Dude, every one of your questions has been rhetorical.
I guess for you, any question that you're unwilling or unable to answer
is a rhetorical one. You're not the only usenet poster to use that
tactic.
> No one should answer those.
If they're really rhetorical, then the answer is contained in the
question or is self-obvious.
So your directive that an answer not be posted indicates that you
actually believe they are not rhetorical questions.
It's clear that one can't have an honest converstation with you, or
trust that what you post isin't some form of deception, misinformation
or obfuscation.
John F. Morse
Go away. You annoy me.
--
John
John F. Morse
You are peggin' my Troll-O-Meter.
--
John
D. Stussy
news:4B2197F1...@Guy.com...
See "managed hierarchy" (Usenet).
D. Stussy
> ...
> Obviously there are many linux/unix advocates that want to see the
> microsoft hierarcy removed from usenet. That much is clear.
No. They want to see Micro$oft removed from the face of the Earth, but
they'll take what they can get.
Julien ÉLIE
> Microsoft could choose tommorrow to end it's participation or peering
> with usenet if that peering does not meet corporate objectives or it
> requires too much oversight/censorship on their part. What would Julien
> �LIE do in that case? Would he issue control messages to delete the
> entire microsoft.* hierarchy from usenet? What purpose would that
> serve?
I would post an article here to mention what happened, as I have just done
for the removal of 535 newsgroups. We would discuss the appropriate way
to make the microsoft.* hierarchy historic/defunct.
I believe there would be rmgroup control messages sent so as to properly
clean the hierarchy.
The purpose? Just to reflect the contents of msnews.microsoft.com;
it is what the control messages I have been sending for more than two years
are for. It will always be their aim. And news administrators who wish
to follow them (that is to say who wish to carry the same groups as the
ones carried by msnews.microsoft.com) can honour them. Otherwise, they
do not. Nothing forces them to honour my control articles. Nothing
prevents them to create the microsoft.* newsgroups they wish on their
news servers.
Ah, another thing: I would have liked to answer a few more of your
messages but I *really* *DISLIKE* seeing that you call me "Julien �LIE"
here, in front of me, while at the same time you call me "dick-wad"
on another newsgroup:
http://groups.google.fr/group/microsoft.public.win98.gen_discussion/browse_frm/thread/5ff9a172081aeaca
<news:4B21A046...@Guy.com>
"Those groups are still being propagated by the wider usenet community
as if nothing happened. This may change if a certain dick-wad issues
control messages that are obeyed by independent server operators
(it's not clear that they will)."
At least, I have just learnt a new word ("wad"), though it is unlikely
I would ever use it in the sense you meant.
--
Julien �LIE
� Aut bibas aut abeas. �
Julien ÉLIE
>> I'm assuming that because you're going to issue control messages for
>> those groups, that some news servers actually obey your messages.
>> Please explain how this situation evolved or came to be.
>
> Microsoft kept saying they were going to provide control messages for
> their hierarchy and then never got around to setting it up. Julien
> volunteered to provide, as a service, control messages that synchronize
> group lists with the groups that are active on Microsoft's servers.
On Monday, April 16th 2007, 16h00 (in France) I also had a phone
conversation with a Microsoft employee (Aymeric Evennou) to speak
about what I was going to do.
No written approval, though. I only have mails I sent afterwards
about the microsoft.* hierarchy and on which he tacitly acknowledged
what I was doing, forwarding a few of them to Microsoft Corp.
The only thing that was clear is the fact that Microsoft prefers
their users to use their web forums.
Well, that's another debate. We already spoke about that two years ago
and there is nothing more to add to this thread about the microsoft.*
hierarchy.
> The newsgroup list on isc.org was already trying (via a largely
> unsuccessful manual process) to track the groups as present on the
> Microsoft servers, so it uses Julien's control messages since that saved
> me a bunch of time and actually works.
It makes me think that it is such a long time since we last heard of
the perl.* hierarchy...
Maybe I will drop them a mail to ask them whether they have changes
to do (perl.cpan.workers seems to be an active list not present in
perl.* current checkgroups).
--
Julien �LIE
� Quod sis, esse uelis nihilque malis ;
Summum nec metuas diem nec optes. � (Martial)
Adam H. Kerman
I see certain characters. When reading messages, I see your last name with
an acute accent over E. When I invoke the composer (just vim), I see an acute
accent over A. I've been changing it to unaccented A.
I have no idea what it is about my setup that causes me to see the same
character code with two different glyphs depending on if I'm reading or
composing.
Undoubtably I've been misspelling your last name for years, so I apologize.
Adam H. Kerman
>Hi Russ,
>>>I'm assuming that because you're going to issue control messages for
>>>those groups, that some news servers actually obey your messages.
