Re: New module proposal: tracker

2009-08-19 Thread Sankar P
Not contributing to the core discussion.

On Wed, Aug 19, 2009 at 3:26 PM, Alan Coxa...@lxorguk.ukuu.org.uk wrote:
 I think there lies a misassumption. The actual indexing has a fairly high
 cost. The cost of extracting metadata while indexing ought to be
 relatively low in comparison. That argues that allowing stuff to plug
 into the indexing based on file type is useful. It's not really function
 creep either given the only interface the indexer needs is

        - who is associated with this file type (which exists)
        - give me your metadata for this file content

One short coming in this approach will be, It will cause a problem
where multiple applications can be associated with a file-type, over a
period of time. For instance, for .mbox files, the applications could
vary like: Evolution, Mutt, Pine, Claws, Thunderbird, etc. And it is
common among some people to switch between applications; not for email
but other applications like PDF-viewer, etc. once in few months.

All these different applications may have different interpretations
for what is a valid meta-data. For instance, Evolution will consider
any thing within ** will be metadata whereas mutt might consider
subject is the metadata etc. So every time the user switches
applications, the earlier collected meta-data might need some brushup.

In my personal biased opinion, the need for meta-data and desktop
search is over-rated. Internet is extra-ordinarily mammoth and is
impossible to reach without a search engine as you won't even know how
many sites exist. For desktop the scale of the things is less,
individual application-provided-search is enough and will satisfy the
needs of most of the users. ctags, mairix etc. can provide specialized
and more effective searching.

-- 
Sankar P
http://psankar.blogspot.com
___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list


Re: Unifying name for Plugins/Extensions/etc.

2007-12-19 Thread Sankar P
On Wed, 2007-12-19 at 17:33 -0600, Shaun McCance wrote:
  (The fact that Plugins is very hard to translate was brought up by
 a
  number of people. Extensions apparently does not have this problem
  (comment #8 on the blog): I vote for Extensions. In Italian you
 can't
  really translate plugin, thus one has got to let it untraslated.
 Every
  time I use the word plugin I have to spend 5 min explaining it.
 So,
  besides being very culturally insensitive, it's time consuming as
  well.)
  
 Since plugin and extension are effectively synonomous,
 what's to stop the Italian translators for using their
 extension word for plugin?

I will vote for continuing the usage of Plugins over any new jargon.
Evolution, gedit, rhythmbox, gvim, anjuta all use the word Plugins.
IIRC, Eclipse, Mono-develop, compiz-fusion as well.

From my past experience, Using a new jargon may not be so well-received
by users. We changed VFolders in Evolution to Search Folders. Still
there are a lots of people using both the terms interchangeably. And all
the code refers as vfolders, so developers end-up using VFolders more
frequently than Search-folders.

If translation is tough in Italian for the word Plugins, as Shaun
said, why can't we use the Italian word for extension in that place ?

Changing the term Plugin means, All the dependent user-docs, wiki pages,
developer-docs, FAQ pages, screenshots should also be updated. Some
distros ship plugins as seperate RPMs/packages. So these rpm/package
names should also be changed. IMHO it is too much of an effort with
lesser benefits. 

-- 
Sankar P
Evolution Plugins (Extensions ?) Maintainer

___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list


Re: Evolution Plugins (Was Re: Rise of the Plugins)

2007-11-15 Thread Sankar P

On Thu, 2007-11-15 at 07:22 +, Philip Withnall wrote:
 We currently have the other extreme: several different applications
 each
 have their own take on the address book (and then each have their own
 different ways of integrating/syncing it with e-d-s). Why not just
 have
 one desktop-wide address book (I'm thinking Soylent here), and have
 the
 mail application use that instead? This would give us the opportunity
 to
 swap out the address book application in favour of other things if
 necessary, such as an application which would use some web service as
 an
 address book.


I am not much aware of what Soylent is. So I cannot comment about using
that atm. However, evolution-data-server is the system-wide address-book
that will answer your needs. IIRC, eds exposes a dbus interface which
can be consumed by any of the desktop applications, say Contacts,
Gnome-calendar, OpenOffice, etc. Evolution-Addressbook (not eds) is just
another client for the evolution-data-server addressbook. Probably Ross
Burton can explain in more detail. 


  IMHO a better approach will be to make the user choose what
 components
  he want to see in his Switcher. Say if someone wants to use only
 Mailer
  and Address-book, do not bother showing the Calendar component in
 the
  switcher.
 
 But that's effectively like turning the components into plugins, and
 then disabling some of them; it comes back to the problem that core
 functionality shouldn't be in plugins, and you might as well split off
 such plugins into separate applications.
 
  Atleast for me, It will be far more useful than launching two or
 three
  applications everyday morning (Mail/Calendar/Tasks)
 
 Isn't that what your startup programs list is for? :-P


Yeah. I should have made it a little elaborate. Everytime I (or any
corporate user) launches evolution, what I want to do with that is:

- Send/receive mails

- Accept/Decline/Schedule appointments

- Mark the tasks that I have completed, so my boss knows if it is worth
paying me for.