>>>Please explain how this situation evolved or came to be.
>>Microsoft kept saying they were going to provide control messages for
>>their hierarchy and then never got around to setting it up. Julien
>>volunteered to provide, as a service, control messages that synchronize
>>group lists with the groups that are active on Microsoft's servers.
>On Monday, April 16th 2007, 16h00 (in France) I also had a phone
>conversation with a Microsoft employee (Aymeric Evennou) to speak
>about what I was going to do.
>No written approval, though. I only have mails I sent afterwards
>about the microsoft.* hierarchy and on which he tacitly acknowledged
>what I was doing, forwarding a few of them to Microsoft Corp.
Does someone at Microsoft have a News administration function? Or is it
simply a task that someone among a large group gets assigned to perform
on an irregular basis?
Till you mentioned the recent changes, I was kind of assuming that Microsoft
had put it on the back burner due to recession layoffs.
D. Stussy
news:hfuld0$9hm$1...@news.trigofacile.com...
> It makes me think that it is such a long time since we last heard of
> the perl.* hierarchy...
> Maybe I will drop them a mail to ask them whether they have changes
> to do (perl.cpan.workers seems to be an active list not present in
> perl.* current checkgroups).
I can tell you that they're alive since I peer with them, but you already
knew that.
I think what would be useful is if ALL managed hierarchies issued a
checkgroups sometime during this month. I note that only 22 hierarchies
(counting the Big-8 as 8) seem to do so on a regular basis. There seem to
be 7 others that have issued one in the past year. However, control.ctl
lists alot more. (At least, this is what I have observed). My written
notes indicate 5 more which have, but not recently.
No date means I saw one during 2009 before October 20:
# grep CHECKGROUPS /news/index/newsgroups
belwue.CHECKGROUPS Checkgroups processed on 2009-11-30 23:45:28 UTC
ch.CHECKGROUPS Checkgroups processed
comp.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
de.CHECKGROUPS Checkgroups processed on 2009-11-03 00:32:28 UTC
fido7.CHECKGROUPS Checkgroups processed on 2009-11-14 21:02:53 UTC
humanities.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
hun.CHECKGROUPS Checkgroups processed on 2009-12-02 05:24:24 UTC
it.CHECKGROUPS Checkgroups processed
ka.CHECKGROUPS Checkgroups processed on 2009-11-15 10:07:27 UTC
kgk.CHECKGROUPS Checkgroups processed on 2009-11-30 17:01:12 UTC
linux.CHECKGROUPS Checkgroups processed
luebeck.CHECKGROUPS Checkgroups processed
microsoft.CHECKGROUPS Checkgroups processed on 2009-12-01 18:35:13 UTC
misc.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
news.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
nippon.CHECKGROUPS Checkgroups processed on 2009-12-01 01:11:39 UTC
nl.CHECKGROUPS Checkgroups processed on 2009-11-21 04:31:36 UTC
no.CHECKGROUPS Checkgroups processed
opennews.CHECKGROUPS Checkgroups processed on 2009-12-01 00:18:13 UTC
openwatcom.CHECKGROUPS Checkgroups processed on 2009-12-04 10:50:05 UTC
rec.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
sci.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
soc.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
szaf.CHECKGROUPS Checkgroups processed on 2009-12-01 07:47:29 UTC
talk.CHECKGROUPS Checkgroups processed on 2009-11-15 16:08:30 UTC
uk.CHECKGROUPS Checkgroups processed
us.CHECKGROUPS Checkgroups processed on 2009-12-01 08:04:10 UTC
z-netz.CHECKGROUPS Checkgroups processed on 2009-12-01 14:25:19 UTC
D. Stussy
> "News Guy" <Ne...@Guy.com> a �crit dans le message de
news:4B219C2F...@Guy.com...
> > Microsoft could choose tommorrow to end it's participation or peering
> > with usenet if that peering does not meet corporate objectives or it
> > requires too much oversight/censorship on their part. What would
Julien
> > �LIE do in that case? Would he issue control messages to delete the
> > entire microsoft.* hierarchy from usenet? What purpose would that
> > serve?
>
> I would post an article here to mention what happened, as I have just
done
> for the removal of 535 newsgroups. We would discuss the appropriate way
> to make the microsoft.* hierarchy historic/defunct.
>
> I believe there would be rmgroup control messages sent so as to properly
> clean the hierarchy.
Actually, I would prefer that an [almost] empty checkgroups message be
issued. If a hierarchy can't be emptied via checkgroups, then it removes
all but one, and that one gets an rmgroup.
Julien ÉLIE
> which implies that only 1772 newsgroups currently remain.
It appears that they sent last month (November, 11th) a message in
a hundred of these groups:
http://groups.google.fr/group/microsoft.public.activex.programming.control.webwiz/browse_frm/thread/a9b54f2bf47e116d
<news:e3zJcJxY...@TK2MSFTNGP06.phx.gbl>
"This news group has been retired based on low volume participation over
time."
A few bits of information:
Microsoft seems to work on "NNTP Bridge", which is an application (Windows-based?)
installed on the client. It uses NNTP authentication to use Windows Live ID
credentials to authenticate on the web forum.
On Microsoft Connect (the link in that page above), I read:
December Update
We will be releasing a more stable version of the beta together with the upcoming
Forum release, targeted at Dec 15. Within a few days after this release, you will
be able to download the newer version of the Bridge from this site. We have made
great progress in terms of improving the stability of the Bridge. Many bugs were
fixed that could have caused the "time out error" or "cannot post error", which
contributes to most of your feedback.
Q: Which newsreaders have been specifically tested for use with the Microsoft
Forums NNTP Bridge Client Beta?
A: The NNTP Bridge Beta has been designed to work with any newsreader. It's been
specifically tested for use with Outlook Express, Windows Mail, and Windows Live Mail.
Q: What is the purpose of the Microsoft Forums NNTP Bridge Client Beta?
A: The NNTP Bridge Beta is a small client application that exists alongside your
newsreader that allows you read/write access to popular Microsoft forums such as
the MSDN and TechNet forums. It's designed to allow you to download current forum
messages into the newsreader of your choice, respond in an offline scenario if you
so choose, and upload responses.
In the installation section, I see that the news server is 127.0.0.1. The bridge
does everything afterwards.
With that sort of bridge, I doubt that peering will go on... Each post is authentified...
And the only reference server is Microsoft's one. (Which is what they initially wanted
with microsoft.* but they underestimated article reinjections outside, and it finally
ended up in microsoft.* as we know it today.)
--
Julien ï¿œLIE
ￜ Bis repetita placent. ￜ (issu de Horace)
Julien ÉLIE
> Julien, there has always been something screwy about my setup that affects how
> I see certain characters. When reading messages, I see your last name with
> an acute accent over E. When I invoke the composer (just vim), I see an acute
> accent over A. I've been changing it to unaccented A.
>
> Undoubtably I've been misspelling your last name for years, so I apologize.
Don't worry; you shouldn't apologize for that.
I don't mind seeing "É" transformed. I'm accustomed to :)
Even on my identity card, it is written "JULIEN [CÉDRIC] ELIE". There is
an accent on my middle name Cédric, but non on my last name Élie. When I renewed
my card in July, they couldn't do anything (either remove all accents, or put
all accents -- I was fine with both these options) because it is the way
I was registered at my birth!
Anyway, I don't mind, really, if people put an accent or not on upper case
letters. The "Académie Française" (our authority for French) says it is
mandatory. However, lots of people do not use it because either they learnt
at school that they shouldn't use accentuated letters, or they do not know
how to write them on a computer.
I often use "Elie" on software when I see that alphabetical sorting is broken
(I then appear after "Z"), or "É" is transformed to another char ("?" or mangled
UTF-8). Sometimes, I see that it is stored as "Élie" in database... Not
good... It might break something elsewhere in the program...
Maybe your program does not work well with ISO-8859-1? I will encode this
article in UTF-8.
And do not worry about misspelling my name. Otherwise I will worry about
misspronouncing yours (and others) :)
--
Julien ÉLIE
« Mais pourquoi courent-ils si vite ? Pour gagner du temps ! Comme le temps,
c'est de l'argent... plus ils courent vite, plus ils en gagnent. » (Raymond Devos)
Julien ÉLIE
> Actually, I would prefer that an [almost] empty checkgroups message be
> issued. If a hierarchy can't be emptied via checkgroups, then it removes
> all but one, and that one gets an rmgroup.
I think a hierarchy can be emptied via checkgroups but that current software
do not handle that.
RFC 5537:
checkgroups-body = *( valid-group CRLF )
We need to provide a scope in the Control: header so that it can be considered
as an empty checkgroups for that hierarchy. (Or rely on the control.ctl entry.)
The problem with that process is that the remaining newsgroup would be the
administrative one (where control articles are fed -- Newsgroups: header).
Then, we remove it. If the checkgroups has not been received (yet), only one
newsgroup will be removed.
So maybe sending rmgroup articles for every newsgroup is better. And it can
be done over several weeks, if needed. It would be proper.
Anyway, we will discuss it further when and if it happens.
Julien ÉLIE
> Does someone at Microsoft have a News administration function? Or is it
> simply a task that someone among a large group gets assigned to perform
> on an irregular basis?