In addition to these basic activities sometimes I may be publishing my
calendar etc. 

All these activities adhere to the Communication aspect of an office
worker. And I do this atleast 4-5 times a day. Rest of the times I spend
on gnome-planner, OO, vi etc. So opening three applications for this
communication aspect, each time, may not be an appealing option.

I do not say that it is bad/evil to split applications but IMHO it may
be a better idea to have Gnome applications aligned/coupled on the lines
user behaviors, bringing similar applications in a shell, but
very-loosely tied. In the same way how OO has a Spreadsheet, Doc-writer,
Presentation-tool. Just my 0.001 Paisa(1). though :)

 
 Regards,
 Philip
-- 
Sankar P
(1) - http://en.wikipedia.org/wiki/Paisa


___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list


Re: Evolution Plugins (Was Re: Rise of the Plugins)

2007-11-15 Thread Sankar P

On Thu, 2007-11-15 at 21:20 +1100, Jeff Waugh wrote:
 quote who=Alexander Larsson
 
  This is kinda a backward question. The reason you want to split it out is
  of course not because you want to use the mailer without the addressbook,
  it is because you want to use the address book without the mailer. 
 
 ... and you can provide a deliciously optimised user experience for both use
 cases. It's worth looking at the user experience of the OS X communications
 tools, and possibly even the Windows Vista ones (although they're definitely
 uglier) to see why this is such a good way to go. It would be *FANTASTIC* to
 see Evolution become a suite of wonderful, focused applications instead of
 the Outlook clone it is today (and, sure, designed to be).

The components of Evolution are loosely ties to the Evolution shell, so
that you can use any component without other components loaded at all.

To answer Alexander, if you have at least one account configured, you
can directly launch Evolution Addressbook without ever having anything
to do with Mail component. If you launch Evolution Addressbook from
your gnome main menu, it should not ask your IMAP password. If it does,
it is a bug, please file that.

The reason for bringing up the account setup wizard when no account is
configured is because it was expected that configuring an account will
be the first thing a user will want to do. I accept that the bringing-up
of account-setup wizard at startup, when no account is configured, needs
to go away, with a Forget It. I want to use Evolution button, in the
wizard-beginning. It is a needless restriction.

 
 I must admit, I've been trying to usher anyone involved in Evolution in this
 direction for years, so I'm very firmly on one side of the fence here. ;-)
 
 - Jeff
 
-- 
Sankar P
Harver's Law: A drunken man's words are a sober man's thoughts

___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list


Evolution Plugins (Was Re: Rise of the Plugins)

2007-11-14 Thread Sankar P
Hi All,

Some of the issues that were discussed (here and in #496839) and my
personal take on them will be:

1) Plugin list being un-manageable 

I am thinking of adding a smart-search-box which will query based on
plugin-names and pre-defined-tags associated with each plugin. This will
make it easy to find out the plugin that you are looking for.

Do you have any suggestion for how you want the Evolution Plugin-manager
to look like ?  Can you send any mockups that you have in mind ?

2) Some plugins should not be shown 

This may not be possible as people may not want a functionality
implemented by a plugin. Disabling this plugin helps them to save
screen-real-estate (menu-items etc) and reduces memory foot-print. 

If any of the core feature is implemented as a plugin, I accept that it
should not be shown in the plugin-manager and will happily remove it
from the plugin-manager list. (It should have been core-code than a
plugin in the first place though) 

HTTP Calendars and IMAP Featuers are not core features since it may not
be needed by everyone. Say a corporate Evolution-GroupWise user.

3) Splitting Evolution into individual applications

I have seen this asked by people a few times. But I wonder what good a
mail-client or Calendar-client will be without an addressbook. 

Evolution shell will not load a component (Tasks or Notes) unless you
click open the component (except for addressbook) explicitly. But some
people have a mis-conception that all components are always loaded (and
hence consumes memory etc.)

IMHO a better approach will be to make the user choose what components
he want to see in his Switcher. Say if someone wants to use only Mailer
and Address-book, do not bother showing the Calendar component in the
switcher.

Atleast for me, It will be far more useful than launching two or three
applications everyday morning (Mail/Calendar/Tasks)

4) Evolution Preferences Dialog is Horrendous

Yes. This is a terrible thing to have especially when you are working in
low resolution. The reason why this was so mammoth was because even the
plugin configurations were added to the General preferences. 

From 2.12 onwards, we have Configure support for plugins and hence
people can configure plugins within the plugin-manager window itself.  

attachment-reminder plugin's configure UI has been already moved from
General preferences to Plugin-Manager-Configure already. Work is on for
Automatic-Contacts and other plugins already.

One another option will be to show the preferences of the current
component alone. Like if you launch preferences from Mailer component
then only Mailer preferences should be shown.

-- 
Sankar
Gnome-Evolution Plugins Maintainer

 Novell, Inc. 
Software for the Open Enterprise™
http://www.novell.com
___
desktop-devel-list mailing list
desktop-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/desktop-devel-list