I don't know. It was not clear at all when I tried to obtain information
about that. It seems to be centralized in the USA; and I believe they
receive local (from other countries) feedback from correspondants or MVPs
(Most Valuable Professionals).
John Eddy would know more, if he reads us.
--
Julien ÉLIE
« L'amateur de plaisir solitaire fait souvent bande à part. » (Pierre Dac)
Adam H. Kerman
>Hi Adam,
>>Julien, there has always been something screwy about my setup that
>>affects how I see certain characters. When reading messages, I see
>>your last name with an acute accent over E. When I invoke the composer
>>(just vim), I see an acute accent over A. I've been changing it to
>>unaccented A.
>>Undoubtably I've been misspelling your last name for years, so I apologize.
>Don't worry; you shouldn't apologize for that. . . .
If it's on your birth registration, then I'll use Elie.
>Maybe your program does not work well with ISO-8859-1? I will encode this
>article in UTF-8.
What I'm using dislikes UTF-8 entirely. If I'm supposed to see an accute
accent over E, then I'm seeing the correct glyphs when I'm reading messages.
This is coincidental, of course, as nothing I'm using is MIME aware.
For some reason, the composer is using the same codes to display other glyphs.
I've never found where the encodings are chosen.
Adam H. Kerman
>http://groups.google.fr/group/microsoft.public.activex.programming.control.webwiz/browse_frm/thread/a9b54f2bf47e116d
> <news:e3zJcJxY...@TK2MSFTNGP06.phx.gbl>
>See:
> December Update
Doesn't Outlook Express authenticate (not that I'd use other than Unix
newsreaders)? I don't understand why existing tools aren't used.
> Q: What is the purpose of the Microsoft Forums NNTP Bridge Client Beta?
> A: The NNTP Bridge Beta is a small client application that exists
> alongside your newsreader that allows you read/write access to popular
> Microsoft forums such as the MSDN and TechNet forums. It's designed
> to allow you to download current forum messages into the newsreader
> of your choice, respond in an offline scenario if you so choose,
> and upload responses.
Does this suggest that certain forums aren't newsgroups? Now I see what's
going on.
Julien ÉLIE
> Doesn't Outlook Express authenticate (not that I'd use other than Unix
> newsreaders)? I don't understand why existing tools aren't used.
Yes, it can authenticate. But here, authentication is made on a web page.
The bridge rewrites your (NNTP) AUTHINFO commands in (HTTP) POST requests,
as well as all other NNTP commands (to retrieve articles, overview, etc.).
They will probably add an extension to Windows Mail so that it can also
retrieve the "score" of an article (5 stars out of 5) or note an article,
the same way the web interface works.
> Does this suggest that certain forums aren't newsgroups? Now I see what's
> going on.
I think that they no longer use newsgroups... But there is a need to have
an NNTP access (for all the advantages we know), so they are bridging it!
--
Julien �LIE
� Je viens de rencontrer Isoc�le : il a une id�e pour un nouveau triangle. �
(Woody Allen)
Adam H. Kerman
>Hi Adam,
>>Doesn't Outlook Express authenticate (not that I'd use other than Unix
>>newsreaders)? I don't understand why existing tools aren't used.
>Yes, it can authenticate. But here, authentication is made on a web page.
Not directly to a server? That's sounds more complicated than it need be.
>The bridge rewrites your (NNTP) AUTHINFO commands in (HTTP) POST requests,
>as well as all other NNTP commands (to retrieve articles, overview, etc.).
>They will probably add an extension to Windows Mail so that it can also
>retrieve the "score" of an article (5 stars out of 5) or note an article,
>the same way the web interface works.
Goodness. I take it "score" is in double quotes, because Windows Mail scoring
is nothing like scoring on strn or slrn.
>>Does this suggest that certain forums aren't newsgroups? Now I see what's
>>going on.
>I think that they no longer use newsgroups... But there is a need to have
>an NNTP access (for all the advantages we know), so they are bridging it!
Hm. That's too bad. Any interfacing should be between a Web forum server
and News server in their network, not on the client side... Well, strike
that. Posting to a Web forum is a rather different style than posting to
a News server that interfacing can be counterproductive to discussion.
I guess learning to post Web forum style while using a newsreader will
entirely eliminate any style advantage of News articles over Web forum posts.
Julien ÉLIE
>> Yes, it can authenticate. But here, authentication is made on a web page.
>
> Not directly to a server? That's sounds more complicated than it need be.
I do not know, as you point that out, why a local bridge is needed. It is
strange that they did not set up a server with an NNTP interface...
They may have their reasons (that I haven't understood yet).
>> They will probably add an extension to Windows Mail so that it can also
>> retrieve the "score" of an article (5 stars out of 5) or note an article,
>> the same way the web interface works.
>
> Goodness. I take it "score" is in double quotes, because Windows Mail scoring
> is nothing like scoring on strn or slrn.
"Scoring" means giving a mark to an article, to see which are the most
interesting to read. It is something people keep asking on their web
interface...
> I guess learning to post Web forum style while using a newsreader will
> entirely eliminate any style advantage of News articles over Web forum posts.
Unfortunately :-/
Julien ÉLIE
>> It makes me think that it is such a long time since we last heard of
>> the perl.* hierarchy...
>> Maybe I will drop them a mail to ask them whether they have changes
>> to do (perl.cpan.workers seems to be an active list not present in
>> perl.* current checkgroups).
>
> I can tell you that they're alive since I peer with them, but you already
> knew that.
Four groups have just been officially created: perl.kiokudb, perl.pod2,
perl.qpsmtpd.dev, and perl.cpan.workers.
That's great!
--
Julien �LIE
� -- Poussez pas derri�re !
-- Pas si vite devant ! � (Ast�rix)
D. Stussy
> > Actually, I would prefer that an [almost] empty checkgroups message be
> > issued. If a hierarchy can't be emptied via checkgroups, then it
removes
> > all but one, and that one gets an rmgroup.
>
> I think a hierarchy can be emptied via checkgroups but that current
software
> do not handle that.
>
> RFC 5537:
>
> checkgroups-body = *( valid-group CRLF )
>
> We need to provide a scope in the Control: header so that it can be
considered
> as an empty checkgroups for that hierarchy. (Or rely on the control.ctl
entry.)
>
> The problem with that process is that the remaining newsgroup would be
the
> administrative one (where control articles are fed -- Newsgroups:
header).
> Then, we remove it. If the checkgroups has not been received (yet), only
one
> newsgroup will be removed.
> So maybe sending rmgroup articles for every newsgroup is better. And it
can
> be done over several weeks, if needed. It would be proper.
>
> Anyway, we will discuss it further when and if it happens.
I say send the checkgroups message first - so that those systems that can
handle it remove all (that is possible) in one batch. Afterwards, follow
up with individual rmgroups messages, which should be ignored on those
systems that processed the checkgroups message.
Julien ÉLIE
> Does this suggest that certain forums aren't newsgroups? Now I see what's
> going on.
Bad news from Microsoft. One of my contact has just told me that they
are in the process of stopping their newsgroups. They will eventually
replace them all with web forums.
As for Windows 7 forums, here are three relevant URLs:
- http://social.answers.microsoft.com/Forums/en-US/category/windows7 for the public
- http://social.technet.microsoft.com/Forums/en-US/category/w7itpro for professionals
- http://social.msdn.microsoft.com/Forums/en-US/categories for developers
I hope they will manage to still gateway them...
--
Julien �LIE
� -- Cet homme qui est sorti du palais, nous renseignera peut-�tre
sur la fa�on d'y entrer. Suivons-le.
-- Mais... Il sait sortir d'accord, mais rien ne prouve qu'il
sache entrer, et... � (Ast�rix)
Adam H. Kerman
>Hi Adam,
>>Does this suggest that certain forums aren't newsgroups? Now I see what's
>>going on.
>Bad news from Microsoft. One of my contact has just told me that they
>are in the process of stopping their newsgroups. They will eventually
>replace them all with web forums.
>As for Windows 7 forums, here are three relevant URLs:
> - http://social.answers.microsoft.com/Forums/en-US/category/windows7
>for the public
> - http://social.technet.microsoft.com/Forums/en-US/category/w7itpro
>for professionals
> - http://social.msdn.microsoft.com/Forums/en-US/categories for developers
>I hope they will manage to still gateway them...
But, as you've explained, they intend to do it on the client end in a
bizarre manner.
Wish they'd make a public announcement of their plans.
Julien ÉLIE
> I say send the checkgroups message first - so that those systems that can
> handle it remove all (that is possible) in one batch. Afterwards, follow
> up with individual rmgroups messages, which should be ignored on those
> systems that processed the checkgroups message.
All right, that's a good idea. I sent a checkgroups just before a first
batch of rmgroup control articles.
However, note that it is not a best practice to honour checkgroups with
so many changes. By default, controlchan will not honour them. It is
/maxdocheckgroups/ in control.ctl -- see
<http://www.eyrie.org/~eagle/software/inn/docs/control.ctl.html>.
So it will normally have to be processed by hand (the ideal would be
before the other batches of rmgroup control articles -- maybe you will
say that I should have waited for a few days before sending the first
batch, but well, the number of days to wait is not easily defined...).
D. Stussy
> > I say send the checkgroups message first - so that those systems that
can
> > handle it remove all (that is possible) in one batch. Afterwards,
follow
> > up with individual rmgroups messages, which should be ignored on those
> > systems that processed the checkgroups message.
>
> All right, that's a good idea. I sent a checkgroups just before a first
> batch of rmgroup control articles.
>
> However, note that it is not a best practice to honour checkgroups with
> so many changes. By default, controlchan will not honour them. It is
> /maxdocheckgroups/ in control.ctl -- see
> <http://www.eyrie.org/~eagle/software/inn/docs/control.ctl.html>.
Noted. Although as this is now past, it's too late to suggest to other
server admins that they raise the level of acceptable changes (like I did).
> So it will normally have to be processed by hand (the ideal would be
> before the other batches of rmgroup control articles -- maybe you will
> say that I should have waited for a few days before sending the first
> batch, but well, the number of days to wait is not easily defined...).
I also suggest that the rmgroup messages NOT be batched. They should be
spaced a few minutes apart. 15 minutes apart will allow 96 per day.
Otherwise, one may get INN servers constantly churning on active file
reloads for a significant time as if they were down. One per 15 minutes
spaces these apart that they can catch up inbetween. (5 or 10 minutes
might be sufficient, but 15 -> ~100/day which seemed to be your target
rate.)
Julien ÉLIE
> I also suggest that the rmgroup messages NOT be batched. They should be
> spaced a few minutes apart. 15 minutes apart will allow 96 per day.
> Otherwise, one may get INN servers constantly churning on active file
> reloads for a significant time as if they were down. One per 15 minutes
> spaces these apart that they can catch up inbetween. (5 or 10 minutes
> might be sufficient, but 15 -> ~100/day which seemed to be your target
> rate.)
The first "batch" of rmgroup messages was spaced 30 seconds apart.
I will increase that time for the other "batches".
--
Julien �LIE
� Pour une personne optimiste, le verre est � moiti� plein.
Pour une personne pessimiste, il est � moiti� vide.
Pour l'informaticien, il est deux fois plus grand que n�cessaire. �
Adam H. Kerman
I finally figured it out where the mismatch is!
I'm using PuTTY to emulate a terminal on a linux host. PuTTY must be told
how to translate received characters, because it doesn't know what is being
used. The linux server uses UTF-8, which I didn't realize.
You've been using a variety of Latin-1, Latin-9 (almost Latin-1 with a few
changes line the euro), and UTF-8. Your encoded-word syntax on your headers
appears to be correct.
So, when I read your message, I have to set the translation to whatever
you are using. But when I followup to your message, I'm using vi to edit
on on the remote host, which is using UTF-8. That's why I'm seeing two
different characters, because I have to manually change the translation
setting to reflect two different character sets in use.
Normally, I don't have the translation set to UTF-8 as I don't know of a
character set with fixed-width characters usable with UTF-8, unless I use
one of the character sets for CJK, which I don't care for. Courrier New
is the easiest to read, but the font I have is for Latin-1, not UTF-8.
When I use it with UTF-8, it doesn't further translate all the character
it includes, just displays a solid box for non-ASCII characters because most
of them are in different positions.
Maybe there's a more encompassing Courrier New out there for UTF-8, but
I've not seen one. At least if I had a more convenient way to translate
a second time between UTF-8 and Latin-1, I could at least display all
the Latin-1 characters correctly.
I don't get a satisfactory result, but nobody is doing anything wrong.
Aargh.
Julien ÉLIE
> You've been using a variety of Latin-1, Latin-9 (almost Latin-1 with a few
> changes line the euro), and UTF-8. Your encoded-word syntax on your headers
> appears to be correct.
You may sometimes also see Windows-1252 in a few headers (encoded in QP).
But not in the Content-Type: header, for the encoding of the article body.
> So, when I read your message, I have to set the translation to whatever
> you are using.
So you are manually converting each article I send? It is not a
user-friendly news reader!
> Normally, I don't have the translation set to UTF-8 as I don't know of a
> character set with fixed-width characters usable with UTF-8, unless I use
> one of the character sets for CJK, which I don't care for. Courrier New
> is the easiest to read, but the font I have is for Latin-1, not UTF-8.
[...]
> Maybe there's a more encompassing Courrier New out there for UTF-8, but
> I've not seen one.
Can't you use Courrier New for the subset of characters it knows?
Courrier New knows more characters than there are in Latin-1. It even
includes Greek.
(I personally use Consolas, which I find better looking that Courrier New.
But, like Courrier New, it is not for all UTF-8 characters.)
--
Julien ï¿œLIE
ￜ Qu'est-ce que je vous sers pour arroser le sanglier bouilli ?
De l'eau chaude, de la cervoise tiￜde ou du vin rouge glacￜ ? ￜ (Astￜrix)
Adam H. Kerman
>Hi Adam,
>>So, when I read your message, I have to set the translation to whatever
>>you are using.
>So you are manually converting each article I send? It is not a
>user-friendly news reader!
The issue isn't the newsreader, which just passes characters along. I
have to choose the appropriate translation in the terminal emulation,
which is PuTTY. PuTTY doesn't know what character set is in use, which
is why I must choose.
>>Normally, I don't have the translation set to UTF-8 as I don't know of a
>>character set with fixed-width characters usable with UTF-8, unless I use
>>one of the character sets for CJK, which I don't care for. Courrier New
>>is the easiest to read, but the font I have is for Latin-1, not UTF-8.
>[...]
>>Maybe there's a more encompassing Courrier New out there for UTF-8, but
>>I've not seen one.
>Can't you use Courrier New for the subset of characters it knows?
Logically, one would think that the two settings, Translation: Received
Data Assumed To Be In Which Character Set and Font: Language Script don't
have to match. The Language Script setting don't use the registered character
set names, but names like Western, which I think is supposed to be Latin-1
but I don't know where it's defined.
The second translation problem is limited to viewing the non-ASCII characters
in a UTF-8 article. I can view them easily enough when the article is in
Latin-1, either ISO-8859-1 or -15 or I assume Windows-1252 when I tell
the emulation which translation to use.
Things are MIME aware, as I'm not seeing the character codes when
encoded-word sequences are used on headers.
>(I personally use Consolas, which I find better looking that Courrier New.
>But, like Courrier New, it is not for all UTF-8 characters.)
I'm not fond of sanserif.
Julien ÉLIE
> The issue isn't the newsreader, which just passes characters along. I
> have to choose the appropriate translation in the terminal emulation,
> which is PuTTY. PuTTY doesn't know what character set is in use, which
> is why I must choose.
Cannot trn be configured to use a "UTF-8 mode" to encode characters to
UTF-8 when it writes to the terminal?
> Logically, one would think that the two settings, Translation: Received
> Data Assumed To Be In Which Character Set and Font: Language Script don't
> have to match. The Language Script setting don't use the registered character
> set names, but names like Western, which I think is supposed to be Latin-1
> but I don't know where it's defined.
I think it refers to:
http://www.unicode.org/standard/supported.html
Unicode defines "Latin", a superset of "Western European". It is indeed better,
because American people use Latin script and are not in Europe :)
http://www.unicode.org/versions/Unicode5.2.0/ch07.pdf
Unicode follows ISO/IEC 8859-1 in the layout of Latin letters up to U+00FF.
[...]
The Latin Extended-A block also covers additional characters from ISO/IEC 6937.
[...]
I do not see a reference to the Western script in the Unicode web site.
Better say Latin, then.
--
Julien �LIE
� Ibi etsi uis te non esse sed es ibi. �
Adam H. Kerman
>Hi Adam,
>>The issue isn't the newsreader, which just passes characters along. I
>>have to choose the appropriate translation in the terminal emulation,
>>which is PuTTY. PuTTY doesn't know what character set is in use, which
>>is why I must choose.
>Cannot trn be configured to use a "UTF-8 mode" to encode characters to
>UTF-8 when it writes to the terminal?
It just passes characters through as is. Lack of bloat is one reason why I
like it so much. I'm not sure why we'd want a newsreader itself to perform
translations.
>>Logically, one would think that the two settings, Translation: Received
>>Data Assumed To Be In Which Character Set and Font: Language Script don't
>>have to match. The Language Script setting don't use the registered character
>>set names, but names like Western, which I think is supposed to be Latin-1
>>but I don't know where it's defined.
>I think it refers to:
> http://www.unicode.org/standard/supported.html
I have a feeling there's no standard script at all, that it varies from
what the author of one font has included to the next. Or it's something
to do with pre-Unicode code pages.
>Unicode defines "Latin", a superset of "Western European". It is indeed
>better, because American people use Latin script and are not in Europe :)
Romanian is a Romance language. Last time I looked, Romania wasn't in
Western Europe either.
Adam H. Kerman
Oops. I see I left in non-ASCII characters without adding appropriate
MIME headers, in a message in which I'm bitching about the troubles I'm
having getting characters to display with the appropriate glyphs.
My bad
Julien ÉLIE
> Oops. I see I left in non-ASCII characters without adding appropriate
> MIME headers, in a message in which I'm bitching about the troubles I'm
> having getting characters to display with the appropriate glyphs.
You know that with Grand-Son-of, MIME headers are mandatory and all
your articles are in fact invalid...
--
Julien �LIE
� En v�rit�, le chemin importe peu, la volont� d'arriver suffit � tout. �
Adam H. Kerman
>You know that with Grand-Son-of, MIME headers are mandatory and all
>your articles are in fact invalid...
I know, but I'm choosing to go naked for the time being.
Peter J Ross
ÉLIE <iul...@nom-de-mon-site.com.invalid> wrote:
> Hi Adam,
>
>> Oops. I see I left in non-ASCII characters without adding appropriate
>> MIME headers, in a message in which I'm bitching about the troubles I'm
>> having getting characters to display with the appropriate glyphs.
>
> You know that with Grand-Son-of,
Please don't call it that. It suggests that Son-of had some kind of
special significance, instead of being totally unofficial and utterly
unadopted.
> MIME headers are mandatory and all
> your articles are in fact invalid...
If that's what the new RFC says, then the new RFC is an ass.
--
PJR :-)
slrn newsreader v0.9.9p1: http://slrn.sourceforge.net/
extra slrn documentation: http://slrn-doc.sourceforge.net/
newsgroup name validator: http://pjr.lasnobberia.net/usenet/validator
Peter J Ross
ÉLIE <iul...@nom-de-mon-site.com.invalid> wrote:
> Hi Adam,
>
>> The issue isn't the newsreader, which just passes characters along. I
>> have to choose the appropriate translation in the terminal emulation,
>> which is PuTTY. PuTTY doesn't know what character set is in use, which
>> is why I must choose.
>
> Cannot trn be configured to use a "UTF-8 mode" to encode characters to
> UTF-8 when it writes to the terminal?
No, trn is obsolete. Or, to be generous, trn is obsolescent.
It's been several years since trn received its last significant update
or bugfix. There's no reason at all why you should have to do anything
to accommodate the defects of an old, buggy newsreader.
<http://sourceforge.net/projects/trn/>
Notice the triviality of all recent changes.
This is off-topic anyway.
Adam H. Kerman
>Julien ÉLIE <iul...@nom-de-mon-site.com.invalid> wrote:
>>Hi Adam,
>>>Oops. I see I left in non-ASCII characters without adding appropriate
>>>MIME headers, in a message in which I'm bitching about the troubles I'm
>>>having getting characters to display with the appropriate glyphs.
>>You know that with Grand-Son-of,
>Please don't call it that. It suggests that Son-of had some kind of
>special significance, instead of being totally unofficial and utterly
>unadopted.
Huh?
Nearly everybody calls it Grandson-of; I never remember the RFC numbers.
Henry Spencer's document reflected current practice, which had moved beyond
RFC 1036, and made a number of recommendations that were widely adopted.
>>MIME headers are mandatory and all your articles are in fact invalid...
>If that's what the new RFC says, then the new RFC is an ass.
IETF has imposed consistent use of MIME across all document formats in use
on the Internet; it's not just for Mail anymore.
Waitaminit... Yer not claiming trn4 is utterly borked and out of date?
Julien ÉLIE
>>>You know that with Grand-Son-of,
>>
>>Please don't call it that. It suggests that Son-of had some kind of
>>special significance, instead of being totally unofficial and utterly
>>unadopted.
>
> Huh?
> Nearly everybody calls it Grandson-of;
"Nearly everybody"? Is it rhetorical? :)
When I used it, it was in fact more because I know you like that term
when you speak of USEFOR. My remark was "private", in the tone of
a joke, because you told "[you] left in non-ASCII characters without
adding appropriate MIME headers" whereas your news reader do not
send such headers.
Adam H. Kerman
>Hi Adam,
>>>Julien wrote:
>>>>You know that with Grand-Son-of,
>>>Please don't call it that. It suggests that Son-of had some kind of
>>>special significance, instead of being totally unofficial and utterly
>>>unadopted.
>>Huh?
>>Nearly everybody calls it Grandson-of;
>"Nearly everybody"? Is it rhetorical? :)
The audience for the several RFCs are a few dozen people, most of whom
we know off the top of our heads. A survey is feasible.
>When I used it, it was in fact more because I know you like that term
>when you speak of USEFOR. My remark was "private", in the tone of
>a joke, because you told "[you] left in non-ASCII characters without
>adding appropriate MIME headers" whereas your news reader do not
>send such headers.
If I weren't lazy, I'd add the correct MIME headers. Actually, I need a parser
to get rid of all those non-ASCII space characters that Google Groups (spit)
adds. If I can't see them in the quote, it's a problem if I don't eliminate
them.
Wish I could take credit for Grandson-Of, but somebody else thought of
it long ago.